[Rypper]

Recently I upgraded the controller on my 4 x 8 3 axis cnc machine, my stepper motors were unipolar on my old controller.

But my new controller running Mach3 and Gecko 201 drivers required me to convert my unipolar to bipolar ( 4 vs 8 wires). No problem since my Vexta PK266-02A motors could run in both, after the conversion everything was working fine for about two weeks but now the x-axis is struggling and occasionally stalling.

My questions are, by doing this conversion did I lose torque and since these were older (about eight years old) and smaller steppers to begin with is it possible that they are failing under the stress.

Thanks,
Tom

[project5k]

in my experience bipolar makes more torque than unipolar.. my first question would be is the drive current the same? and my second is are you running in full step, 1/2 or what? also, you need to know if you are running full step, are you running 1 coil on at a time, or 2?

as for your existing motors.... a stepper will live forever, well almost, the bearings/bushings would be the first to fail, assuming that your not cooking them with too much current... so long as your not cooking them, and the bearings/bushings are good, then theres really nothing else that can fail.. its just a bunch of wires wound up, and some magnets attached to the shaft..... i suppose there is a remote possibility that the magnets have lost some of thier pull, but your conversion should have no effect on that one way or the other.... i would check your step mode, and how many coils your firing at a time... simplest fix to start with...

[pminmo]

Practically you shouldn't have an issue running as unipolar for awhile then bipolar. Something has probably gone wrong. Most common place to start is your wiring, check your wiring connections that they are all tight, and that you didn't partially damage a wire. I assume you set the G201's to the correct current? Which is?

Is it a 4, 6 or 8 wire motor?

[candle]

well
if you have universal (8 wire) motors then you two ways of wireing them to bipolar the right way, and one the wrong :]
you may put two coils of each winding in series for more torque on higher rpms or paralelly to get more on low speeds
the other possibility is you connected just 4 wires, and left 4 dangling
and running in this configuration with unchanged current limit will easly run you into problem with overheated rotor and partially demagnetization of it
thus - poor torque

[Madclicker]

well
if you have universal (8 wire) motors then you two ways of wireing them to bipolar the right way, and one the wrong :]
you may put two coils of each winding in series for more torque on higher rpms or paralelly to get more on low speeds
the other possibility is you connected just 4 wires, and left 4 dangling
and running in this configuration with unchanged current limit will easly run you into problem with overheated rotor and partially demagnetization of it
thus - poor torque

Here's the way I understand possible 8 wire connections:

Low end torque is the same for parallel or series.

High end torque falls off faster for series connected because the inductance is much higher (4 times?) than parallel connected coils.

I don't think connecting just one coil of each side is necessarily 'wrong'. It depends on the application...ie available current and voltage.

Voltage and current values will be different for each case.

Correct me if I'm wrong. Wouldn't be the first time.

[candle]

low end toruqe is lower for parrallel, but its more constant on higher rpm
inductance is 2 times lower with parrallel, but there is close corelation between torque and inductance - the more inductance winding have, the more stall torque motor is able to produce
im not telling that one coil is utterly wrong and the one who tryies should be hang, but it is a bit pointless (why not to start with smaller steppers at all?)

[Madclicker]

I stand by my original statements.

If inductance in 1 coil = L

Inductance in 2 coils in series = 2L

Inductance in 2 coils parallel = L/2

Therefore series inductance = 4 X parallel inductance.

Simple circuits. Resistance is figured the same way.

The low end torque is the same for series and parallel connected because when connected parallel you are allowed twice the current.

Take a look at this data sheet for real numbers for real motors and you will see that the numbers are right. Just pick an 8 wire motor and follow across.

Superior stepper motor data sheet

[pminmo]

Madclicker you are correct inductance will be 4X or 1/4th between the two wiring configurations. Torque is the same bipolar series or bipolar parallel as long as the current is correct for both wiring schemes. The speed curve will be much better for the bipolar parallel because of the lower inductance.

[project5k]

um, im not nit picking, just making sure i understand, speed curve???? didnt you mean torque curve?

[Madclicker]

Yep, that would be nit picking. Technically it's called the "performance envelope".

[pminmo]

Speed vs. torque.......... Torque on one axis, speed on the other. I should have been more explicit........

[project5k]

performance envelope huh??? well lets just get all technical then

[Rypper]

Thanks for all the replies, checked all the wiring everything was fine. Then as a last ditch effort I opened up my new controller and the wiring was reversed on the A and b coils to all the motors. The puzzling thing is that the y ans z axis work (at least by all appearances) fine with the wires the way they were. Only the X axis showed problems.

Thanks again,
Tom

[candle]

madclicker: you're right, i didn't noticed you're speaking about this in comprasion to parallel topology inductance, thus mine .5 rather than .25

[project5k]

wait a min, if the wiring was reversed, wouldnt that invert the axis? making +x be -x and so on? it would be moving in the wrong direction, just like the kids in the movie the explorers.. they reversed thier x axis, and instead of going straight up, the computer was still thinking they were going up, but in reality they were going down?!?!?!?!?

[bunalmis]

I stand by my original statements.

If inductance in 1 coil = L

Inductance in 2 coils in series = 2L

Inductance in 2 coils parallel = L/2

Dont forget the magnetic couplage.

If inductance in 1 coil = L

Inductance in 2 coils in series = 4L

Inductance in 2 coils parallel = L

How?

L=(N^2)K

If one inductance has N wire (spin) two series inductance has 2N wire (spin).

Ls=((2N)^2)K=4(N^2)K=4L

-----------------

Total inductance of parallel connected coils equal to one coil inductance.
Because coils use same magnetic core.

L=(N^2)K, N not change at the parallel coils.

Lp=L

Also zero inductance possible at the serial/parallel connection.

[Madclicker]

I don't think the coils use the same core in a stepper, but I'll defer to a motor person. Your "proof" isn't supported by the data sheets.

[bunalmis]

I am true also i test it by LCR meter.

[candle]

well
all coils use the same core, because the stator is constructed that way (many sheets o metal), and zero inductance is easly accomplished by winding up one coil in one direction, and windig the other one in opposite - this is easy having 8 wire motor - just swap connections of one half in parrallel mode

[Madclicker]

So, Polish and Turkish motors are wound differently than the rest?

Would someone that REALLY knows how these motors are made please clear this up before there is hopeless confusion.

All I know is the data sheets I have don't support all this hand waving and psuedo proofs.