Stepper motor slipping = damaged?

[walkingcube]

Hi folks,


I have a strange noobs problem that I can not wrap my head around. I have a plastic tubing stretcher that uses a stepper motor (Minebea 23KM-K709-07) to precisely do the stretching without going through any gearboxes etc.

It was working just fine for a long time, until one day, the motor slipped as we were experimenting on a new tougher plastic. Ever since, the motor has been slipping from time to time, even though we've not used the tough plastic ever since.

I thought that overloading a stepper motor once or twice shouldn't result in any permanent, adverse effect. Can anyone share their experience on how to best deal with this?


Thanks a lot!

 

[MikeHalloran]

Perhaps it was slipping before and you didn't notice.

You didn't by chance disassemble the stepper for inspection, did you?
That destroys their permanent magnets.



Mike Halloran
Pembroke Pines, FL, USA

 

[waross]

The original "slip" may have partially demagnetized the magnets.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[walkingcube]

Thanks a lot for your reponses.

I really thought that the reason why the original design used steppers what that overloading them will not cause damage.

What can I do to verify the intergrity of the magnets?

 

[Skogsgurra]

One way is to rotate the rotor at known speed and measure induced voltage. If it is much lower than same voltage in an undamaged motor, the magnets are weaker.

Much lower? Say 20 percent lower. If same, no probs.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[MikeHalloran]

I didn't think a few missed steps would kill them, either.
Never noticed anything like that happening with them.

There remains the possibility that the stiffer tubing pulled something out of line, and the possibility that some lubrication point needs attention.

In particular, steppers often ship with only a preservative oil in their ball bearings. One (only) drop of sewing machine oil applied to the outside face of the bearing won't hurt. The bearings are almost always shielded but not sealed, so the oil will wick right in.



Mike Halloran
Pembroke Pines, FL, USA

 

[electricpete]

5.3 – Stepper Motor Drives…

The permanent-magnet motor has a laminated, slot wound, stator, usually with two, three or four phases. The rotor mounted on a bearing in each end frame is usually a solid cylinder magnetised in a two, four, six or eight-pole configuration….

Care must be taken that the current rating of PM steppers is not exceeded, otherwise demagnetisation of the rotor can occur.

Seems to support the idea that demagnetization can occur. According to this ref, the solid rotor is permanently magnetized, rather than attaching separate magnets. I don’t know what is common construction.


=====================================
(2B)+(2B)' ?

 

[Skogsgurra]

Yes, no doubt that they can be demagnetized when current gets high. Same problem with pan-cake motors. It all depends on what driver type you have. This stepper is not the type you find in office machines, but a bit more brutal. If the drivers are not current limiting, things do happen when the torque demand (and thence current) gets too high.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[cswilson]

Typical stepper motor/drive systems do not act at all like induction motor drives. The simplest ones simply output a fixed voltage magnitude in a rotating time sequence around the phases. The current is higher at lower speeds due to reduced back EMF, but the same voltage is usually applied at zero speed to provide holding torque. I can't see how a stall/slip condition could push the current into demagnetization range.

More sophisticated drives, such as microstepping drives, use current feedback to achieve a fixed current magnitude in a rotating time sequence (sinusoidal or pseudo-sinusoidal) around the phases. This has even less chance of demagnetization, as the voltage is automatically adjusted to maintain the current level.

Unless you get into the "sensorless vector" style of stepper drives, the drive will not know that the load has increased, and will not try to dump in more current to overcome the load.

Curt Wilson
Delta Tau Data Systems

 

[potteryshard]

Could it be that the problem lies in a drive output, rather than in the motor itself?

 

[TurbineGen]

In my experience it has been heat over extended periods of time that leads to demagnetization of the stepper motor. Has the ambient temperature changed with the new material? If so, that could have something to do with what you are seeing.

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.

 

[walkingcube]

Hi folks,


Thanks for all the tips and ideas.

potteryshard: That's something I was reading up too. Is there a way to test that?

TurbineGen: No, the temperature is relatively constant and the motor is not contantly in use (> 10% duty rate).

I called the distributor last week - not only do they have no clue, but the motor has been discontinued without an equivalent!

 

[potteryshard]

The obvious way to check the drive outputs is to look at them with an oscilliscope or voltmeter when the motor is slipping. If the outputs are present when slipping, the problem likely lies within the motor.

It isn't always easy to get an intermittent problem to cooperate however. It may not choose to act up while you are measuring outputs. It may be possible to temporary in a couple of low output lights matching the motor voltage parallel with the motor leads. When slipping an operator can glance at the lights to see if all outputs are firing.

 

[MikeHalloran]

Step motor voltage waveforms as seen on an oscilloscope get very strange when it's slipping. I'd like to think some kind of slip detector is possible, but I never figured out how to do it with any kind of simple/passive circuit.

I usually attach LEDs to the drivers in lab setups, but they don't add much information; when a driver isn't firing, the motor may twitch, but it doesn't rotate.




Mike Halloran
Pembroke Pines, FL, USA