Set up using AMC AB15A100 drives, brushed DC motor Prototrak Plus with Encoders

Looking much better, even the 60ipm jog doesn't look too bad.

Once you are satisfied that it is as good as you can get it. I think you should be able to proceed with trying to tune the the PID loop in Linuxcnc.

You might also be able to experiment with increasing the acceleration, to see if the acceleration limit has improved with better tuning. But stick with making one change at a time. So you can see what effect your change has made. For example if you try increasing the acceleration before tuning the PID in Linuxcnc, you will need to watch for the acceleration limit causing the slope of the encoder velocity trace not match the open loop commanded velocity's slope. If you test this after tuning the PID in Linuxcnc, you will need to watch out for a spike in following error combined with saturation of the PID output (commanding a full 10v to the drive's analog input.) Any time the drive's analog input gets sent the full 10v, it means you are exceeding the systems capabilities for the current settings in some way, and you will need to make some adjustment to stop it from doing that, either better tuning to increase the performance capability or limiting output to stay within those capabilities.

Todd, thanks for all your help. The new 24vDC 5 amp power supply (replaced a 24vDC 2 amp) has been installed. I was expecting to start back to tuning as per next chart requirements. It was difficult to find a set of setting to get any trace close to acceptable, so I went back to the most recent trace posted at #248269, (take notice the trace has a (-.375 Pot 2) a lower number to allow the Pot 1 to be 12.875, one of the highest in my data set). But was unable to duplicate the trace with settings adjusted a lot and trace was not good at all. Will the 5 amp vs. the 2 amp cause my issues?? As I was adjusting around the prior set (in the #248269 trace) of settings, occasionally the motor seemed like the perfect set of settings (motor accelerated real fast (as fast as I believe it would go) to get up to speed) had occurred, but the white/purple trace would go so high it was off the screen. I spend quality time trying to tune to each set of settings as good as possible. Will this new 24vDC, 5 amp power supply, work ok?

I am confident that the settings for the trace at post #248269 were able to be duplicated, around this time is when the power supply stopped working.

Many thanks!

You're images didn't show up,

What exactly is this 24v power supply powering?

Todd, this is a pic of the trace, as shows moving ok (what I expected, looked 20"/min) then the motor speeds up quickly. The other pic starting out this goes past 20"/min; notice the units for the white trace are different from others (are at 200/) units and have offset (the baseline) to show complete trace, (notice max velocity is set to 0.0 on the slider). The file names are the settings: PS=new power supply s1-1.617; s3=110; p1=Pot1=10.0; Pot3=8.0; Pot2=11.0 (-.375)=10.625CW(as per equation, set to 11.0 CW).
I was thinking the new power supply would not change tuning at all, or very small amount.
The traces are attached.
Many thanks
Edit: The 24vDC power supply goes to the (long orange plug on 7i77 card, 2/3 the way down on right side, (I do not have my folder with me the proper ID)
 

That would be the field power connector for the 7i77. Yes, that should have no effect on drive tuning. Is that the only thing it is connected to?

How is it connected? Positive leads go where, negative leads go where? Are chassis grounds involved?

Todd, many thanks, if this is ok to use the 5amp PS & I just need to start tuning with fresh & new approach, I can do that.
The only place the 24vdc PS is used is the 7i77 card. Power (110v) from the wall goes to a terminal block inside the electrical cabinet to a terminal block (separate place for each of the 3 wires, GND, White & Black); keeping the black & white consistence to where they attach 5vDC PS is connected, a wire from the terminal block goes to the 24 v PS & then to the 5 v PS. Keeping the GND (green) is separate & goes to the cabinet as where the green from wall attaches to cabinet. I tested the 24 v (power supply) with digital meter (measured 24.3 on 200 scale) before attaching to 7i77 card. There is a 110 volt small fan inside the cabinet that gets power from the terminal block, but the fan is not plugged in yet, because of the noise it makes. If a separate wire from the 110 volt terminal block to each power supply I can do that, but when it was "back emf" the power supply was as I describe, first to 1 (PS) then to other PS.

Many thanks

Todd, Many thanks! I found a Rigol DS1054Z Digital Oscilloscopes - Bandwidth 50 MHz Channels 4, w/o probes. What probes are required to work with to do testing required? This is a link to accessories at Rigol: www.rigolna.com/products/accessories/ or just if it is a common accessory, just describe well enough, so I understand.
Please, If aftermarket will work probes would work, I am ok with that. Just need the correct tool for the job.
I am working through the chart otherwise, Pot 1 & 3, the number of turns to get the best trace seem to be low by 2 turns (like the best trace provided with old PS, was 12.875 CW on Pot 1), now it seems to be in the 10 CW zone. I have only a very small part of the chart tested with new 24vPS, I am spending quality time to tune each setting, so I am confident of the data. It is for this reason I am looking at o-scopes again.
The times when the motor will just accelerate at great speed, at the start or mid of the trace I believe happens more often in the early, just starting the tuning of that particular SW2 & 3 settings. After refining the Pot adjustments, it does not happen.

Many thanks, I am not shore if the seller of the scope is going to work with me on price or not.

Almost any set of 1x/10x pasive probes should work fine for your use and are fairly generic (they dont' have to come from Rigol) and should be all you should need. (Unless you want to spend a lot of money on a current probe, which I don't think is a must for this application.

Here is a pair of passive probes for $12 on ebay.
www.ebay.com/itm/175358457829

Todd, Many thanks!! A new o scope will be here early next week. What should I prepare to set up for? I have reread your post at (page 13) #245270 (forum.linuxcnc.org/12-milling/44454-set-...ers?start=120#245270), 16 Jun 2022 15:58. This post has this statement:
"Attached is a small test configuration with a hal file set up for your x-axis motor with a simple user interface for Siggen to create a square wave output to your drive."

I will review this post, try to set up the "test.zip" file as you describe. Take belts off motors. I need to keep the motor from spinning, when doing the test. What else should I be thinking about over the weekend, to be ready for the Rigol scope to arrive?

Many thanks.

Edit: I have read the App Note 15 (attached) and viewed the A-M-C YouTube video, link is below.

(link:https://www.youtube.com/redirect?event=video_description&redir_token=QUFFLUhqa1JHMjRlMlhJajhmU0kxUE9UanlqUUlxdnJYQXxBQ3Jtc0tsamhhWDFiU1Y5LUotN2FqSEloRVQ2bWJSSFByV3YydzVfbDROWmZqOWl1UHNjM1k1Rm05UmQtU3pmdEJOUGdidUNiQ2dHc2treHhGc3h6WGFWUllRSl8yaW5RZlFQMzNRQUlabXV2SVBGNmVqLVEwMA&q=https%3A%2F%2Fwww.a-m-c.com%2Fdownloads%2F&v=X3OC448VSts

I do not see or have read the motor shaft needs to be locked. But I am ready with a plan if required. I do see on page 9 of 13 in the app note 15 a section on brushed motors.
From what I read to prepare for testing the as per the data sheet AB15A100 the (set to current mode) SW1-1=OFF & SW1-3=ON; unplug hall sensing (I do not have hall sensing, but set to OFF; and SW1-4=ON (for testing).

Currently, the test .zip file is on the desktop, and will as per Todd's suggestion:
"Run it by copying the dir to your Linuxcnc configs dir.  Then make the "test.demo" file executable, and run it (while no other Linuxcnc apps are running.)"

Many thanks, the scope last update was Aug 5, 2022, could take until the 10th for delivery. I see from the YouTube that several setting on the o-scope are required, after I have something in hand this will be clearer to understand. Many thanks.



 

Todd, thanks. Attached is a pic of the .zip file extracted in the Linuxcnc configs dir. Does this appear correct, or what am I looking for, a screen like the one in the post at (page 13) #245270? Will there be an icon to pick from the desktop to start the signal? I have not seen an icon for this.

Many thanks!