Pedersen VP2000 810M retrofit, severe servo hunting

Hey all. I am doing a retrofit of a milling machine that ran a Siemens 810M system previously using a Mesa 7i92 and a 7i77 for running all the I/O. We have gotten to the point where movement mostly works but we are having an issue where the Z axis specifically hunts back and forth about 0.02mm. This is unlike the other axes which hunt an expected 0.001mm since we are using 1 micron linear scales and analog velocity mode servos. Can anyone think of why this is happening? It decreased after adding shielding to the signal cable but is still not satisfactory.

The original system never showed any significant hunting (not noticeable). I suspect it might have disabled individual axes after reaching the set point, but it seems that all the analog enable signals only show up as a single pin in HAL. This is just speculation though. Thoughts?

Can you post your hal and ini files.

Z axis are often harder to tune because they typically have a static load

Here you go.

Looks like you have 2 hal files and no .ini file

Crap, sorry. Here's the INI.

Use a smartphone with whatever software for sound analysing, check if the frequency of that hunting is 50Hz ( or 60 Hz if you are in USA ). If yes, there is something wrong with the cabling or shielding. If any other frequency, do some tuning to the Z axis, namely lower the P value a bit and check if it subsides.

Okay thanks. The frequency is more like a fraction of a hertz. I'll try some further tuning.

OK, fraction of Hz definitely sounds like gravity pulling the Z down. For machines with heavy spindle heads, you need a counterbalance for the weight. So, check this first, it may need adjustment of air pressure or something. (On a Mazak, could even be hydraulic.) If not counterbalance, then you need to adjust the BIAS in the axis tuning.
This adds a steady force in the servo to counter the head's weight.

Jon

If these are velocity mode drives, its actually better to add some offset to the torque
loop in the drive itself (to counteract gravity) if this is possible.

PCW wrote: If these are velocity mode drives, its actually better to add some offset to the torque
loop in the drive itself (to counteract gravity) if this is possible.

Yes, if this is available in the drive, and IF the pots can be set fine enough for this, then that is quite right.
But, some servo amps just have one-turn pots, and you can't get anywhere near fine enough adjustment.

Jon