The E-bike battery and a tale of perseverance...

Batteries, Chargers, and Battery Management Systems.
Post Reply
emr   100 mW

100 mW
Posts: 37
Joined: Apr 22 2018 5:37am

The E-bike battery and a tale of perseverance...

Post by emr » Aug 05 2019 9:17am

************
Warning, this is a long story, only for the very bored. Read it at your own risk.
************
I recently had an interesting learning adventure with my e-bike. The (long) story begins some time ago...
A year or more ago, I built my 52v TSDZ2 with 52v 17Ah 14s battery. Now, I didn't really need an e-bike, but as I have always enjoyed being a 'maker', the e-bike was an interesting project for the time. As I am sure you have also found, there is much reading and learning and sifting and absorbing of information, from multiple forums and other sources, to locate and identify the parts of the puzzle that is 'e-bike making' . I reckon the e-bike project learning curve, is like an 18% climb on the 50/14.
Anyway, I did complete the initial build and it did work. Hooray. I didn't ride the e-bike much and for quite a long time, it sat idle and unloved in the corner of the garage. One day though, after some months, I decided to fire it up, and to my disappointment, the LCD came to life only briefly, and with a blink it was asleep again. Turning the battery off and on produced the same result. Hmmm! I thought, is this due to the battery having lost it's charge. Ok. lets plug it into the charger (2A) and make sure it is fully charged. More hmmm, the charger indicator light remains green, indicating the battery is fully charged. Putting the multimeter across the battery output indicated over 52v, but only briefly, as it quickly faded away to a few volts within a few seconds. Double hmmm and bugger. Did this mean some cells in the battery have failed and the BMS (Battery Management System) is preventing the battery from outputting the required current.

A couple of options presented themselves..
Find someone who knows about batteries and BMS, package the battery up, and courier it away and out of my concern (sensible option).
Or
Pick up that phillips head and investigate (dangerous and brave/silly option)

So, being an inquisitive person, I chose the second option (of course).
Upon opening the case and revealing the cells, BMS and wiring, I identified the battery negative and positive terminals. Measuring across these terminals indicated over 52 volts, and no voltage dropping off as was experienced when checking via the BMS output. This was useful information, but still didn't prove where the issue lay. I next identified groups of cells that were each in the 12v range and one 12v group was a little low. I decided to force the charger to charge the battery by overriding the BMS (I am not recommending you do this). I unplugged the multi pin connecting socket from the BMS, and added a wire connecting the negative of the charger terminal on the BMS, directly to the negative of the battery. This did indeed cause the charger indicator light to turn red and I watched the voltage gradually rise (multimeter connected) over an hour or so to 57v.

After reconnecting the BMS wiring, the battery output now maintained the 57v. Hooray, looking good. I plugged the battery back into the e-bike and the LCD powered up, and the motor ran (using the throttle with bike on work stand). But oh dear, only for a few minutes and then back to the previous behaviour of the voltage fading away. And once again, the charger indicator light failed to turn red. I repeated the previous process of taking the BMS out of the circuit, and once again, after a short charge the battery was operational. At this stage I started to be convinced that some of the cells were failing and causing the BMS to turn off the battery output.

I repeated this process a number of times and eventually the battery was able to run the motor for more than an hour. It seemed that my suspicion of failing cells was supported, and that the forced charging was resurrecting them.
Disappointment set in the next day when the same symptoms were present once again. But not one to be defeated, I persevered and it appeared to improve. I was even able to ride it for 40km without issue. After the ride, I put the battery on charge and it did charge, but only for an hour and then turned green with sporadic flashes of red light. This continued for many hours, and to my chagrin, the dreaded power fading presented itself again. I opened the case and checked for any dry joints or other dodgy connections, but none found.

At this stage I did a google search, and after considerable research, located a suitable BMS, which I purchased just in case it was the cause of the issue I was experiencing. The BMS arrived, but some months went by before my energy for playing with e-bikes was recharged enough to once again attack to problem.
Upon re entering the field of battle (between the battery and me), I attacked using the previous strategy. Force charge and then discharge and repeat. Again, this appeared to work and I was able to ride a significant distance. The next day I once again rode off, only to experience a complete shutdown while at the furthest point of the ride (of course). So now I had no option but to pedal all the way home without any assistance, and on a 23kg bike, and up the nasty 18% hill to my house. Well I did make it. Who said e-bikes make you soft!

Upon puffing into the garage, I immediately ripped the battery apart, and after careful checking of the differently placed connections, I replaced the BMS with the item I had in stock (I love having things in stock). When I connected the charger, it turned red and stayed red for a number hours. This looked very hopeful. The final charge was 58.5v. I connected to bike and worked, Yay. More testing the next day and still worked Yay Yay. A 50km ride the next day and no problems. Charging also went for many hours as it should. Yay yay yay.

It is unknown why the BMS would have failed while the bike was just sitting in the corner, but it is a great relief to have avoided the need to replace expensive cells. It was also very satisfying to have resolved the issue without assistance, and just with logic, perseverance and a new BMS.

If you have arrived at this part of the story, you should also be congratulated on your perseverance!!!

User avatar
neptronix   100 GW

100 GW
Posts: 13984
Joined: Jun 15 2010 5:56pm
Location: California refugee living in Utah, USA
Contact:

Re: The E-bike battery and a tale of perseverance...

Post by neptronix » Aug 05 2019 10:01am

Hate to say it but you never ruled out a bad cell as being a problem. What happens when one cell goes kaputz is that the nearby cells often overcompensate by being of higher voltage. So your multimeter may have missed a key piece of information by measuring 3 cells in series.

Hopefully you are right about the BMS being the problem.
BMS is a common failure point, as many are cheaply made. I am extremely picky about BMSed battery packs for this reason..

Kudos on learning instead of outsourcing :) if you are as crazy for ebikes as i am, knowing the ins and outs of the entire system will save your ass when you have a failure on the field.
"Love and compassion are necessities, not luxuries. Without them, humanity cannot survive." - Dalai Lama

My first major build: 1.6kW 8T MAC motor on a Trek 4500.
The new all-arounder: Leafmotor 1500w @ 4kW on a Turner O2 full suspension.
The monster scooter: 20" eZee on a Cannondale Semi Recumbent.
Whipper-snapper: ? on a lightweight BikeE Semi Recumbent

emr   100 mW

100 mW
Posts: 37
Joined: Apr 22 2018 5:37am

Re: The E-bike battery and a tale of perseverance...

Post by emr » Aug 05 2019 4:10pm

neptronix wrote:
Aug 05 2019 10:01am
Hate to say it but you never ruled out a bad cell as being a problem. What happens when one cell goes kaputz is that the nearby cells often overcompensate by being of higher voltage. So your multimeter may have missed a key piece of information by measuring 3 cells in series.
You are correct of course, but to confirm and fix that issue will be a much more difficult process. I will wait to see if it does prove to be a bad cell, and take action at that time. It may be that a bad cell can be resurrected to some degree if kept within happy charge/discharge limits.

Cheers from Emma

emr   100 mW

100 mW
Posts: 37
Joined: Apr 22 2018 5:37am

Re: The E-bike battery and a tale of perseverance...

Post by emr » Aug 06 2019 3:23am

neptronix wrote:
Aug 05 2019 10:01am
BMS is a common failure point, as many are cheaply made. I am extremely picky about BMSed battery packs for this reason..
@neptronix...

Please advise what constitutes a 'good' BMS, and where do you suggest is a good source to purchase.

Post Reply