Hi
I'm using the following circuit to charge a single LIPO cell. VIN is 5V. The Mosfet is there to stop the load drawing current while the battery is charging but I don't think it's relevant. My problem is that the LED does not light. The STAT pin remains at about 4V, unless it's disconnected, in which case it's at about 1V. Even when the battery is fully charged to 4.2V the STAT pin does not change. The battery does however charge as expected.

Weirdly, I had this exact circuit made up by a company in China a while back, and the modules from them do work. If I swap round the MCP73831 between the two sets of boards, the China module no longer works and the one I'm making does. It seems unlikely but the only thing I can think is that my batch of chips is different from the ones the Chinese people used, and that mine are faulty. Any other ideas?

 

Did you check the easy stuff? Is the LED installed in the correct direction?

 

Thanks. Yes, this is shown by the fact that when I swap the chips round, the LED that used to light no longer does. and the one that didn't, now does. Everything I can think of - eg a design problem with the new PCB - seems to be ruled out by the result of switching the chips round.

 

The programming resistor, R3, is different between the two circuits shown. That might matter. It is also possible that the two ICs are quite different. I have observed that with some logic IC devices. DIFFERENT BRANDS WORK MUCH DIFFERENTLY EVEN THOUGH THE MODEL NUMBERS ARE THE SAME. So it may be that you need to change brands of ICs.
It could also be that you have one IC that has failed partially. That also happens.

 

Yes the programming resistor is different because I'm charging different capacity cells. I've tried two of my current batch of ICs with the same result so ... yes I think it might be a brand issue.