Centroid Community CNC Support Forum

slodat wrote: ↑Sat Jan 04, 2020 12:03 am Are you running 240v to the DYN4 logic (the red/black twisted to the two glass fuses) as well?

Yes. Current is very low for logic power. But I use 18 gauge wire.

Marty did you ever get any explanation of the GEAR_NUM/LINE_NUM parameters from DMM. I notice you are using 2000 for both in lieu of the 8192/4095 they recommended.

BASE wrote: ↑Sun Jan 05, 2020 12:20 pm Marty did you ever get any explanation of the GEAR_NUM/LINE_NUM parameters from DMM. I notice you are using 2000 for both in lieu of the 8192/4095 they recommended.

The numbers they published didn't seem right to me and I can't get them to respond. They have been VERY SLOW.
Watch their DYN4 DMMDRV setup video.

This was a preliminary benchtest. At least it shows that things appear to be working correctly. I will be doing the FADAL TRM soon. That will be the acid test.

The components I tested went with the Bridgeport TC22 I recently sold. Customer will install. I wanted to make sure motors were turning properly.
And I wanted to report back since this has been a VERY long journey.

As always, I am only showing my personal experience....
Marty

Yeah those numbers don’t seem like they could be right.

I have been chipping away at my retrofit for a while now and am to the point where I need to test my servos and drives, so your video caught my interest.

I did find that DMM video last night. I should be able to try mine out with the Oak sometime in the next week.

Just remember, unless your drives are new in the last month or two, you may have to send them and the cables back for updating

I’m nervous about that. Drives and cables are about 2 months old and the servos just got here a couple weeks ago. I had them build some 1kW servos that were stretched out versions of the 750W 86M model so it took a while to get them built.

At least I’ll know what to look for if they aren’t working right, so thanks for that.

BASE wrote: ↑Sun Jan 05, 2020 3:29 pm I’m nervous about that. Drives and cables are about 2 months old and the servos just got here a couple weeks ago. I had them build some 1kW servos that were stretched out versions of the 750W 86M model so it took a while to get them built.

At least I’ll know what to look for if they aren’t working right, so thanks for that.

Call them with the firmware info and detail numbers on the sides of the drives

Question about the parameter 256 , is that just for one axis or all 3. Reason i ask is im putting an oak in my fadal 4020, and also building a 4th axis with dmm ac servo drive. But im reusing the glenteck drives and those have to be velocity driven, not position.

occustoms wrote: ↑Wed Jan 08, 2020 12:34 am Question about the parameter 256 , is that just for one axis or all 3. Reason i ask is im putting an oak in my fadal 4020, and also building a 4th axis with dmm ac servo drive. But im reusing the glenteck drives and those have to be velocity driven, not position.

All Axis I believe: Match up your drives/motors

Marty

Explanation of suggested settings from DMM:

GEAR_NUM is the scaling of the pulse INPUT from the controller (OAK) to the servo drive,

GEAR_NUMx4=number of counts per rev

So we recommend setting GEAR_NUM to 8192=32768counts per rev from the OAK, so set 32768 into OAK motor encoder count

LINE_NUM is the encoder output resolution

LINE_NUM=4095 is a special setting we made for the OAK, setting LINE_NUM to 4095 will output same 32768count/rev to the OAK


You can set GEAR_NUM and LINE_NUM to different settings in our drive, but the OAK requires these two parameters to be the same. So the above setting as in the manual will satisfy this.

We talked to Centroid before and they said they recommend running at this resolution since this is the resolution that they run with the Yaskawa drives, so we just matched that,

You can set GEAR_NUM to 2000 and 8000 in OAK encoder count, you will also need to set LINE_NUM to 2000 to match. This will also work, but the control resolution will be at 8000 instead of 32768count/rev. It will still work, but won't be as high resolution, although 8000 is already pretty high and will satisfy almost all mechanical resolution requirements.

If you recall, we run the Acorn with 8000counts/rev and that's been no problem with everyone. But the OAK being a higher end controller, we increase the resolution up to 32768counts/rev for higher performance,