Alt-BEAM Archive

Message #03092



To: "'Wouter Brok'" w.j.m.brok@stud.tue.nl
From: Wilf Rigter Wilf.Rigter@powertech.bc.ca
Date: Sat, 8 May 1999 12:37:59 -0700
Subject: [alt-beam] Re: non-block-wave signal for steppermotor



Good point Wouter,

Most commercial steppers controllers use some kind of bi-level current
drive. That is usually done by first applying a two step pulse. During the
first part the current must build up in the winding and the rotor/load must
be accelerated and this current is sourced from a high voltage supply with
a series resistor or current limiting linear driver or with a current
limiting PWM driver. Then after the first part of the pulse this is followed
with a drop to a lower current source for constant velocity and sometimes
followed with even lower current for holding current. The reason for using a
high voltage (compliant) current source is to build flux quickly in the
inductive windings. If this is not done then the high rotor inertia would
severely limit the stepping rate. The voltage across the coils would be a
double/triple slope dropping waveform and the current waveform waveform is
more complex first rising, then flat and then dropping and then flat. Both
cases are limited by the effect of the winding resistance. I believe that
for steady state conditions the most efficient waveform for a bipolar
stepper would be two quadrature sinewaves. Afaik this is sometimes referred
to as synchronous motor operation. To keep things simple, for low stepping
rates, you can use short duration pulses which terminate when the rotor has
moved to the next position. Since it generally take a lot less current to
hold position compared to moving the rotor and load, then for low load
(holding) conditions (ie a head) you can remove all current (ie tristate)
and rely on the permanent magnets to provide sufficient detent to hold the
position or else use low current sources (of the right polarity) to maintain
position of the rotor between steps.

Here is a recent email to Bram van Zoelen on this subject:

Hi Bram,

If you like to use these 4 wire steppers for a true Beam application you can
use a reversing master-slave BiCore to drive them. These of the steppers
have 2 separate independent windings and you can connect one winding to the
master and the other winding to the slave. The stepper will continue to
rotate in one direction one step at a time and will rotate in the reverse
direction when you reverse the slave outputs. The speed of rotation is
controlled with the frequency of the BiCore oscillation usually 36 steps for
one rotation and you should change the capacitor values to 0.01 for
reasonable speed, say 60 RPM. If you want continuous adjustment of the
speed you can adjust both BiCore resistor values together. Anyway that is a
start and probably easy for you to try out if you have some exiting BiCore
circuits hanging around.

regards


Wilf Rigter mailto:wilf.rigter@powertech.bc.ca
tel: (604)590-7493
fax: (604)590-3411

> -----Original Message-----
> From: Wouter Brok [SMTP:w.j.m.brok@stud.tue.nl]
> Sent: Saturday, May 08, 1999 11:08 AM
> To: beam@corp.sgi.com
> Subject: non-block-wave signal for steppermotor
>
> Hello,
>
> just an idea I was thinking about:
>
> In general stepper motors are driven with block-form waveforms at the
> inputs. I don't think this is favorable from the energetic point of view;
> what I mean is this: When one input of a bipolar stepper-motor is made
> high
> (suppose this is the driving voltage level, so all the others are low) the
> motor-axis will turn in the appropriate direction because it is attracted
> by the powered coil. At a certain point however the axis is at the
> position
> from where on the powered coil will slow the axis down. At that point the
> input which was high should be made low and the next input should be made
> high as to power the next coil so that the axis is attracted to the next
> position. However, I do think it doesn't happen this way generally: the
> input will be made low just to late, so the motor-axis will be slowed
> down.
> This results in a waste of energy.
> My thought was to feed the steppermotor with another waveform to increase
> its efficiency, but what is this waveform? another possibility that came
> up
> thinking about it is: the motor will add an induction-voltage on top of
> the
> driving voltage and maybe this signal can be used to make an addaptive
> motor-driving-circuit. The motordriver could 'decide' from some
> characteristic of the induction-voltage if the driving signal should be
> altered: feedback; the driving circuit can create the best waveform
> itself.
> My guess is that this will result in a more efficient steppermotor-motion
> and that the angular velocity of the motor can be higher as possible with
> a
> normal block-wave; the motor will work more smoothly.
>
> Any comments on this thought?
>
> Wouter Brok.
>

------------------------------------------------------------------------
eGroup home: http://www.eGroups.com/group/alt-beam
http://www.eGroups.com - Simplifying group communications

Home