PDA

View Full Version : 3 wire stepper?



Philldapill
05-09-2008, 04:17 AM
Well, I haven't posted anything in a while so here it is.

I've been into controlling steppers with the propeller lately and it's gotten old. However, recently, I came upon a BIG motor of some kind that looks like a long NEMA 34 size stepper. I applied power to a pair of leads(after finding which ones) and the thing kicked VERY hard into a single position. I've found that only 3 wires form pairs inside the motor by taking it apart a bit. Anyone know anything about this? It looks like it used to have an encoder or something on it as there is a ground to the case, and two clipped wires inside that do nothing. The remaining three are what apparently energize the coils of some kind inside. Very odd.

Once I understand how the coils are configured, then the question is, building a circuit for the prop to control. I have a feeling it's not going to be simple.

Phil Pilgrim (PhiPi)
05-09-2008, 05:47 AM
It's probably not a stepper, but a three-phase brushless DC motor.

-Phil

Philldapill
05-09-2008, 07:02 AM
Isn't a brushless DC motor essentially a stepper? I mean, the principal of multiple poles is the same, right?

I did some testing on it, and it sems that I can get it to step by alternating the series of connections, just like a stepper. Basically, I got it to work like this: Let's call each wire L1, L2, L3. By alternating like the following, it rotates with a VERY strong holding torque.

L1+, L2-, L3-
L1-, L2+, L3-
L1-, L2-, L3+
repeat

I haven't determined the step angle yet, but it seems to be on the order of about 1.8 degrees, just like my other steppers...

So, Phil, back to the question of it being a brushless DC motor... Aren't they kinda the same?

James Long
05-09-2008, 07:09 AM
Philldapill said...
Isn't a brushless DC motor essentially a stepper? I mean, the principal of multiple poles is the same, right?

I did some testing on it, and it sems that I can get it to step by alternating the series of connections, just like a stepper. Basically, I got it to work like this: Let's call each wire L1, L2, L3. By alternating like the following, it rotates with a VERY strong holding torque.

L1+, L2-, L3-
L1-, L2+, L3-
L1-, L2-, L3+
repeat

I haven't determined the step angle yet, but it seems to be on the order of about 1.8 degrees, just like my other steppers...

So, Phil, back to the question of it being a brushless DC motor... Aren't they kinda the same?
Phil,

Yes, that is true to a point. But a brush-less motor never gets to rotor lock position, the poles are changed before the armature reaches that point. So fromˇthe control stand point theyˇare a little different.

The primary way of control is the same, but the timing of steppers vs. 3 phase brush-less is different.ˇ Also, steppers from a voltage stand point are usually controlled all from positive. 3 Phase brush-less on the other hand actually swap the polarity of the windings.

So there are some differences which are subtle, but create a different animal.

James L

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James L

Partner/Designer
Lil Brother LLC (SMT Assembly Services) (http://www.lilbro.mosquito.net.nz/index.html)

Philldapill
05-09-2008, 08:07 AM
That's not too bad then. I'm use to 2 phase bipolar, so I understand how the polarity is swapped here and there. Is there some kind of feedback, such as an encoder, ALWAYS on a brushless? I'm just fine with using this monster as a stepper motor, which it seems to do just fine. Any pointers on doing just that? I imagine the propeller control aspect should be a snap if my thinking is correct.

James Long
05-09-2008, 09:05 AM
Philldapill said...
That's not too bad then. I'm use to 2 phase bipolar, so I understand how the polarity is swapped here and there. Is there some kind of feedback, such as an encoder, ALWAYS on a brushless? I'm just fine with using this monster as a stepper motor, which it seems to do just fine. Any pointers on doing just that? I imagine the propeller control aspect should be a snap if my thinking is correct.
Not all brush-less motors have an encoder. Most these days use a feedback signal for zero cross (back emf).ˇ This reduces the amount of parts in the motors and makes them cheaper.

If you want to use it as a stepper, I don't see why you couldn't just control it as such. It should work just fine as a stepper. I'm not sure what the angle would be per step (depends on the motor internal design), but I think it would work fine.

I figure you could control with just standard half bridge type setup......

James L

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
James L

Partner/Designer
Lil Brother LLC (SMT Assembly Services) (http://www.lilbro.mosquito.net.nz/index.html)

Graham Stabler
05-09-2008, 03:08 PM
Phil, you can think of it as a stepper motor if you want but it only has 6 steps so you would be better considering it as a 3ph servo motor. It is designed to spin quickly so control needs feedback via back emf, hall sensors or encoder as already mentioned.

Unless you have a use for driving the motor I would use it as a paper weight.

Graham

ErNa
05-09-2008, 04:06 PM
The main difference to a stepper is what is called "cogging". A stepper is designed to have as much cogging as possible, we call it "step". A brushless dc motor is designed to show a low cogging effect, but it can not be completely omitted. To drive the motor, current has to flow e.g. from Phase A to B. The motor will do one step (60° electrical, may be 30° or 20° mechanical, depending on number off pole pairs. Then you have to commutate, that is, Phase B is switched off, but current flows now to Phase C. And so on, next Current flows out of C an in to B. Next: in to B, out A. Now this is invers current direction to step 1. After six commutations, the cycle is closed and starts over. The moment of commutation is very critical: if switched to early, the motor generates no driving force and stops immediately. If to late, the motor generates no EMF and the current increases very fast! Really very fast. With hall sensors it is no problem to commute, just switch the moment the hall pattern changes. sensorless control is a difficult task, especially if you want to start from zero and with load. Old CD-players have such motors, normaly equipped with hall sensors, so the are a cheap, ideal source to build a test environment.

Philldapill
05-09-2008, 08:28 PM
Wow, 6 steps per revolution??? That doesn't sound like this one... After switching the phases manually, it seems to have at least 7.5 degrees/step, maybe even 1.8.

Regardless, this thing is a 6Nm motor of some kind and it sure does produce that much torque. If I can just control it, I can sure get alot of use from it.