Oak controller with ACDC-60s on Fadal VMC3016; control issue questions (Resolved)

All things related to Centroid Oak, Allin1DC, MPU11 and Legacy products

Moderator: cnckeith

q4stix
Posts: 10
Joined: Sun Sep 20, 2020 8:47 pm
Acorn CNC Controller: No
Allin1DC CNC Controller: No
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
CNC Control System Serial Number: 1114190882

Oak controller with ACDC-60s on Fadal VMC3016; control issue questions (Resolved)

Post by q4stix »

I'm converting a Fadal 3016 with original electronics (in other words a mostly stock machine meaning Glentek RM6060-50 servos) to use a new Oak controller with Centroid provided ACDC-60 drives, PLC 1616, and touch screen console with built in PC.

Eventually I'll have it as a fully running machine but since being a structures design engineer in product development and prototyping and it being a retrofit project only when I have free hours at work, this may be a bit slower than the usual retrofit.

So far I have the OAK board, ACDCs, and spindle VFD and gearbox (controlled through M3 and M4 custom code) hooked up. I only have the X-axis DC motor and Centroid 10k line encoder hooked up.

I'll get the report exported when I'm back at work but I wanted to ask some questions that I *hope* don't require a report to answer or recognize.


Questions...
1: When I turn on the system or after triggering the e-stop and then jog the axis, it jumps the equivalent of 15-20 degrees servo rotation. It doesn't do that after the first jog. Clearly this isn't ideal and shouldn't happened. Is there something obvious I'm missing? The response time on the motor is the default 1.75ms so it shouldn't be a matter of delay. I even turned the feed override down to 5% and it's still a quick 20 degree rotation when jogging after an e-stop release. This is really the only time the servo jumps.

2: I'm confused about the PID setup. From what I see there's a drive current PID and a motor position PID which seems like a PID stacked on a PID. I get some of the values with inertia, etc and can work with the position PID but the current PID seems dependent on position. I see starting values for the servos in the documentation but not the drive PID. Is there documentation on this?

3: I'm sure more to come...

Thanks in advance!
-Matt
Attachments
20200910_151254.jpg
eng199
Posts: 148
Joined: Fri Jan 10, 2014 11:29 am
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: Yes
DC3IOB: Yes
CNC12: Yes
CNC11: Yes
CPU10 or CPU7: Yes
CNC Control System Serial Number: none
Location: Howard, PA

Re: Oak controller with ACDC-60s on Fadal VMC3016; control issue questions

Post by eng199 »

The RM6060-50 is not on the list of motors tested with AC/DC. Actually, the part number is likely an error, or special for Fadal. Normally Glentek numbers start with GM or GMR. Back to the point... it is essentially impossible to tune the current PID properly without a special setup. Normally, the sales guys will steer you toward sending in unknown motors for setup, or buying new motors if shipping is not feasible.

I will guess the jump is due to AC/DC being set for a brushless 8 pole motor, while the one you are using is DC with brushes (poles must be set to 0). There is almost certainly a error message displayed, which you have omitted from your post.

Drive PID controls the motor current, while position PID controls positioning. Drive or current PID is set to match the motor and does not change. Position PID is what you will need to tune to match the machine. It is recommended to contact Centroid and send a motor in for current tuning.
q4stix
Posts: 10
Joined: Sun Sep 20, 2020 8:47 pm
Acorn CNC Controller: No
Allin1DC CNC Controller: No
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
CNC Control System Serial Number: 1114190882

Re: Oak controller with ACDC-60s on Fadal VMC3016; control issue questions

Post by q4stix »

Hi,
You are correct, it's GM6060-50 (listed generically by Fadal as a GM6000 but based on the current ratings, resistance, inductance, etc identical).

I do already have the poles setting at 0 but that was a good check. No error messages have been displayed.
I was able to start with values for the Glentek GM4030-41 and GM3340-30 which are listed in the ACDC drive documentation (AC/DC 130307 User Guide). This is a case of being able to convince management to make funds for current tuning after I prove the new setup can work but needs tuned vs. spending unallocated funds on tuning before I show the machine can move and I've otherwise set it up correctly.

I've attached a report now that I'm by the machine.
Attachments
20200921_112013_resized_1.jpg
20200921_112028_resized_1.jpg
report_1114190882_2020-09-21_14-21-08.zip
(3.56 MiB) Downloaded 6 times
cncsnw
Posts: 1604
Joined: Wed Mar 24, 2010 5:48 pm

Re: Oak controller with ACDC-60s on Fadal VMC3016; control issue questions

Post by cncsnw »

The Kg and Kv1 values are suspicious. How does it behave with Kg=0 and Kv1=20?

Does it still jump if you enable it without immediately commanding a move (e.g. by pressing F3/MDI, instead of by pressing a jog button)?
eng199
Posts: 148
Joined: Fri Jan 10, 2014 11:29 am
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: Yes
DC3IOB: Yes
CNC12: Yes
CNC11: Yes
CPU10 or CPU7: Yes
CNC Control System Serial Number: none
Location: Howard, PA

Re: Oak controller with ACDC-60s on Fadal VMC3016; control issue questions

Post by eng199 »

Kg and Kv1 should be 0 on AC/DC systems. They do not make any improvement like they used to on older drives.

The high Kv1 could actually be the cause of the jump. It is a factor added in proportion to requested velocity. If it is too high, a movement may occur due to the Kv1 contribution that the PID calculation then has to correct.
q4stix
Posts: 10
Joined: Sun Sep 20, 2020 8:47 pm
Acorn CNC Controller: No
Allin1DC CNC Controller: No
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
CNC Control System Serial Number: 1114190882

Re: Oak controller with ACDC-60s on Fadal VMC3016; control issue questions

Post by q4stix »

Well shoot, that's interesting! Can you point me to where it says not to use those values for the AC/DC? I'd like to be able to point to it if someone else has to take over one day.

I started and reduced the Kg to 0 and that eliminated the quick jump at the first jog after power on or releasing the e-stop. Big thank you for that! cncsnw, to answer your question, it would also jump when going into MDI mode. I did some playing and found the jump offset is nearly proportional to the Kg value. Now with Kg=0, the first jog input doesn't do anything but every jog thereafter responds correctly.

Kv1 was set based on the back EMF value (V/Krpm) from the servo specifications which is why it was 50. After reducing it to 20 the electrical noise dropped and now having to set it to 0 based on eng199's input, the noise has been eliminated. Another thanks!

These two things also fixed setting my home position which I figured was related to the jump on first jog or jump when going to MDI.
Last edited by q4stix on Mon Sep 21, 2020 7:12 pm, edited 1 time in total.
martyscncgarage
Posts: 6686
Joined: Tue Mar 28, 2017 12:01 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: No
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: Yes
CPU10 or CPU7: Yes
CNC Control System Serial Number: none
Location: Mesa, AZ

Re: Oak controller with ACDC-60s on Fadal VMC3016; control issue questions

Post by martyscncgarage »

Glad this got posted here. Great information. Let us know how you get along with the project. I would like to see pictures of it. Post here?
Keep us updated.
Marty
Reminder, for support please follow this post: viewtopic.php?f=20&t=383
We can't "SEE" what you see...
Mesa, AZ
eng199
Posts: 148
Joined: Fri Jan 10, 2014 11:29 am
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: Yes
DC3IOB: Yes
CNC12: Yes
CNC11: Yes
CPU10 or CPU7: Yes
CNC Control System Serial Number: none
Location: Howard, PA

Re: Oak controller with ACDC-60s on Fadal VMC3016; control issue questions

Post by eng199 »

q4stix, after looking at the user's guide again, I realized it does not state that Kg and Kv1 are not used anymore, they are just not mentioned.

GM6090-50 has a continuous stall rating of 26.7A. DC brush motors typically work best with drive current set to 2 times continuous. This would make the current setting 89 instead of the 30 or 20 in your report. There are heating and cooling parameters to prevent over working the motor. It is unlikely you would work that motor too hard, but we can check them if you like. If you post the weight of the motor, I will have all the data to calculate the parameters.

Parameter 365 may be causing your missed first jog if you are quick on the buttons. It is primarily for third party drives. There should be no problem setting it to 0 with AC/DC.
q4stix
Posts: 10
Joined: Sun Sep 20, 2020 8:47 pm
Acorn CNC Controller: No
Allin1DC CNC Controller: No
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
CNC Control System Serial Number: 1114190882

Re: Oak controller with ACDC-60s on Fadal VMC3016; control issue questions

Post by q4stix »

eng199, thanks for the feedback on the current limits. I knew they could be higher than the continuous but didn't know how high and figured starting lower was better than too high.

The motors are GM6060-50, not GM6090-50 (I got caught too looking at the wrong one a couple times because they're closely numbered). I've attached the spec sheets I found and was referring to. Since the 6060 is rated at 17.8A continuous that works to 29.6%. I'll double that to be 60% based on your feedback. Peak stall is listed at 89.0A so even at 100% it hopefully wouldn't be a problem.

From the spec sheet, the weight of the whole motor is listed at 41 lbs so I'm not sure if that's what you're looking for or not since that includes the case.

Looks like the lowest I can set parameter 365 to is 250.0. Setting it higher works but anything lower than 250 reverts back to 250. Not a huge deal at this point since I have so many other things to finish setting up first.

Thanks for the continued help
Attachments
GM6000%20Datasheet.pdf
(666.23 KiB) Downloaded 7 times
eng199
Posts: 148
Joined: Fri Jan 10, 2014 11:29 am
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: Yes
DC3IOB: Yes
CNC12: Yes
CNC11: Yes
CPU10 or CPU7: Yes
CNC Control System Serial Number: none
Location: Howard, PA

Re: Oak controller with ACDC-60s on Fadal VMC3016; control issue questions

Post by eng199 »

60% looks correct to me. You want to stay below the peak current, but that is no problem since it is usually very high. The 2x continuous rule has worked best in testing for brush motors.

Here are the numbers I got for heating and cooling. The parameter numbers are for the first axis. The heating coefficient changes based on the units the control is set to use.

Imperial
p21 1.0491801859906429
p25 2.7777777777777777

Metric
p21 0.5828778811059128
p25 2.7777777777777777

If you are using the "Centroid AC/DC Setup Wizard" program, you can verify the values by filling in the data fields and pressing "Calculate Parameters".
Post Reply