--- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@...> wrote:
>
> Hi Russ,
>  
> Russ:Well just looked again and, did not catch the "V", "A", and "J" under
> profile were for those items. It does not show units so is that in steps? For a
> motor with a max velocity of 3000RPM and an encoder of 2000/rev do you just say
> 3000*2000 = 6,000,000? What is definition of JERK? Please don't say me.. ha ha
>
> TK: The units are counts (or steps) and seconds.  So you forgot to divide by 60
> secs/minute.    3000 / 60 * 2000 = 100,000 cnts/sec.  We may be off by a factor
> of 4 if your encoder is 2000 cycles/rev instead of 2000 quadrature counts/rev.
>
> Yes driving your amps in analog velocity mode would offer more control but more
> of a learning curve :}
>
> "Jerk" is the rate of change of acceleration.  Applying the brakes a small
> amount quickly is more likely to make you spill your coffee than applying the
> brakes harder but more gradually.   When using Mach3 it generates all the motion
> trajectories with instant acceleration (infinite Jerk).  With our KMotionCNC
> rapids (G0) and Jogging uses jerk controlled independent motions.  Coordinated
> motions G1,G2,G3 still use infinite Jerk though.   If you set the Jerk value
> 100X bigger than the Acceleration value then the acceleration will be applied in
> 0.01 seconds which will be similar to what Mach3 does. 
>
> Our definition of "jog" is to move at constant velocity forever.  So Jog0=250
> means to move at 250 step/sec which would be quite slow.  Move commands move to
> the specified position as fast as possible within the specified Velocity,
> Acceleration, and Jerk limits.  
>
> Regards
> TK
>
>
>
>
> ________________________________
> From: cncman172 <rdlarson@...>
> To: DynoMotion@yahoogroups.com
> Sent: Sun, November 7, 2010 12:06:52 PM
> Subject: [DynoMotion] Re: KFLOP+KANALOG
>
>  
> Tom,
>
> Wow you are AWESOME. Amazing to get responses from you on Saturday and Sunday
> now that is nothing short of first class. I am still pounding away working the
> issues one at a time.
>
> I have tried to respond to some of your comments and questions below.
>
> Russ
>
> Since you are using a driver chip anyway you could eliminate the need for the
> pull up resistor by driving the outputs as 3.3V LVTTL signals. LVTTL instead of
> "open collector" is selected by adding 8 to the Step/Dir channel number (so you
> would specify 12 instead of 4)
>
> Tom
>
> Yes, I figured the LVTTL would probably work without the using the pullups but
> this is just sitting on a proto board so it took two seconds to add the pullup
> resistors to ensure the open collector option would work.
> Russ
>
> Yes there is little to do in the Step Response Screen for a system like yours
> where KFlop does not close the servo loop. The only thing to do is to set the
> max Velocity, Acceleration, and Jerk and to verify proper operation as a
> diagnostic tool.
> Tom
>
> After reading through lots of documentation on the KFLOP it looks like I could
> put my amplifiers in velocity mode and have the KFLOP close the loop using +-10V
> inputs and encoder outputs from my amplifiers, but figured I would try this mode
> first because I know the Panasonic tuning was very good and had nice response.
>
> When I look at the config page and the Step Response page I did not see where
> you can set the max velocity, acceleration, and jerk. I have seen on some of the
> C programs where that gets set. Did I overlook something somewhere?
> Russ
>
> Motion profile should be the only thing that has an effect, which will define
> the velocity, acceleration, and jerk.
> Tom
>
> Well just looked again and, did not catch the "V", "A", and "J" under profile
> were for those items. It does not show units so is that in steps? For a motor
> with a max velocity of 3000RPM and an encoder of 2000/rev do you just say
> 3000*2000 = 6,000,000? What is definition of JERK? Please don't say me.. ha ha
>
> Russ
>
> Do you have encoder feedback connected to KFlop??
> Tom
>
> No, so that explains why the RED actual postion never plots because it can not
> see my encoder changing. Makes sense to me.
> Russ
>
> OK, A little more progress in the last hour. I tracked down the issue of the
> motor not reversing. One of my wires from the differential driver on the
> direction output was broken inside the insulation. It was only getting half the
> signal. The motor now reverses during the test in step response.
>
> Things are starting to become a little more stable, but still have lots of
> questions. When I go to the console I can now tell it to jog0=-250 and it moves
> but very slowly. When I tell it to Move0=0 it goes very fast back to the zero
> position. The jog speed must be defined somewhere I overlooked.
>
> I tried setting the max velicoity using the calculations I provided above but it
> does not actually seem anything close to 3000 rpm. At least I have a motor
> moving still need to understand lots of items before I attempt to start putting
> the C-code together. Wow big learning curve. :) Tom thanks for all you help it
> is making this effort go easier.
>
>
> Russ
>