RS485 For Automation Tech 1100W Spindle Motor

[ThreeDJ16]

So do pulley ratios not work with step/direction? Or are the just used for speed actual information? I'm commanding 1000 RPM and get exactly 2000 RPM on the spindle using a tach. The step and direction are dead on with the DMM 750w servo motor. But I need to be able to command 1000 rpm and have 1000 rpm at the spindle (ratio is 1 to 2). But using the pulley has no affect. So should I just trim the electronic gearing in the DYN4 down to get 1/2 of the steps per rev? Curious how others handled this situation.

Thanks.
-Jasen

[ThreeDJ16]

Well, in case someone happens on this thread later and has an issue like I did, the answer is simply change the gear number in the DMM setup. Unfortunately it seems my spindle was a bit to much for the 750W DYN4 in step / direction / position servo mode and faulted out. I've since switch to speed servo mode and using 0-10v analog and it's working fine. It also using the electronic gearing for scaling the 0-10v, but doesn't seem as straight forward logic as step/dir was to figure out.

Also the lowest PPR setting for encoder pass through into a 4Mhz input board to get encoder feedback is 500. Anyway, it saturates the board around 2k RPM. I didn't calculate it out, but the numbers go crazy about that. But realistically you only need the feedback at low RPM for rigid tapping anyway.

Nice thing is this 750W servo makes quick work of the OPs my 550W Consew was struggling to keep up. So overall, happy with the change. Still needs some tweaking though.

[Battwell]

step/direction works fine for spindle if set correctly
you do need to set the m3,m4 acceleration/deceleration times in spindle setup or your spindle will be trying to start/stop instantaniously and make it fault out.
(m30 at the moment doesnt obey this- but thats being sorted. i just added m5 before m30 as a get around)

[ThreeDJ16]

step/direction works fine for spindle if set correctly
you do need to set the m3,m4 acceleration/deceleration times in spindle setup or your spindle will be trying to start/stop instantaniously and make it fault out.
(m30 at the moment doesnt obey this- but thats being sorted. i just added m5 before m30 as a get around)

I don't know what else that could have been done as a DMM Engineer manually tuned it via Team Viewer as I have a high inertia spindle (according to them and autotune refused to get through all cycles without faulting). But as soon as I started a reasonable milling op it faulted using position servo and step / direction. Switched to speed servo and analog 0-10v, no issues.

[Battwell]

they didnt recommend adding a braking resistor?

[ThreeDJ16]

they didnt recommend adding a braking resistor?

They sent a regenerative resistor but said it shouldn't need it.

[ThreeDJ16]

Actually the two main faults I received were mainly "Lost Phase" and then "High Current", never once got an "Over Voltage" fault. I didn't have accelerate set too high when I tried step/direction. It actually ran fine until it started into the materials, then it mainly had "Lost Phase". I may mess with it again, as they sent me a new DYN4 with a special firmware for higher current / torque last week. But I've been a little nervous to mess with position servo too much as the previous drive must have gotten roasted as it got to the point as soon as it was powered up, it immediately faulted on "High Current". Everything is wired per their manual, wire gauge, fuses and filters. They're thoughts were it's due to my R8 spindle and using a 1 to 2 ratio so I can get a reasonable spindle speed. Really I see no disadvantages to speed servo and analog. I could probably still rigid tap since the encoder feedback is there at lower RPM. Really I could care less about rigid tapping. Mainly just wanted a more powerful motor than what I had.

But, if I decide to change up the control, I'll post the results. Just right now would rather have a working motor. Plus I've got to get in gear working out the ATC stuff for my groups next project. As soon as we swap to linear rails, this mill will be completely DIY (within our group).

[Battwell]

in most servo drives- if using a braking resistor you must turn it on in the drive parameters- or it wont be used. (totally different to a vfd in operation)
i had to add them to my big machine axis servos as it couldnt stop a 2 ton gantry many times without overheating the internal regen resistor. it actually took a while to show this fault up- when i wrote some bad code that was sending the gantry rapiding long distances between drilling holes.

[ThreeDJ16]

in most servo drives- if using a braking resistor you must turn it on in the drive parameters- or it wont be used. (totally different to a vfd in operation)
i had to add them to my big machine axis servos as it couldnt stop a 2 ton gantry many times without overheating the internal regen resistor. it actually took a while to show this fault up- when i wrote some bad code that was sending the gantry rapiding long distances between drilling holes.

Odd, never saw this post. With DMM, there isn't a function to turn on the flyback current resistor (as my servo doesn't have a brake, different model number). Anyway, I never did get step/direction to work properly as under heavy load (really not even heavy) it would always give me a lost phase error. So instead I just build a spindle encoder with an AMT102 to get the PPR low enough for my input card to handle up to 8K RPM or more. Works like a champ with the built in PID feedback. The DYN4 feedback was fine for keeping the motor in tune with what the motor thought the actual was supposed to be, but in fact I never could get the calibration good enough to my satisfaction. So decided to quit screwing with calibration and just use software feedback and now actual and commanded at within 1 RPM unloaded. Not that it matters, but one day I'll grow a pair and try some rigid tapping which should make much better use of the direct spindle feedback....

Also, still extremely happy I returned that POS motor I got at the start of this thread and bought the DMM 750w instead.

-=J