Converter In-Rush Thermistors etc, UPDATE -4

Salvo wrote:
I don't believe the thermister was at fault, nor do I believe the cause of the converter failure is due to not being connected to the battery.

The thermister blew because one of the Al capacitors shorted. The cap shorted because of poor quality parts as well as a bad design (I'm speculating). The converter is under greatest stress when outputting 100A. At 100A, the Al caps see the greatest ripple voltage and ripple current. The combination of the capacitor's ESR (internal resistance) and ripple current heat the capacitors. They get hot and the electrolyte dries out. At some point in time they will short.

What's the rated voltage stamped on those caps, 200V or 250V? If they are rated 200V, then there's little margin since the working voltage is 170V. I worked in very high reliability electronic designs. We always derated capacitor voltage by 50%. If working voltage is 170V, then cap voltage must be greater than 2 * 170V = 340V.

I also suspect ripple current is too high because there's not enough capacitance. The rule of thumb for the front end caps is greater than 1.5 uF/Watt. This converter has a 1600W rating. That means the capacitance should be greater than:

C = 1.5uF/Watt * 1600Watt = 2400 uF

Take a look inside to verify cap voltage rating and capacitance.

Sal

the code on the "original" thermistor is SCK 202R5 it is a little smaller than a 25 cent piece but thicker. I found that on the internet somewhere with the company name that makes them and the specs for that thermistor, but don't have that now. I must have Googled it.

The "beefier" thermistor (I expect it is standard in the new PM4 series that last year replaced this PM3 version I have) is bigger than a 25 cent piece and much fatter. Code SL 5R020

The capacitors if they are the ones in question, are two big black ones the size of flashlight batteries, marked Nippon Chemi-Con SMM 85C 250V 1200 uF(M)