Strange (rebuilt) battery behavior. Thoughts?

[gfunk]

Sorry this is so long, but the context is important.

I have two Hurley Mini Swells. They are less than a year old and have less than 200 miles. One of the batteries crapped out and I sent it to be rebuilt. When I received the rebuilt battery I was told that the original Chinese Sinc 2500 mah cells were worn out and severely out of balance. They were replaced with LG 3200 mah cells..and a new 20A BMS.

After receiving the rebuilt battery, I charged it from 80 to 100%. The bike doesn’t operate correctly with the new battery. I tried it on both bikes and the behavior is the same. It shows a charge and powers up the control panel. When engaging the pedal assist there is a slight “grinding” sound in the hub. Once the bike hits 10 mph it “grinds” again and stops. Note that the "grinding" likely isn't a literal physical grinding, but it is audible. Sometimes the pedal assist and controller power cut out, and sometimes just the power assist stops.

Both bikes work as expected with the old battery. Volt meter shows readings - both fully charged (i.e., green light when plugged into charger).

OEM - 54.5
Rebuilt - 54.2

The vendor's came back with this:
"My theory is that the 100% charge voltage on the rebuilt battery of 54.6v is causing a problem with the controller. 54.6v is the fully charged voltage for a 48v battery. After we rebuilt the battery with the original BMS it would only charge to 51v or around 80% charge. This is not normal for 99% of ebikes. So we replaced the original BMS so it would charge up to the full 100% charge of 54.6v. I believe if we can get the voltage of the rebuilt battery down to 51v it will start behaving normally. Then we will have to figure out a way of limiting the charge voltage to just 80%."

This was BEFORE I sent the charge voltage readings to the vendor. Is it possible that the new BMS is somehow incompatible with the controller? Or that there is a limiter or some sort that kills things once exceeded?

Interested in anybody's thoughts?

 

[CloneWerks]

My theory... you have one (or more) bad cells in the "newly rebuilt" battery packs or at minimum at least one cell has a poor weld that isn't carrying voltage well.

 

[gfunk]

My theory... you have one (or more) bad cells in the "newly rebuilt" battery packs or at minimum at least one cell has a poor weld that isn't carrying voltage well.

Thanks, and this wouldn't have any impact on the battery being charged to 54.2 volts?

Edit: I'm guessing a poor cell would be reflected in the "full charge" voltage? And that a poor weld would not?

 

[CloneWerks]

If it were a single weak cell in the mix somewhere you might get a slightly longer-than-usual charge time but it will still go up to a "full charge". However the minute you start stressing it that weak cell will sag down hard. This can actually be dangerous because a Li-ION cell discharged hard enough can actually reverse polarity and that can create a thermal runaway situation (poof...fire...outgassing... horror).

If there is a poor weld someplace a pack will usually charge okay (charging is generally pretty low stress) but if you pull amps hard the connections heat up and expand/contract. This can lead to a cascading event where resistance keeps increasing to the point that the BMS freaks out because volts/amps/watts go way out of spec. But then a few seconds/minutes later everything seems fine again because the offending connection has cooled down again. Eventually that kind of a bad connection -will- fail due to the expansion/contraction stresses.

No matter the cause, if the battery pack isn't functioning as expected, but the bike is fine with the old packs then I would strongly recommend that you do two things.
1) stop using that pack immediately and put it somewhere fireproof!
2) contact the company/rebuilder/whomever and let them know they need to make it right WITHOUT any "workarounds" or band-aids!

Please do not work on convincing yourself that "it's not that bad" and you can keep fooling around with that pack. If it's not behaving as expected, then it's suspect. Suspect packs should be treated as a time bomb unless you have the training/knowledge to actually disassemble and diagnose them (not a job for the untrained, you can get really badly hurt).

This is a demo of an R/C battery pack fire. Your eBike is about 3x bigger than this. I'm not saying be afraid, but I am saying respect the power crammed into that package.