A4984 Steppers

Hey Folks,

I've been banging my head against a wall for a while. I've been trying to build my own board using the A4984 stepper. I have essentially built the same circuit as the application example in the datasheet.

I've made some nice little boards that allow me to connect power, micro, and stepper to the A4984. On my breadboard, I can get the steppers to work beautifully. The steppers run with 5V logic side and 9V supply for the motor.

However, when I plug the same board into my actual rig with a 24V power supply and 5V logic, it seems to burn out the driver. I stick my finger in after powering down and find that the chip is quite warm. This is without actually sending any actual steps; just powering up the driver on the logic and motor sides.

Afterwards, I bring it back to my breadboard and it doesnt work, burnt out. I check the Vref voltage that I had set with a voltage divider to be 2V and it now reads 0.68V, indicating that the Vref pin is now sucking back current. If I check many of the other pins such as the step, dir, enable, sleep pins they are all around the same level 0.68-0.71V. The Vreg pin which is supposed to be ~7V normally, is ~0.1V. The way it looks to me, something has damaged the protection zener diodes in the chip on these pins and they are now essentially a forward biased diode to ground.

I've been trying to rack my brain to figure out what might be causing this. The 24V supply should be well within the 35V range of the driver, and it should not be a thermal problem because the chip is supposed to contain thermal protection that shuts down the chip when the temperature rises too high.

The only thing I can think of that may be a problem is the exposed pad beneath the chip. This pad is supposed to be grounded, and indeed in my design it is. However, since I am soldering this up by hand, it may be a poor solder connection. The way I soldered the pad was to first tin both the pad on the chip and the pad on the PCB. Next, I fluxed each of them with a flux pen and then positioned the chip on the PCB with tweezers. Next, I take my iron and put it on the hottest setting (~800F) and touch the tip to the exposed copper right next to the chip. I wait until I can see that the solder has liquified about half an inch from the tip on the board (indicating to me that the solder under the chip should too be liquid). Next, I repeat this from the opposite side. Finally, I check that the chip is soldered down by the pad by attempting to wiggle it to break the solder connection, but it seems good.

It may be possible that the problem is the poor solder between the pad and ground which may be "good enough" for my low voltage test, but not good enough for the high voltage test. My next step that I think I may try is to dremel out a small hole in the board under the IC which I can connect with a physical wire to ground.

Please let me know if anybody here has any experience with this chip and might be able to help me figure out what's going on.

Cheers.
Mark

Maybe 800F is frying the chip. It is way above max re-flow temperature.

---

[www.hydraraptor.blogspot.com]

It is possible, but it is unlikely the chip itself is getting that hot. Also, it doesn't explain why it would work fine at the lower voltage on my breadboard.

[edit]
I just figured out the problem. On my setup with the 24V supply, the board I am connecting it to has a 12V and a 5V regulator. I mistakingly connected it to the 12V instead of the 5V, thus frying it. I feel like an idiot lol, but hopefully I just need to change the chip once more and it should work.

Edited 1 time(s). Last edit at 10/28/2012 11:41AM by stellarcore.