JK BMS: 100A vs 200A

[RCinFLA]

This is a pretty flawed test, but it does have some merit. I coated the mosfet base and thermistor with whiteout and set my emissivity to .95. I scrapped the whiteout off a little bit where soldering iron made contact with the mosfet base. I left the soldering iron on for a while, then removed and took some pics with the thermal camera.

In the thermal camera view you can see about a 4 deg F difference between the thermistor location (white pointer) and mosfet base (red pointer).

There is a lot of copper here (heavy trace, high current jumpers, etc). I think it helps to give the thermistor a fighting chance of getting a somewhat representative reading. Obviously this 4 deg F difference will change with higher temps, but still gives us an idea.

I believe the default MOS overtemp setting on the JK BMS is 100C. Does it seem realistic that the mosfet base temp would be within an acceptable range at this setting?

View attachment 308210View attachment 308211View attachment 308212View attachment 308213View attachment 308214

Not sure what you are trying to accomplish by heating center MOSFET's with a soldering pencil.

More insight will be obtained by having variable current pass through BMS and having a multi-channel thermocouple meter with small thermocouples attached to metal tabs of MOSFET's, at least on two ends and middle MOSFET's will give their temps vs BMS current. I would not expect the temps at various MOSFET positions to be the same for several reasons.

Although fairly good, the thermal resistance between MOSFET die and its package metal pedestal is not zero and the die will be several degrees C hotter than package pedestal.

When I did this a couple of years ago, I got this result.

 

[bestconcreteblock]

Not sure what you are trying to accomplish by heating center MOSFET's with a soldering pencil.

I was just trying to get an idea of the difference in temp between the MOSFET metal tab and the built-in chip thermistor. Soldering iron simulated heat generated at the MOSFET.

Thanks for the chart. What I don't see in your test is the JK BMS reported "MOS temp"? I assume lower than your thermocouple readings, but how much?

I don't mean to beat this to death, I'm just having a hard time understanding the vast difference between your chart and my discharge tests. In the first test I posted earlier, I am seeing a reported "MOS temp" of about 24 deg C @ 115-135A. Test was about 26 min. Ambient was about 15 deg C.

I understand the JK BMS chip thermistor is not giving an actual MOSFET temp, but I am having trouble understanding how it could be 75+ deg C lower than the MOSFET metal tab? That is what I was trying to show in my test.

I'm wondering if it has something to do with me re-soldering the 7AWG silicone pigtails. If you look closely at a new JK BMS, they are using insufficient heat when soldering these wires to the board/copper current spreader bars. Only the outer strands of the wire is actually soldered. The center is loose and you can see it move when you twist the wire. I made a special "super duty" soldering iron using a large chunk of copper heated by a MAPP gas torch. With this and flux, I can get enough heat to fully solder these fine strand wires. Not sure if the heat from this poorly soldered connection would be enough to explain the discrepancy though...

 

[robbob2112]

For soldering stuff like that you need something like a 100w iron like is used for stain glass.... you need the ability to deliver high enough heat to raise the temperature of the wire, solder and spreader bar quickly.... do it slow and you risk lifting the pad or damaging the surrounding bits... and use thermal gel around it keep the temp on the rest of the board from getting to hot.

 

[bestconcreteblock]

For soldering stuff like that you need something like a 100w iron like is used for stain glass.... you need the ability to deliver high enough heat to raise the temperature of the wire, solder and spreader bar quickly.... do it slow and you risk lifting the pad or damaging the surrounding bits... and use thermal gel around it keep the temp on the rest of the board from getting to hot.

Yeah it takes a very large and hot copper tip to apply heat quickly in the area needed. This is an especially difficult soldering job because there is a thick spreader bar on both sides of the board... I don't have any experience with thermal gel, but it sounds like a good idea.