JK-BMS : Temporary MOSFET Temperature Increase on JK BMS Without Load – Normal Behavior or Hidden Issue?

[Kta22]

Hello everyone,


I’ve noticed an interesting behavior with my JK BMS:

• The MOSFET temperature briefly rose to around 5°C, even though there was no loadconnected, and then it dropped back to normal ambient temperature (~5°C).
• There were no other anomalies: current was zero, and cell voltages were stable.
• This is happening time to time(couple of weeks).

My questions are:

• Has anyone else observed this behavior on JK BMS units?
• Is it possible that the BMS periodically tests the MOSFETs internally, even without load, to check their functionality?
• Or is this just a normal fluctuation caused by internal electrical noise or sensor measurement quirks?

P.S. graph - at 9PM started to charge my car with 7.2kWh

Any experiences, technical insights, or official information would be highly appreciated!


Thanks in advance!

Attachments​

• 
  jk bms MOS temperatue .png
  27.3 KB · Views: 2

 

[amberwolf]

How are you measuring the temperature? What specific test equipment, leads, how is the sensor fastened to the FETs, how is it logged, etc?

Unless they have special circuitry and extra switches in there to block the FETs from being able to pass charge current into the battery or discharge current out of the battery during any kind of internal auto-test (which would mean having additonal sets of FETs in series with the existing ones, making it twice as expensive and half as efficient), there isn't any way for the BMS to test the FETs by passing current thru them, unless you have a load on the discharge side (or a current source on the charge side) that it knows what it should be, so it knows what the results shoudl be and can compare them to actual test results.

If it just turned on the charge or discharge FETs to "test" them, it would be powering up your vehicle (or whatever) by connecting the battery to it when it shouldn't be (if the FETs are turned off at test time), or charging the battery when it shouldn't be (if the charge FETs were turned off because the battery was full, for instance, or if a cell was found problematic, etc).


However, if you are testing the discharge FET temperature when your vehicle / etc is turned off, but the vehicle turns itself on periodically to do its own internal whatever, *then* there will be current draw thru the FETs, and a current rise from the energy wasted within the FETs from internal resistance.

A 5C rise is fairly hefty for a "nothing" load, unless the BMS is enclosed in something where it cannot shed heat, in which case you could expect it to fail from normal usage that would cause higher current and much higher temperature rises.


So I would recommend setting up a logging wattmeter setup on the current path of the FETs you are temperature testing, so that you can see, at the same time as the logging of temperature, preferablly all wihtin the same logging device, what current is flowing to cause this temperature rise.

Then you can troulbeshoot why current is flowing when it shouldn't be, and fix that problem.

 

[Kta22]

With internal monitor MOS temperature sensor and Home Assistant. At that time - not balancing, no current consumption reported by HA or VRM and no variation in cell voltages (dropping). Afterwards everything going back to normal (I have 4 battery packs that I can data compare with). I can monitor each battery individually and there were no sharing currents between banks.

 

[Kta22]

Well guys, same hour but different day, mystery finally solved. Yesterday I was just checking remotely. Today I’m standing right next to the battery banks. And guess what… it’s exactly what it looks like...Verry funny.

 

[amberwolf]

Sometimes in science, direct observations are the only way to see the whole picture.




(where I work, we sometimes get someone with an aquarium that gets heavy algae only across one wierd-shaped streak near one end, and can't figure out why. I ask them if it's near a window and they say no...then one day they come back in to tell me that 'well, there's a window across the room but the sun only gets in when the cat's sitting in the curtain (or some similar thing happens) :lol: ).