Re: RE: stepper driver-transformer sizing question

Couple of comments:

1) Backdriving a stepper with its windings shorted is the reverse of
having a stepper delivering mechanical power to a load while being
driven by a drive. The motor is a bi-directional transducer that
converts electrical power into mechanical power and visa versa.

2) The motor does not "suck up lots of shaft power when you turn them
slowly"; the torque is high exactly because the RPM is low. A step
motor is a constant power trasducer. This means the product of torque
and RPM stays constant. As you turn it faster, the torque required to
turn it drops because the absorbed power (converted to electrical
power and dissipated in the winding resistance) stays constant.

3) You can try to measure the AC RMS input power to supply but that
is going to a lot of unecessary trouble. You can safely assume an
unregulated supply has a very high efficiency (>95%) and that
efficiency is relatively constant. Simply measure the DC amps from
the supply and multiply it by the DC voltage to get the Watts
supplied to the load. It will be a very close approximation of the AC
power drawn.

4) Assume the power a step motor can deliver to a load at a given
supply voltage is constant when the motor is operating past
the "knee" of its speed-torque curve. Safely assume the motor has a
certain conversion efficiency. The conclusion is the current drawn
from the supply will be independent of speed at a stall load (motor
delivers 100% of its available power just before stalling).

You can empirically test this hypothesis by stalling your motor at a
variety of speeds while noting the supply current at the moment of
stall. The current will be the same in every case past the knee speed.

Mariss



--- In CAD_CAM_EDM_DRO@yahoogroups.com, "Greg Jackson" <greg@t...>
wrote:

> I did not record the speed, but it was more a problem with the load

than

> with the speed/torque profile. I was using stepper motors with

shorted

> windings as the load. If you short the windings of a stepper motor

you will

> find they can suck up lots of shaft power when you turn them slowly

because

> they will pump lots of current through the internal resistance. If

you turn

> them fast you will get the equivalent of "missed steps" and they are
> actually easier to turn at higher speeds.
>
> The speed/torque profile of any stepper is flat until the current

starts to

> get limited by the inductance, and then it slopes down. Max

theoretical

> power is close to the knee. The speed/power profile ramps up from

the start

> but then turns down or goes flat at the knee. Of course power is

speed x

> torque. Before the knee, power is low because speed is low; after

the knee,

> power is low because available torque is low.
>
> I think that practical sizing of the power supply ends up much

smaller than

> prevailing wisdom because it is impractical to use steppers

anywhere near

> their power delivery capacity. Everybody knows that the torque

demands you

> place on a stepper should be a really long way from the holding

torque. The

> peak holding torque is really the failure point, since a missed

step is

> functionally a failure. If you run a system with torque demands

which are

> 50% below the holding torque, then your power supply will only need

50% of

> the theoretical power. Sizing a transformer like that would be

risky if the

> load of the system were viscous in nature, like stirring paint with

a paddle

> wheel, since the whole system would crash when the power demands

exceeded

> the delivery capacity of the transformer. However, the load of

most of our

> systems is usually not viscous, but more subject to short term

peaks from

> accelerations or impact forces. For those types of short term

loads the

> capacitor, not the transformer, are the key to carrying the system

through

> without loss of steps.
>
> In other words, size the transformer for actual RMS power but size

the

> capacitor for peak torque demands.
>
> It would be very interesting if some others would hang a clamp on

current

> probe on the AC input to their power supply transformer. I suspect

that

> most of them are way oversized. The factor that I do not know the

answer to

> is how screwed up the reading will be on an AC current meter given

the

> nature of current flow in a rectifier application. Current only

flows in

> the transformer at the points where the voltage exceeds the bus

voltage.

> This is a very nasty way to use a transformer, and a long way from

a sine

> wave that most cheap meters expect to see. It's a good excuse to

buy a true

> RMS meter.
>
> Greg
>
> -----Original Message-----
> From: Ray Henry [mailto:rehenry@u...]
> Sent: Thursday, January 01, 2004 11:29 AM
> To: CAD_CAM_EDM_DRO@yahoogroups.com
> Subject: [CAD_CAM_EDM_DRO] Re: RE: stepper driver-transformer sizing
> question
>
>
> I find this report fascinating because the result of the empirical,
> real-world tests seem to run counter to the prevailing wisdom. The

only

> thing that you did not mention was the rotational speed of the

stepper

> motors at what you saw as max AC draw.
>
> Others mention running the steppers as fast as possible and I think

that

> this is not the speed which would cause max current draw from a

power

> supply. Max velocity under load would be for servos but not for
> steppers.
>
> I had a discussion about some of this with Mariss a while back. My
> thinking was that you needed to factor the motor voltage into the

draw

> question as well.
>
> Ray