Stepper Driver 2.3 Nonresponsive

Summary:

An electrical engineering master's student can't figure out hobbyist-level driver kits; what is he doing wrong?

Stepper Driver 2.3 Nonresponsive

PAPPP 30 Apr 2009 02:58

I picked up three Stepper Motor Driver v2.3 Kits to use in a little benchtop CNC project I have going, and I just can't seem to get them to work.

I put them together in a proper fabrication lab (I'm an EE grad student, that sort of thing is readily available for use), including a nice rotary reflow oven.

When I powered each one (smoke test, no external components, limiter pot to minimum, fed from a known-good ATX supply with the sense wire bridged to ground), two had the power light come on as indicated, and one had every single LED come on shortly before the IC smoked. I'm guessing I bridged something in the last one, I haven't done a lot of surface mount in the past so it wouldn't be terribly surprising.

The two that appeared to work however will not energize the output phases; I removed all external uncertainty by feeding the step input (Pin3 on the header) with a low frequency (period » 1uS, like the datasheet says) square wave from a function generator, with enable (pin 5) and direction (pin 4) grounded back to some of the grounds on the header, and tried both a motor and a DMM on the output phases; absolutely no output to the phase terminals, regardless of the setting on the current limiter pot. I also checked the motor windings by manually advancing the shaft and watching the back EMF on a DMM to verify that the motors weren't the problem.

I've visually inspected and continuity probed all the likely paths for bad connections, and found a few: Pin4 on one board's A3982 was intermittent, I reheated the joint with an iron to fix, and the 100uF caps acted as nice big heat sinks and didn't "cook" fully: heated those joints back up with an iron as well, and all seems fine on that front now, but the boards are still totally non-responsive.

I've checked the .22uF and 1nF caps with a scope and resistor (no LCR meter on hand, so I just measured the time constants) to make sure they aren't swapped, they appear to be correct.

Any suggestions for a next debugging step? Suggestions for what I might be doing wrong? I'm beginning to fear the A3982s got fried somehow along the way.

Re: Stepper Driver 2.3 Nonresponsive

PAPPP 02 May 2009 03:48

Answering my own question: The leads on the A3982s were ever so slightly bent, so some of the connections on the chip were mechanically but not electrically sound. Probably should have put a little more solder paste in and it would have hidden the problem. It was hard to diagnose because the pressure from a probe usually pushed it in to place, but I eventually spotted the problem by visual inspection with a magnifying glass. After pressing down each lead with a hot iron two of three boards are working great. If anyone else gets totally nonresponsive boards that seem to power, that is definitely something to check on early in the debug process.