[thedudeabidez88]

Hi,

I am modernizing an old Anorad XY servo positioning stage for a laser trimming system. The electronics control station is very primitive and late 70s or 80's technology. I worked on trying to get it up and running. The small CRT monitor doesn't work very well, parts don't work and it looks like it was built by an armature. I did get as far as moving one of the servos with the old work station but the whole system has issues. So I'm going to build a custom CNC/Mach3 box. I'm trying to locate new servo drives for these CMC MT2130-071BE PM Field servo motors. The positioning stage has linear encoders, but I suspect they don't work so I could switch to quadrature rotary encoders. I have only built stepper CNC systems before so this is my first servo system.

The only information about the motor that I could find is in here.
http://www.electromate.com/db_suppor...011Catalog.pdf

I was thinking about the Gecko G320X servo controller, but I'm not sure if it will work since this is a PM field servo motor which I think means its a permanent magnet motor...Please correct me if I'm wrong. Would this drive work with this motor? Does anyone have any other recommendations on drives? I have called Cleveland Motion Controls three times but I've only ever been able to just leave a message and someone will call me back but no one ever does...

[Al_The_Man]

CMC brushed motor have a good reputation and have been used extensively in the past for motion control applications.
Most DC motors are P.M. now, you would be hard pressed to find a wound field DC servo anymore.
I have the CMC spec's on most of their motors.
Any of the servo drives that offer step/dir should be able to accommodate this motor.
The trick is is if the motor has a rear shaft to take a encoder.
Al.

[thedudeabidez88]

Okay thank you. I'd be interested in any other data sheets or info that you have if you don't mind. So the G320X sounds like it will work then for this motor? ( http://www.geckodrive.com/geckodrive...ves/g320x.html) Yep there is a rear shaft in the back of the motor. I think the servo motor has a tach on it as well. There are 4 connections. Two of them are the armature + and - . Here are some attached images of the whole Anorad system. I was going to change the whole system out with steppers but then I discovered that the NEMA 23 stepper that had is to big and wont clear under the table saddle.

[Al_The_Man]

My book shows MT2130 as Cont Torque 53oz-in: 10v/krpm: 13.6 oz-in/amp
You can remove the tach brushes as they will not be needed for modern drives.
One thing you should do is replace the aluminum coupler with a stainless steel version.
Al.

[shaftalignment]

My book shows MT2130 as Cont Torque 53oz-in: 10v/krpm: 13.6 oz-in/amp
You can remove the tach brushes as they will not be needed for modern drives.
One thing you should do is replace the aluminum coupler with a stainless steel version.
Al.

Al,
Why would you replace the aluminum coupling? The coupling appears to be a 6 beam ROCOM coupling which is every bit as strong as a single beam stainless Helical coupling of the same size. It also has less wind up.

Max

[thedudeabidez88]

Yeah I'm going to keep the helical coupler that it had, it was a very high precision CNC laser cutting station at one time and will be, no reason to change it. Just to share an update and request for some help on interfacing the original linear glass encoders that was already in place. I was wondering if these linear encoders are even capable of being interfaced to the G320x drives...I'm trying to figure out how I would interface these older linear encoders to the new system im putting together. If I cant use these linear encoders with these drives, im just going to go with quadrature rotary encoders on the rear shaft. Which is what I'd like to do, but it just makes sense to use what I have available if I can.

[Al_The_Man]

These ARE quadrature encoders, just that they take the two quadrature signals and use the low level sin/cos signal to calculate a co-tangent value for position, you could take the signals and square them up to TTL or RS485 level if you felt inclined, and essentially convert to simple pulse/dir type.
They also are old technology that uses an incandescent lamp.
Al.

[thedudeabidez88]

I knew these were quadrature, but how would I go about doing that? I don't have access to an oscilloscope to look at the signals. Some sort of TTL converter to convert the sin/cos signals to TTL? Any recommendations.

[Al_The_Man]

There is some that are made by Heidenhain you may pick up on ebay etc, otherwise a level translator such as a schmitt trigger or an Opamp circuit could be configured.
Al.

[Al_The_Man]

Also a simple comparitor IC such as LM311 may work.
Voltage Comparators
Al.

[thedudeabidez88]

Since I've last posted, I finished the electronics for this project and powered up everything. However I've been having issues. I have the US digital E5 encoders with the the differential line driver that came with the encoder as well as a differential to single ended signal converter both from CNC4PC all connected using cat5 patch cables to a C35S BOB. Well at first I had issues with the G320X drive faulting after power up and not giving power to the motor or encoder. But I solved this by flicking the HEDS switch on the drive to ON and jumping from the ENC+ to the ERR/RES terminal to clear any faults. After I did this, power was then flowing to the encoder and motor and it was holding position if I used my thumb to move the shaft, however the torque was not there. I was using a 24 volt power supply at 15 amps I've been using the power supply's that came out of the Anorad system that powered these motors. I've been trying to find the right one that was actually used for these motors. There is one that is going up to 97 VDC, but that is to much for the drive to handle and there is one that I measured at 75 volts and I hooked that up to the drivers and the motor still isn't getting any torque. The original drives that came out of the old system were 30 volt drivers. I can moved the motor in mach3, but yeah no torque. I realized that these motors potential can go up to 65 volt dc correct me if I'm wrong. I'm still trying to make sense of the specs on the motor. But anyways, I'm not sure what I'm doing wrong. I've tried adjusting the PID and current limit pots and although that seems to make the motor more responsive and quiet, but no torque. I can turn the motor shaft with only slight resistance, but I only do that for a few seconds to prevent doing something to the motor. I had the motor disconnected from the lead screw on the machine for testing so I don't crash the table if it did suddenly receive the torque if something happens. I'm stumped...I have personally come to the conclusion that servo systems are more difficult than stepper systems, but for this machine I had no choice. I've wasted to much time trying to diagnose and fix this problem, any help would be very appreciated.

[thedudeabidez88]

Any help on this problem would be very appreciated.

[mike_Kilroy]

sounds like time to send gecko ur wiring diagram to see. does enc go to gecko at all or just mach3? sounds like just to mach3 and ur gecko things are only current loop drive and so you are missing a velocity loop; r sending gecko command for speed instead of command for torque. this would produce exact results u describe.where is ur velocity loop? mia?

[thedudeabidez88]

These have tach brushes as well, and they were wired to the motor on the old system. However I was told that I could remove those for the G320X. I'm using a 500 line rotary encoder on the end of the motor. I have just the Pos and Neg armature leads going to the driver where they should go and the encoder going to the driver.

[Al_The_Man]

I usually remove the tach brushes these are not needed on modern drives.
Al.

[jfong]

If the specs that Al gave is correct, these are not high torque servo motors. I use similar size 43oz-in Cliftons with Gecko320. Yes you can turn the shaft with just two fingers. Does the motor spring back into position when you move the shaft? If so then is it probably wired correctly. My servos are directly coupled to 20tpi leadscrews. Because of the mechanical advantage the leadscrew give, with just 43oz-in torque, you cannot stop the table from moving.

Just wanted to make sure your not expecting lots of torque like a stepper motor.

FYI. It's hard to correctly tune a servo motor with out some kind of load on the shaft. You may need to retune with the leadscrew attached.

[thedudeabidez88]

Thank you.
Yes the motor does bounce back, but sometimes oscillates back and forth. I think that is called dithering, and I think I just need to refine the tuning on the drivers. I don't have a scope, so all I can do is tune the motor by hearing and looking at the motor shaft. With the low amount of torque that is actually coming from the motor, it's hard to imagine the motor not stalling when I couple it back on the Ballscrew. These are the original motors that came with the laser trimmer, So these have ran with the ballscrews. If I put the motor back on the screw, and tuned the motor for that, would the driver sense that resistance and give more current to the motor to compensate? I can turn the Ballscrew by hand, but it's certainly more resistance to turn the screw than it takes to turn the motor shaft when powered. I'm just wondering if I have the current limit pot on the driver set correctly for these motors.

[jfong]

I don't bother to use a scope even though I have a couple TEk''s. To big to lug downstairs. It's called dithering where the encoder jumps between positions. There will always be some and I tune to try minimize it. The oscillation should be so small that you will not see it but only hear it. If you see oscillation, lower your P term.

Yes connect your ballscrew to the motor. For the 320x, I think full current is CCW. But I would probably set the current adjust to 50% (middle) since these are small motors and unlikely to handle more than 10amps or so.

Don't worry about torque, it should be enough. Unless the motor is defective, it will be fine. Check brushes to see if worn down.

I'll try to help as much as I can. I have about a dozen gecko servo drives so I've done tuning

Also I would use the original 24volt supply. The others are to much for the motors.

[thedudeabidez88]

I have tried 3 of the 4 servos that I have, all are the same model number. I attached one of two encoders that I have on each motor and then attached them one at a time on the ball screw and fiddled with the PID settings. I turned the current limit on the drive CCW to half way. All with the same result, only slight resistance on the shaft uncoupled and nothing when the motor is attached. When I tried this, I did what you said and used the 24v supply. My observation the more voltage I put to the motor, the more responsive they are, but no torque. For example, in mach3 if I have the 24v supply hooked up, I try to move and the shaft only slightly turns, maybe a quarter rotation intermittently it doesn't work at all sometimes. This is when the motor is unattached to the ball screw. When I attach it and try to move it, no movement happens with the screw. I haven't tried at higher voltages attached to the ball screw, but unattached at higher voltages, the motor shaft seems rotate better and respond better, but hardly any torque is there. Should I try again with the 75 volt power supply and attach it to the ball screw, tune it, and see what happens?

[thedudeabidez88]

I'm really starting to give up on these old motors....