Stepper Motor Question

A stepper motor is a synchronous motor. If you are careful about accelerating the rotor, you can make it spin at any speed short of the point where the motor will fly apart, though you may need to apply hundreds of volts to the windings to get enough current through them to generate useful torque.

Many years ago (1978) I was apply 60V to the coils of a "5V" motor to get it to spin at some ten times the natural resonance frequency of the rotor (determined by the moment of interia of the rotor and the stiffness of the magnetic field holding the rotor at a static step position and typically in the region of 100Hz for small stepper motors).

Getting the rotor up to the speed required programmed acceleration, starting with a step of a quarter of the resonant period (roughly

2.5msec), and followed by steps that were progressively closer together, each one injecting progessively less extra rotational energy into the rotor.

We had total control over the load which was the scanning carriage in a mechanical ultrasound scanner. A more open system would have been trickier.

-- Bill Sloman, Nijmegen

=A0 =A0 ...Jim Thompson

Here's a data point: 1,000 rpm is pretty fast for most NEMA-23 sized hybrid stepper applications. 10,000 rpm is really screaming, with zip torque left. For a 200-step motor, 10,000 rpm takes 2M steps per minute, a 33KHz step rate.

-- Cheers, James Arthur

Depends on the maximum dissipation. I did manage to get a big stepper (2.5A) to about 6000 steps/s (1800 rpm) with a single ended driver. But it was necessary to ramp the speed up & down. Nowadays I think I would use a servo motor with a position encoder or at least drive the steppers with a full bridge PWM and feed the motor sine waves unless halted at some point. The biggest problem with steppers is that you can miss a step and you have no way of knowing unless you add the encoder.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

OK. I'd guess my thoughts of adding 100ns safety against H-bridge crowbar-overlap wouldn't be a speed-limiting idea :-)

(This is an RFQ with less than marvelous explanations :-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |

               I can see November from my house :-)

Just a passing thought but seems to me it should be theoretically possible, at least, to detect step movement by the inactive coil's back EMF.

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Depends on the motor, inertial load, damping, type of drive...

I work mostly with size 17, 1.8 degree/full step motors. We designed our own microstep drive and typically run at 30 rev/sec. A lot depends on ramp rate, start/stop velocity and avoidance of any speed where resonance can become a problem. Most 1.8 degree steppers experience the worst resonance at about 1 revolution/second.

What do you have in mind?

Oppie

My 3-phase motor driver for Bosch/Mercedes did exactly that... and used it for commutation. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |

               I can see November from my house :-)

Well, that's considerably past being 'theory', then ;)

It's commonly done for 3-phase brushless PM motors. I don't know if it's been done much for steppers, although I'm sure that _someone's_ tried it.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

Do you want normal max or far-out-craxzy max?

With a big of googling, I'f found 55,000 RPM and 1.8 degree steps. That's 200 steps per revolution. Round 55 to 60 and we have 1000 RPS. At 200 steps per revolution, that's 200 KHz.

Round up some if you want to be conservative. Round down some if you think I'm crazy.

[I've never played with stepper motors.]

--
These are my opinions, not necessarily my employer's.  I hate spam.

That will only work at moderate stepping rates. And consider the situation while microstepping.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

It's a well-known technique, but not all that reliable. I think it has been patented, but long enough ago that the patent has probably expired. IIRR Inland Motor - now part of Kollmorgen - owned the patent.

but I can't find any reference to it on the Kollmorgen web-site.

-- Bill Sloman, Nijmegen

In fact you are trying to separate the back-emf being generated in the winding from the the emf that your drive current is generating across the winding - the resistive, the inductive and the capacitative components of the winding impedance. Since the inductance of the coil changes as it moves through a step, working out the voltages that you are generating is tricky, and your confidence in the back emf that you finally extract isn't always overwhelming.

-- Bill Sloman, Nijmegen

YEs, it is possible, I've done it working with eddy current type brakes/clutches..

When the unit is not in slip, induction of the coils are a constant how ever, when they slip, you get an effective induction change which can be detected..

The same basic theories hold with other types of devices on the same line, DC motors, steppers etc..

We were able to create a slip detector on a machine that employs a tape dispense that wraps around the shield of cable as you pass it through the center of the mount that holds the tape pad. The pad is mounted on a disc that has a eddy current brake for tension control and gets it's current via a set of slip rings (collector rings for some people). With 2 wires as the supply we were able to detect slip (movement) using a pulsed power supplied and detector circuit on this pulse.. This was to detect if the tape broke/ran out in the process and stop the machine.

Since these tapes are clear and on moving mast, it's kind of hard otherwise.

Hey! It's a topic for Another Useless Doctorate, then!

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

Do you think he can handle two, in one lifetime?

--
Politicians should only get paid if the budget is balanced, and there is
enough left over to pay them.

hase?

per

.

nk I

the

ss

u e u

Easy. Finding a supervisor and funding would be the tricky part.

-- Bill Sloman, Nijmegen

Slowman's understanding of back emf is under-whelming :-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |

               I can see November from my house :-)

What do you expect? The only thing he learned about it was when he pissed on an electric fence, with a mechanical fence charger.

--
Politicians should only get paid if the budget is balanced, and there is
enough left over to pay them.