Erratic & Erroneous Cell Voltages Whilst Charging - JKBMS B2A24S15P

[BipedalPrimate]

JKBMS B2A24S15P

Readings for Cell 9 in the attached video are the culprit.

Whilst charging with above 35 amps to 55.2v (3.45v/cell), the voltage is 'reported' to be widely fluctuating. Other cells also fluctuate but not to the same degree.

'Reported' because 2 different multi-meters across the cell during charging shows a stable voltage just below the average cell voltages.

The fluctuations are above 50mv at times.

Have tried:

• swapping cell 9 to another position - the problem remains with voltage readings for the replacement in cell 9 position.
• re-terminated voltage sense lead on cell 9 - if it was loose connection, I would have expected the voltage to be reported low - not high.
• changed voltage sense lead resistance values in app - no impact.
• changing the charge amps - at below 10 amps, issue not observed. As amps increased, the fluctuations increase proportionally.
• top balanced entire pack (again) to 3.55v - no impact

Conjecture:
As charging current increases in JKBMS electronics, a larger/stronger magnetic field is induced which interferes with the voltage sense readings.

Impacts:
As cell 9 is misreporting high voltages, it is the first cell to be balanced (at 3.45v) which reduces the AH of that cell. After absorption and during discharge, cell 9 has consistently the lowest voltage.

Apart from getting a replacement BMS, are there any other suggestions on how to resolve?

 

[Checkthisout]

That's a 24S BMS, right? I assume you're using 16 out of the 24 balance lead ports.

Is there a reason you cannot make cell 9 use the balance port for cell 17 and re-test?

 

[Skypower]

JKBMS B2A24S15P

Readings for Cell 9 in the attached video are the culprit.

Whilst charging with above 35 amps to 55.2v (3.45v/cell), the voltage is 'reported' to be widely fluctuating. Other cells also fluctuate but not to the same degree.

'Reported' because 2 different multi-meters across the cell during charging shows a stable voltage just below the average cell voltages.

The fluctuations are above 50mv at times.

Have tried:

• swapping cell 9 to another position - the problem remains with voltage readings for the replacement in cell 9 position.
• re-terminated voltage sense lead on cell 9 - if it was loose connection, I would have expected the voltage to be reported low - not high.
• changed voltage sense lead resistance values in app - no impact.
• changing the charge amps - at below 10 amps, issue not observed. As amps increased, the fluctuations increase proportionally.
• top balanced entire pack (again) to 3.55v - no impact

Conjecture:
As charging current increases in JKBMS electronics, a larger/stronger magnetic field is induced which interferes with the voltage sense readings.

Impacts:
As cell 9 is misreporting high voltages, it is the first cell to be balanced (at 3.45v) which reduces the AH of that cell. After absorption and during discharge, cell 9 has consistently the lowest voltage.

Apart from getting a replacement BMS, are there any other suggestions on how to resolve?

The most often problem is a bad terminal connection. I’m assuming that you have two rows of 8 for 16 cells. Are you using a smaller conductor that jumps across from 8 to cell 9? All busses/jumpers must have the same resistance/connectivity. If you swap around cells did you use the same buss/jumper? Is the bottom of that connector clean, not just visually but buffed? Is there any temperature difference felt in any terminals under higher loads? Don’t get shocked!

 

[BipedalPrimate]

Is there a reason you cannot make cell 9 use the balance port for cell 17 and re-test?

I am fairly sure the BMS would throw a tantrum with that. It is expecting 16 cells in sequence 1-16.

 

[BipedalPrimate]

All busses/jumpers must have the same resistance/connectivity.

Is the reason for this statement that Cell 9 voltage is calculated by the BMS using measurement points on Cell 8 +ve terminal & Cell9 +ve terminal?

Why would the reported voltage be erratic if they were not the same resistance/conductivity? I could understand it if the voltage was consistently under actual should there be a higher resistance bus/jumper between cell 8 & 9.

 

[Checkthisout]

I am fairly sure the BMS would throw a tantrum with that. It is expecting 16 cells in sequence 1-16.

So I dunno. Are you only allowed a standard number of cells?

I would try it and see what happens. Would take 5 seconds.

 

[Skypower]

Is the reason for this statement that Cell 9 voltage is calculated by the BMS using measurement points on Cell 8 +ve terminal & Cell9 +ve terminal?

Why would the reported voltage be erratic if they were not the same resistance/conductivity? I could understand it if the voltage was consistently under actual should there be a higher resistance bus/jumper between cell 8 & 9.

Is the conductor between cells 8 & 9 indeed different from all the others?

 

[Skypower]

Wildly fluctuations with a steady load would indicate a bad connection in that area. If a jumper, it could indicate a bad crimp. Suspect everything from cell 8 to 9 and 9 to 10. Under load, are you able to probe the actual cell terminal with a voltmeter not just measuring from the tops of the busses? Measure any difference between different cells (under the busses) with this method under load. Compare to known good areas.

 

[BipedalPrimate]

Suspect everything

You are correct.

I had a 350A Class T Fuse between cells 8 & 9 because the AS/NZS 5139 standard indicates the Arc Flash Incident Energy hazard can be reduced by 50% with an over current protection device in series in middle of battery string.

The AS/NZS 5139 Arc Flash calculations are based on NFPA 70E Annex D.

Have now rebuilt the pack to relocate the 350A Class T Fuse to between cell 16 and battery disconnect circuit breaker - the large cell 9 voltage fluctuations have disappeared.