Built my mom an ebike and it's destroying batteries

miataman2000

10 µW
Joined
Feb 19, 2019
Messages
5
So I built my mother an ebike with a cheap eBay 36v 1000w hub motor and it worked perfectly at first. I used 3 36v 4ah hoverboard batteries wired in parallel for the battery. After a few uses it seemed like the battery would drain after just one or two minutes. After testing the voltage none of the 36v batteries read anywhere near 36 one is reading about 12v and the other two are reading about 4v. I tested the bike with another perfectly good 36v battery charged to about 39v and the bike throttle turns on but as soon as I twist the throttle to ride it shorts out and the lights go off. What gives? Is the controller bad somehow?
 
I haven't ever had the electronic brakes hooked up. I had them on another build and didn't care for them. The battery was fully charged to 41v before riding but now that's no longer possible as the batteries no longer hold a charge.
 
Does it work with the wheel off-ground?

If so, then the load while riding is too high for hte batteries, and the BMS in the batteries are shutting down to protect them.

If there is no BMS, then that means the cells can't output enough current so they drop in voltage so far that the controller shuts down to try to protect them.

1000w at 36v is about 28A. That is a lot of current for a 12Ah battery of most likely very cheap cells, and can be hard on them, so they may not last long. Most likely it doesn't see that kind of current for more than a few seconds at startup from a stop, but it can still affect them over time. If it's used on hills it could see that current for significant time.


If it does not work with wheel off-ground either, then something might be causing friction on the motor wheel, such as brakes rubbing on the rim, or a bent wheel causing tire to rub on frame. On a rear wheel it can even be the sprocket or freewheel pushing against the frame, and the hub or it's plastic "protector" rubbing on the freewheel.

It's also possible something is wrong with the controller itself, blown FETs, etc. If so, you'll find the motor is hard to turn by hand, usually, as the FETs typically fail shorted.

The motor itself could also be shorted--this usually happens when the phase wires coming out of the axle get damaged in a crash or similar, and tend to get cut against hte axle, and may intermittently short to the axle and each other. Again, the motor is hard to turn by hand.
 
These hoverboard batteries, how old? They can be toasted just by storing them wrong awhile, not keeping them topped up all winter could do er.
 
I am with the other guys here. It sounds like your battery pack wasn't up to the challenge of powering your moms bike. A 36v battery at 12v is something like 1v per cell/cell group. Sounds like that battery is toast, sorry to say. Those cells had to deliver something like 7C and couldn't do it. If you can test the individual cells see if some are low. You might be able to use the good ones for something else.

I think it is time to get a new battery pack.

:D :bolt:
 
spinningmagnets said:
Those are weak generic cells in those hover board packs. You can use them at 1A per cell

I have over 30x of those hoverboard batteries and 90% of them have Samsung cells rated at 10a, some have LG cells rated at 10a and there are some with generic blue China cells but not many at all. The BMS are rated for 20a

I would still draw the current at half the cells ratings which for 3x packs in parallel that's 6p so 30a max draw especially if they are used

This doesn't mean the batteries cannot fail, I have packs with dead cells and they are useless unless you pull them apart

I power two ebikes with these packs right now and have no issues. One is at 72v with 2s3p and draws 50a no prob

jMawnHd4XghzIqTIwOccisDwQHPJfeKRoTF-3s023kdL2nfNDIsHnM_INDdXJ-U3iO5hrhp6mWGtb9XVYAMbVVG-HHr-gC3UsyucSB93qWmgstGPYWpNIF2RR_f6rYZGqNtMJBrjpYC_S3VYEkD1cN1XdSCEDH_gF0l6JgRX7BIVvMdGuBK91g5_JijcIe1FUOAlN4kpHI-uX6lrvBX4M_uoCqa_Fg1c24cdeLcsFuZQV83LT68b4-0tkZ-ezThqYO9WoJbC3yyJcWfrQYO6iCb8m2oM9kUZhhOhDhCcjgGvrJjYMJ4MOA2j4luHARxFLc07fK_NQz-MV-eKPvPcjl--xx8szmTJB4pbcBHBYXB-r3i4hUrjgKb5cyt5hOIOKENN_Kl0Y3OtRbh31pzE4KW8G029XYePwJmSUurIgabnSq9rFx-6UaU-voAteyWGBJzc528ps2T79BC2SzzhHB3LzMK_HQ8cMzHztKkHggzBWvmMa9Gvnvxcola5Eg8NzLD10VElIn1pWA2j3byuAqtmByjlDBdHIkRHzSKKOVQhw9G-ncuuh8EUpfZf5e3_bKpVCbSHBcz-5stknL_PC8ywf13VVXEdbOF6yp37ZhdACa7_DSGdB4ScrsO3qFsWbDTjoMPlMyiHbuUDloYFAl0olG8bCtk=w1195-h896-no

0eqn3WAm-adgsjPGXxTayefq5UVXjy6kkQ-_6f5aacI9ybOnl0O0a8Hhbkn3tlnn7y1GUBTev8KsfCs2zJ3YIeBsJdx5amAlplnXEdzZySd_ojfPNiLcPKUNxqduJaca8aAX-adZJBcTyM-griiP79ojSVWh4OKRuUQhLV2GM91zkv6-29uiY0QLQoh5XbDDoKyAzh4zhvZdSWPVuVLbV82S9WwE679xP4Xbidvb04QEpV8ipZI_YD2IovVmq8wk08uIm4Y63pru9KtmQcUTU9GYjYYI4wxD6aF2F1Cz66dFt7bbZNvcY01WHn6SGh4lD7MPWHtzjsyFdnxREf9rjxPwWUWVHgdFPe7-of5KmA2SE_zb9z3bafIkXgmfiZI_LV1TLQebutXAkwn8OFOs7_RcoVtnXjAJWokA0ADCXMmQgseRe-ec0xhsgFrn6Ga48eq5K4m4zW93tiXstRdt2rpNu5ZWpRBL1EIbZBCgl4XVtgRlYlgOnbylX1m0u6yl8ltM5BDQta8OBgwE-eAP8F4vq_WB2y8AEPrHFW_jTg7udbnop2hm4zg8_uikHQWzUNUbOKTTg_zrp8qdeNScctbklyoO_d-Ny24BUVzidPbqXrN4bxxGKwKCKVy0rrUKHBzsTeFYIdbL7xye9cQ6KqhX5uJjZ5Q=w672-h896-no

qh6MGJvLlUCI1JEOlorUnRO22S32Fo9wAnkkMcR7exdAFow3yauw11fK-JlebOhQBXQDsTm61b9aT5D4I-1-2ag3q5mQqINGCFkGq7U9unkGUbuNr9VoX0dtd2VQ0ropEsZ5bEjBT8ChlN8IQWJYRUTNOqoUgY89R--z4O9DJlY7MUqIDOEf0KpPaV6rlWWAKDtwpu6Eoayogo7WLpCfrNgTYJsReAipKpxIXgNMbuaqZcnyhUV-btXjCyFRjkIzbbLG5aTKIMC3TWeT2rYgTd0l7XTC_MUyuaNVacnMsezNJw2jwcQ4eEy0zlMauU9eB06wjvF9ExrpFi1w2tkN_MEl3r9dBdaCL2zSTfh-L4SQ77n3Q3F94NVJp6Wdcr_8Aag9jqdLnxB3YWaZ9tbjuyePVkF5X-CSODsX01i3Y4ctCj1o5GNYZ_r4fek7EUrcxtLU6-e1ToThs-3EakhSX__O5VRRX51Bw3P-ov-VSJynawd1rGoPLgobDdSqGWQIvM6nN8rGkkOdbHRyfBXdc5PbLyrA-l7XSSTZGNRo5_LbibZq0NXVydGSX6Dya6VOY8_PNm3Tws20LlI1l4gPCT7raknbCj8teYYHVaqmP1WnqNZQB68DhCTo4XGn2hTnu4430AnWGQAR4qeNe1waUYa79I0ohME=w1195-h896-no

DCODNvsLHVLT8T7_e2Nd_8qCpyG7tM1Xvuo8AwoRT7AIGvLLfusnyfR_CX95apLVjKh4lIQQKHbRJHPIU20H7TBQCQ3rSubPv7G-PUGT_4Md8HEo771WmPFqE_99PYVOjZkYFdxgLBEkdRxYbOlMW_GA5DhzGYISdJRxLDrXt-FIW5IePd_d0Y86MBL3PsoHAFP-0hRayOhcgak94Az1qE89iQiOEuXV6PA9cjpvQOBS4Bc6hGh7gqE2YFqjH0XbWKdqimWMIjAnd5QMqC3wmyKHfHnpM0uSWGh5ObiSKLhhvgnsS5_YGOrSeEMnbN0vxaUSHoCLHp2TRzQpCxvSAOTi5NkfVTal7Ipl8-FJmMP1ZDJh6YpAiiT_pyqkqpu21vEXf2JphC8EUiZStyTJoOfm5ZTanrVusHFJbL81_6VLxFM2rwxrDW4ci1ielUvt7CmxEA4r-d41p1FK4aELwP8NtcD1bIutJtYiVsz5Lm-7BhYKlZRvct-Cl1SktTYi6NXFFWHroDq0ZPUnXXvRoKhw-HZZ76HrxTckHukKxquF7goqmZRzvUNmeHFiOWu7_KyV17K2PJHXGJkwq9bMg8VA2-7vhIWSJOUh35idFyO5QFAl6EsyJEbfYMkWYh0noxXDuS2Slc5-PkIknA5OWEM1fCmrxtI=w1195-h896-no
 
My apologies. I have had bad luck in the past with counterfeit cells that looked very authentic. It was just a guess based on their performance. Perhaps it's a problem with the controller or motor...
 
spinningmagnets said:
..... Perhaps it's a problem with the controller or motor...

I am still thinking that the 4a pack was under-rated or the amp draw of the controller the OP built in his moms bike.

Somebody check my math but with the Samsung 22P cells at 36v wouldn't that pack be 12s 2p? That would put them at 20 amps continuous discharge. And if that bike draws 28-30 amps max, how how long could a used 20 amp battery pack last?

Edit: Sorry my bad! I didn't read the original post right. Using 3 of these packs should have been just enough to power the bike provided the batteries were in new condition Maybe the battery packs were just bad. However, without knowing how many amps the bike's controller pulls then it is hard to know for sure.

I still think a new battery is in order. Something that is able to take more then 30a continuous.

:D :bolt:
 
Are these the original load of hoverboard pack sold by alarmhookup? Those were 22P equivalent cells and delivered 20A per pack. AT least mine do.

Since then, the low cost 10S2P hoverboard packs on ebay appear to be junk to me. Wouldn't buy them.
 
They may have been good packs once. I was saying they likely are not now. They might charge full, but be totally unable to carry a load of any kind, for long. A couple years in a warehouse would kill em, but they could be recharged and still appear ok for sale.
 
Sorry I have been pretty busy with school but I did test another much larger battery with the bike and I was still experiencing problems. So I ordered a replacement controller and sure enough problem solved. As for the batteries they were the original alarm hookup with the Samsung cells and I have been using 6 of them wired up to 72v with my cyclone build for about 2 years now and they work great. Thanks for all the help here.
 
Glad you got it going!

Edit: What are the specs on the new controller?

:D :bolt:
 
Those packs have BMSs in them. If you put three of them in parallel, you must disconnect them from each other for charging, otherwise two get charged through the discharge wires, which bybasses the charge control functions in the BMS. You can avoid all that if you parallel the sense wires and use a single BMS with a suitable rating.
 
If miataman2000 is using the alarmhookup packs, the BMS charge ports are not used in these packs. The maker decided they would be charged thru the discharge port. Consequently, I charge mine in parallel.

What's the main risk? Charger fault and over voltage? That's why I charge them outdoors.
 
The specs of the new controller is for 36v and only 500w. It's got plenty of pick up for my mom and should be easier on the batteries.

As for charging the packs individually, this is interesting, I haven't thought much about it. I assumed that as far as each bms was concerned they would be seeing the same current regardless. Every once in a while I'll charge up each one individually just to monitor for a weak pack but so far they are staying well ballanced.
 
If a BMS has just a charge/disharge port it doesn't matter.

But...

If a BMS has both a charge port and a separate discharge port, then:

If charging, and the BMS shuts off the charge port due to a high cell, to keep charge current from flowing, but it is still connected via the discharge port to another pack that's still charging, then it will still receive charge current and thus overcharge that high cell; it has no way to stop the current.

So the discharge ports must be disconnected during charge, for paralleled packs like this.

Similarly, the charge ports must be disconnected while discharging, because if a pack disconnects the discharge port due to a low cell, and the charge port is still connected, current wil continue to flow out of the pack, and thus overdischarge that low cell. The BMS has no way to stop that current flow.
 
THe alarmhookup batteries only have the single XT60 plug for both charging and power, but the BMS circuits were designed for a separate charge port..

I took one apart, and peeked under the shrink wrap at another that had a different BMS. Both had unused C- pads, intended for the return leg of a charge socket. They didn't have balance circuits either.

That would be an explanation of why a few dozen packs caught fire, out of the millions that were sold. Bad chargers.

I admit that I went to the trouble of buying a BMS cards with balance circuits, and I "upgraded" the pack that was opened up, but I replicated the error of not using a separate charge port. That was the first time I worked with a BMS.
 
Back
Top