Help Designing a 18650 charging circuit - MCP73831

[MabumisHuge]

Hi there!

I recently started designing some pcbs for a project I'm working on that contains a circuit to charge 2 18650 batteries in parallel. However when my assembled PCBs arrived, the MCP73831 started to smoke and died. I'm fairly sure I followed the example schematic pretty closely, besides some capacitors which i assumed were for noise.

Below is a photo of the schematic as it was on the PCB, as well as the recommended layout from the datasheet. If someone can find a glaring cause for what I'm assuming was a short of the chip please let me know (or if the problem may be down stream).

Cheers 

[ataradov]

The datasheet has the whole section about capacitors and why they are necessary. Whether it would lead to smoking - hard to tell. It might go into some weird oscillation mode and die, for sure.

[Peabody]

Any chance you had the battery installed backwards?

[MabumisHuge]

The datasheet has the whole section about capacitors and why they are necessary. Whether it would lead to smoking - hard to tell. It might go into some weird oscillation mode and die, for sure.

Yeah I defs missed that part when reading initially, complacency on my part from assuming they were only for filtering. Upon reading datasheet does say MAY be necessary, however I would be quite surprised if this was the sole cause.

Any chance you had the battery installed backwards?

Battery is installed correctly, further down in the circuit the battery voltage is converted to 3.3V to power an esp32, which I was able to flash.

[ataradov]

however I would be quite surprised if this was the sole cause.

Well, LM1117 is infamous for its instability with incorrect capacitors, which leads to overheating. I don't think it overheats enough to smoke, but still.

It is hard to predict how this device may behave. Your best test is to try it in isolation without a battery or any other circuits attached.

[MabumisHuge]

Well, LM1117 is infamous for its instability with incorrect capacitors, which leads to overheating

Hmm maybe the capacitor is the cause then, downstream in the circuit i had a 22nf instead of the intended 4.7uf.
Being a factor of over 200 off on the value might have had an effect if they are temperamental 😅

Thanks for the advice, ill update the circuit and see how it goes