Big number on the BMS. What about the rest of the battery?

Uplorry on the long road

Illustrative shop scene and virtual chat.

Friday, 6:40pm. School was out. Jing Kai was riding his Fiido home, and his girlfriend Xin Yi was on the back in her school uniform. Her bus was late. Jing Kai believed in tompang.

The pack under the seat was a 60V 30Ah. The sticker said Panasonic. The builder, a homebase guy found online, very cheap and very friendly, said it was the real thing.

Jing Kai had paid him extra for a stronger BMS. 80A. Bigger number is stronger, right?

“BMS eighty amp. Nothing can go wrong.”

Then two LTA officers stepped out from behind a lamp post.

Jing Kai twisted the throttle all the way. The Fiido gave one small, tired sigh and stopped in front of them. No bang, no smoke. The pack had cut out at about half charge.

The officers called for the lorry. Xin Yi held her bag over her face.

That night, a photo went up on Facebook. Faces blurred. School uniforms sharp. The group chat started before dinner.

A CHAT WE SEE A LOT

Jing Kai came to our shop two days later, alone, with the pack in a plastic bag. He asked if the BMS was spoilt. We told him it was probably fine. That was the problem.

We opened it.

Three numbers, three parts

The pack, as sold: “60V 30Ah”. We checked each part.

• The sticker says 40A. That’s most likely what this battery was originally safely rated for.
• We found a cheap ass 80A BMS inside. Some homebase builder swapped it out.
• The cells say LG. Six in parallel. Real cells like these give about 10A each, so about 60A constant and 90A peak at max.

Cell, BMS, sticker. All fighting who is right.

If the cells are only good for 40A, no competent battery shop will put an 80A board on them (cost reasons, obviously). So the BMS seems to have been swapped out by some random home-based builder who never bothered to swap out the sticker.

Pair it with an 80A controller and you overload the cells.

What happens when you overload the cells? Best case, a damaged battery that cuts off early. Worst case, a fire.

Real cell, fake cell

Those cells just looked off to us. So we plucked one out and inspected them

• Real LG cells: 14.7, 14.9, 15.7 mΩ.
• This pack’s “LG” cells: 18.3 and 17.9 mΩ.
• The positive terminal cap have some discrepancies too. Real LG cells have three legs. Those that we found have four.

Higher resistance, more heat for the same current.

these definately does not seem to be real LG cells anyway, IR doesnt match, either a fake rewrapped cell, or a dismantled real LG cell near its end of life, either way its not looking good

Real ones can do about 60A. These start getting hot above 40A, maybe 50A. So 40A on the sticker is probably the honest number.

Why it cuts out early

Weak cells don’t only get hot. They sag.

• Push current, the voltage drops. Higher resistance, bigger drop.
• The BMS sees the voltage fall too low and cuts power. Even with charge left.
• That’s the half-charge cut-out. Right when you twist the throttle.

Keep pushing and the cells heat up. One cell goes, the one next to it follows. That’s how a pack goes from hot to fire.

Before you pay

Ask for these, in this order:

Two more we’ve seen

Same mistake, two more packs. Both went wrong at the BMS.

1. A BMS way too big for the pack

• The cells are real Tesla/Panasonic 21700s.
• The BMS is far bigger than this pack needs.
• A bigger BMS does not make the cells stronger. It just steps in later. The cells still only do what they can.

2. A cheap BMS that blew up

• He ran it on a Fardriver ND72360 controller. 190A line current. That is what the controller can pull from the battery.
• 190A asked. A 90A BMS. More than double.
• The BMS blew up. Soot all over the board.
• He’s lucky. Not a good way for a BMS to do its job, but at least it blew up instead of letting the current through.
• Check the controller’s line current against the BMS. Then the BMS against the cells. Before you ride.

What now

Sources (1)

• Singapore Consumer Product Safety Office: lithium-ion battery safety