Dynomotion

Group: DynoMotion Message: 3811 From: himykabibble Date: 2/8/2012
Subject: Gecko Tuning Success!
Just got done re-re-re-tuning my X/Y/Z axis Gecko G320s/G320X, with assistance from KMotion, and I'm happy to say the result is a substantial improvement in performance. I've been running for the last three years with them tuned "by ear", and they've performed very well, but by hooking the encoders back to the KFlop, and using the Step Response dialog in KMotion to actually see what's really happening, I've been able to improve the tuning significantly. My previous limits were set at 200 IPM velocity and 25in/sec2 acceleration. I can now easily, and reliably, reach 350 IPM, with acceleration of 40 in/sec2, and Jerk cranked up to 1E12! The sound is quite different - much more "purposeful". It just sounds like it really means business, and you can see and hear how much stiffer it is.

One thing I can't explain though - in all these years, working with Mach3 and the old settings, I've never, ever lost position. But for some reason, jogging with the KFlop it was quite easy to cause servo faults, even with the axes significantly "detuned" in the KFlop configuration. Too early to be 100% certain, but so far, with the new tuning, that problem seems to be completely gone. doesn't make a lot of sense to me....

The A axis (knee) is a different story, for now. It has a mechanical problem, brought on by the cold weather. In the next day or two I should have some new gas springs that I hope will resolve the problems there, and get that going.

Regards,
Ray L.
Group: DynoMotion Message: 3813 From: Tom Kerekes Date: 2/8/2012
Subject: Re: Gecko Tuning Success!
Hi Ray,
 
I'd enjoy seeing a plot of the final Gecko performance.
Regards
TK

Group: DynoMotion Message: 3815 From: himykabibble Date: 2/8/2012
Subject: Re: Gecko Tuning Success!
Tom,

I'll try to capture a few in a few days when I put the knee back together. I actually found the step mode useful to do the first pass gain/damping adjustment. Set it for 100 steps, and see how much ringing there is. A small amount (a few cycles) seems about right. That seems to get things pretty close to right. The input/output plots are kinda boring - it just goes where you tell it to, with very little deviation. Even at 350 IPM and 40 in/sec^2, I'm seeing following errors on the order of 15-20 counts. That seems pretty darned good to me!

Regards,
Ray L.

--- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@...> wrote:
>
> Hi Ray,
>  
> I'd enjoy seeing a plot of the final Gecko performance.
> Regards
> TK
>
>
> ________________________________
> From: himykabibble <jagboy@...>
> To: DynoMotion@yahoogroups.com
> Sent: Wednesday, February 8, 2012 9:40 AM
> Subject: [DynoMotion] Gecko Tuning Success!
>
>
>  
> Just got done re-re-re-tuning my X/Y/Z axis Gecko G320s/G320X, with assistance from KMotion, and I'm happy to say the result is a substantial improvement in performance. I've been running for the last three years with them tuned "by ear", and they've performed very well, but by hooking the encoders back to the KFlop, and using the Step Response dialog in KMotion to actually see what's really happening, I've been able to improve the tuning significantly. My previous limits were set at 200 IPM velocity and 25in/sec2 acceleration. I can now easily, and reliably, reach 350 IPM, with acceleration of 40 in/sec2, and Jerk cranked up to 1E12! The sound is quite different - much more "purposeful". It just sounds like it really means business, and you can see and hear how much stiffer it is.
>
> One thing I can't explain though - in all these years, working with Mach3 and the old settings, I've never, ever lost position. But for some reason, jogging with the KFlop it was quite easy to cause servo faults, even with the axes significantly "detuned" in the KFlop configuration. Too early to be 100% certain, but so far, with the new tuning, that problem seems to be completely gone. doesn't make a lot of sense to me....
>
> The A axis (knee) is a different story, for now. It has a mechanical problem, brought on by the cold weather. In the next day or two I should have some new gas springs that I hope will resolve the problems there, and get that going.
>
> Regards,
> Ray L.
>