A4983 w/1.5A peak rails power supply

I am using an A4983 with a 0.6V setting for current limit. The chip has heat sink with silver bonding and the chip itself never fails. The problem is, as I increase the peak held load on the stepper motor, it reaches a point where the current draw on the power supply (24V, 175W) exceeds the maximum and the power supply goes into hiccup mode. In the on-period of hiccup mode the motor runs normally. In the off-mode it doesn’t (as one would expect). The maximum current rating of the power supply should be 7.3A. The current limit on the A4983 is 1.5A. Why would the A4983 appear, essentially, as a short circuit to the power supply at any time?


I don’t know the details of your power supply and why it might be failing, but it might help to put a decent capacitor (at least a few hundred uF) across the motor driver supply. Also, have you verified separately that the power supply can actually deliver something like 5A without shutting off?

- Jan

The power supply is an XP CLC175 24V http://www.xppower.com/pdfs/SF_CLC175.pdf. It can source 7A without issues. As for a capacitor across the outputs, wouldn’t that be slightly redundant given the output caps for the power supply itself?

There is a well-known bi-polar stepper driver problem when the output DMOS is driven using two-phase chopping. It is described relatively well here: http://notes-application.abcelectronique.com/005/5-10419.pdf (pages 6-7). Essentially, this issue will cause the output current to runaway under specific circumstances. Do you know if the A4983 uses a two-phase chopping setup?

The cap would not necessarily be redundant; I don’t know what’s on the supply and how it decides when there’s a problem. If you put on the cap at the motor driver, it will smooth out what the supply has to deliver.

I don’t think the A4983 is going to have the trouble you’re concerned about since it has fairly sophisticated current control on there. The datasheet covers what they do in quite a bit of detail; the “fast decay” they talk about, with the synchronous rectification, is equivalent to the two-phase chopping in your app note, but it’s supposed to be done in a safe way (only limited time, not near 0 current, etc.).

You can be pretty sure the issue is with your power supply: if more than a few amps were actually coming out of it and through the motor driver, the driver would be toast.

- Jan

I’ll try the addition of the cap and, barring that solution, try a bigger power supply. Thanks for the help.

I don’t think you need a bigger supply; I just don’t trust that the one you have can output what it’s supposed to. Have you separately confirmed that you can get several amps out of it?

- Jan