Hi Mattias,
What is the resolution of your axis? (how many counts per inch?) So we can determine what size errors we are seeing.
With a High D gain
of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). You can smooth that with a filter and the D gain will probably become more effective. Set the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4. Then see what P and D gains work best for you. Maybe something like P=5 D=100 might be a starting point. Then try adding I gain. Try increasing it until it has some effect. Possibly I=0.01 or more.
Regards TK
| Group: DynoMotion |
Message: 6397 |
From: maba_sweden |
Date: 1/9/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Tom
Thanks for your answer.
When i move 2000 steps it´s close to 1mm so that would give ~50800 counts per inch
I just tried the filter and did one test run with P5 I0 D100
I uploaded some new pictures.
I will continue my tests tomorrow and se what differens I´ll get now with the filter.
Thanks!
Regards Mattias
--- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
>
> Hi Mattias,
>
> What is the resolution of your axis? Â (how many counts per inch?) Â So we can determine what size errors we are seeing.
>
> With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). Â You can smooth that with a filter and the D gain will probably become more effective. Â Set the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4. Â Then see what P and D gains work best for you. Â Maybe something like P=5 D=100 might be a starting point. Â Then try adding I gain. Â Try increasing it until it has some effect. Â Possibly I=0.01 or more.
>
> Regards
> TK
>
>
> ________________________________
> From: maba_sweden
> To: DynoMotion@yahoogroups.com
> Sent: Wednesday, January 9, 2013 5:10 PM
> Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
>
>
> Â
> Hi
> I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
>
> I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
>
> I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and couldn´t go any further.
>
> I also tried some 0.001 I gain but i didn´t though it maked it any better.
>
> I have posted some pictures under "photos" and "Mattias Combimill"
> They are named as the setting i tested
>
> Any tip on how I should proceed?
>
> Regards Mattias
>
|
|
| Group: DynoMotion |
Message: 6399 |
From: Tom Kerekes |
Date: 1/9/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias,
Notice huge spikes in the output are now gone.
But there is servo oscillation moving in the plus direction (but not in the negative). Probably
different friction moving one way vs the other?
The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
What type of amplifiers do you have? Is there tachometer feedback? Do you have the velocity loop in the amplifier optimized?
If you could get a reasonable Bode plot it would help understand things.
Regards TK
| Group: DynoMotion |
Message: 6401 |
From: maba_sweden |
Date: 1/9/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Yes the output looked better
I noticed a sound in the plus direction but not in the minus so that observation is spot on.
My Amplifiers are the original ones from GME systems sweden 1985
tha label says 10KHz SW FERQ
And it have tacho feedback
I have uploaded a picture of the label and the schematics of the amplifier
To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
And I haven´t read the part about the Bode Plot yet, guess i have some work to do.
Regards Mattias
--- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
>
> Hi Mattias,
>
> Notice huge spikes in the output are now gone.
>
> But there is servo oscillation moving in the plus direction (but not in the negative). Â Probably different friction moving one way vs the other?
>
> The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
>
> What type of amplifiers do you have? Â Is there tachometer feedback? Â Do you have the velocity loop in the amplifier optimized?
>
> If you could get a reasonable Bode plot it would help understand things.
>
> Regards
> TK
>
>
> Â
>
>
> ________________________________
> From: maba_sweden
> To: DynoMotion@yahoogroups.com
> Sent: Wednesday, January 9, 2013 6:14 PM
> Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
>
>
> Â
> Hi Tom
> Thanks for your answer.
>
> When i move 2000 steps it´s close to 1mm so that would give ~50800 counts per inch
>
> I just tried the filter and did one test run with P5 I0 D100
> I uploaded some new pictures.
> I will continue my tests tomorrow and se what differens I´ll get now with the filter.
>
> Thanks!
> Regards Mattias
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> >
> > Hi Mattias,
> >
> > What is the resolution of your axis? ÃÂ (how many counts per inch?) ÃÂ So we can determine what size errors we are seeing.
> >
> > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). ÃÂ You can smooth that with a filter and the D gain will probably become more effective. ÃÂ Set the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4. ÃÂ Then see what P and D gains work best for you. ÃÂ Maybe something like P=5 D=100 might be a starting point. ÃÂ Then try adding I gain. ÃÂ Try increasing it until it has some effect. ÃÂ Possibly I=0.01 or more.
> >
> > Regards
> > TK
> >
> >
> > ________________________________
> > From: maba_sweden
> > To: DynoMotion@yahoogroups.com
> > Sent: Wednesday, January 9, 2013 5:10 PM
> > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> >
> >
> > ÃÂ
> > Hi
> > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> >
> > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> >
> > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and couldnôt go any further.
> >
> > I also tried some 0.001 I gain but i didnôt though it maked it any better.
> >
> > I have posted some pictures under "photos" and "Mattias Combimill"
> > They are named as the setting i tested
> >
> > Any tip on how I should proceed?
> >
> > Regards Mattias
> >
>
|
|
| Group: DynoMotion |
Message: 6403 |
From: Tom Kerekes |
Date: 1/9/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias,
If the system was working ok before then the amplifier is probably set up reasonably well. Your schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. That is probably the tachometer gain setting. The higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position. But of course too high of gain will make the speed
regulation in the amplifier go unstable and make things bad. I would leave that alone for now.
Regards TK
| Group: DynoMotion |
Message: 6406 |
From: maba_sweden |
Date: 1/10/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Thanks for the explanation Tom
Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
I still got the few count difference when it stops after the + move but i corrected by the system.
And then i always get a few counts wrong when it stops after the - move.
I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesn´t go away before i disable the axle. The output signal gets green over the whole screen when it happens.
Am I on the right way?
Any tip what to try next?
Regards Mattias
--- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
>
> Hi Mattias,
>
> If the system was working ok before then the amplifier is probably set up reasonably well. Â Your schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. Â That is probably the tachometer gain setting. Â The higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position. Â But of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad. Â I would leave that alone for now.
>
> Regards
> TK
>
>
> ________________________________
> From: maba_sweden
> To: DynoMotion@yahoogroups.com
> Sent: Wednesday, January 9, 2013 7:01 PM
> Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
>
>
> Â
> Yes the output looked better
>
> I noticed a sound in the plus direction but not in the minus so that observation is spot on.
>
> My Amplifiers are the original ones from GME systems sweden 1985
> tha label says 10KHz SW FERQ
> And it have tacho feedback
>
> I have uploaded a picture of the label and the schematics of the amplifier
>
> To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> And I haven´t read the part about the Bode Plot yet, guess i have some work to do.
>
> Regards Mattias
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> >
> > Hi Mattias,
> >
> > Notice huge spikes in the output are now gone.
> >
> > But there is servo oscillation moving in the plus direction (but not in the negative). ÃÂ Probably different friction moving one way vs the other?
> >
> > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> >
> > What type of amplifiers do you have? ÃÂ Is there tachometer feedback? ÃÂ Do you have the velocity loop in the amplifier optimized?
> >
> > If you could get a reasonable Bode plot it would help understand things.
> >
> > Regards
> > TK
> >
> >
> > ÃÂ
> >
> >
> > ________________________________
> > From: maba_sweden
> > To: DynoMotion@yahoogroups.com
> > Sent: Wednesday, January 9, 2013 6:14 PM
> > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> >
> >
> > ÃÂ
> > Hi Tom
> > Thanks for your answer.
> >
> > When i move 2000 steps itôs close to 1mm so that would give ~50800 counts per inch
> >
> > I just tried the filter and did one test run with P5 I0 D100
> > I uploaded some new pictures.
> > I will continue my tests tomorrow and se what differens Iôll get now with the filter.
> >
> > Thanks!
> > Regards Mattias
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > >
> > > Hi Mattias,
> > >
> > > What is the resolution of your axis? Ãâà(how many counts per inch?) ÃâàSo we can determine what size errors we are seeing.
> > >
> > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). ÃâàYou can smooth that with a filter and the D gain will probably become more effective. ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4. ÃâàThen see what P and D gains work best for you. ÃâàMaybe something like P=5 D=100 might be a starting point. ÃâàThen try adding I gain. ÃâàTry increasing it until it has some effect. ÃâàPossibly I=0.01 or more.
> > >
> > > Regards
> > > TK
> > >
> > >
> > > ________________________________
> > > From: maba_sweden
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > >
> > >
> > > ÃâÃÂ
> > > Hi
> > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > >
> > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > >
> > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and couldn̫̉t go any further.
> > >
> > > I also tried some 0.001 I gain but i didn̫̉t though it maked it any better.
> > >
> > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > They are named as the setting i tested
> > >
> > > Any tip on how I should proceed?
> > >
> > > Regards Mattias
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6407 |
From: Tom Kerekes |
Date: 1/10/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias, Your errors are only several um so that is probably as good as one can expect. I don't understand why the small I gain causes a problem with long moves. It might be totally unrelated and you are just very marginal to begin with. It would be helpful to capture a plot of it going unstable. I gain will make the system more unstable but a tiny amount should not make a significant difference. Try even smaller (1e-6) and increase until you see an effect. Any amount whatsoever should eventually drive any residual error to zero but just take a long time. Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent. Regards TK
| Group: DynoMotion |
Message: 6408 |
From: maba_sweden |
Date: 1/10/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
I don´t know what I was doing wrong the last time i tested I-gain but I went out and tested again and know it worked.
Both with step size 2000 and 20000
It worked best with I 0.025 with P4 D120
new pictures under photos named *_P4I0025D120
Is the position error ok, I think it looks very blue?
I´m having some trouble to understand the scales in the graph window, can I get the scales to follow the zoom?
Thanks
Regards Mattias
--- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
>
> Hi Mattias,
>
> Your errors are only several um so that is probably as good as one can expect.
>
> I don't understand why the small I gain causes a problem with long moves. It might be totally unrelated and you are just very marginal to begin with. It would be helpful to capture a plot of it going unstable. I gain will make the system more unstable but a tiny amount should not make a significant difference. Try even smaller (1e-6) and increase until you see an effect. Any amount whatsoever should eventually drive any residual error to zero but just take a long time.
>
> Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent.
>
> Regards
> TK
>
>
>
>
> ________________________________
> From: maba_sweden
> To: DynoMotion@yahoogroups.com
> Sent: Thursday, January 10, 2013 1:59 PM
> Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
>
>
> Â
> Thanks for the explanation Tom
> Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
>
> I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
>
> I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
>
> I still got the few count difference when it stops after the + move but i corrected by the system.
> And then i always get a few counts wrong when it stops after the - move.
>
> I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesn´t go away before i disable the axle. The output signal gets green over the whole screen when it happens.
>
> Am I on the right way?
> Any tip what to try next?
>
> Regards Mattias
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> >
> > Hi Mattias,
> >
> > If the system was working ok before then the amplifier is probably set up reasonably well. ÃÂ Your schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. ÃÂ That is probably the tachometer gain setting. ÃÂ The higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position. ÃÂ But of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad. ÃÂ I would leave that alone for now.
> >
> > Regards
> > TK
> >
> >
> > ________________________________
> > From: maba_sweden
> > To: DynoMotion@yahoogroups.com
> > Sent: Wednesday, January 9, 2013 7:01 PM
> > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> >
> >
> > ÃÂ
> > Yes the output looked better
> >
> > I noticed a sound in the plus direction but not in the minus so that observation is spot on.
> >
> > My Amplifiers are the original ones from GME systems sweden 1985
> > tha label says 10KHz SW FERQ
> > And it have tacho feedback
> >
> > I have uploaded a picture of the label and the schematics of the amplifier
> >
> > To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> > And I havenôt read the part about the Bode Plot yet, guess i have some work to do.
> >
> > Regards Mattias
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > >
> > > Hi Mattias,
> > >
> > > Notice huge spikes in the output are now gone.
> > >
> > > But there is servo oscillation moving in the plus direction (but not in the negative). ÃâàProbably different friction moving one way vs the other?
> > >
> > > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> > >
> > > What type of amplifiers do you have? ÃâàIs there tachometer feedback? ÃâàDo you have the velocity loop in the amplifier optimized?
> > >
> > > If you could get a reasonable Bode plot it would help understand things.
> > >
> > > Regards
> > > TK
> > >
> > >
> > > ÃâÃÂ
> > >
> > >
> > > ________________________________
> > > From: maba_sweden
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Wednesday, January 9, 2013 6:14 PM
> > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > >
> > >
> > > ÃâÃÂ
> > > Hi Tom
> > > Thanks for your answer.
> > >
> > > When i move 2000 steps it̫̉s close to 1mm so that would give ~50800 counts per inch
> > >
> > > I just tried the filter and did one test run with P5 I0 D100
> > > I uploaded some new pictures.
> > > I will continue my tests tomorrow and se what differens Ỉ̫ll get now with the filter.
> > >
> > > Thanks!
> > > Regards Mattias
> > >
> > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > >
> > > > Hi Mattias,
> > > >
> > > > What is the resolution of your axis? ÃÆ'ââ¬Å¡Ãâà(how many counts per inch?) ÃÆ'ââ¬Å¡ÃâàSo we can determine what size errors we are seeing.
> > > >
> > > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). ÃÆ'ââ¬Å¡ÃâàYou can smooth that with a filter and the D gain will probably become more effective. ÃÆ'ââ¬Å¡ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4. ÃÆ'ââ¬Å¡ÃâàThen see what P and D gains work best for you. ÃÆ'ââ¬Å¡ÃâàMaybe something like P=5 D=100 might be a starting point. ÃÆ'ââ¬Å¡ÃâàThen try adding I gain. ÃÆ'ââ¬Å¡ÃâàTry increasing it until it has some effect. ÃÆ'ââ¬Å¡ÃâàPossibly I=0.01 or more.
> > > >
> > > > Regards
> > > > TK
> > > >
> > > >
> > > > ________________________________
> > > > From: maba_sweden
> > > > To: DynoMotion@yahoogroups.com
> > > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > > >
> > > >
> > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > Hi
> > > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > > >
> > > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > > >
> > > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and couldnÃÆ'ââ¬Å¡Ãâôt go any further.
> > > >
> > > > I also tried some 0.001 I gain but i didnÃÆ'ââ¬Å¡Ãâôt though it maked it any better.
> > > >
> > > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > > They are named as the setting i tested
> > > >
> > > > Any tip on how I should proceed?
> > > >
> > > > Regards Mattias
> > > >
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6412 |
From: Tom Kerekes |
Date: 1/10/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias,
Those look pretty good. But there does seem to be some small (several um) oscillation when moving in the +X direction. It is hard to see what frequency it is without zooming in. A Bode Plot might be helpful to understand things.
But realistically that is pretty close to what can be expected. If the oscillation bothers you then you might reduce the gains enough to get rid of it and accept a bit larger errors. If the oscillations don't go away with lower gains then it might be some sort of vibration or chatter. Any ideas on why motion one way is different from the other?
It is probably time to try higher speeds and accelerations. Have you tried those? What are your goals?
Regards TK
| Group: DynoMotion |
Message: 6657 |
From: maba_sweden |
Date: 2/5/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Again
The machine specs are X Y 5000mm/min Z 2500mm/min
I got ~2000steps/mm so I´m aiming for V:160000 that should be~5m/min
and A:666664 witch should be ~0,25s
If my calcylations are right :-)
I tested to increase the speeds of the X-axle and it runs really smooth with almost no errors at all. I have added a picture called "X-axle_finish.png" with settings V160000 A600000 running 20000 steps.
I have started to test my Y-axle and it seems like the same gain settings as the X-axle works best.
But as soon as I increases the speed on Y-axle, I get problems with oscillation.
I have added the same 500Hz 2nd low pass filter as for X-axle.
The ballscrew seems to run smooth without any excessive play.
Should i reduce my gains or should I use an other type of filter?
I have added pictures under "Photos" "Y-axle V9000 5000steps" AND "Y-axle V9000 40000steps"
Regards// Mattias
--- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
>
> Hi Mattias,
>
> Those look pretty good. But there does seem to be some small (several um) oscillation when moving in the +X direction. It is hard to see what frequency it is without zooming in. A Bode Plot might be helpful to understand things.
>
> But realistically that is pretty close to what can be expected. If the oscillation bothers you then you might reduce the gains enough to get rid of it and accept a bit larger errors. If the oscillations don't go away with lower gains then it might be some sort of vibration or chatter. Any ideas on why motion one way is different from the other?
>
>
> It is probably time to try higher speeds and accelerations. Have you tried those? What are your goals?
>
> Regards
> TK
>
>
>
> ________________________________
> From: maba_sweden
> To: DynoMotion@yahoogroups.com
> Sent: Thursday, January 10, 2013 3:35 PM
> Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
>
>
> Â
> I don´t know what I was doing wrong the last time i tested I-gain but I went out and tested again and know it worked.
> Both with step size 2000 and 20000
> It worked best with I 0.025 with P4 D120
>
> new pictures under photos named *_P4I0025D120
>
> Is the position error ok, I think it looks very blue?
>
> I´m having some trouble to understand the scales in the graph window, can I get the scales to follow the zoom?
>
> Thanks
> Regards Mattias
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> >
> > Hi Mattias,
> >
> > Your errors are only several um so that is probably as good as one can expect.
> >
> > I don't understand why the small I gain causes a problem with long moves.ÃÂ It might be totally unrelated and you are just very marginal to begin with.ÃÂ It would be helpful to capture a plot of it going unstable.ÃÂ I gain will make the system more unstable but a tiny amount should not make a significant difference.ÃÂ Try even smaller (1e-6) and increase until you see an effect.ÃÂ Any amount whatsoever should eventually drive any residual error to zero but just take a long time.
> >
> > Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent.
> >
> > Regards
> > TK
> >
> >
> >
> >
> > ________________________________
> > From: maba_sweden
> > To: DynoMotion@yahoogroups.com
> > Sent: Thursday, January 10, 2013 1:59 PM
> > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> >
> >
> > ÃÂ
> > Thanks for the explanation Tom
> > Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
> >
> > I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
> >
> > I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
> >
> > I still got the few count difference when it stops after the + move but i corrected by the system.
> > And then i always get a few counts wrong when it stops after the - move.
> >
> > I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesnôt go away before i disable the axle. The output signal gets green over the whole screen when it happens.
> >
> > Am I on the right way?
> > Any tip what to try next?
> >
> > Regards Mattias
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > >
> > > Hi Mattias,
> > >
> > > If the system was working ok before then the amplifier is probably set up reasonably well. ÃâàYour schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. ÃâàThat is probably the tachometer gain setting. ÃâàThe higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position. ÃâàBut of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad. ÃâàI would leave that alone for now.
> > >
> > > Regards
> > > TK
> > >
> > >
> > > ________________________________
> > > From: maba_sweden
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Wednesday, January 9, 2013 7:01 PM
> > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > >
> > >
> > > ÃâÃÂ
> > > Yes the output looked better
> > >
> > > I noticed a sound in the plus direction but not in the minus so that observation is spot on.
> > >
> > > My Amplifiers are the original ones from GME systems sweden 1985
> > > tha label says 10KHz SW FERQ
> > > And it have tacho feedback
> > >
> > > I have uploaded a picture of the label and the schematics of the amplifier
> > >
> > > To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> > > And I haven̫̉t read the part about the Bode Plot yet, guess i have some work to do.
> > >
> > > Regards Mattias
> > >
> > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > >
> > > > Hi Mattias,
> > > >
> > > > Notice huge spikes in the output are now gone.
> > > >
> > > > But there is servo oscillation moving in the plus direction (but not in the negative). ÃÆ'ââ¬Å¡ÃâàProbably different friction moving one way vs the other?
> > > >
> > > > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> > > >
> > > > What type of amplifiers do you have? ÃÆ'ââ¬Å¡ÃâàIs there tachometer feedback? ÃÆ'ââ¬Å¡ÃâàDo you have the velocity loop in the amplifier optimized?
> > > >
> > > > If you could get a reasonable Bode plot it would help understand things.
> > > >
> > > > Regards
> > > > TK
> > > >
> > > >
> > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > >
> > > >
> > > > ________________________________
> > > > From: maba_sweden
> > > > To: DynoMotion@yahoogroups.com
> > > > Sent: Wednesday, January 9, 2013 6:14 PM
> > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > >
> > > >
> > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > Hi Tom
> > > > Thanks for your answer.
> > > >
> > > > When i move 2000 steps itÃÆ'ââ¬Å¡Ãâôs close to 1mm so that would give ~50800 counts per inch
> > > >
> > > > I just tried the filter and did one test run with P5 I0 D100
> > > > I uploaded some new pictures.
> > > > I will continue my tests tomorrow and se what differens IÃÆ'ââ¬Å¡Ãâôll get now with the filter.
> > > >
> > > > Thanks!
> > > > Regards Mattias
> > > >
> > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > >
> > > > > Hi Mattias,
> > > > >
> > > > > What is the resolution of your axis? ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâà(how many counts per inch?) ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSo we can determine what size errors we are seeing.
> > > > >
> > > > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYou can smooth that with a filter and the D gain will probably become more effective. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen see what P and D gains work best for you. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàMaybe something like P=5 D=100 might be a starting point. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen try adding I gain. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry increasing it until it has some effect. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàPossibly I=0.01 or more.
> > > > >
> > > > > Regards
> > > > > TK
> > > > >
> > > > >
> > > > > ________________________________
> > > > > From: maba_sweden
> > > > > To: DynoMotion@yahoogroups.com
> > > > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > > > >
> > > > >
> > > > > ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > Hi
> > > > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > > > >
> > > > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > > > >
> > > > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and couldnÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go any further.
> > > > >
> > > > > I also tried some 0.001 I gain but i didnÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt though it maked it any better.
> > > > >
> > > > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > > > They are named as the setting i tested
> > > > >
> > > > > Any tip on how I should proceed?
> > > > >
> > > > > Regards Mattias
> > > > >
> > > >
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6660 |
From: Tom Kerekes |
Date: 2/5/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias, Your calculations look correct. I would lower gains to improve stability. Your Max I is set at 200. To go at high speed the I term would need more than that so increase to more than what is needed as Output to fo at your max speed. Regards TK
| Group: DynoMotion |
Message: 6663 |
From: maba_sweden |
Date: 2/5/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Thanks for your inputs Tom!
I will test to increase the Max I and decrease the gain settings to see if it will be more stable.
One more thing,
if I disable the axis in "step resonse" the Y-axle will slowly start to move and as soon as I enable the axis it stops.
This don´t happens on the X-axis, it´s still all the time.
Is this normal?
Regards Mattias
--- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
>
> Hi Mattias,
>
> Your calculations look correct.
>
> I would lower gains to improve stability.
>
> Your Max I is set at 200. To go at high speed the I term would need more than that so increase to more than what is needed as Output to fo at your max speed.
>
> Regards
> TK
>
>
>
>
> ________________________________
> From: maba_sweden
> To: DynoMotion@yahoogroups.com
> Sent: Tuesday, February 5, 2013 8:15 AM
> Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
>
>
> Â
> Hi Again
>
> The machine specs are X Y 5000mm/min Z 2500mm/min
> I got ~2000steps/mm so I´m aiming for V:160000 that should be~5m/min
> and A:666664 witch should be ~0,25s
> If my calcylations are right :-)
>
> I tested to increase the speeds of the X-axle and it runs really smooth with almost no errors at all. I have added a picture called "X-axle_finish.png" with settings V160000 A600000 running 20000 steps.
>
> I have started to test my Y-axle and it seems like the same gain settings as the X-axle works best.
> But as soon as I increases the speed on Y-axle, I get problems with oscillation.
> I have added the same 500Hz 2nd low pass filter as for X-axle.
> The ballscrew seems to run smooth without any excessive play.
>
> Should i reduce my gains or should I use an other type of filter?
>
> I have added pictures under "Photos" "Y-axle V9000 5000steps" AND "Y-axle V9000 40000steps"
>
> Regards// Mattias
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> >
> > Hi Mattias,
> >
> > Those look pretty good.ÃÂ But there does seem to be some small (several um) oscillation when moving in the +X direction.ÃÂ It is hard to see what frequency it is without zooming in.ÃÂ A Bode Plot might be helpful to understand things.
> >
> > But realistically that is pretty close to what can be expected.ÃÂ If the oscillation bothers you then you might reduce the gains enough to get rid of it and accept a bit larger errors.ÃÂ If the oscillations don't go away with lower gains then it might be some sort of vibration or chatter.ÃÂ Any ideas on why motion one way is different from the other?
> >
> >
> > It is probably time to try higher speeds and accelerations.ÃÂ Have you tried those?ÃÂ What are your goals?
> >
> > Regards
> > TK
> >
> >
> >
> > ________________________________
> > From: maba_sweden
> > To: DynoMotion@yahoogroups.com
> > Sent: Thursday, January 10, 2013 3:35 PM
> > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> >
> >
> > ÃÂ
> > I donôt know what I was doing wrong the last time i tested I-gain but I went out and tested again and know it worked.
> > Both with step size 2000 and 20000
> > It worked best with I 0.025 with P4 D120
> >
> > new pictures under photos named *_P4I0025D120
> >
> > Is the position error ok, I think it looks very blue?
> >
> > Iôm having some trouble to understand the scales in the graph window, can I get the scales to follow the zoom?
> >
> > Thanks
> > Regards Mattias
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > >
> > > Hi Mattias,
> > >
> > > Your errors are only several um so that is probably as good as one can expect.
> > >
> > > I don't understand why the small I gain causes a problem with long moves.ÃâàIt might be totally unrelated and you are just very marginal to begin with.ÃâàIt would be helpful to capture a plot of it going unstable.ÃâàI gain will make the system more unstable but a tiny amount should not make a significant difference.ÃâàTry even smaller (1e-6) and increase until you see an effect.ÃâàAny amount whatsoever should eventually drive any residual error to zero but just take a long time.
> > >
> > > Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent.
> > >
> > > Regards
> > > TK
> > >
> > >
> > >
> > >
> > > ________________________________
> > > From: maba_sweden
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Thursday, January 10, 2013 1:59 PM
> > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > >
> > >
> > > ÃâÃÂ
> > > Thanks for the explanation Tom
> > > Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
> > >
> > > I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
> > >
> > > I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
> > >
> > > I still got the few count difference when it stops after the + move but i corrected by the system.
> > > And then i always get a few counts wrong when it stops after the - move.
> > >
> > > I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesn̫̉t go away before i disable the axle. The output signal gets green over the whole screen when it happens.
> > >
> > > Am I on the right way?
> > > Any tip what to try next?
> > >
> > > Regards Mattias
> > >
> > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > >
> > > > Hi Mattias,
> > > >
> > > > If the system was working ok before then the amplifier is probably set up reasonably well. ÃÆ'ââ¬Å¡ÃâàYour schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. ÃÆ'ââ¬Å¡ÃâàThat is probably the tachometer gain setting. ÃÆ'ââ¬Å¡ÃâàThe higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position. ÃÆ'ââ¬Å¡ÃâàBut of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad. ÃÆ'ââ¬Å¡ÃâàI would leave that alone for now.
> > > >
> > > > Regards
> > > > TK
> > > >
> > > >
> > > > ________________________________
> > > > From: maba_sweden
> > > > To: DynoMotion@yahoogroups.com
> > > > Sent: Wednesday, January 9, 2013 7:01 PM
> > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > >
> > > >
> > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > Yes the output looked better
> > > >
> > > > I noticed a sound in the plus direction but not in the minus so that observation is spot on.
> > > >
> > > > My Amplifiers are the original ones from GME systems sweden 1985
> > > > tha label says 10KHz SW FERQ
> > > > And it have tacho feedback
> > > >
> > > > I have uploaded a picture of the label and the schematics of the amplifier
> > > >
> > > > To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> > > > And I havenÃÆ'ââ¬Å¡Ãâôt read the part about the Bode Plot yet, guess i have some work to do.
> > > >
> > > > Regards Mattias
> > > >
> > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > >
> > > > > Hi Mattias,
> > > > >
> > > > > Notice huge spikes in the output are now gone.
> > > > >
> > > > > But there is servo oscillation moving in the plus direction (but not in the negative). ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàProbably different friction moving one way vs the other?
> > > > >
> > > > > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> > > > >
> > > > > What type of amplifiers do you have? ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIs there tachometer feedback? ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàDo you have the velocity loop in the amplifier optimized?
> > > > >
> > > > > If you could get a reasonable Bode plot it would help understand things.
> > > > >
> > > > > Regards
> > > > > TK
> > > > >
> > > > >
> > > > > ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > >
> > > > >
> > > > > ________________________________
> > > > > From: maba_sweden
> > > > > To: DynoMotion@yahoogroups.com
> > > > > Sent: Wednesday, January 9, 2013 6:14 PM
> > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > >
> > > > >
> > > > > ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > Hi Tom
> > > > > Thanks for your answer.
> > > > >
> > > > > When i move 2000 steps itÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôs close to 1mm so that would give ~50800 counts per inch
> > > > >
> > > > > I just tried the filter and did one test run with P5 I0 D100
> > > > > I uploaded some new pictures.
> > > > > I will continue my tests tomorrow and se what differens IÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôll get now with the filter.
> > > > >
> > > > > Thanks!
> > > > > Regards Mattias
> > > > >
> > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > >
> > > > > > Hi Mattias,
> > > > > >
> > > > > > What is the resolution of your axis? ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâà(how many counts per inch?) ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSo we can determine what size errors we are seeing.
> > > > > >
> > > > > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYou can smooth that with a filter and the D gain will probably become more effective. ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4. ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen see what P and D gains work best for you. ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàMaybe something like P=5 D=100 might be a starting point. ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen try adding I gain. ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry increasing it until it
> has some effect. ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàPossibly I=0.01 or more.
> > > > > >
> > > > > > Regards
> > > > > > TK
> > > > > >
> > > > > >
> > > > > > ________________________________
> > > > > > From: maba_sweden
> > > > > > To: DynoMotion@yahoogroups.com
> > > > > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > > > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > > > > >
> > > > > >
> > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > Hi
> > > > > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > > > > >
> > > > > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > > > > >
> > > > > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and couldnÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go any further.
> > > > > >
> > > > > > I also tried some 0.001 I gain but i didnÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt though it maked it any better.
> > > > > >
> > > > > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > > > > They are named as the setting i tested
> > > > > >
> > > > > > Any tip on how I should proceed?
> > > > > >
> > > > > > Regards Mattias
> > > > > >
> > > > >
> > > >
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6664 |
From: Tom Kerekes |
Date: 2/5/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias,
This is normal. It is almost impossible to command analog velocity of exactly zero because there will always be some analog noise and offsets. If this is a problem then you will need to wire a disable signal to your amplifier. Then add C statement to disable the Amplifier whenever the KFLOP Axis Channel is disabled. See the WatchEnable.c example. This is not a problem when the servo is enabled because the feedback will command whatever voltage is
necessary to hold position.
HTH Regards TK
| Group: DynoMotion |
Message: 6667 |
From: maba_sweden |
Date: 2/5/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Ok, one thing that i don´t have to worry about :-)
I tested some high speeds and longer runs and ended up lowering the gain pretty much on both X and Y.
I moved on to my Z-axle and it seems like there´s not enough output to go along with the Command.
I have max output 2000 intergration 1000
Velocity 4000
Acceleration 40000
I also added 2nd low pass filter 500hz 1,4q as for the X and Y axle
When i moves 2000 steps it´s ok but when i test 4000steps or more it gets really bad.
I have added a picture under photos named "Z-axle_first_run"
Regards Mattias
--- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
>
> Hi Mattias,
>
> This is normal. It is almost impossible to command analog velocity of exactly zero because there will always be some analog noise and offsets. If this is a problem then you will need to wire a disable signal to your amplifier. Then add C statement to disable the Amplifier whenever the KFLOP Axis Channel is disabled. See the WatchEnable.c example. This is not a problem when the servo is enabled because the feedback will command whatever voltage is necessary to hold position.
>
>
> HTH
> Regards
> TK
>
>
>
> ________________________________
> From: maba_sweden
> To: DynoMotion@yahoogroups.com
> Sent: Tuesday, February 5, 2013 9:36 AM
> Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
>
>
> Â
> Thanks for your inputs Tom!
>
> I will test to increase the Max I and decrease the gain settings to see if it will be more stable.
>
> One more thing,
> if I disable the axis in "step resonse" the Y-axle will slowly start to move and as soon as I enable the axis it stops.
> This don´t happens on the X-axis, it´s still all the time.
> Is this normal?
>
> Regards Mattias
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> >
> > Hi Mattias,
> >
> > Your calculations look correct.
> >
> > I would lower gains to improve stability.
> >
> > Your Max I is set at 200.ÃÂ To go at high speed the I term would need more than that so increase to more than what is needed as Output to fo at your max speed.
> >
> > Regards
> > TK
> >
> >
> >
> >
> > ________________________________
> > From: maba_sweden
> > To: DynoMotion@yahoogroups.com
> > Sent: Tuesday, February 5, 2013 8:15 AM
> > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> >
> >
> > ÃÂ
> > Hi Again
> >
> > The machine specs are X Y 5000mm/min Z 2500mm/min
> > I got ~2000steps/mm so Iôm aiming for V:160000 that should be~5m/min
> > and A:666664 witch should be ~0,25s
> > If my calcylations are right :-)
> >
> > I tested to increase the speeds of the X-axle and it runs really smooth with almost no errors at all. I have added a picture called "X-axle_finish.png" with settings V160000 A600000 running 20000 steps.
> >
> > I have started to test my Y-axle and it seems like the same gain settings as the X-axle works best.
> > But as soon as I increases the speed on Y-axle, I get problems with oscillation.
> > I have added the same 500Hz 2nd low pass filter as for X-axle.
> > The ballscrew seems to run smooth without any excessive play.
> >
> > Should i reduce my gains or should I use an other type of filter?
> >
> > I have added pictures under "Photos" "Y-axle V9000 5000steps" AND "Y-axle V9000 40000steps"
> >
> > Regards// Mattias
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > >
> > > Hi Mattias,
> > >
> > > Those look pretty good.ÃâàBut there does seem to be some small (several um) oscillation when moving in the +X direction.ÃâàIt is hard to see what frequency it is without zooming in.ÃâàA Bode Plot might be helpful to understand things.
> > >
> > > But realistically that is pretty close to what can be expected.ÃâàIf the oscillation bothers you then you might reduce the gains enough to get rid of it and accept a bit larger errors.ÃâàIf the oscillations don't go away with lower gains then it might be some sort of vibration or chatter.ÃâàAny ideas on why motion one way is different from the other?
> > >
> > >
> > > It is probably time to try higher speeds and accelerations.ÃâàHave you tried those?ÃâàWhat are your goals?
> > >
> > > Regards
> > > TK
> > >
> > >
> > >
> > > ________________________________
> > > From: maba_sweden
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Thursday, January 10, 2013 3:35 PM
> > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > >
> > >
> > > ÃâÃÂ
> > > I don̫̉t know what I was doing wrong the last time i tested I-gain but I went out and tested again and know it worked.
> > > Both with step size 2000 and 20000
> > > It worked best with I 0.025 with P4 D120
> > >
> > > new pictures under photos named *_P4I0025D120
> > >
> > > Is the position error ok, I think it looks very blue?
> > >
> > > Ỉ̫m having some trouble to understand the scales in the graph window, can I get the scales to follow the zoom?
> > >
> > > Thanks
> > > Regards Mattias
> > >
> > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > >
> > > > Hi Mattias,
> > > >
> > > > Your errors are only several um so that is probably as good as one can expect.
> > > >
> > > > I don't understand why the small I gain causes a problem with long moves.ÃÆ'ââ¬Å¡ÃâàIt might be totally unrelated and you are just very marginal to begin with.ÃÆ'ââ¬Å¡ÃâàIt would be helpful to capture a plot of it going unstable.ÃÆ'ââ¬Å¡ÃâàI gain will make the system more unstable but a tiny amount should not make a significant difference.ÃÆ'ââ¬Å¡ÃâàTry even smaller (1e-6) and increase until you see an effect.ÃÆ'ââ¬Å¡ÃâàAny amount whatsoever should eventually drive any residual error to zero but just take a long time.
> > > >
> > > > Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent.
> > > >
> > > > Regards
> > > > TK
> > > >
> > > >
> > > >
> > > >
> > > > ________________________________
> > > > From: maba_sweden
> > > > To: DynoMotion@yahoogroups.com
> > > > Sent: Thursday, January 10, 2013 1:59 PM
> > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > >
> > > >
> > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > Thanks for the explanation Tom
> > > > Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
> > > >
> > > > I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
> > > >
> > > > I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
> > > >
> > > > I still got the few count difference when it stops after the + move but i corrected by the system.
> > > > And then i always get a few counts wrong when it stops after the - move.
> > > >
> > > > I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesnÃÆ'ââ¬Å¡Ãâôt go away before i disable the axle. The output signal gets green over the whole screen when it happens.
> > > >
> > > > Am I on the right way?
> > > > Any tip what to try next?
> > > >
> > > > Regards Mattias
> > > >
> > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > >
> > > > > Hi Mattias,
> > > > >
> > > > > If the system was working ok before then the amplifier is probably set up reasonably well. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYour schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThat is probably the tachometer gain setting. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThe higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad. ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI would leave that alone for now.
> > > > >
> > > > > Regards
> > > > > TK
> > > > >
> > > > >
> > > > > ________________________________
> > > > > From: maba_sweden
> > > > > To: DynoMotion@yahoogroups.com
> > > > > Sent: Wednesday, January 9, 2013 7:01 PM
> > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > >
> > > > >
> > > > > ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > Yes the output looked better
> > > > >
> > > > > I noticed a sound in the plus direction but not in the minus so that observation is spot on.
> > > > >
> > > > > My Amplifiers are the original ones from GME systems sweden 1985
> > > > > tha label says 10KHz SW FERQ
> > > > > And it have tacho feedback
> > > > >
> > > > > I have uploaded a picture of the label and the schematics of the amplifier
> > > > >
> > > > > To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> > > > > And I havenÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt read the part about the Bode Plot yet, guess i have some work to do.
> > > > >
> > > > > Regards Mattias
> > > > >
> > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > >
> > > > > > Hi Mattias,
> > > > > >
> > > > > > Notice huge spikes in the output are now gone.
> > > > > >
> > > > > > But there is servo oscillation moving in the plus direction (but not in the negative). ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàProbably different friction moving one way vs the other?
> > > > > >
> > > > > > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> > > > > >
> > > > > > What type of amplifiers do you have? ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIs there tachometer feedback? ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàDo you have the velocity loop in the amplifier optimized?
> > > > > >
> > > > > > If you could get a reasonable Bode plot it would help understand things.
> > > > > >
> > > > > > Regards
> > > > > > TK
> > > > > >
> > > > > >
> > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > >
> > > > > >
> > > > > > ________________________________
> > > > > > From: maba_sweden
> > > > > > To: DynoMotion@yahoogroups.com
> > > > > > Sent: Wednesday, January 9, 2013 6:14 PM
> > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > >
> > > > > >
> > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > Hi Tom
> > > > > > Thanks for your answer.
> > > > > >
> > > > > > When i move 2000 steps itÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôs close to 1mm so that would give ~50800 counts per inch
> > > > > >
> > > > > > I just tried the filter and did one test run with P5 I0 D100
> > > > > > I uploaded some new pictures.
> > > > > > I will continue my tests tomorrow and se what differens IÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôll get now with the filter.
> > > > > >
> > > > > > Thanks!
> > > > > > Regards Mattias
> > > > > >
> > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > >
> > > > > > > Hi Mattias,
> > > > > > >
> > > > > > > What is the resolution of your axis? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâà(how many counts per inch?) ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSo we can determine what size errors we are seeing.
> > > > > > >
> > > > > > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYou can smooth that with a filter and the D gain will probably become more effective. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen see what P and D gains work best for you.
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàMaybe something like P=5 D=100 might be a starting point. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen try adding I gain. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry increasing it until it
> > has some effect. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàPossibly I=0.01 or more.
> > > > > > >
> > > > > > > Regards
> > > > > > > TK
> > > > > > >
> > > > > > >
> > > > > > > ________________________________
> > > > > > > From: maba_sweden
> > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > > > > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > > > > > >
> > > > > > >
> > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > Hi
> > > > > > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > > > > > >
> > > > > > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > > > > > >
> > > > > > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and couldnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go any further.
> > > > > > >
> > > > > > > I also tried some 0.001 I gain but i didnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt though it maked it any better.
> > > > > > >
> > > > > > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > > > > > They are named as the setting i tested
> > > > > > >
> > > > > > > Any tip on how I should proceed?
> > > > > > >
> > > > > > > Regards Mattias
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6668 |
From: Tom Kerekes |
Date: 2/5/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias, It looks like you probably have a lot of weight and friction. Notice as it goes positive (I assume this is lifting weight) the position (red) lags way behind moving at a varying speed even though the output (green) is fully saturated at 2000 counts (~10V). Your Motor/Amplifier/Supply probably just doesn't have enough power. But notice going negative (down?) it is able to go very fast even with almost zero command to the amplifier. Is there supposed to be a counter balance or something? Or you are just trying to move too fast. How did the machine perform originally? Regards TK
| Group: DynoMotion |
Message: 6669 |
From: maba_sweden |
Date: 2/5/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Yes positive is lifting.
The machine got no counter weight what I´m aware, and according to the manual it should be able to travel 2500mm/min half the speed of the X and Y.
Probably it´s just the acceleration that is to aggressive, if i remember correctly it accelerated pretty smoth with the original Sinumerik system.
I will test with a lower setting and see if it gets any better.
Thanks
Regards//Mattias
--- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
>
> Hi Mattias,
>
> It looks like you probably have a lot of weight and friction. Notice as it goes positive (I assume this is lifting weight) the position (red) lags way behind moving at a varying speed even though the output (green) is fully saturated at 2000 counts (~10V). Your Motor/Amplifier/Supply probably just doesn't have enough power.   But notice going negative (down?) it is able to go very fast even with almost zero command to the amplifier. Is there supposed to be a counter balance or something? Or you are just trying to move too fast. How did the machine perform originally?
>
> Regards
> TK
>
> Â
>
>
>
>
> ________________________________
> From: maba_sweden
> To: DynoMotion@yahoogroups.com
> Sent: Tuesday, February 5, 2013 3:53 PM
> Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
>
>
> Â
> Ok, one thing that i don´t have to worry about :-)
> I tested some high speeds and longer runs and ended up lowering the gain pretty much on both X and Y.
>
> I moved on to my Z-axle and it seems like there´s not enough output to go along with the Command.
> I have max output 2000 intergration 1000
> Velocity 4000
> Acceleration 40000
>
> I also added 2nd low pass filter 500hz 1,4q as for the X and Y axle
>
> When i moves 2000 steps it´s ok but when i test 4000steps or more it gets really bad.
> I have added a picture under photos named "Z-axle_first_run"
>
> Regards Mattias
>
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> >
> > Hi Mattias,
> >
> > This is normal.ÃÂ It is almost impossible to command analog velocity of exactly zero because there will always be some analog noise and offsets.ÃÂ If this is a problem then you will need to wire a disable signal to your amplifier.ÃÂ Then add C statement to disable the Amplifier whenever the KFLOP Axis Channel is disabled.ÃÂ See the WatchEnable.c example.ÃÂ This is not a problem when the servo is enabled because the feedback will command whatever voltage is necessary to hold position.
> >
> >
> > HTH
> > Regards
> > TK
> >
> >
> >
> > ________________________________
> > From: maba_sweden
> > To: DynoMotion@yahoogroups.com
> > Sent: Tuesday, February 5, 2013 9:36 AM
> > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> >
> >
> > ÃÂ
> > Thanks for your inputs Tom!
> >
> > I will test to increase the Max I and decrease the gain settings to see if it will be more stable.
> >
> > One more thing,
> > if I disable the axis in "step resonse" the Y-axle will slowly start to move and as soon as I enable the axis it stops.
> > This donôt happens on the X-axis, itôs still all the time.
> > Is this normal?
> >
> > Regards Mattias
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > >
> > > Hi Mattias,
> > >
> > > Your calculations look correct.
> > >
> > > I would lower gains to improve stability.
> > >
> > > Your Max I is set at 200.ÃâàTo go at high speed the I term would need more than that so increase to more than what is needed as Output to fo at your max speed.
> > >
> > > Regards
> > > TK
> > >
> > >
> > >
> > >
> > > ________________________________
> > > From: maba_sweden
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Tuesday, February 5, 2013 8:15 AM
> > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > >
> > >
> > > ÃâÃÂ
> > > Hi Again
> > >
> > > The machine specs are X Y 5000mm/min Z 2500mm/min
> > > I got ~2000steps/mm so Ỉ̫m aiming for V:160000 that should be~5m/min
> > > and A:666664 witch should be ~0,25s
> > > If my calcylations are right :-)
> > >
> > > I tested to increase the speeds of the X-axle and it runs really smooth with almost no errors at all. I have added a picture called "X-axle_finish.png" with settings V160000 A600000 running 20000 steps.
> > >
> > > I have started to test my Y-axle and it seems like the same gain settings as the X-axle works best.
> > > But as soon as I increases the speed on Y-axle, I get problems with oscillation.
> > > I have added the same 500Hz 2nd low pass filter as for X-axle.
> > > The ballscrew seems to run smooth without any excessive play.
> > >
> > > Should i reduce my gains or should I use an other type of filter?
> > >
> > > I have added pictures under "Photos" "Y-axle V9000 5000steps" AND "Y-axle V9000 40000steps"
> > >
> > > Regards// Mattias
> > >
> > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > >
> > > > Hi Mattias,
> > > >
> > > > Those look pretty good.ÃÆ'ââ¬Å¡ÃâàBut there does seem to be some small (several um) oscillation when moving in the +X direction.ÃÆ'ââ¬Å¡ÃâàIt is hard to see what frequency it is without zooming in.ÃÆ'ââ¬Å¡ÃâàA Bode Plot might be helpful to understand things.
> > > >
> > > > But realistically that is pretty close to what can be expected.ÃÆ'ââ¬Å¡ÃâàIf the oscillation bothers you then you might reduce the gains enough to get rid of it and accept a bit larger errors.ÃÆ'ââ¬Å¡ÃâàIf the oscillations don't go away with lower gains then it might be some sort of vibration or chatter.ÃÆ'ââ¬Å¡ÃâàAny ideas on why motion one way is different from the other?
> > > >
> > > >
> > > > It is probably time to try higher speeds and accelerations.ÃÆ'ââ¬Å¡ÃâàHave you tried those?ÃÆ'ââ¬Å¡ÃâàWhat are your goals?
> > > >
> > > > Regards
> > > > TK
> > > >
> > > >
> > > >
> > > > ________________________________
> > > > From: maba_sweden
> > > > To: DynoMotion@yahoogroups.com
> > > > Sent: Thursday, January 10, 2013 3:35 PM
> > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > >
> > > >
> > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > I donÃÆ'ââ¬Å¡Ãâôt know what I was doing wrong the last time i tested I-gain but I went out and tested again and know it worked.
> > > > Both with step size 2000 and 20000
> > > > It worked best with I 0.025 with P4 D120
> > > >
> > > > new pictures under photos named *_P4I0025D120
> > > >
> > > > Is the position error ok, I think it looks very blue?
> > > >
> > > > IÃÆ'ââ¬Å¡Ãâôm having some trouble to understand the scales in the graph window, can I get the scales to follow the zoom?
> > > >
> > > > Thanks
> > > > Regards Mattias
> > > >
> > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > >
> > > > > Hi Mattias,
> > > > >
> > > > > Your errors are only several um so that is probably as good as one can expect.
> > > > >
> > > > > I don't understand why the small I gain causes a problem with long moves.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt might be totally unrelated and you are just very marginal to begin with.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt would be helpful to capture a plot of it going unstable.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI gain will make the system more unstable but a tiny amount should not make a significant difference.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry even smaller (1e-6) and increase until you see an effect.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny amount whatsoever should eventually drive any residual error to zero but just take a long time.
> > > > >
> > > > > Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent.
> > > > >
> > > > > Regards
> > > > > TK
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > ________________________________
> > > > > From: maba_sweden
> > > > > To: DynoMotion@yahoogroups.com
> > > > > Sent: Thursday, January 10, 2013 1:59 PM
> > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > >
> > > > >
> > > > > ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > Thanks for the explanation Tom
> > > > > Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
> > > > >
> > > > > I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
> > > > >
> > > > > I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
> > > > >
> > > > > I still got the few count difference when it stops after the + move but i corrected by the system.
> > > > > And then i always get a few counts wrong when it stops after the - move.
> > > > >
> > > > > I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesnÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go away before i disable the axle. The output signal gets green over the whole screen when it happens.
> > > > >
> > > > > Am I on the right way?
> > > > > Any tip what to try next?
> > > > >
> > > > > Regards Mattias
> > > > >
> > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > >
> > > > > > Hi Mattias,
> > > > > >
> > > > > > If the system was working ok before then the amplifier is probably set up reasonably well. ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYour schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThat is probably the tachometer gain setting. ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThe higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position. ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad.
> ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI would leave that alone for now.
> > > > > >
> > > > > > Regards
> > > > > > TK
> > > > > >
> > > > > >
> > > > > > ________________________________
> > > > > > From: maba_sweden
> > > > > > To: DynoMotion@yahoogroups.com
> > > > > > Sent: Wednesday, January 9, 2013 7:01 PM
> > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > >
> > > > > >
> > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > Yes the output looked better
> > > > > >
> > > > > > I noticed a sound in the plus direction but not in the minus so that observation is spot on.
> > > > > >
> > > > > > My Amplifiers are the original ones from GME systems sweden 1985
> > > > > > tha label says 10KHz SW FERQ
> > > > > > And it have tacho feedback
> > > > > >
> > > > > > I have uploaded a picture of the label and the schematics of the amplifier
> > > > > >
> > > > > > To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> > > > > > And I havenÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt read the part about the Bode Plot yet, guess i have some work to do.
> > > > > >
> > > > > > Regards Mattias
> > > > > >
> > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > >
> > > > > > > Hi Mattias,
> > > > > > >
> > > > > > > Notice huge spikes in the output are now gone.
> > > > > > >
> > > > > > > But there is servo oscillation moving in the plus direction (but not in the negative). ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàProbably different friction moving one way vs the other?
> > > > > > >
> > > > > > > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> > > > > > >
> > > > > > > What type of amplifiers do you have? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIs there tachometer feedback? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàDo you have the velocity loop in the amplifier optimized?
> > > > > > >
> > > > > > > If you could get a reasonable Bode plot it would help understand things.
> > > > > > >
> > > > > > > Regards
> > > > > > > TK
> > > > > > >
> > > > > > >
> > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > >
> > > > > > >
> > > > > > > ________________________________
> > > > > > > From: maba_sweden
> > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > Sent: Wednesday, January 9, 2013 6:14 PM
> > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > >
> > > > > > >
> > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > Hi Tom
> > > > > > > Thanks for your answer.
> > > > > > >
> > > > > > > When i move 2000 steps itÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôs close to 1mm so that would give ~50800 counts per inch
> > > > > > >
> > > > > > > I just tried the filter and did one test run with P5 I0 D100
> > > > > > > I uploaded some new pictures.
> > > > > > > I will continue my tests tomorrow and se what differens IÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôll get now with the filter.
> > > > > > >
> > > > > > > Thanks!
> > > > > > > Regards Mattias
> > > > > > >
> > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > >
> > > > > > > > Hi Mattias,
> > > > > > > >
> > > > > > > > What is the resolution of your axis? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâà(how many counts per inch?) ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSo we can determine what size errors we are seeing.
> > > > > > > >
> > > > > > > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYou can smooth that with a filter and the D gain will probably become more effective.
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen see what P and D gains work best for you.
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàMaybe something like P=5 D=100 might be a starting point. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen try adding I gain.
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry increasing it until it
> > > has some effect. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàPossibly I=0.01 or more.
> > > > > > > >
> > > > > > > > Regards
> > > > > > > > TK
> > > > > > > >
> > > > > > > >
> > > > > > > > ________________________________
> > > > > > > > From: maba_sweden
> > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > > > > > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > > > > > > >
> > > > > > > >
> > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > Hi
> > > > > > > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > > > > > > >
> > > > > > > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > > > > > > >
> > > > > > > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and couldnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go any further.
> > > > > > > >
> > > > > > > > I also tried some 0.001 I gain but i didnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt though it maked it any better.
> > > > > > > >
> > > > > > > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > > > > > > They are named as the setting i tested
> > > > > > > >
> > > > > > > > Any tip on how I should proceed?
> > > > > > > >
> > > > > > > > Regards Mattias
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6670 |
From: Tom Kerekes |
Date: 2/5/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias, Now that I look at the plot again the velocity is very low 4000 counts/sec and the acceleration to go that speed is insignificant. There must be something major wrong. Maybe you are missing a power supply voltage or something so it can't drive positive. But if the car is going too slow and the gas pedal is all the way to the floor - it isn't the driver's fault :} Regards TK
| Group: DynoMotion |
Message: 6671 |
From: Tom Kerekes |
Date: 2/5/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias, Now that I look at the plot again the velocity is very low 4000 counts/sec and the acceleration to go that speed is insignificant. There must be something major wrong. Maybe you are missing a power supply voltage or something so it can't drive positive. But if the car is going too slow and the gas pedal is all the way to the floor - it isn't the driver's fault :} Regards TK
| Group: DynoMotion |
Message: 6679 |
From: maba_sweden |
Date: 2/6/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Tom
Just wanted you to know that it wasn´t the drivers fault, if it only could have been that :-)
The Z motor is bad. One of the static magnets was loose, that may explains the uneven motion of the motor, the rear bearing must also bad because it´s heavy to rotate.
Probably is this because of coolant that have came in the motor and it have started to rust.
I have to see if i can rebuild the motor, some says the magnets can loose there strength if you take it apart.
Or else I have to try to find a new/used/rebuild Contraves 402 117V 3,4nm DC servo motor before I can continue my work...
Thanks for all your help.
Hope I will be back soon!
Regards Mattias
--- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
>
> Hi Mattias,
>
> Now that I look at the plot again the velocity is very low 4000 counts/sec and the acceleration to go that speed is insignificant. There must be something major wrong.
>
> Maybe you are missing a power supply voltage or something so it can't drive positive.
>
> But if the car is going too slow and the gas pedal is all the way to the floor - it isn't the driver's fault :}
>
> Regards
> TK
>
>
>
>
>
>
> ________________________________
> From: maba_sweden
> To: DynoMotion@yahoogroups.com
> Sent: Tuesday, February 5, 2013 4:30 PM
> Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
>
>
> Â
> Yes positive is lifting.
> The machine got no counter weight what I´m aware, and according to the manual it should be able to travel 2500mm/min half the speed of the X and Y.
> Probably it´s just the acceleration that is to aggressive, if i remember correctly it accelerated pretty smoth with the original Sinumerik system.
> I will test with a lower setting and see if it gets any better.
>
> Thanks
> Regards//Mattias
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> >
> > Hi Mattias,
> >
> > It looks like you probably have a lot of weight and friction.ÃÂ Notice as it goes positive (I assume this is lifting weight) the position (red) lags way behind moving at a varying speed even though the output (green) is fully saturated at 2000 counts (~10V).ÃÂ Your Motor/Amplifier/Supply probably just doesn't have enough power. ÃÂ ÃÂ But notice going negative (down?) it is able to go very fast even with almost zero command to the amplifier.ÃÂ Is there supposed to be a counter balance or something?ÃÂ Or you are just trying to move too fast.ÃÂ How did the machine perform originally?
> >
> > Regards
> > TK
> >
> > ÃÂ
> >
> >
> >
> >
> > ________________________________
> > From: maba_sweden
> > To: DynoMotion@yahoogroups.com
> > Sent: Tuesday, February 5, 2013 3:53 PM
> > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> >
> >
> > ÃÂ
> > Ok, one thing that i donôt have to worry about :-)
> > I tested some high speeds and longer runs and ended up lowering the gain pretty much on both X and Y.
> >
> > I moved on to my Z-axle and it seems like thereôs not enough output to go along with the Command.
> > I have max output 2000 intergration 1000
> > Velocity 4000
> > Acceleration 40000
> >
> > I also added 2nd low pass filter 500hz 1,4q as for the X and Y axle
> >
> > When i moves 2000 steps itôs ok but when i test 4000steps or more it gets really bad.
> > I have added a picture under photos named "Z-axle_first_run"
> >
> > Regards Mattias
> >
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > >
> > > Hi Mattias,
> > >
> > > This is normal.ÃâàIt is almost impossible to command analog velocity of exactly zero because there will always be some analog noise and offsets.ÃâàIf this is a problem then you will need to wire a disable signal to your amplifier.ÃâàThen add C statement to disable the Amplifier whenever the KFLOP Axis Channel is disabled.ÃâàSee the WatchEnable.c example.ÃâàThis is not a problem when the servo is enabled because the feedback will command whatever voltage is necessary to hold position.
> > >
> > >
> > > HTH
> > > Regards
> > > TK
> > >
> > >
> > >
> > > ________________________________
> > > From: maba_sweden
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Tuesday, February 5, 2013 9:36 AM
> > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > >
> > >
> > > ÃâÃÂ
> > > Thanks for your inputs Tom!
> > >
> > > I will test to increase the Max I and decrease the gain settings to see if it will be more stable.
> > >
> > > One more thing,
> > > if I disable the axis in "step resonse" the Y-axle will slowly start to move and as soon as I enable the axis it stops.
> > > This don̫̉t happens on the X-axis, it̫̉s still all the time.
> > > Is this normal?
> > >
> > > Regards Mattias
> > >
> > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > >
> > > > Hi Mattias,
> > > >
> > > > Your calculations look correct.
> > > >
> > > > I would lower gains to improve stability.
> > > >
> > > > Your Max I is set at 200.ÃÆ'ââ¬Å¡ÃâàTo go at high speed the I term would need more than that so increase to more than what is needed as Output to fo at your max speed.
> > > >
> > > > Regards
> > > > TK
> > > >
> > > >
> > > >
> > > >
> > > > ________________________________
> > > > From: maba_sweden
> > > > To: DynoMotion@yahoogroups.com
> > > > Sent: Tuesday, February 5, 2013 8:15 AM
> > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > >
> > > >
> > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > Hi Again
> > > >
> > > > The machine specs are X Y 5000mm/min Z 2500mm/min
> > > > I got ~2000steps/mm so IÃÆ'ââ¬Å¡Ãâôm aiming for V:160000 that should be~5m/min
> > > > and A:666664 witch should be ~0,25s
> > > > If my calcylations are right :-)
> > > >
> > > > I tested to increase the speeds of the X-axle and it runs really smooth with almost no errors at all. I have added a picture called "X-axle_finish.png" with settings V160000 A600000 running 20000 steps.
> > > >
> > > > I have started to test my Y-axle and it seems like the same gain settings as the X-axle works best.
> > > > But as soon as I increases the speed on Y-axle, I get problems with oscillation.
> > > > I have added the same 500Hz 2nd low pass filter as for X-axle.
> > > > The ballscrew seems to run smooth without any excessive play.
> > > >
> > > > Should i reduce my gains or should I use an other type of filter?
> > > >
> > > > I have added pictures under "Photos" "Y-axle V9000 5000steps" AND "Y-axle V9000 40000steps"
> > > >
> > > > Regards// Mattias
> > > >
> > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > >
> > > > > Hi Mattias,
> > > > >
> > > > > Those look pretty good.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut there does seem to be some small (several um) oscillation when moving in the +X direction.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt is hard to see what frequency it is without zooming in.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàA Bode Plot might be helpful to understand things.
> > > > >
> > > > > But realistically that is pretty close to what can be expected.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIf the oscillation bothers you then you might reduce the gains enough to get rid of it and accept a bit larger errors.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIf the oscillations don't go away with lower gains then it might be some sort of vibration or chatter.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny ideas on why motion one way is different from the other?
> > > > >
> > > > >
> > > > > It is probably time to try higher speeds and accelerations.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàHave you tried those?ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàWhat are your goals?
> > > > >
> > > > > Regards
> > > > > TK
> > > > >
> > > > >
> > > > >
> > > > > ________________________________
> > > > > From: maba_sweden
> > > > > To: DynoMotion@yahoogroups.com
> > > > > Sent: Thursday, January 10, 2013 3:35 PM
> > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > >
> > > > >
> > > > > ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > I donÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt know what I was doing wrong the last time i tested I-gain but I went out and tested again and know it worked.
> > > > > Both with step size 2000 and 20000
> > > > > It worked best with I 0.025 with P4 D120
> > > > >
> > > > > new pictures under photos named *_P4I0025D120
> > > > >
> > > > > Is the position error ok, I think it looks very blue?
> > > > >
> > > > > IÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôm having some trouble to understand the scales in the graph window, can I get the scales to follow the zoom?
> > > > >
> > > > > Thanks
> > > > > Regards Mattias
> > > > >
> > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > >
> > > > > > Hi Mattias,
> > > > > >
> > > > > > Your errors are only several um so that is probably as good as one can expect.
> > > > > >
> > > > > > I don't understand why the small I gain causes a problem with long moves.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt might be totally unrelated and you are just very marginal to begin with.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt would be helpful to capture a plot of it going unstable.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI gain will make the system more unstable but a tiny amount should not make a significant difference.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry even smaller (1e-6) and increase until you see an effect.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny amount whatsoever should eventually drive any residual error to zero but just take a long time.
> > > > > >
> > > > > > Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent.
> > > > > >
> > > > > > Regards
> > > > > > TK
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > ________________________________
> > > > > > From: maba_sweden
> > > > > > To: DynoMotion@yahoogroups.com
> > > > > > Sent: Thursday, January 10, 2013 1:59 PM
> > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > >
> > > > > >
> > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > Thanks for the explanation Tom
> > > > > > Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
> > > > > >
> > > > > > I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
> > > > > >
> > > > > > I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
> > > > > >
> > > > > > I still got the few count difference when it stops after the + move but i corrected by the system.
> > > > > > And then i always get a few counts wrong when it stops after the - move.
> > > > > >
> > > > > > I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesnÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go away before i disable the axle. The output signal gets green over the whole screen when it happens.
> > > > > >
> > > > > > Am I on the right way?
> > > > > > Any tip what to try next?
> > > > > >
> > > > > > Regards Mattias
> > > > > >
> > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > >
> > > > > > > Hi Mattias,
> > > > > > >
> > > > > > > If the system was working ok before then the amplifier is probably set up reasonably well. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYour schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThat is probably the tachometer gain setting. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThe higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position.
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad.
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI would leave that alone for now.
> > > > > > >
> > > > > > > Regards
> > > > > > > TK
> > > > > > >
> > > > > > >
> > > > > > > ________________________________
> > > > > > > From: maba_sweden
> > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > Sent: Wednesday, January 9, 2013 7:01 PM
> > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > >
> > > > > > >
> > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > Yes the output looked better
> > > > > > >
> > > > > > > I noticed a sound in the plus direction but not in the minus so that observation is spot on.
> > > > > > >
> > > > > > > My Amplifiers are the original ones from GME systems sweden 1985
> > > > > > > tha label says 10KHz SW FERQ
> > > > > > > And it have tacho feedback
> > > > > > >
> > > > > > > I have uploaded a picture of the label and the schematics of the amplifier
> > > > > > >
> > > > > > > To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> > > > > > > And I havenÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt read the part about the Bode Plot yet, guess i have some work to do.
> > > > > > >
> > > > > > > Regards Mattias
> > > > > > >
> > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > >
> > > > > > > > Hi Mattias,
> > > > > > > >
> > > > > > > > Notice huge spikes in the output are now gone.
> > > > > > > >
> > > > > > > > But there is servo oscillation moving in the plus direction (but not in the negative). ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàProbably different friction moving one way vs the other?
> > > > > > > >
> > > > > > > > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> > > > > > > >
> > > > > > > > What type of amplifiers do you have? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIs there tachometer feedback? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàDo you have the velocity loop in the amplifier optimized?
> > > > > > > >
> > > > > > > > If you could get a reasonable Bode plot it would help understand things.
> > > > > > > >
> > > > > > > > Regards
> > > > > > > > TK
> > > > > > > >
> > > > > > > >
> > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > >
> > > > > > > >
> > > > > > > > ________________________________
> > > > > > > > From: maba_sweden
> > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > Sent: Wednesday, January 9, 2013 6:14 PM
> > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > >
> > > > > > > >
> > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > Hi Tom
> > > > > > > > Thanks for your answer.
> > > > > > > >
> > > > > > > > When i move 2000 steps itÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôs close to 1mm so that would give ~50800 counts per inch
> > > > > > > >
> > > > > > > > I just tried the filter and did one test run with P5 I0 D100
> > > > > > > > I uploaded some new pictures.
> > > > > > > > I will continue my tests tomorrow and se what differens IÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôll get now with the filter.
> > > > > > > >
> > > > > > > > Thanks!
> > > > > > > > Regards Mattias
> > > > > > > >
> > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > >
> > > > > > > > > Hi Mattias,
> > > > > > > > >
> > > > > > > > > What is the resolution of your axis? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâà(how many counts per inch?)
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSo we can determine what size errors we are seeing.
> > > > > > > > >
> > > > > > > > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYou can smooth that with a filter and the D gain will probably
> become more effective.
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4.
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen see what P and D gains work best for you.
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàMaybe something like P=5 D=100 might be a starting point.
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen try adding I gain.
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry increasing it until it
> > > > has some effect. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàPossibly I=0.01 or more.
> > > > > > > > >
> > > > > > > > > Regards
> > > > > > > > > TK
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ________________________________
> > > > > > > > > From: maba_sweden
> > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > > > > > > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > Hi
> > > > > > > > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > > > > > > > >
> > > > > > > > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > > > > > > > >
> > > > > > > > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and couldnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go any further.
> > > > > > > > >
> > > > > > > > > I also tried some 0.001 I gain but i didnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt though it maked it any better.
> > > > > > > > >
> > > > > > > > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > > > > > > > They are named as the setting i tested
> > > > > > > > >
> > > > > > > > > Any tip on how I should proceed?
> > > > > > > > >
> > > > > > > > > Regards Mattias
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6771 |
From: maba_sweden |
Date: 2/19/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hello, now I´m back :-)
I have changed the bearings and re-glued my static magnets and the servo motor seems to work fine.
I have begun to tune my Z axle and it seems to respond really well.
The only problem I got is that I get cycle oscillations while resting at the target.
I tried to ad dead band but it wont stop.
I tested range from 0-8 with 0 Gain but it won´t stop.
I have also tried to ad some gain from 0.1-1
I´m not really sure how the Dead Band works.
Probably I´m just doing something wrong and would like some tip what settings to start with.
Or should i use something else then the dead band?
I have added a piture under Photos called "Z-axle 1902"
It´s possible to the oscillations there before the movements starts, sometimes the oscillations are less after the movement.
Regards Mattias
--- In DynoMotion@yahoogroups.com, "maba_sweden" <maba@...> wrote:
>
> Hi Tom
>
> Just wanted you to know that it wasn´t the drivers fault, if it only could have been that :-)
> The Z motor is bad. One of the static magnets was loose, that may explains the uneven motion of the motor, the rear bearing must also bad because it´s heavy to rotate.
> Probably is this because of coolant that have came in the motor and it have started to rust.
> I have to see if i can rebuild the motor, some says the magnets can loose there strength if you take it apart.
> Or else I have to try to find a new/used/rebuild Contraves 402 117V 3,4nm DC servo motor before I can continue my work...
>
> Thanks for all your help.
> Hope I will be back soon!
> Regards Mattias
>
>
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> >
> > Hi Mattias,
> >
> > Now that I look at the plot again the velocity is very low 4000 counts/sec and the acceleration to go that speed is insignificant. There must be something major wrong.
> >
> > Maybe you are missing a power supply voltage or something so it can't drive positive.
> >
> > But if the car is going too slow and the gas pedal is all the way to the floor - it isn't the driver's fault :}
> >
> > Regards
> > TK
> >
> >
> >
> >
> >
> >
> > ________________________________
> > From: maba_sweden
> > To: DynoMotion@yahoogroups.com
> > Sent: Tuesday, February 5, 2013 4:30 PM
> > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> >
> >
> > Â
> > Yes positive is lifting.
> > The machine got no counter weight what I´m aware, and according to the manual it should be able to travel 2500mm/min half the speed of the X and Y.
> > Probably it´s just the acceleration that is to aggressive, if i remember correctly it accelerated pretty smoth with the original Sinumerik system.
> > I will test with a lower setting and see if it gets any better.
> >
> > Thanks
> > Regards//Mattias
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > >
> > > Hi Mattias,
> > >
> > > It looks like you probably have a lot of weight and friction.ÃÂ Notice as it goes positive (I assume this is lifting weight) the position (red) lags way behind moving at a varying speed even though the output (green) is fully saturated at 2000 counts (~10V).ÃÂ Your Motor/Amplifier/Supply probably just doesn't have enough power. ÃÂ ÃÂ But notice going negative (down?) it is able to go very fast even with almost zero command to the amplifier.ÃÂ Is there supposed to be a counter balance or something?ÃÂ Or you are just trying to move too fast.ÃÂ How did the machine perform originally?
> > >
> > > Regards
> > > TK
> > >
> > > ÃÂ
> > >
> > >
> > >
> > >
> > > ________________________________
> > > From: maba_sweden
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Tuesday, February 5, 2013 3:53 PM
> > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > >
> > >
> > > ÃÂ
> > > Ok, one thing that i donôt have to worry about :-)
> > > I tested some high speeds and longer runs and ended up lowering the gain pretty much on both X and Y.
> > >
> > > I moved on to my Z-axle and it seems like thereôs not enough output to go along with the Command.
> > > I have max output 2000 intergration 1000
> > > Velocity 4000
> > > Acceleration 40000
> > >
> > > I also added 2nd low pass filter 500hz 1,4q as for the X and Y axle
> > >
> > > When i moves 2000 steps itôs ok but when i test 4000steps or more it gets really bad.
> > > I have added a picture under photos named "Z-axle_first_run"
> > >
> > > Regards Mattias
> > >
> > >
> > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > >
> > > > Hi Mattias,
> > > >
> > > > This is normal.ÃâàIt is almost impossible to command analog velocity of exactly zero because there will always be some analog noise and offsets.ÃâàIf this is a problem then you will need to wire a disable signal to your amplifier.ÃâàThen add C statement to disable the Amplifier whenever the KFLOP Axis Channel is disabled.ÃâàSee the WatchEnable.c example.ÃâàThis is not a problem when the servo is enabled because the feedback will command whatever voltage is necessary to hold position.
> > > >
> > > >
> > > > HTH
> > > > Regards
> > > > TK
> > > >
> > > >
> > > >
> > > > ________________________________
> > > > From: maba_sweden
> > > > To: DynoMotion@yahoogroups.com
> > > > Sent: Tuesday, February 5, 2013 9:36 AM
> > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > >
> > > >
> > > > ÃâÃÂ
> > > > Thanks for your inputs Tom!
> > > >
> > > > I will test to increase the Max I and decrease the gain settings to see if it will be more stable.
> > > >
> > > > One more thing,
> > > > if I disable the axis in "step resonse" the Y-axle will slowly start to move and as soon as I enable the axis it stops.
> > > > This don̫̉t happens on the X-axis, it̫̉s still all the time.
> > > > Is this normal?
> > > >
> > > > Regards Mattias
> > > >
> > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > >
> > > > > Hi Mattias,
> > > > >
> > > > > Your calculations look correct.
> > > > >
> > > > > I would lower gains to improve stability.
> > > > >
> > > > > Your Max I is set at 200.ÃÆ'ââ¬Å¡ÃâàTo go at high speed the I term would need more than that so increase to more than what is needed as Output to fo at your max speed.
> > > > >
> > > > > Regards
> > > > > TK
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > ________________________________
> > > > > From: maba_sweden
> > > > > To: DynoMotion@yahoogroups.com
> > > > > Sent: Tuesday, February 5, 2013 8:15 AM
> > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > >
> > > > >
> > > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > Hi Again
> > > > >
> > > > > The machine specs are X Y 5000mm/min Z 2500mm/min
> > > > > I got ~2000steps/mm so IÃÆ'ââ¬Å¡Ãâôm aiming for V:160000 that should be~5m/min
> > > > > and A:666664 witch should be ~0,25s
> > > > > If my calcylations are right :-)
> > > > >
> > > > > I tested to increase the speeds of the X-axle and it runs really smooth with almost no errors at all. I have added a picture called "X-axle_finish.png" with settings V160000 A600000 running 20000 steps.
> > > > >
> > > > > I have started to test my Y-axle and it seems like the same gain settings as the X-axle works best.
> > > > > But as soon as I increases the speed on Y-axle, I get problems with oscillation.
> > > > > I have added the same 500Hz 2nd low pass filter as for X-axle.
> > > > > The ballscrew seems to run smooth without any excessive play.
> > > > >
> > > > > Should i reduce my gains or should I use an other type of filter?
> > > > >
> > > > > I have added pictures under "Photos" "Y-axle V9000 5000steps" AND "Y-axle V9000 40000steps"
> > > > >
> > > > > Regards// Mattias
> > > > >
> > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > >
> > > > > > Hi Mattias,
> > > > > >
> > > > > > Those look pretty good.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut there does seem to be some small (several um) oscillation when moving in the +X direction.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt is hard to see what frequency it is without zooming in.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàA Bode Plot might be helpful to understand things.
> > > > > >
> > > > > > But realistically that is pretty close to what can be expected.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIf the oscillation bothers you then you might reduce the gains enough to get rid of it and accept a bit larger errors.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIf the oscillations don't go away with lower gains then it might be some sort of vibration or chatter.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny ideas on why motion one way is different from the other?
> > > > > >
> > > > > >
> > > > > > It is probably time to try higher speeds and accelerations.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàHave you tried those?ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàWhat are your goals?
> > > > > >
> > > > > > Regards
> > > > > > TK
> > > > > >
> > > > > >
> > > > > >
> > > > > > ________________________________
> > > > > > From: maba_sweden
> > > > > > To: DynoMotion@yahoogroups.com
> > > > > > Sent: Thursday, January 10, 2013 3:35 PM
> > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > >
> > > > > >
> > > > > > ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > I donÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt know what I was doing wrong the last time i tested I-gain but I went out and tested again and know it worked.
> > > > > > Both with step size 2000 and 20000
> > > > > > It worked best with I 0.025 with P4 D120
> > > > > >
> > > > > > new pictures under photos named *_P4I0025D120
> > > > > >
> > > > > > Is the position error ok, I think it looks very blue?
> > > > > >
> > > > > > IÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôm having some trouble to understand the scales in the graph window, can I get the scales to follow the zoom?
> > > > > >
> > > > > > Thanks
> > > > > > Regards Mattias
> > > > > >
> > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > >
> > > > > > > Hi Mattias,
> > > > > > >
> > > > > > > Your errors are only several um so that is probably as good as one can expect.
> > > > > > >
> > > > > > > I don't understand why the small I gain causes a problem with long moves.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt might be totally unrelated and you are just very marginal to begin with.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt would be helpful to capture a plot of it going unstable.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI gain will make the system more unstable but a tiny amount should not make a significant difference.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry even smaller (1e-6) and increase until you see an effect.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny amount whatsoever should eventually drive any residual error to zero but just take a long time.
> > > > > > >
> > > > > > > Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent.
> > > > > > >
> > > > > > > Regards
> > > > > > > TK
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > ________________________________
> > > > > > > From: maba_sweden
> > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > Sent: Thursday, January 10, 2013 1:59 PM
> > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > >
> > > > > > >
> > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > Thanks for the explanation Tom
> > > > > > > Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
> > > > > > >
> > > > > > > I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
> > > > > > >
> > > > > > > I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
> > > > > > >
> > > > > > > I still got the few count difference when it stops after the + move but i corrected by the system.
> > > > > > > And then i always get a few counts wrong when it stops after the - move.
> > > > > > >
> > > > > > > I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesnÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go away before i disable the axle. The output signal gets green over the whole screen when it happens.
> > > > > > >
> > > > > > > Am I on the right way?
> > > > > > > Any tip what to try next?
> > > > > > >
> > > > > > > Regards Mattias
> > > > > > >
> > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > >
> > > > > > > > Hi Mattias,
> > > > > > > >
> > > > > > > > If the system was working ok before then the amplifier is probably set up reasonably well. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYour schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThat is probably the tachometer gain setting. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThe higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position.
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad.
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI would leave that alone for now.
> > > > > > > >
> > > > > > > > Regards
> > > > > > > > TK
> > > > > > > >
> > > > > > > >
> > > > > > > > ________________________________
> > > > > > > > From: maba_sweden
> > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > Sent: Wednesday, January 9, 2013 7:01 PM
> > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > >
> > > > > > > >
> > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > Yes the output looked better
> > > > > > > >
> > > > > > > > I noticed a sound in the plus direction but not in the minus so that observation is spot on.
> > > > > > > >
> > > > > > > > My Amplifiers are the original ones from GME systems sweden 1985
> > > > > > > > tha label says 10KHz SW FERQ
> > > > > > > > And it have tacho feedback
> > > > > > > >
> > > > > > > > I have uploaded a picture of the label and the schematics of the amplifier
> > > > > > > >
> > > > > > > > To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> > > > > > > > And I havenÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt read the part about the Bode Plot yet, guess i have some work to do.
> > > > > > > >
> > > > > > > > Regards Mattias
> > > > > > > >
> > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > >
> > > > > > > > > Hi Mattias,
> > > > > > > > >
> > > > > > > > > Notice huge spikes in the output are now gone.
> > > > > > > > >
> > > > > > > > > But there is servo oscillation moving in the plus direction (but not in the negative). ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàProbably different friction moving one way vs the other?
> > > > > > > > >
> > > > > > > > > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> > > > > > > > >
> > > > > > > > > What type of amplifiers do you have? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIs there tachometer feedback? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàDo you have the velocity loop in the amplifier optimized?
> > > > > > > > >
> > > > > > > > > If you could get a reasonable Bode plot it would help understand things.
> > > > > > > > >
> > > > > > > > > Regards
> > > > > > > > > TK
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ________________________________
> > > > > > > > > From: maba_sweden
> > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > Sent: Wednesday, January 9, 2013 6:14 PM
> > > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > Hi Tom
> > > > > > > > > Thanks for your answer.
> > > > > > > > >
> > > > > > > > > When i move 2000 steps itÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôs close to 1mm so that would give ~50800 counts per inch
> > > > > > > > >
> > > > > > > > > I just tried the filter and did one test run with P5 I0 D100
> > > > > > > > > I uploaded some new pictures.
> > > > > > > > > I will continue my tests tomorrow and se what differens IÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôll get now with the filter.
> > > > > > > > >
> > > > > > > > > Thanks!
> > > > > > > > > Regards Mattias
> > > > > > > > >
> > > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > > >
> > > > > > > > > > Hi Mattias,
> > > > > > > > > >
> > > > > > > > > > What is the resolution of your axis? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâà(how many counts per inch?)
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSo we can determine what size errors we are seeing.
> > > > > > > > > >
> > > > > > > > > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot). ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYou can smooth that with a filter and the D gain will probably
> > become more effective.
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4.
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen see what P and D gains work best for you.
> > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàMaybe something like P=5 D=100 might be a starting point.
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen try adding I gain.
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry increasing it until it
> > > > > has some effect. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàPossibly I=0.01 or more.
> > > > > > > > > >
> > > > > > > > > > Regards
> > > > > > > > > > TK
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > ________________________________
> > > > > > > > > > From: maba_sweden
> > > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > > > > > > > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > Hi
> > > > > > > > > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > > > > > > > > >
> > > > > > > > > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > > > > > > > > >
> > > > > > > > > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and couldnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go any further.
> > > > > > > > > >
> > > > > > > > > > I also tried some 0.001 I gain but i didnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt though it maked it any better.
> > > > > > > > > >
> > > > > > > > > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > > > > > > > > They are named as the setting i tested
> > > > > > > > > >
> > > > > > > > > > Any tip on how I should proceed?
> > > > > > > > > >
> > > > > > > > > > Regards Mattias
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6774 |
From: Tom Kerekes |
Date: 2/19/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias,
It is a fairly common characteristic of Servos to "hunt" or dither when at rest. It is just their nature to always be correcting. Especially if there is any Integrator gain. It usually is only an encoder count or two. It can also depend on whether the target position is a fractional position or not. For example commanding to a position of 100.25 will always have an error whether the encoder reads either 100 or 101. Any Integrator will force the
average error to be zero so 25% of the time will be spent at 101 and 75% at 100.
Sometimes you can get things to sit still using dead band, but it is very complicated and depends on the dynamics of your system. Friction and compliance and so forth.
Regards TK
| Group: DynoMotion |
Message: 6775 |
From: maba_sweden |
Date: 2/19/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
I made some adjustments on the Z-brake and the belt between the servo motor and ball screw and I got rid of the oscillation when resting at target.
But know I got another problem.
Every time I start a move the servo motor twiches (i think it´s called so) and then it makes the move smooth.
I have tried to zero the servo before the move but it looks strange on the tuning window, the position doesn´t starts where the command are.
I´we added i picture called "Z start of move"
Are there any settings I should test or should i continue to make adjustments on the hardware?
Regards Mattias
--- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@...> wrote:
>
> Hi Mattias,
>
> It is a fairly common characteristic of Servos to "hunt" or dither when at rest. It is just their nature to always be correcting. Especially if there is any Integrator gain. It usually is only an encoder count or two. It can also depend on whether the target position is a fractional position or not. For example commanding to a position of 100.25 will always have an error whether the encoder reads either 100 or 101. Any Integrator will force the average error to be zero so 25% of the time will be spent at 101 and 75% at 100.
>
> Sometimes you can get things to sit still using dead band, but it is very complicated and depends on the dynamics of your system. Friction and compliance and so forth.
>
> Regards
> TKÂ
>
>
>
> ________________________________
> From: maba_sweden <maba@...>
> To: DynoMotion@yahoogroups.com
> Sent: Tuesday, February 19, 2013 7:28 AM
> Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
>
>
> Â
> Hello, now I´m back :-)
>
> I have changed the bearings and re-glued my static magnets and the servo motor seems to work fine.
> I have begun to tune my Z axle and it seems to respond really well.
> The only problem I got is that I get cycle oscillations while resting at the target.
> I tried to ad dead band but it wont stop.
> I tested range from 0-8 with 0 Gain but it won´t stop.
> I have also tried to ad some gain from 0.1-1
> I´m not really sure how the Dead Band works.
>
> Probably I´m just doing something wrong and would like some tip what settings to start with.
> Or should i use something else then the dead band?
>
> I have added a piture under Photos called "Z-axle 1902"
> It´s possible to the oscillations there before the movements starts, sometimes the oscillations are less after the movement.
>
> Regards Mattias
>
> --- In DynoMotion@yahoogroups.com, "maba_sweden" wrote:
> >
> > Hi Tom
> >
> > Just wanted you to know that it wasn´t the drivers fault, if it only could have been that :-)
> > The Z motor is bad. One of the static magnets was loose, that may explains the uneven motion of the motor, the rear bearing must also bad because it´s heavy to rotate.
> > Probably is this because of coolant that have came in the motor and it have started to rust.
> > I have to see if i can rebuild the motor, some says the magnets can loose there strength if you take it apart.
> > Or else I have to try to find a new/used/rebuild Contraves 402 117V 3,4nm DC servo motor before I can continue my work...
> >
> > Thanks for all your help.
> > Hope I will be back soon!
> > Regards Mattias
> >
> >
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > >
> > > Hi Mattias,
> > >
> > > Now that I look at the plot again the velocity is very low 4000 counts/sec and the acceleration to go that speed is insignificant.ÃÂ There must be something major wrong.
> > >
> > > Maybe you are missing a power supply voltage or something so it can't drive positive.
> > >
> > > But if the car is going too slow and the gas pedal is all the way to the floor - it isn't the driver's fault :}
> > >
> > > Regards
> > > TK
> > >
> > >
> > >
> > >
> > >
> > >
> > > ________________________________
> > > From: maba_sweden
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Tuesday, February 5, 2013 4:30 PM
> > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > >
> > >
> > > ÃÂ
> > > Yes positive is lifting.
> > > The machine got no counter weight what Iôm aware, and according to the manual it should be able to travel 2500mm/min half the speed of the X and Y.
> > > Probably itôs just the acceleration that is to aggressive, if i remember correctly it accelerated pretty smoth with the original Sinumerik system.
> > > I will test with a lower setting and see if it gets any better.
> > >
> > > Thanks
> > > Regards//Mattias
> > >
> > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > >
> > > > Hi Mattias,
> > > >
> > > > It looks like you probably have a lot of weight and friction.ÃâàNotice as it goes positive (I assume this is lifting weight) the position (red) lags way behind moving at a varying speed even though the output (green) is fully saturated at 2000 counts (~10V).ÃâàYour Motor/Amplifier/Supply probably just doesn't have enough power. ÃâàÃâàBut notice going negative (down?) it is able to go very fast even with almost zero command to the amplifier.ÃâàIs there supposed to be a counter balance or something?ÃâàOr you are just trying to move too fast.ÃâàHow did the machine perform originally?
> > > >
> > > > Regards
> > > > TK
> > > >
> > > > ÃâÃÂ
> > > >
> > > >
> > > >
> > > >
> > > > ________________________________
> > > > From: maba_sweden
> > > > To: DynoMotion@yahoogroups.com
> > > > Sent: Tuesday, February 5, 2013 3:53 PM
> > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > >
> > > >
> > > > ÃâÃÂ
> > > > Ok, one thing that i don̫̉t have to worry about :-)
> > > > I tested some high speeds and longer runs and ended up lowering the gain pretty much on both X and Y.
> > > >
> > > > I moved on to my Z-axle and it seems like therẻ̫s not enough output to go along with the Command.
> > > > I have max output 2000 intergration 1000
> > > > Velocity 4000
> > > > Acceleration 40000
> > > >
> > > > I also added 2nd low pass filter 500hz 1,4q as for the X and Y axle
> > > >
> > > > When i moves 2000 steps it̫̉s ok but when i test 4000steps or more it gets really bad.
> > > > I have added a picture under photos named "Z-axle_first_run"
> > > >
> > > > Regards Mattias
> > > >
> > > >
> > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > >
> > > > > Hi Mattias,
> > > > >
> > > > > This is normal.ÃÆ'ââ¬Å¡ÃâàIt is almost impossible to command analog velocity of exactly zero because there will always be some analog noise and offsets.ÃÆ'ââ¬Å¡ÃâàIf this is a problem then you will need to wire a disable signal to your amplifier.ÃÆ'ââ¬Å¡ÃâàThen add C statement to disable the Amplifier whenever the KFLOP Axis Channel is disabled.ÃÆ'ââ¬Å¡ÃâàSee the WatchEnable.c example.ÃÆ'ââ¬Å¡ÃâàThis is not a problem when the servo is enabled because the feedback will command whatever voltage is necessary to hold position.
> > > > >
> > > > >
> > > > > HTH
> > > > > Regards
> > > > > TK
> > > > >
> > > > >
> > > > >
> > > > > ________________________________
> > > > > From: maba_sweden
> > > > > To: DynoMotion@yahoogroups.com
> > > > > Sent: Tuesday, February 5, 2013 9:36 AM
> > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > >
> > > > >
> > > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > Thanks for your inputs Tom!
> > > > >
> > > > > I will test to increase the Max I and decrease the gain settings to see if it will be more stable.
> > > > >
> > > > > One more thing,
> > > > > if I disable the axis in "step resonse" the Y-axle will slowly start to move and as soon as I enable the axis it stops.
> > > > > This donÃÆ'ââ¬Å¡Ãâôt happens on the X-axis, itÃÆ'ââ¬Å¡Ãâôs still all the time.
> > > > > Is this normal?
> > > > >
> > > > > Regards Mattias
> > > > >
> > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > >
> > > > > > Hi Mattias,
> > > > > >
> > > > > > Your calculations look correct.
> > > > > >
> > > > > > I would lower gains to improve stability.
> > > > > >
> > > > > > Your Max I is set at 200.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTo go at high speed the I term would need more than that so increase to more than what is needed as Output to fo at your max speed.
> > > > > >
> > > > > > Regards
> > > > > > TK
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > ________________________________
> > > > > > From: maba_sweden
> > > > > > To: DynoMotion@yahoogroups.com
> > > > > > Sent: Tuesday, February 5, 2013 8:15 AM
> > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > >
> > > > > >
> > > > > > ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > Hi Again
> > > > > >
> > > > > > The machine specs are X Y 5000mm/min Z 2500mm/min
> > > > > > I got ~2000steps/mm so IÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôm aiming for V:160000 that should be~5m/min
> > > > > > and A:666664 witch should be ~0,25s
> > > > > > If my calcylations are right :-)
> > > > > >
> > > > > > I tested to increase the speeds of the X-axle and it runs really smooth with almost no errors at all. I have added a picture called "X-axle_finish.png" with settings V160000 A600000 running 20000 steps.
> > > > > >
> > > > > > I have started to test my Y-axle and it seems like the same gain settings as the X-axle works best.
> > > > > > But as soon as I increases the speed on Y-axle, I get problems with oscillation.
> > > > > > I have added the same 500Hz 2nd low pass filter as for X-axle.
> > > > > > The ballscrew seems to run smooth without any excessive play.
> > > > > >
> > > > > > Should i reduce my gains or should I use an other type of filter?
> > > > > >
> > > > > > I have added pictures under "Photos" "Y-axle V9000 5000steps" AND "Y-axle V9000 40000steps"
> > > > > >
> > > > > > Regards// Mattias
> > > > > >
> > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > >
> > > > > > > Hi Mattias,
> > > > > > >
> > > > > > > Those look pretty good.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut there does seem to be some small (several um) oscillation when moving in the +X direction.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt is hard to see what frequency it is without zooming in.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàA Bode Plot might be helpful to understand things.
> > > > > > >
> > > > > > > But realistically that is pretty close to what can be expected.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIf the oscillation bothers you then you might reduce the gains enough to get rid of it and accept a bit larger errors.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIf the oscillations don't go away with lower gains then it might be some sort of vibration or chatter.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny ideas on why motion one way is different from the other?
> > > > > > >
> > > > > > >
> > > > > > > It is probably time to try higher speeds and accelerations.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàHave you tried those?ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàWhat are your goals?
> > > > > > >
> > > > > > > Regards
> > > > > > > TK
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > ________________________________
> > > > > > > From: maba_sweden
> > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > Sent: Thursday, January 10, 2013 3:35 PM
> > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > >
> > > > > > >
> > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > I donÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt know what I was doing wrong the last time i tested I-gain but I went out and tested again and know it worked.
> > > > > > > Both with step size 2000 and 20000
> > > > > > > It worked best with I 0.025 with P4 D120
> > > > > > >
> > > > > > > new pictures under photos named *_P4I0025D120
> > > > > > >
> > > > > > > Is the position error ok, I think it looks very blue?
> > > > > > >
> > > > > > > IÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôm having some trouble to understand the scales in the graph window, can I get the scales to follow the zoom?
> > > > > > >
> > > > > > > Thanks
> > > > > > > Regards Mattias
> > > > > > >
> > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > >
> > > > > > > > Hi Mattias,
> > > > > > > >
> > > > > > > > Your errors are only several um so that is probably as good as one can expect.
> > > > > > > >
> > > > > > > > I don't understand why the small I gain causes a problem with long moves.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt might be totally unrelated and you are just very marginal to begin with.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt would be helpful to capture a plot of it going unstable.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI gain will make the system more unstable but a tiny amount should not make a significant
> difference.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry even smaller (1e-6) and increase until you see an effect.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny amount whatsoever should eventually drive any residual error to zero but just take a long time.
> > > > > > > >
> > > > > > > > Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent.
> > > > > > > >
> > > > > > > > Regards
> > > > > > > > TK
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > ________________________________
> > > > > > > > From: maba_sweden
> > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > Sent: Thursday, January 10, 2013 1:59 PM
> > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > >
> > > > > > > >
> > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > Thanks for the explanation Tom
> > > > > > > > Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
> > > > > > > >
> > > > > > > > I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
> > > > > > > >
> > > > > > > > I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
> > > > > > > >
> > > > > > > > I still got the few count difference when it stops after the + move but i corrected by the system.
> > > > > > > > And then i always get a few counts wrong when it stops after the - move.
> > > > > > > >
> > > > > > > > I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go away before i disable the axle. The output signal gets green over the whole screen when it happens.
> > > > > > > >
> > > > > > > > Am I on the right way?
> > > > > > > > Any tip what to try next?
> > > > > > > >
> > > > > > > > Regards Mattias
> > > > > > > >
> > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > >
> > > > > > > > > Hi Mattias,
> > > > > > > > >
> > > > > > > > > If the system was working ok before then the amplifier is probably set up reasonably well. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYour schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> That is probably the tachometer gain setting. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThe higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position.
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad.
> > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI would leave that alone for now.
> > > > > > > > >
> > > > > > > > > Regards
> > > > > > > > > TK
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ________________________________
> > > > > > > > > From: maba_sweden
> > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > Sent: Wednesday, January 9, 2013 7:01 PM
> > > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > Yes the output looked better
> > > > > > > > >
> > > > > > > > > I noticed a sound in the plus direction but not in the minus so that observation is spot on.
> > > > > > > > >
> > > > > > > > > My Amplifiers are the original ones from GME systems sweden 1985
> > > > > > > > > tha label says 10KHz SW FERQ
> > > > > > > > > And it have tacho feedback
> > > > > > > > >
> > > > > > > > > I have uploaded a picture of the label and the schematics of the amplifier
> > > > > > > > >
> > > > > > > > > To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> > > > > > > > > And I havenÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt read the part about the Bode Plot yet, guess i have some work to do.
> > > > > > > > >
> > > > > > > > > Regards Mattias
> > > > > > > > >
> > > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > > >
> > > > > > > > > > Hi Mattias,
> > > > > > > > > >
> > > > > > > > > > Notice huge spikes in the output are now gone.
> > > > > > > > > >
> > > > > > > > > > But there is servo oscillation moving in the plus direction (but not in the negative). ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàProbably different friction moving one way vs the other?
> > > > > > > > > >
> > > > > > > > > > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> > > > > > > > > >
> > > > > > > > > > What type of amplifiers do you have? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIs there tachometer feedback?
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàDo you have the velocity loop in the amplifier optimized?
> > > > > > > > > >
> > > > > > > > > > If you could get a reasonable Bode plot it would help understand things.
> > > > > > > > > >
> > > > > > > > > > Regards
> > > > > > > > > > TK
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > ________________________________
> > > > > > > > > > From: maba_sweden
> > > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > > Sent: Wednesday, January 9, 2013 6:14 PM
> > > > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > Hi Tom
> > > > > > > > > > Thanks for your answer.
> > > > > > > > > >
> > > > > > > > > > When i move 2000 steps itÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôs close to 1mm so that would give ~50800 counts per inch
> > > > > > > > > >
> > > > > > > > > > I just tried the filter and did one test run with P5 I0 D100
> > > > > > > > > > I uploaded some new pictures.
> > > > > > > > > > I will continue my tests tomorrow and se what differens IÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôll get now with the filter.
> > > > > > > > > >
> > > > > > > > > > Thanks!
> > > > > > > > > > Regards Mattias
> > > > > > > > > >
> > > > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > > > >
> > > > > > > > > > > Hi Mattias,
> > > > > > > > > > >
> > > > > > > > > > > What is the resolution of your axis?
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâà(how many counts per inch?)
> > >
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSo we can determine what size errors we are seeing.
> > > > > > > > > > >
> > > > > > > > > > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot).
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYou can smooth that with a filter and the D gain will probably
> > > become more effective.
> > > >
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4.
> > >
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen see what P and D gains work best for you.
> > > > >
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàMaybe something like P=5 D=100 might be a starting point.
> > >
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen try adding I gain.
> > > >
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry increasing it until it
> > > > > > has some effect.
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàPossibly I=0.01 or more.
> > > > > > > > > > >
> > > > > > > > > > > Regards
> > > > > > > > > > > TK
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > ________________________________
> > > > > > > > > > > From: maba_sweden
> > > > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > > > > > > > > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > > Hi
> > > > > > > > > > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > > > > > > > > > >
> > > > > > > > > > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > > > > > > > > > >
> > > > > > > > > > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and
> couldnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃ
> Â¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go any further.
> > > > > > > > > > >
> > > > > > > > > > > I also tried some 0.001 I gain but i
> didnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãââ
> ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt though it maked it any better.
> > > > > > > > > > >
> > > > > > > > > > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > > > > > > > > > They are named as the setting i tested
> > > > > > > > > > >
> > > > > > > > > > > Any tip on how I should proceed?
> > > > > > > > > > >
> > > > > > > > > > > Regards Mattias
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6800 |
From: maba_sweden |
Date: 2/23/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
I have tried to change the servomotor and used the servodriver and cables from the X-axle to eliminate hardware faults.
But I still get the same problem.
The Position don´t starts where the command are so the motor makes makes a quick move to correct this in the beginning of a move.
And then during + move (Z table down) I get oscillations, but - (Z table up) move is smooth and fine.
I have tried alot of different PID settings but I can´t get it to work.
I also tried to make the move with the brake engaged and then it worked really nice, I guess this is because I get another type of friction so it´s easier for the regulator to cope with it.
Should I try some kind of filter or how do i proceed?
I have uploaded picture "Z with break" and "Z without break" and "Z start of move" under photos mattias combimill
Regards Mattias
--- In DynoMotion@yahoogroups.com, "maba_sweden" <maba@...> wrote:
>
> I made some adjustments on the Z-brake and the belt between the servo motor and ball screw and I got rid of the oscillation when resting at target.
> But know I got another problem.
>
> Every time I start a move the servo motor twiches (i think it´s called so) and then it makes the move smooth.
> I have tried to zero the servo before the move but it looks strange on the tuning window, the position doesn´t starts where the command are.
> I´we added i picture called "Z start of move"
>
> Are there any settings I should test or should i continue to make adjustments on the hardware?
>
> Regards Mattias
>
>
>
> --- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@> wrote:
> >
> > Hi Mattias,
> >
> > It is a fairly common characteristic of Servos to "hunt" or dither when at rest. It is just their nature to always be correcting. Especially if there is any Integrator gain. It usually is only an encoder count or two. It can also depend on whether the target position is a fractional position or not. For example commanding to a position of 100.25 will always have an error whether the encoder reads either 100 or 101. Any Integrator will force the average error to be zero so 25% of the time will be spent at 101 and 75% at 100.
> >
> > Sometimes you can get things to sit still using dead band, but it is very complicated and depends on the dynamics of your system. Friction and compliance and so forth.
> >
> > Regards
> > TKÂ
> >
> >
> >
> > ________________________________
> > From: maba_sweden <maba@>
> > To: DynoMotion@yahoogroups.com
> > Sent: Tuesday, February 19, 2013 7:28 AM
> > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> >
> >
> > Â
> > Hello, now I´m back :-)
> >
> > I have changed the bearings and re-glued my static magnets and the servo motor seems to work fine.
> > I have begun to tune my Z axle and it seems to respond really well.
> > The only problem I got is that I get cycle oscillations while resting at the target.
> > I tried to ad dead band but it wont stop.
> > I tested range from 0-8 with 0 Gain but it won´t stop.
> > I have also tried to ad some gain from 0.1-1
> > I´m not really sure how the Dead Band works.
> >
> > Probably I´m just doing something wrong and would like some tip what settings to start with.
> > Or should i use something else then the dead band?
> >
> > I have added a piture under Photos called "Z-axle 1902"
> > It´s possible to the oscillations there before the movements starts, sometimes the oscillations are less after the movement.
> >
> > Regards Mattias
> >
> > --- In DynoMotion@yahoogroups.com, "maba_sweden" wrote:
> > >
> > > Hi Tom
> > >
> > > Just wanted you to know that it wasn´t the drivers fault, if it only could have been that :-)
> > > The Z motor is bad. One of the static magnets was loose, that may explains the uneven motion of the motor, the rear bearing must also bad because it´s heavy to rotate.
> > > Probably is this because of coolant that have came in the motor and it have started to rust.
> > > I have to see if i can rebuild the motor, some says the magnets can loose there strength if you take it apart.
> > > Or else I have to try to find a new/used/rebuild Contraves 402 117V 3,4nm DC servo motor before I can continue my work...
> > >
> > > Thanks for all your help.
> > > Hope I will be back soon!
> > > Regards Mattias
> > >
> > >
> > >
> > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > >
> > > > Hi Mattias,
> > > >
> > > > Now that I look at the plot again the velocity is very low 4000 counts/sec and the acceleration to go that speed is insignificant.ÃÂ There must be something major wrong.
> > > >
> > > > Maybe you are missing a power supply voltage or something so it can't drive positive.
> > > >
> > > > But if the car is going too slow and the gas pedal is all the way to the floor - it isn't the driver's fault :}
> > > >
> > > > Regards
> > > > TK
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > ________________________________
> > > > From: maba_sweden
> > > > To: DynoMotion@yahoogroups.com
> > > > Sent: Tuesday, February 5, 2013 4:30 PM
> > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > >
> > > >
> > > > ÃÂ
> > > > Yes positive is lifting.
> > > > The machine got no counter weight what Iôm aware, and according to the manual it should be able to travel 2500mm/min half the speed of the X and Y.
> > > > Probably itôs just the acceleration that is to aggressive, if i remember correctly it accelerated pretty smoth with the original Sinumerik system.
> > > > I will test with a lower setting and see if it gets any better.
> > > >
> > > > Thanks
> > > > Regards//Mattias
> > > >
> > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > >
> > > > > Hi Mattias,
> > > > >
> > > > > It looks like you probably have a lot of weight and friction.ÃâàNotice as it goes positive (I assume this is lifting weight) the position (red) lags way behind moving at a varying speed even though the output (green) is fully saturated at 2000 counts (~10V).ÃâàYour Motor/Amplifier/Supply probably just doesn't have enough power. ÃâàÃâàBut notice going negative (down?) it is able to go very fast even with almost zero command to the amplifier.ÃâàIs there supposed to be a counter balance or something?ÃâàOr you are just trying to move too fast.ÃâàHow did the machine perform originally?
> > > > >
> > > > > Regards
> > > > > TK
> > > > >
> > > > > ÃâÃÂ
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > ________________________________
> > > > > From: maba_sweden
> > > > > To: DynoMotion@yahoogroups.com
> > > > > Sent: Tuesday, February 5, 2013 3:53 PM
> > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > >
> > > > >
> > > > > ÃâÃÂ
> > > > > Ok, one thing that i don̫̉t have to worry about :-)
> > > > > I tested some high speeds and longer runs and ended up lowering the gain pretty much on both X and Y.
> > > > >
> > > > > I moved on to my Z-axle and it seems like therẻ̫s not enough output to go along with the Command.
> > > > > I have max output 2000 intergration 1000
> > > > > Velocity 4000
> > > > > Acceleration 40000
> > > > >
> > > > > I also added 2nd low pass filter 500hz 1,4q as for the X and Y axle
> > > > >
> > > > > When i moves 2000 steps it̫̉s ok but when i test 4000steps or more it gets really bad.
> > > > > I have added a picture under photos named "Z-axle_first_run"
> > > > >
> > > > > Regards Mattias
> > > > >
> > > > >
> > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > >
> > > > > > Hi Mattias,
> > > > > >
> > > > > > This is normal.ÃÆ'ââ¬Å¡ÃâàIt is almost impossible to command analog velocity of exactly zero because there will always be some analog noise and offsets.ÃÆ'ââ¬Å¡ÃâàIf this is a problem then you will need to wire a disable signal to your amplifier.ÃÆ'ââ¬Å¡ÃâàThen add C statement to disable the Amplifier whenever the KFLOP Axis Channel is disabled.ÃÆ'ââ¬Å¡ÃâàSee the WatchEnable.c example.ÃÆ'ââ¬Å¡ÃâàThis is not a problem when the servo is enabled because the feedback will command whatever voltage is necessary to hold position.
> > > > > >
> > > > > >
> > > > > > HTH
> > > > > > Regards
> > > > > > TK
> > > > > >
> > > > > >
> > > > > >
> > > > > > ________________________________
> > > > > > From: maba_sweden
> > > > > > To: DynoMotion@yahoogroups.com
> > > > > > Sent: Tuesday, February 5, 2013 9:36 AM
> > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > >
> > > > > >
> > > > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > Thanks for your inputs Tom!
> > > > > >
> > > > > > I will test to increase the Max I and decrease the gain settings to see if it will be more stable.
> > > > > >
> > > > > > One more thing,
> > > > > > if I disable the axis in "step resonse" the Y-axle will slowly start to move and as soon as I enable the axis it stops.
> > > > > > This donÃÆ'ââ¬Å¡Ãâôt happens on the X-axis, itÃÆ'ââ¬Å¡Ãâôs still all the time.
> > > > > > Is this normal?
> > > > > >
> > > > > > Regards Mattias
> > > > > >
> > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > >
> > > > > > > Hi Mattias,
> > > > > > >
> > > > > > > Your calculations look correct.
> > > > > > >
> > > > > > > I would lower gains to improve stability.
> > > > > > >
> > > > > > > Your Max I is set at 200.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTo go at high speed the I term would need more than that so increase to more than what is needed as Output to fo at your max speed.
> > > > > > >
> > > > > > > Regards
> > > > > > > TK
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > ________________________________
> > > > > > > From: maba_sweden
> > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > Sent: Tuesday, February 5, 2013 8:15 AM
> > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > >
> > > > > > >
> > > > > > > ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > Hi Again
> > > > > > >
> > > > > > > The machine specs are X Y 5000mm/min Z 2500mm/min
> > > > > > > I got ~2000steps/mm so IÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôm aiming for V:160000 that should be~5m/min
> > > > > > > and A:666664 witch should be ~0,25s
> > > > > > > If my calcylations are right :-)
> > > > > > >
> > > > > > > I tested to increase the speeds of the X-axle and it runs really smooth with almost no errors at all. I have added a picture called "X-axle_finish.png" with settings V160000 A600000 running 20000 steps.
> > > > > > >
> > > > > > > I have started to test my Y-axle and it seems like the same gain settings as the X-axle works best.
> > > > > > > But as soon as I increases the speed on Y-axle, I get problems with oscillation.
> > > > > > > I have added the same 500Hz 2nd low pass filter as for X-axle.
> > > > > > > The ballscrew seems to run smooth without any excessive play.
> > > > > > >
> > > > > > > Should i reduce my gains or should I use an other type of filter?
> > > > > > >
> > > > > > > I have added pictures under "Photos" "Y-axle V9000 5000steps" AND "Y-axle V9000 40000steps"
> > > > > > >
> > > > > > > Regards// Mattias
> > > > > > >
> > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > >
> > > > > > > > Hi Mattias,
> > > > > > > >
> > > > > > > > Those look pretty good.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut there does seem to be some small (several um) oscillation when moving in the +X direction.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt is hard to see what frequency it is without zooming in.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàA Bode Plot might be helpful to understand things.
> > > > > > > >
> > > > > > > > But realistically that is pretty close to what can be expected.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIf the oscillation bothers you then you might reduce the gains enough to get rid of it and accept a bit larger errors.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIf the oscillations don't go away with lower gains then it might be some sort of vibration or chatter.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny ideas on why motion one way is different from the other?
> > > > > > > >
> > > > > > > >
> > > > > > > > It is probably time to try higher speeds and accelerations.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàHave you tried those?ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàWhat are your goals?
> > > > > > > >
> > > > > > > > Regards
> > > > > > > > TK
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > ________________________________
> > > > > > > > From: maba_sweden
> > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > Sent: Thursday, January 10, 2013 3:35 PM
> > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > >
> > > > > > > >
> > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > I donÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt know what I was doing wrong the last time i tested I-gain but I went out and tested again and know it worked.
> > > > > > > > Both with step size 2000 and 20000
> > > > > > > > It worked best with I 0.025 with P4 D120
> > > > > > > >
> > > > > > > > new pictures under photos named *_P4I0025D120
> > > > > > > >
> > > > > > > > Is the position error ok, I think it looks very blue?
> > > > > > > >
> > > > > > > > IÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôm having some trouble to understand the scales in the graph window, can I get the scales to follow the zoom?
> > > > > > > >
> > > > > > > > Thanks
> > > > > > > > Regards Mattias
> > > > > > > >
> > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > >
> > > > > > > > > Hi Mattias,
> > > > > > > > >
> > > > > > > > > Your errors are only several um so that is probably as good as one can expect.
> > > > > > > > >
> > > > > > > > > I don't understand why the small I gain causes a problem with long moves.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt might be totally unrelated and you are just very marginal to begin with.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt would be helpful to capture a plot of it going unstable.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI gain will make the system more unstable but a tiny amount should not make a significant
> > difference.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry even smaller (1e-6) and increase until you see an effect.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny amount whatsoever should eventually drive any residual error to zero but just take a long time.
> > > > > > > > >
> > > > > > > > > Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent.
> > > > > > > > >
> > > > > > > > > Regards
> > > > > > > > > TK
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ________________________________
> > > > > > > > > From: maba_sweden
> > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > Sent: Thursday, January 10, 2013 1:59 PM
> > > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > Thanks for the explanation Tom
> > > > > > > > > Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
> > > > > > > > >
> > > > > > > > > I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
> > > > > > > > >
> > > > > > > > > I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
> > > > > > > > >
> > > > > > > > > I still got the few count difference when it stops after the + move but i corrected by the system.
> > > > > > > > > And then i always get a few counts wrong when it stops after the - move.
> > > > > > > > >
> > > > > > > > > I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go away before i disable the axle. The output signal gets green over the whole screen when it happens.
> > > > > > > > >
> > > > > > > > > Am I on the right way?
> > > > > > > > > Any tip what to try next?
> > > > > > > > >
> > > > > > > > > Regards Mattias
> > > > > > > > >
> > > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > > >
> > > > > > > > > > Hi Mattias,
> > > > > > > > > >
> > > > > > > > > > If the system was working ok before then the amplifier is probably set up reasonably well. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYour schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > That is probably the tachometer gain setting. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThe higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position.
> > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad.
> > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI would leave that alone for now.
> > > > > > > > > >
> > > > > > > > > > Regards
> > > > > > > > > > TK
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > ________________________________
> > > > > > > > > > From: maba_sweden
> > > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > > Sent: Wednesday, January 9, 2013 7:01 PM
> > > > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > Yes the output looked better
> > > > > > > > > >
> > > > > > > > > > I noticed a sound in the plus direction but not in the minus so that observation is spot on.
> > > > > > > > > >
> > > > > > > > > > My Amplifiers are the original ones from GME systems sweden 1985
> > > > > > > > > > tha label says 10KHz SW FERQ
> > > > > > > > > > And it have tacho feedback
> > > > > > > > > >
> > > > > > > > > > I have uploaded a picture of the label and the schematics of the amplifier
> > > > > > > > > >
> > > > > > > > > > To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> > > > > > > > > > And I havenÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt read the part about the Bode Plot yet, guess i have some work to do.
> > > > > > > > > >
> > > > > > > > > > Regards Mattias
> > > > > > > > > >
> > > > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > > > >
> > > > > > > > > > > Hi Mattias,
> > > > > > > > > > >
> > > > > > > > > > > Notice huge spikes in the output are now gone.
> > > > > > > > > > >
> > > > > > > > > > > But there is servo oscillation moving in the plus direction (but not in the negative). ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàProbably different friction moving one way vs the other?
> > > > > > > > > > >
> > > > > > > > > > > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> > > > > > > > > > >
> > > > > > > > > > > What type of amplifiers do you have? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIs there tachometer feedback?
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàDo you have the velocity loop in the amplifier optimized?
> > > > > > > > > > >
> > > > > > > > > > > If you could get a reasonable Bode plot it would help understand things.
> > > > > > > > > > >
> > > > > > > > > > > Regards
> > > > > > > > > > > TK
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > ________________________________
> > > > > > > > > > > From: maba_sweden
> > > > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > > > Sent: Wednesday, January 9, 2013 6:14 PM
> > > > > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > > Hi Tom
> > > > > > > > > > > Thanks for your answer.
> > > > > > > > > > >
> > > > > > > > > > > When i move 2000 steps itÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôs close to 1mm so that would give ~50800 counts per inch
> > > > > > > > > > >
> > > > > > > > > > > I just tried the filter and did one test run with P5 I0 D100
> > > > > > > > > > > I uploaded some new pictures.
> > > > > > > > > > > I will continue my tests tomorrow and se what differens IÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôll get now with the filter.
> > > > > > > > > > >
> > > > > > > > > > > Thanks!
> > > > > > > > > > > Regards Mattias
> > > > > > > > > > >
> > > > > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > > > > >
> > > > > > > > > > > > Hi Mattias,
> > > > > > > > > > > >
> > > > > > > > > > > > What is the resolution of your axis?
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâà(how many counts per inch?)
> > > >
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSo we can determine what size errors we are seeing.
> > > > > > > > > > > >
> > > > > > > > > > > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot).
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYou can smooth that with a filter and the D gain will probably
> > > > become more effective.
> > > > >
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4.
> > > >
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen see what P and D gains work best for you.
> > > > > >
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàMaybe something like P=5 D=100 might be a starting point.
> > > >
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen try adding I gain.
> > > > >
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry increasing it until it
> > > > > > > has some effect.
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàPossibly I=0.01 or more.
> > > > > > > > > > > >
> > > > > > > > > > > > Regards
> > > > > > > > > > > > TK
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > ________________________________
> > > > > > > > > > > > From: maba_sweden
> > > > > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > > > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > > > > > > > > > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > >
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > > > Hi
> > > > > > > > > > > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > > > > > > > > > > >
> > > > > > > > > > > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > > > > > > > > > > >
> > > > > > > > > > > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and
> > couldnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃ
> > Â¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go any further.
> > > > > > > > > > > >
> > > > > > > > > > > > I also tried some 0.001 I gain but i
> > didnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãââ
> > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt though it maked it any better.
> > > > > > > > > > > >
> > > > > > > > > > > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > > > > > > > > > > They are named as the setting i tested
> > > > > > > > > > > >
> > > > > > > > > > > > Any tip on how I should proceed?
> > > > > > > > > > > >
> > > > > > > > > > > > Regards Mattias
> > > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6806 |
From: maba_sweden |
Date: 2/24/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
I read something interesting in the manual regarding the step response screen:
"If the axis is currently enabled, it will be disabled, all parameters from all screens will be downloaded, the axis will be enabled, the move or step will be performed "'
Is it possible that this is what i see?
The short time the axis is diasabled before it start it´s move the gravitation makes the table move down a little and then when it´s enabled it starts to correct the fault that happend when it was diasabled?
//Mattias
--- In DynoMotion@yahoogroups.com, "maba_sweden" <maba@...> wrote:
>
> I have tried to change the servomotor and used the servodriver and cables from the X-axle to eliminate hardware faults.
> But I still get the same problem.
>
> The Position don´t starts where the command are so the motor makes makes a quick move to correct this in the beginning of a move.
> And then during + move (Z table down) I get oscillations, but - (Z table up) move is smooth and fine.
>
> I have tried alot of different PID settings but I can´t get it to work.
>
> I also tried to make the move with the brake engaged and then it worked really nice, I guess this is because I get another type of friction so it´s easier for the regulator to cope with it.
>
> Should I try some kind of filter or how do i proceed?
>
> I have uploaded picture "Z with break" and "Z without break" and "Z start of move" under photos mattias combimill
>
> Regards Mattias
>
>
>
>
> --- In DynoMotion@yahoogroups.com, "maba_sweden" <maba@> wrote:
> >
> > I made some adjustments on the Z-brake and the belt between the servo motor and ball screw and I got rid of the oscillation when resting at target.
> > But know I got another problem.
> >
> > Every time I start a move the servo motor twiches (i think it´s called so) and then it makes the move smooth.
> > I have tried to zero the servo before the move but it looks strange on the tuning window, the position doesn´t starts where the command are.
> > I´we added i picture called "Z start of move"
> >
> > Are there any settings I should test or should i continue to make adjustments on the hardware?
> >
> > Regards Mattias
> >
> >
> >
> > --- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@> wrote:
> > >
> > > Hi Mattias,
> > >
> > > It is a fairly common characteristic of Servos to "hunt" or dither when at rest. It is just their nature to always be correcting. Especially if there is any Integrator gain. It usually is only an encoder count or two. It can also depend on whether the target position is a fractional position or not. For example commanding to a position of 100.25 will always have an error whether the encoder reads either 100 or 101. Any Integrator will force the average error to be zero so 25% of the time will be spent at 101 and 75% at 100.
> > >
> > > Sometimes you can get things to sit still using dead band, but it is very complicated and depends on the dynamics of your system. Friction and compliance and so forth.
> > >
> > > Regards
> > > TKÂ
> > >
> > >
> > >
> > > ________________________________
> > > From: maba_sweden <maba@>
> > > To: DynoMotion@yahoogroups.com
> > > Sent: Tuesday, February 19, 2013 7:28 AM
> > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > >
> > >
> > > Â
> > > Hello, now I´m back :-)
> > >
> > > I have changed the bearings and re-glued my static magnets and the servo motor seems to work fine.
> > > I have begun to tune my Z axle and it seems to respond really well.
> > > The only problem I got is that I get cycle oscillations while resting at the target.
> > > I tried to ad dead band but it wont stop.
> > > I tested range from 0-8 with 0 Gain but it won´t stop.
> > > I have also tried to ad some gain from 0.1-1
> > > I´m not really sure how the Dead Band works.
> > >
> > > Probably I´m just doing something wrong and would like some tip what settings to start with.
> > > Or should i use something else then the dead band?
> > >
> > > I have added a piture under Photos called "Z-axle 1902"
> > > It´s possible to the oscillations there before the movements starts, sometimes the oscillations are less after the movement.
> > >
> > > Regards Mattias
> > >
> > > --- In DynoMotion@yahoogroups.com, "maba_sweden" wrote:
> > > >
> > > > Hi Tom
> > > >
> > > > Just wanted you to know that it wasn´t the drivers fault, if it only could have been that :-)
> > > > The Z motor is bad. One of the static magnets was loose, that may explains the uneven motion of the motor, the rear bearing must also bad because it´s heavy to rotate.
> > > > Probably is this because of coolant that have came in the motor and it have started to rust.
> > > > I have to see if i can rebuild the motor, some says the magnets can loose there strength if you take it apart.
> > > > Or else I have to try to find a new/used/rebuild Contraves 402 117V 3,4nm DC servo motor before I can continue my work...
> > > >
> > > > Thanks for all your help.
> > > > Hope I will be back soon!
> > > > Regards Mattias
> > > >
> > > >
> > > >
> > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > >
> > > > > Hi Mattias,
> > > > >
> > > > > Now that I look at the plot again the velocity is very low 4000 counts/sec and the acceleration to go that speed is insignificant.ÃÂ There must be something major wrong.
> > > > >
> > > > > Maybe you are missing a power supply voltage or something so it can't drive positive.
> > > > >
> > > > > But if the car is going too slow and the gas pedal is all the way to the floor - it isn't the driver's fault :}
> > > > >
> > > > > Regards
> > > > > TK
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > ________________________________
> > > > > From: maba_sweden
> > > > > To: DynoMotion@yahoogroups.com
> > > > > Sent: Tuesday, February 5, 2013 4:30 PM
> > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > >
> > > > >
> > > > > ÃÂ
> > > > > Yes positive is lifting.
> > > > > The machine got no counter weight what Iôm aware, and according to the manual it should be able to travel 2500mm/min half the speed of the X and Y.
> > > > > Probably itôs just the acceleration that is to aggressive, if i remember correctly it accelerated pretty smoth with the original Sinumerik system.
> > > > > I will test with a lower setting and see if it gets any better.
> > > > >
> > > > > Thanks
> > > > > Regards//Mattias
> > > > >
> > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > >
> > > > > > Hi Mattias,
> > > > > >
> > > > > > It looks like you probably have a lot of weight and friction.ÃâàNotice as it goes positive (I assume this is lifting weight) the position (red) lags way behind moving at a varying speed even though the output (green) is fully saturated at 2000 counts (~10V).ÃâàYour Motor/Amplifier/Supply probably just doesn't have enough power. ÃâàÃâàBut notice going negative (down?) it is able to go very fast even with almost zero command to the amplifier.ÃâàIs there supposed to be a counter balance or something?ÃâàOr you are just trying to move too fast.ÃâàHow did the machine perform originally?
> > > > > >
> > > > > > Regards
> > > > > > TK
> > > > > >
> > > > > > ÃâÃÂ
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > ________________________________
> > > > > > From: maba_sweden
> > > > > > To: DynoMotion@yahoogroups.com
> > > > > > Sent: Tuesday, February 5, 2013 3:53 PM
> > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > >
> > > > > >
> > > > > > ÃâÃÂ
> > > > > > Ok, one thing that i don̫̉t have to worry about :-)
> > > > > > I tested some high speeds and longer runs and ended up lowering the gain pretty much on both X and Y.
> > > > > >
> > > > > > I moved on to my Z-axle and it seems like therẻ̫s not enough output to go along with the Command.
> > > > > > I have max output 2000 intergration 1000
> > > > > > Velocity 4000
> > > > > > Acceleration 40000
> > > > > >
> > > > > > I also added 2nd low pass filter 500hz 1,4q as for the X and Y axle
> > > > > >
> > > > > > When i moves 2000 steps it̫̉s ok but when i test 4000steps or more it gets really bad.
> > > > > > I have added a picture under photos named "Z-axle_first_run"
> > > > > >
> > > > > > Regards Mattias
> > > > > >
> > > > > >
> > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > >
> > > > > > > Hi Mattias,
> > > > > > >
> > > > > > > This is normal.ÃÆ'ââ¬Å¡ÃâàIt is almost impossible to command analog velocity of exactly zero because there will always be some analog noise and offsets.ÃÆ'ââ¬Å¡ÃâàIf this is a problem then you will need to wire a disable signal to your amplifier.ÃÆ'ââ¬Å¡ÃâàThen add C statement to disable the Amplifier whenever the KFLOP Axis Channel is disabled.ÃÆ'ââ¬Å¡ÃâàSee the WatchEnable.c example.ÃÆ'ââ¬Å¡ÃâàThis is not a problem when the servo is enabled because the feedback will command whatever voltage is necessary to hold position.
> > > > > > >
> > > > > > >
> > > > > > > HTH
> > > > > > > Regards
> > > > > > > TK
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > ________________________________
> > > > > > > From: maba_sweden
> > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > Sent: Tuesday, February 5, 2013 9:36 AM
> > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > >
> > > > > > >
> > > > > > > ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > Thanks for your inputs Tom!
> > > > > > >
> > > > > > > I will test to increase the Max I and decrease the gain settings to see if it will be more stable.
> > > > > > >
> > > > > > > One more thing,
> > > > > > > if I disable the axis in "step resonse" the Y-axle will slowly start to move and as soon as I enable the axis it stops.
> > > > > > > This donÃÆ'ââ¬Å¡Ãâôt happens on the X-axis, itÃÆ'ââ¬Å¡Ãâôs still all the time.
> > > > > > > Is this normal?
> > > > > > >
> > > > > > > Regards Mattias
> > > > > > >
> > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > >
> > > > > > > > Hi Mattias,
> > > > > > > >
> > > > > > > > Your calculations look correct.
> > > > > > > >
> > > > > > > > I would lower gains to improve stability.
> > > > > > > >
> > > > > > > > Your Max I is set at 200.ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTo go at high speed the I term would need more than that so increase to more than what is needed as Output to fo at your max speed.
> > > > > > > >
> > > > > > > > Regards
> > > > > > > > TK
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > ________________________________
> > > > > > > > From: maba_sweden
> > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > Sent: Tuesday, February 5, 2013 8:15 AM
> > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > >
> > > > > > > >
> > > > > > > > ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > Hi Again
> > > > > > > >
> > > > > > > > The machine specs are X Y 5000mm/min Z 2500mm/min
> > > > > > > > I got ~2000steps/mm so IÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôm aiming for V:160000 that should be~5m/min
> > > > > > > > and A:666664 witch should be ~0,25s
> > > > > > > > If my calcylations are right :-)
> > > > > > > >
> > > > > > > > I tested to increase the speeds of the X-axle and it runs really smooth with almost no errors at all. I have added a picture called "X-axle_finish.png" with settings V160000 A600000 running 20000 steps.
> > > > > > > >
> > > > > > > > I have started to test my Y-axle and it seems like the same gain settings as the X-axle works best.
> > > > > > > > But as soon as I increases the speed on Y-axle, I get problems with oscillation.
> > > > > > > > I have added the same 500Hz 2nd low pass filter as for X-axle.
> > > > > > > > The ballscrew seems to run smooth without any excessive play.
> > > > > > > >
> > > > > > > > Should i reduce my gains or should I use an other type of filter?
> > > > > > > >
> > > > > > > > I have added pictures under "Photos" "Y-axle V9000 5000steps" AND "Y-axle V9000 40000steps"
> > > > > > > >
> > > > > > > > Regards// Mattias
> > > > > > > >
> > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > >
> > > > > > > > > Hi Mattias,
> > > > > > > > >
> > > > > > > > > Those look pretty good.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut there does seem to be some small (several um) oscillation when moving in the +X direction.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt is hard to see what frequency it is without zooming in.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàA Bode Plot might be helpful to understand things.
> > > > > > > > >
> > > > > > > > > But realistically that is pretty close to what can be expected.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIf the oscillation bothers you then you might reduce the gains enough to get rid of it and accept a bit larger errors.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIf the oscillations don't go away with lower gains then it might be some sort of vibration or chatter.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny ideas on why motion one way is different from the other?
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > It is probably time to try higher speeds and accelerations.ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàHave you tried those?ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàWhat are your goals?
> > > > > > > > >
> > > > > > > > > Regards
> > > > > > > > > TK
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ________________________________
> > > > > > > > > From: maba_sweden
> > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > Sent: Thursday, January 10, 2013 3:35 PM
> > > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > I donÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt know what I was doing wrong the last time i tested I-gain but I went out and tested again and know it worked.
> > > > > > > > > Both with step size 2000 and 20000
> > > > > > > > > It worked best with I 0.025 with P4 D120
> > > > > > > > >
> > > > > > > > > new pictures under photos named *_P4I0025D120
> > > > > > > > >
> > > > > > > > > Is the position error ok, I think it looks very blue?
> > > > > > > > >
> > > > > > > > > IÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôm having some trouble to understand the scales in the graph window, can I get the scales to follow the zoom?
> > > > > > > > >
> > > > > > > > > Thanks
> > > > > > > > > Regards Mattias
> > > > > > > > >
> > > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > > >
> > > > > > > > > > Hi Mattias,
> > > > > > > > > >
> > > > > > > > > > Your errors are only several um so that is probably as good as one can expect.
> > > > > > > > > >
> > > > > > > > > > I don't understand why the small I gain causes a problem with long moves.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt might be totally unrelated and you are just very marginal to begin with.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIt would be helpful to capture a plot of it going unstable.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI gain will make the system more unstable but a tiny amount should not make a significant
> > > difference.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry even smaller (1e-6) and increase until you see an effect.ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàAny amount whatsoever should eventually drive any residual error to zero but just take a long time.
> > > > > > > > > >
> > > > > > > > > > Setting a smaller Max Following error can be used to automatically disable the axis if it goes unstable and before things get too violent.
> > > > > > > > > >
> > > > > > > > > > Regards
> > > > > > > > > > TK
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > ________________________________
> > > > > > > > > > From: maba_sweden
> > > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > > Sent: Thursday, January 10, 2013 1:59 PM
> > > > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > Thanks for the explanation Tom
> > > > > > > > > > Yes there are a pot there, but as you wrote. The system worked before I pulled out the old Sinumerik 3m so it should be ok.
> > > > > > > > > >
> > > > > > > > > > I continued tuning today and i lowered the P-gain to P4 I0 D120 and the oscillation went away.
> > > > > > > > > >
> > > > > > > > > > I also added "Feed forward" 0.02 so the possition during move got closer to the command. the +move got almost spot on and the -move got better but not perfect.
> > > > > > > > > >
> > > > > > > > > > I still got the few count difference when it stops after the + move but i corrected by the system.
> > > > > > > > > > And then i always get a few counts wrong when it stops after the - move.
> > > > > > > > > >
> > > > > > > > > > I have tried to ad I-gain 0.0001 but as soon as I make a longer move say 40000 steps I get a very annoying high frequenc sound that doesnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go away before i disable the axle. The output signal gets green over the whole screen when it happens.
> > > > > > > > > >
> > > > > > > > > > Am I on the right way?
> > > > > > > > > > Any tip what to try next?
> > > > > > > > > >
> > > > > > > > > > Regards Mattias
> > > > > > > > > >
> > > > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > > > >
> > > > > > > > > > > Hi Mattias,
> > > > > > > > > > >
> > > > > > > > > > > If the system was working ok before then the amplifier is probably set up reasonably well. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYour schematic is tiny and hard to read, but there seems to be a pot near where it shows the tachometer is coming in. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > That is probably the tachometer gain setting. ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThe higher the gain the tighter the speed regulation will be, and the better the speed regulation is the better KFLOP will be able to control the position.
> > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàBut of course too high of gain will make the speed regulation in the amplifier go unstable and make things bad.
> > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàI would leave that alone for now.
> > > > > > > > > > >
> > > > > > > > > > > Regards
> > > > > > > > > > > TK
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > ________________________________
> > > > > > > > > > > From: maba_sweden
> > > > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > > > Sent: Wednesday, January 9, 2013 7:01 PM
> > > > > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > > Yes the output looked better
> > > > > > > > > > >
> > > > > > > > > > > I noticed a sound in the plus direction but not in the minus so that observation is spot on.
> > > > > > > > > > >
> > > > > > > > > > > My Amplifiers are the original ones from GME systems sweden 1985
> > > > > > > > > > > tha label says 10KHz SW FERQ
> > > > > > > > > > > And it have tacho feedback
> > > > > > > > > > >
> > > > > > > > > > > I have uploaded a picture of the label and the schematics of the amplifier
> > > > > > > > > > >
> > > > > > > > > > > To be honest, I have no idea how to optimze the Velocity loop in the amplifier??
> > > > > > > > > > > And I havenÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt read the part about the Bode Plot yet, guess i have some work to do.
> > > > > > > > > > >
> > > > > > > > > > > Regards Mattias
> > > > > > > > > > >
> > > > > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > > > > >
> > > > > > > > > > > > Hi Mattias,
> > > > > > > > > > > >
> > > > > > > > > > > > Notice huge spikes in the output are now gone.
> > > > > > > > > > > >
> > > > > > > > > > > > But there is servo oscillation moving in the plus direction (but not in the negative). ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàProbably different friction moving one way vs the other?
> > > > > > > > > > > >
> > > > > > > > > > > > The max errors are fairly small 0.4 mil (10um) but you are moving very slow.
> > > > > > > > > > > >
> > > > > > > > > > > > What type of amplifiers do you have? ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàIs there tachometer feedback?
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàDo you have the velocity loop in the amplifier optimized?
> > > > > > > > > > > >
> > > > > > > > > > > > If you could get a reasonable Bode plot it would help understand things.
> > > > > > > > > > > >
> > > > > > > > > > > > Regards
> > > > > > > > > > > > TK
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > ________________________________
> > > > > > > > > > > > From: maba_sweden
> > > > > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > > > > Sent: Wednesday, January 9, 2013 6:14 PM
> > > > > > > > > > > > Subject: [DynoMotion] Re: Tuning Combimill DAC servo DC brushed motors
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > > > Hi Tom
> > > > > > > > > > > > Thanks for your answer.
> > > > > > > > > > > >
> > > > > > > > > > > > When i move 2000 steps itÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôs close to 1mm so that would give ~50800 counts per inch
> > > > > > > > > > > >
> > > > > > > > > > > > I just tried the filter and did one test run with P5 I0 D100
> > > > > > > > > > > > I uploaded some new pictures.
> > > > > > > > > > > > I will continue my tests tomorrow and se what differens IÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôll get now with the filter.
> > > > > > > > > > > >
> > > > > > > > > > > > Thanks!
> > > > > > > > > > > > Regards Mattias
> > > > > > > > > > > >
> > > > > > > > > > > > --- In DynoMotion@yahoogroups.com, Tom Kerekes wrote:
> > > > > > > > > > > > >
> > > > > > > > > > > > > Hi Mattias,
> > > > > > > > > > > > >
> > > > > > > > > > > > > What is the resolution of your axis?
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâà(how many counts per inch?)
> > > > >
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSo we can determine what size errors we are seeing.
> > > > > > > > > > > > >
> > > > > > > > > > > > > With a High D gain of 270 a change of 1 encoder cout will generate a 270 DAC Count spike in the output (spikes in green plot).
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàYou can smooth that with a filter and the D gain will probably
> > > > > become more effective.
> > > > > >
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàSet the 3rd IIR Filter to a 2nd order low pass at 500Hz and Q=1.4.
> > > > >
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen see what P and D gains work best for you.
> > > > > > >
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàMaybe something like P=5 D=100 might be a starting point.
> > > > >
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàThen try adding I gain.
> > > > > >
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàTry increasing it until it
> > > > > > > > has some effect.
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâàPossibly I=0.01 or more.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Regards
> > > > > > > > > > > > > TK
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > > > ________________________________
> > > > > > > > > > > > > From: maba_sweden
> > > > > > > > > > > > > To: DynoMotion@yahoogroups.com
> > > > > > > > > > > > > Sent: Wednesday, January 9, 2013 5:10 PM
> > > > > > > > > > > > > Subject: [DynoMotion] Tuning Combimill DAC servo DC brushed motors
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'
> > > Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ
> > > > > > > > > > > > > Hi
> > > > > > > > > > > > > I finally got everything hooked up and ready for test on my Storebro Combimill with DAC servos and Brushed DC motors.
> > > > > > > > > > > > >
> > > > > > > > > > > > > I started to add P gain and when I came up to P gain 3 I noticed the beginning of a vibration like sound, I suspect that this is the first steps of ocsillaition?
> > > > > > > > > > > > >
> > > > > > > > > > > > > I then started to ad some D Gain and came up to P gain 9 and D gain 270 and then I gotted the same sound and
> > > couldnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃ
> > > Â¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt go any further.
> > > > > > > > > > > > >
> > > > > > > > > > > > > I also tried some 0.001 I gain but i
> > > didnÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Å¡ÃâÃÂ¦ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâà'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãââ
> > > ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬ÃâÃÂ¦ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'âââ¬Ã 'ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Å¡ÃâÃÂ¢ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'ââ¬Å¡ÃâÃÂ¬ÃÆ'Ã'âââ¬ÃÂ¦ÃÆ'ââ¬Å¡ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââ¬Â 'ÃÆ'Ã'ÃâÃÂ¢ÃÆ'ââââ¬Ã
¡ÃâÃÂ¬ÃÆ'ââ¬Â¦ÃâÃÂ¡ÃÆ'Ã'Ãâ 'ÃÆ'ââââ¬Å¡Ã¬Ãâ¦ÃÂ¡ÃÆ'Ã'âââ¬Ã
Â¡ÃÆ'ââ¬Å¡Ãâôt though it maked it any better.
> > > > > > > > > > > > >
> > > > > > > > > > > > > I have posted some pictures under "photos" and "Mattias Combimill"
> > > > > > > > > > > > > They are named as the setting i tested
> > > > > > > > > > > > >
> > > > > > > > > > > > > Any tip on how I should proceed?
> > > > > > > > > > > > >
> > > > > > > > > > > > > Regards Mattias
> > > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
|
|
| Group: DynoMotion |
Message: 6808 |
From: Tom Kerekes |
Date: 2/24/2013 |
| Subject: Re: Tuning Combimill DAC servo DC brushed motors |
Hi Mattias, Yes you are probably correct with the brake off the Z falls a little bit while disabled and changing settings. Simple solution - keep the brake on all the time - just kidding :} Viscus Friction (not stiction) can help stabilize a servo. It provides damping much like D Gain. Your system is nonlinear and unstable. Obviously fix any mechanical problems first. Then you might try more D gain or a Lead/Lag filter to get more phase margin and higher stability. If you could get a good Bode Plot that would help understand where you are. You didn't include which filters you are using if any. Also remove any Feed Forward setting until after you optimize the Servo loop. HTH Regards TK
| | | | | | | | | | | | | | | | | | | | | | | |