[alan_3301]

Hi I haven't seen anything in manuals for various servo systems about if it is ok or not for the servo to be driven by another motor while disabled, but still connected to the powered up drive.
In another thread I'm trying to figure out a diy C axis situation with a dmm 750w servo. Because of low encoder resolution, I would probably have to gear it 10:1 to have reasonable resolution at the spindle.
One option is gearing the servo 10:1 and only using for positioning, while running a VFD and motor for regular turning operations.

I need to know if I can leave the servo connected to the belt drive, but disabled on the drive (will free spin, but I'm guessing will act like a generator) while the vfd is running the main motor. This would be nice because I could use the servo encoder to feedback for threading.
Otherwise, if I have to mechanically disconnect the servo from the main drive, I would have to have yet another encoder on the spindle.

Anyone tried this or know how if any manufacturer specifies if this is ok or not?

[alan_3301]

ok I realized that it wouldn't work given than if I ran the servo 10:1 at the spindle, when the vfd had the spindle running at full speed, the servo would be going 15,000rpms. obviously that wouldn't work.

But my question still stands, would an ac servo be ok to be driven, when the drive is not enabled, and the motor can freespin. given the speed limits are followed?

[bhurts]

It would be a bad idea. It would act as a generator and inject that voltage back into your drive possibly causing damage.

Ben

[mike_Kilroy]

Yes you can on most drives as u say not exceeding max rpm limit

[alan_3301]

Thanks for both replies.

My delta drives will display DC bus voltage, and alarm if it goes too high.
DMM drive has over voltage alarm also, but no display of voltage.

On the deltas, I will try to rig up a variable speed drill to the servo and monitor the voltage through the software. Just for fun, they will never be backdriven like that.
On the DMM I am waiting for an email back from the mfg. to be sure, since I cant monitor the voltage.

I'm curious to find out if the dc bus voltage will actually increase at all when the motor is back driven.

[mike_Kilroy]

I'm curious to find out if the dc bus voltage will actually increase at all when the motor is back driven.

GOOD QUESTION! Never thought of this before.... I will do Monday at office to verify, but seems the answer is yes, you will develop the Kb*rpm voltage on the dc bus by backdriving with drive off.... thru the flyback diodes across each IGBT.... I think it will go right up to peak of the sinewaves being created on the drive....

This is why with new fangled spindle brushless pm motors running 10s of 1000 of rpms via torque angle control, if power is lost while spinning at 20-30,000rpm, the dc bus will instantly try to go to 5-10,000Vdc (depending on the motor).... so all those spindle systems need a BIG FAST CROWBAR dc bus dumper (DB resistor type gidget)....

[mike_Kilroy]

BTW, we run those same ac induction motors (spindle motors) all the time with almost similar response to servos for C axis use.... The ONLY gotcha is that the induction motor has higher inertia than the PM brushless motor, so it must respond that ratio slower (if same current applied to both).... Other than that, spindle motors today on high performance drives do C axis function very well without the old fashion method of adding a PM motor on its rear end....

[Al_The_Man]

A common idea as per another recent thread is to zener detect an overvoltage value and at the point switch in a dump resistor via a high power Mosfet.
Al.

[alan_3301]

Thanks Mike, I can't say that I understood most of what you wrote, but I appreciate you replying.
I did try to drive the motor with a drill, using a rubber hose as a coupling, and the motor has considerable resistance the faster it is spun.
The coupling would start to slip past 50rpms, but the voltage didn't seem to increase any.

I guess if I was to go this route I would definitely need some sort of clutch to decouple the servo when running another motor.

[Al_The_Man]

and the motor has considerable resistance the faster it is spun.
The coupling would start to slip past 50rpms, but the voltage didn't seem to increase any.

I guess if I was to go this route I would definitely need some sort of clutch to decouple the servo when running another motor.

Either that or disconnect/open circuit the motor, the reason for the resistance is the BEMF has a load source, just connect the three stator connections and try and spin it by hand to see it in effect.
Al..

[alan_3301]

just connect the three stator connections and try and spin it by hand to see it in effect.

was that supposed to say "just disconnect...."?

Yes I was trying it with motor + logic power connected, but the drive disabled, and it was too stiff.
I tried again with only logic power, but it was still stiff. I can disconnect the motor wires with only logic power and not do any damage, it spins easy there.
It also spins easy with the drive powered down, and the motor connected, but no access to encoder in that situation.

I guess I could go the clutch route, or try to find a normally closed 3 pole relay for the motor, and have commands in place to cut drive motor power, and then open the relay contacts.
I would have to be certain it never opened under power as I understand this will damage the drive.

I wonder if I can find an a/c clutch from a tiny compressor from a car. I don't think I've seen one tiny enough.
I bet there is a off the shelf part somewhere that is way out of my price range.

Thanks.

[Al_The_Man]

I meant to get an idea of the resistance caused by BEMF is to disconnect the 3 stator connections and then connect all 3 together and turn the motor.
Al.

[alan_3301]

Oh ok, yes it has a good amount of resistance when all 3 leads are shorted together.