Simodrive 610 Analog drive torque or velo mode

Hey guys me again,

I am unsure about the command that my siemens simodrive 610 needs, I am currently running it with the velocity command of linux, but is that correct?
The performance is not the best, the PID tuning doesn't seem to work 100% correct. The accel. is set at 750, so not really high for the machine. The motors seem to act a bit aggressive on jog commands, so they jerk ( is that the correct word for it? :-D) pretty hard in X and Y even with slow motions- about 850mm/min.

I've got an example of another machine running the same drives with linux, but in Ini it's different to my output_min /max.
My Ini is set to +-10, the example has much higher ratios, about 400 or so- that seems to be the torque mode, correct?

I hope you can help me, that would be great!

Thanks in advance

You can tell if a drive is running in velocity mode rather than torque mode
if it holds position or drifts slowly when enabled but resists external applied torque

Single PID loop torque mode tuning is very different as it it uses D to close the velocity loop and typically uses 0 or close to 0 FF1

Velocity mode tuning is 99% getting FF1 correct (scaled properly) since the velocity command to the drive must be correct before you even worry about feedback to correct position errors

Can you post your existing hal/ini files?

Hey, thanks for that fast reply!

Here we go, attached the hal and ini.

What the simodrive does is : transfer the input voltage -+0-10v into speed,and it tries to hold that speed with internal algorithms.
All Motors have encoders and tachos- the encoders go back to the mesa 7i77, the tachos feed the simodrive with information about the speed.

What actually is a problem, is that the machine can't hold tight tolerances of 0.01mm's. If I mill a circle of lets say 35.000mm the circe is about 34.980.

If I tune the PID more agressive, the F error gets less, but it starts shaking the whole machine because of the aggresive acc and dec. of the motors.

Thanks guys you are a big help!

Those look like reasonable values for a velocity mode servo.
Assuming FF1 is set properly (~400 mm/s at 10V) I doubt that you can use
much higher P gain than 6 since this is a velocity correction of 240 mm/s
with a 1 mm error which is pretty aggressive. That is, its likely to oscillate at
higher P settings

Can you plot the following error when running your circle?
That would help pinpoint the source of the error

Thanks for the answer!

FF1 at 24.000mm/min ? Didn't check that yet- the fastet I adjustet was 10.000 with 0.0008-0.001 F Error.
If I try higher P ratios, the Motor starts oscillating reeeeeaaaalllly ugly.

How can I plot a circle? I used a piece of aluminum and a twopoint gauge to check the roundness of it.

Thanks again and again for your help )

What I was saying is that if your FF1 setting is correct, the scaling of you PID loop is 400 mm/s at 10V, so I can work backwards from that to figure out the P scaling

A G2 or G3 move with a huge P value could be used for long term circular motions
then plotting the commanded position, commanded velocity and following error can be plotted with halscope

Okay, sorry I didn't quite understand it directly :-D

I will do that in a few minutes and post the result here.