Battery Dies with full charge, then works after cycling key

[Evolvedbicycles]

Hi Everyone in the Ebike world

I have this problem with a 10 Ah 48 volt battery. It's was new in August 215 and has about 60-80 charging cycles. Here is the issue. It charges fine and reaches 54.1 volts when charged. It holds a charge for months and months. I ride a mile or less and the controller just goes dark and the power get cut. I turn the key off, then back on and power is restored. This happens no matter how much load the motors is demanding. It happens with PAS only and throttle only. Sometimes I get 100 feet, sometimes a mile. It is totally random. Pot holes don't cause the problem. It happens on my service lift too.

So I started a technical help request with our thin-skinned ebike friends down under. They can only sell parts as good as they can buy. I don't blame the suppliers. They do try, I will give them that. Fair enough I say. I blame US Customers for buying inexpensive crap systems. Anyway they had me reset my P setting which I did. The controller for this 48v system with 10 Ah battery is discontinued I am told. After 8-12 emails back and forth I can't fix the problem. So I cut my whole loss on the down under system, and hope to reuse the battery. I rip that bike up, and it's now getting an Xtracycle Leap kit and I am trying Golden out. Our friends to the north are known for being very polite.

So I buy a new kit from the hip guys in So. Cal whos names rhymes with Kahuna. I buy a new hot all black 2017 Adventure Hybrid with remote lockouts WTB tubless rims from Bikes Direct perfect for my height and guess what. BUT HOLY CRAP the 48 volt BBS02 acts the same way. (I am reusing the old battery of course) Randomly the battery shuts down and then works find when the key is cycles off then on. Right before it cuts out the power goes into mini surges of on-off before it goes dark. This new BBS02 mid-drive has so much more torque than my hub motor, I did not notice the power flutter right before it went dark.

It was never the controller, it is my battery!

So ebike geniuses of the world, any ideas. The cells have to be fine. It charges fine and hold a charge and does not get warm or act weird otherwise.

Maybe the switch is bad? Maybe the BMS is faking out? I will greatly appreciate any help.

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

UPDATE ON AUGUST 10TH 9:16 AM PST

I disassembled the battery out the case retaining the yellow plastic insulating box. This is a 13s4p 48 Volt 10 Ah with HH 22650-3M cells.

The problem is not with the switch. With a fresh charge and zip-tied the throttle on with the bike in the stand and the battery would not trip off. So I slipped the battery into my new triangle bag and had to ride it 1/2 mile to get the battery to stop. When I tested it, the voltage started off around 24 volts and seemed to rise slowly like up to 30, then 32 moving up a tenth at a time. I repeated the ride and the battery "tripping" off 3 times. As soon as I either cycle the switch or disconnected the XT60 the voltage will sprang right back to a little above 52 volts when reconnected. It "trips into this half voltage state randomly. 3 times I could go 1/2 mile and it failed exactly in front of the same house, then later it would trip in under 100 feet. So the more I pushed it the faster it tripped. Nothing got hot or looked weird or smelled like over heating.

Tonight I will pull the sleeve off and test all the cell groups before and after it goes onto the "tripped" state.

If I replace the BMS, does anyone know if I can I use any BMS set up for 48v's with 13 series wiring? Or does it have to be the exact same as the original?

Thanks very much for this help.

 

[GalFisk]

Check out the BMS first. The key switch is probably not it since the issue comes up independently of mechanical vibrations.
If you can provoke the issue with the battery disassembled, you can measure the BMS input and output voltage (and the key switch too). You can also inspect them visually, and feel and smell for overheating.

 

[dogman dan]

The bms is doing what it is supposed to do, cut off power when one cell group in the battery drops voltage. it holds voltage, but has near zero capacity. So under load, voltage drops huge, and the pack shuts off.

3 years old, its done its reasonable expected lifespan now. you did not use it up, it timed out.

On the bright side, you got two usable motor systems now.

 

[Evolvedbicycles]

The bms is doing what it is supposed to do, cut off power when one cell group in the battery drops voltage. it holds voltage, but has near zero capacity. So under load, voltage drops huge, and the pack shuts off.

3 years old, its done its reasonable expected lifespan now. you did not use it up, it timed out.

On the bright side, you got two usable motor systems now.

Thanks Dogman Dan for the reply. Just for the learning curve advance I will open it and let the bike rip on the stand. When it fails, with the battery open I can measure into-output on the BMS. I am into batteries recently because I am building my second one using the Vruzend kit with some very good cells 3500 mAh 12.95Wh LGDBMJ11865

 

[Evolvedbicycles]

Check out the BMS first. The key switch is probably not it since the issue comes up independently of mechanical vibrations.
If you can provoke the issue with the battery disassembled, you can measure the BMS input and output voltage (and the key switch too). You can also inspect them visually, and feel and smell for overheating.

Thanks GalFisk for the reply. Your suggestion will be my next move. I could replace switch or BMS if needed. I am building a kit battery presently just for the learning fun of it, and most of the big packs I wanted were out of stock everywhere due to summer buying surges.

 

[fechter]

I'd measure all the individual cell voltages at the connector going to the BMS. You might just have one or more low cell groups that need balancing. Some BMS circuits will drain the bottom 3 cells more than the rest and cause an imbalance over time. The low cells can be brought up by individually charging them with a single cell charger.

 

[Evolvedbicycles]

I took the battery out the case then out of the insulating jacket. All the cells test out 4.16 volts. Right now the pack measures 54.2 volts.

Again, the problem is it runs great then randomly dies. If you unplug it or cycle the key switch you are good to go for 1/4 mile or 100 feet, then bang, no power.

All in favor of replacing the BMS signify by say Aye. :wink:

 

[Telemachus]

Check out my thread:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=90097

You need to test voltages of the cells under load. The BMS is doing its job.

 

[kocetoMR]

this cell are very old chenge the pack

 

[Evolvedbicycles]

Check out my thread:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=90097

You need to test voltages of the cells under load. The BMS is doing its job.

QUESTION to Telemachus: (1) How would you put the bike/battery "under load" without riding it. Riding it is ruled out because I can't ride with an open disassembled battery in the first place let alone test. I assume you mean on an indoor lift/stand. I tried using the two ways I could think of (1) using brakes but that got hot very fast and (2) I used a stick to create a fulcrum against the tire but I stopped that quickly too because it was killing the my nice new NANO 40x700 tubless tire.

So how would you Telemachus test the voltage under a load?

 

[Evolvedbicycles]

this cell are very old chenge the pack

"I am doing this testing will a fully charged pack"

 

[Evolvedbicycles]

Check out my thread:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=90097

You need to test voltages of the cells under load. The BMS is doing its job.

QUESTION to Telemachus: (1) How would you put the bike/battery "under load" without riding it. Riding it is ruled out because I can't ride with an open disassembled battery in the first place let alone test. I assume you mean on an indoor lift/stand. I tried using the two ways I could think of (1) using brakes but that got hot very fast and (2) I used a stick to create a fulcrum against the tire but I stopped that quickly too because it was killing the my nice new NANO 40x700 tubless tire.

So how would you Telemachus test the voltage under a load?

"Updated question after reading your related post. I see it is done with connecting external loads to the battery". Thanks.

 

[Evolvedbicycles]

Check out my thread:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=90097

You need to test voltages of the cells under load. The BMS is doing its job.

QUESTION to Telemachus: (1) How would you put the bike/battery "under load" without riding it. Riding it is ruled out because I can't ride with an open disassembled battery in the first place let alone test. I assume you mean on an indoor lift/stand. I tried using the two ways I could think of (1) using brakes but that got hot very fast and (2) I used a stick to create a fulcrum against the tire but I stopped that quickly too because it was killing the my nice new NANO 40x700 tubless tire.

So how would you Telemachus test the voltage under a load?

"Updated question after reading your related post. I see it is done with connecting external loads to the battery". Thanks.

Update: 9/2/17 @ 1:57 PM PST: When I shifted my gears into 24th speed (I have BBS02) then when I pulsed the brake I could get the battery to fail here in my living room on the stand. So all 13 series group tested the same right after the time of the fail. But maybe the point is I need an assistant to pulse the brakes 13 times after I have my probes relocated and I have to test all the BMS connections WHILE THE TIRE IS FLYING AROUND FAST and the motor is ACTUALLY RUNNING. Is that it? Only THEN after I compare those readings will I see if one (or more) series of 4 cells is different.

 

[Telemachus]

The assistant is your multimeter hooked up to a each cell sequentially. Put it on a stand where you can see it.

Test each module (S) at full charge. They should all be about 4.2V. If one is way off (>0.15V) than likely that one is the culprit. You might just need to charge that one infldividually with a hobby charger.

If they all are farely close wheb fully charged, you next want to ride the bike until it cuts out. Now check all thr voltages again. If one array is much more discharged than the others, then it is a likely problem.

Finnaly, when the bike is already dying on you, do some under load tests.

Under load on your high gear you will watch what the voltages do. Hit the throttle and note the voltage. Hit the break and note the voltage. Then move the probes to the next module looking for the one that dips the lowest. Then, keep the probe on that one while you break until the power dies. You will be able to see approximately what the LVC of the BMS is.

If all three rests point to the same array, then it's the first candidate for replacement.

 

[fechter]

I agree it is really challenging to measure individual cells when using the motor as a load. Some other kind of load just for testing might be worth figuring out. I use an old hair dryer but that only draws about 5A on a 52v pack. Something like that might be enough to at least identify if there is one or two cells that sag more than the others. If you find a weak one, then maybe test with the bike motor.

 

[999zip999]

A oil heater with make a bigger load.

 

[999zip999]

Check your brake cut-offs either at the brake handles themselves or just disconnect the brake cut off and see if that could have some relevance.