Re: Digest Number 1042

On the stepper topic, manufacturers like to
rate the HOLDING torque, as it sounds the highest.
They also rate it at 2 phase on, both phases full
rated current. This gives them the biggest figure.

For microstepping, you work with the one-phase-on
current, as this is your worst case torque.
Then all the microsteps are current adjusted
to give this same torque value. A novice might
jump to think "reduce torque?!" but once the motor
is turning things behave very differently and
you get a lot more torque at speed.
You DO want the motor to turn? ;o)

A turning motor has a rotating magnetic field.
In full stepping you are feeding it with square
wave shaped current. Very inefficient. There
are large losses in flyback needing big diodes
which get hot. Part of the energy you are
feeding in is coming back out to "fight" the
motor if that simple analogy makes sense.
And worse, this fighting has a resonant effect
where at some speeds it's not too bad, and
at other speeds you will get 100% torque loss.

Once you run microstepping of 8 or more steps
per full step, the motor is getting a pretty
good sine wave. (4TH/n)x{1-cos(fe)} Excitation
theory shows that there is 50 times less
excitation energy in 8th stepping than in full
stepping, in laymans terms means there is about
50 times less noise, vibration, "fighting" etc
etc. Much more than a 8 times effect.

And it's very nice to be able to decelerate
your motor to stop at any one of 1600 positions.

I like linear microstepping best, this is not
common but works the nicest, especially if
current slewing is used. I have old motors
here that are 34mH (very slow!) and using linear
8th stepping I can rev them from stopped to
30rps (1800rpm) in about a second. With full
stepping I would be pushing to get 4 rps. With
traditional chopper 8th stepping I might get
15 rps, and with a lot of heat loss. :o)
-Roman

> > One thing that I hear from time to time is that the holding torque

of

> > some of these in-between steps is quite a bit less than in other
> > steps, resulting in the increased likelyhood of missed steps. From

> > what you've just explained, it seems to me that there isn't that

much

> > of a decrease... Am I correct in assuming that?
>
> Well, there's in between and there's in between.
>
> Some of the early half-step controllers ( and perhaps some of the
> current controllers as well ) did not increase the current in the

"one

> winding on" state, and under these circumstances, there is about 1/2
> the torque available every second step.
>
> Until the advent of "chopping controllers", it was a bit
> more difficult to control the current in a stepper. SOME means of
> current control was required, if only to prevent the motor from
> melting down. Most of the manufacturers used a simple resistor for
> this task, thereby supplying the same current under all

conditions.

>
> Quarter step and finer controllers by their very nature require
> adjustable current control, and once some sort of current control
> electronics are in place, it is not too hard at all to pick the

CORRECT

> current values.
>
> It is fundamental in the math of all but the four half step

positions

> that the winding currents required to get to a given position are

also

> the currents required for equal torque - or so the theory implies.
>
> Given controllers supplying appropriate currents, I would not expect

any

> substantial change in torque from one step to the next.
>
> There is a "however" to all of this, though.
>
> Not every chopping controller ( which is to say, just about every

modern

> controller ) will work optimally with every motor. There is an
> interaction between motor inductance and chopping frequency. It is
> possible to use a motor of too high or too low an inductance for the
> chopping frequency. In these cases, the current the controller

"thinks"

> it is supplying may not be what is actually being supplied. Under

these

> conditions, steps could be missed.
>
> Unipolar steppers run in bipolar mode are somewhat prone to this
> phenomena, as they typically have 4 times the inductance of an
> equivalently sized bipolar motor. This is somewhat solveable by

lowering

> the chopping frequency. The lower frequency can result in the motor
> overheating, so a decrease in the peak current supplied is

adviseable.

>
> Alan
>
>
> --
>
> Alan Rothenbush | The Spartans do not ask the number

of the

> Academic Computing Services | enemy, only where they are.
> Simon Fraser University |
> Burnaby, B.C., Canada | Agix of

Sparta