Subject: L297/L298 Bipolar chopper stepper drive problem
Date: Mon, 28 Oct 2002 16:59:53 +0000 (UTC)
Organization: BT Openworld
NNTP-Posting-Date: Mon, 28 Oct 2002 16:59:53 +0000 (UTC)
X-Newsreader: Microsoft Outlook Express 5.50.4807.1700
We are using the MAE stepper motor( 6 lead HY200-1713-040-B6 0.4A/phase,
15.6R phase resistance) for development, previously we were using a unipolar
driver, namely a UCN5804B device from Allegro, but with the documented lack
of power efficiency, we decided to develop a bipolar drive. I used the
L297/L298 combination, as I felt this would give me a better grounding in
how the bipolar chopper functioned, layout requirements, power dissipation
that sort of thing.
The issue...when the motor is in a static position, a holding torque
applied, the motor is starting to overheat, when the motor is moving
(typically at 600 steps/second) the motor behaves as it should and does not
exhibit any overheating. How can I prevent the overheating in the static
position, is there something I'm missing (likely)?
We are using a 15V power supply for the motor. The controller from start is
always enabled, we're in half step mode using Enable Chopping (not phase
chopping). For the current sense, a pair of 0R5 resistors are employed with
a reference voltage for the comparator set at >0.2V (max phase current =
0.4A). The chopping frequency is about 20KHz.
At the static position, if I decrease the Vref to near or lower than the
sensed voltage (current), then the Vref and current sense inputs become
'noisy' and although the motor maintains a holding torque, albeit with some
vibration, the phases are now being turned on and off. The datasheets talk
about chopper stability and that if spikes on the sensing inputs exceed the
Vref, then the comparator can be fooled into resetting the flip-flop; it
describes including an RC on the sense input to the comparator or to use the
inherent set dominance of the flip-flop to mask out the spike.
I've tried increasing the chopper frequency and including an RC at the
inputs to the L297 from the sense resistors, without any noticeable effect.
I hope someone can help.