DG4S-08020 servo drive

This Section is for users to discuss hardware

Re: DG4S-08020 servo drive

Postby Derek » Mon Jan 15, 2018 1:10 pm

What encoders are you using. I seem to get a ton of buzz with AMT encoders when the resolution is high. My US Digital encoders seem to provide a more stable output at high counts.
Derek
 
Posts: 341
Joined: Mon Sep 05, 2016 9:57 am

Re: DG4S-08020 servo drive

Postby Chuck767 » Mon Jan 15, 2018 2:17 pm

I’m using the original encoders from the FANUC servos. They are 2500 count per revolution with 4 outputs but I’m only using two. I’m using the CNCdrive differential line drivers specified for the DG4S drivers. I tried switching the A B encoder outputs with -A-B but the loud Humming noise stayed the same.

Regards,
Chuck
Chuck767
 
Posts: 23
Joined: Mon May 02, 2016 2:14 pm

Re: DG4S-08020 servo drive

Postby cncdrive » Mon Jan 15, 2018 4:13 pm

The buzzing happens when the motor is stall?
If so then the PID loop is "upset".
It is because the resolution is high and there is some elasticness/cogging in the system and probably also the loop is tuned hard.

You could try to resolve this with tuning the PID loop "softer", however if there is a significant elasticness/cogging in the system then you might not be able to resolve this in software, but only removing the cogging from the mechanics.
The issue is that if there is a cogging/elasticness and the resolution is so high then one encoder tick is a very short movement and so a minor cogging in the system will cause the motor to get a low torque on that point, so when it moves over the point then it accelerates up fast to high speed, because the torque requirement gets low quickly at the cogging and so it will overrun the setpoin and will back compensate and so on and so on, the fast moving back and forth and so the buzzing happens.

So, check if there is any backlash in the system and also check if there is any loosen belts and if the encoder is connected stiff enough to the motor shaft.
cncdrive
Site Admin
 
Posts: 4695
Joined: Tue Aug 12, 2014 11:17 pm

Re: DG4S-08020 servo drive

Postby Chuck767 » Mon Jan 15, 2018 4:29 pm

Thanks for the detailed and fast response Derek.

Relative to “The buzzing happens when the motor is stall?”
No, The buzzing occurs when the motor is just setting there idle. What I mean by idle is if I command the X axis to move to inches, it moves precisely two inches but then the motor just buzzes after that.

As for backlash in the system, I get the same buzzing when the motors are simply sitting on a table. As far as backlash between the servo motor and The encoder, I don’t believe that is an issue since the set up, servo/encoder is all original FANUC

Thanks again,
Chuck
Chuck767
 
Posts: 23
Joined: Mon May 02, 2016 2:14 pm

Re: DG4S-08020 servo drive

Postby cncdrive » Mon Jan 15, 2018 5:43 pm

Yes, by "stall" I ment "stands still", in other words when there is no motion command, so I ment the same what you just described.
cncdrive
Site Admin
 
Posts: 4695
Joined: Tue Aug 12, 2014 11:17 pm

Re: DG4S-08020 servo drive

Postby Chuck767 » Mon Jan 15, 2018 5:49 pm

OK, thanks.
I thought you were referring to the motor coming up against a heavy load and stalling.

Do you have any thoughts as to what I can try in the way of PID tuning?

Chuck
Chuck767
 
Posts: 23
Joined: Mon May 02, 2016 2:14 pm

Re: DG4S-08020 servo drive

Postby cncdrive » Mon Jan 15, 2018 9:16 pm

Well, if there is a backlash or a belt is ellastic that is exactly that the axis is going from and to a heavy load, I mean that is exactly a situation like that,
because when the axis starts rotating the screw it requires a much larger torque than when the axis is on the backlash when the nut rotating about freely when overcoming the backlash.
So, however it is that same situation as if the axis quickly gets a load and then the load is removed in an infinate cycle which can upset the PID loop especially if the encoder resolution is high and/or if the PID loop is tuned hard.

You could for start try lowering parameters, especially the P and I, however you also have to lower the D. And you can try to limit the integral term too, set the I low and try to limit with Li.
cncdrive
Site Admin
 
Posts: 4695
Joined: Tue Aug 12, 2014 11:17 pm

Re: DG4S-08020 servo drive

Postby Chuck767 » Mon Jan 15, 2018 9:40 pm

OK thanks, I will see if I can do some PID tuning as you specified.

As for backlash or elasticity of the belt, I am getting this problem with just having the servo sitting on a bench with no belts or ballscrews attached.

These are my current PID Settings.

Thanks for your help,
Chuck
Attachments
85A0473C-936C-4524-A326-17D76CF75F6C.jpeg
Settings.pid settings
Chuck767
 
Posts: 23
Joined: Mon May 02, 2016 2:14 pm

Re: DG4S-08020 servo drive

Postby Chuck767 » Tue Jan 16, 2018 4:50 am

Derek,
To help me get in the ballpark with my PID tuning could you provide the settings that you are using on your servos.
I’m having trouble getting to a point where I don’t have this loud hum and vibration when the servos are just sitting there.

My concern is that I’m going to Fry the servo. There was a post on CNCzone where somebody had this problem and destroyed their servo, and I really don’t want to do that.

Thanks,
Chuck
Chuck767
 
Posts: 23
Joined: Mon May 02, 2016 2:14 pm

Re: DG4S-08020 servo drive

Postby Derek » Tue Jan 16, 2018 12:08 pm

Chuck767 wrote:Derek,
To help me get in the ballpark with my PID tuning could you provide the settings that you are using on your servos.
I’m having trouble getting to a point where I don’t have this loud hum and vibration when the servos are just sitting there.

My concern is that I’m going to Fry the servo. There was a post on CNCzone where somebody had this problem and destroyed their servo, and I really don’t want to do that.

Thanks,
Chuck


Hi Chuck.
I'm using 16035 drives so this may not be applicable. I just use the recommended starting points from CNC4PC http://cnc4pc.com/Files/DG4S08020-34-180-72.PID and tweaked them. The only way for me to get rid of the at rest hum was to reduce the encoder count. The AMT encoders are open in the back. I could place my finger against the center disk and stop the buzzing.

Unless the buzz is heating up the servo or the drive it shouldn't be a problem. Balazs may have more to say on that.
Derek
 
Posts: 341
Joined: Mon Sep 05, 2016 9:57 am

PreviousNext

Return to Hardware

Who is online

Users browsing this forum: No registered users and 2 guests