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

Yes, At A-M-C (P1 connector, pins 6, 7 & 11). Many thanks!

OK if you set all PID parameters to 0, do you still get oscillations?

If all including P have 0.0 & (DEADBAND 3e-05), the x axes barely moves (.001"/ second or less) using the arrow key. If P = 1.0 all others 0.0 (DEADBAND 3e-05) feed rate 15"/ min, all moves smooth. If go to 60" / min it takes a bit of time to wind up to speed, but smooth. many thanks!

This pic is of hal trace, with the values as low as it seems to go, 23.? feed (look at the rate in pic). The trace I need to watch is the "axes f error" and make this horizontal w/o over shoot. Please take a look, no "I" at all.

I was thinking about the "I" variable, to calculate the rpm I can back into the feed rate, it would seem. Use the hal scope to monitor the voltage, while adjusting POT 3.

Many thanks!

If the following error is in the same direction as the velocity (which it is here)
It means you don't have enough FF1

It also look like you have very low velocity gain in the motor drive
Did you tune this? (That needs help from AMC)

PCW and others, Many thanks for all your help! Attached is a scan of email from A-M-C engineer with details of tuning velocity gain. I have ball screw (.200"/rev) with timing belt pulley (16 to 44 teeth; 2.75 to 1) info thinking that was the way to find RPM, but after review it may not require that approach. Motor max RPM is 1350; 6.25 amp (data sheet for motor & encoder see, 1st post).

 

Many thanks

Edit:
The encoders (www.usdigital.com: modules EM1-2-500-N & HUBDISK-2-500-500-NE) are 500 line, using Quadrature Detection techniques = 2000. The correct value is line count of 500.

The "scale factor" is 22,000 HzV.

Vmonitor = (135RPM*resolution)/(Scalling Factor*60

So V monitor = (135 RPM * 500 Lines)/ 22,000HzV * 60 = 0.051VDC

Many thanks!

Edit: This 0.051 VDC is the appropriate voltage to read between Pin1-15 relative to signal ground, adjusting POT 3 to set Reference Gain.

 

Many thanks for all the support and help. I have used the .051 as the value of "I" as shown in the attached traces of the hal scope. The full screen traces show when one of the variables are zeroed out. I have used the tommylight Servo Tuning as a guide & the attached pics show the best I have been able to tune by starting low with incremental changes. Feed rate is 35"/min, read the file name to understand what changes show in the screen-shot. There is 2 traces of the same settings, 1 pic shows the screen with feed rate & other info seen, & the full screen trace of  that.
I welcome any suggestions.
Many thanks

The drive itself seems to not be tuned
(if it is setup for velocity mode)

There is not much point in tuning from the LinuxCNC
side until the drive itself is tuned

Does AMC have drive tuning instructions?

Are you sure the drive is setup for velocity mode?

PCW, Many thanks for all you do. I believe the drive to be setup for velocity mode. Is it possible that the value of "I" (0.051) is off by dismal point & needs to be 0.51? A-M-C engineer is completely in the loop.

PCW, In tommylight's servo tuning you offer this:
"Added 22.08.2019
-PCW mentioned several times that the FF1 value can be:

You either calculate FF1 (10/velocity@10V)
or
scale the analog output in velocity in which case FF1 = 1.0

Explanation: that is in case you know the maximum velocity the drive/motor combo can do at 10V analog input to the drive.
An initial P value of 1 is probably OK
Remove the maxerror lines in the hal file, if you dont have the maxerror lines removed, you will no be able to tune."

What I do not see if the "maxerror" lines are removed in the .hal file? See page 3, post: 12 May 2022 13:20 #242795 Or is that the "#" that was used to omit those lines, just after the PWM generator setup signal area of the hal file?

When I tuned the POT3 to .051 this was not a steady signal, it varied from .045 to .051; took (around) 3 CW turns.

Always open to helpful suggestions. I will try other tuning over the weekend.

Many thanks

Did AMC give you any instructions for tuning the velocity loop in the drive?

This is something that must be done external to the tuning in Linuxcnc (within the drive itself). But that doesn't mean that Linuxcnc can't be a useful tool for tuning the drive. For example you will probably need to use an oscilloscope to tune the drive. But you can probably use Halscope in Linuxcnc as that oscilloscope. Also if a pulse or signal needs to be generated to the drive's command input. That can be done using hal components in Linuxcnc.

But instructions for guidance for the procedures from AMC would be very helpful.

The procedure will likely be something like generating a specific command input pulse while monitoring the motor's velocity feedback. While twiddling pots and switching different jumpers on the drive to turn on/off resistors and capacitors in the drive to adjust the velocity response of the motor for the quickest response with minimal overshoot or oscillation. But specific instructions from AMC will likely be essential.