Battery failure question

[morphix]

One of my little 5Ah LiFePo4 batteries has failed after 7 months of usage and I'm carrying out some diagnostics on it to see whether the problem is the battery cells or the BMS and have a question I'm hoping someone might be able to help with..

First I'll describe the status and behaviour of the battery... Until last week it was working fine and then suddenly it just cut-out a few minutes into a ride. There was no warning sign prior to this, and the battery has performed consistently as good as the first day, no degrading in power, and no shortening of the charging cycle.

After it cut-out the first time, it did power up again but immediately cut-out again. This happened a few times. I tried charging it again and it did commence charging and charged for about 30 minutes or so, but on testing the battery again it cut-out once again a couple of minutes into the ride.

Now I have two of these 5Ah batteries both identical from the same supplier in China (purchased off eBay), so I'm wondering whether it might be worth checking if the BMS is dead by switching them over.

The only problem with this is that the BMS is hard wired into the battery cells by way of the 36v power leads. This means I would probably have to cut the leads and use a connector block or make some power connectors. That's not a problem as I'm a former electronics technician and handy with a soldering iron. However I'm wondering if it's a good idea or not to just switch over the BMS circuits (which are wrapped in silver electrical duct tape as are the battery cells themselves)? Logically one would assume the BMS would be identical and interchangeable, but is there a possibility each could be uniquely adjusted to the battery cells in some way?

 

[banbury frank]

Hi
I would go onto the suppler and order a new set with 2 x spare bms to keep all three battery's going

Frank

 

[flecc]

Those symptoms are typical of a battery that has a dud cell, causing voltage to collapse when put under load, triggering the low voltage protection shutdown. It's fairly unlikely to be BMS caused, since the normal BMS function is monitoring the charging, not the current usage.

 

[morphix]

I forgot to mention..

I took a battery voltage reading with a meter yesterday immediately after charging and testing the battery on the bike..it was reading 38.4v. Today I took a second reading, it's now 37.7v.

 

[banbury frank]

Hi

As Flecc says sounds like Dud cell what is the voltage off the good battery

Frank

 

[morphix]

Hi

As Flecc says sounds like Dud cell what is the voltage off the good battery

Frank

It's giving 39.5v off the good battery. I wonder should I try and isolate the bad cell and replace it, or just replace the entire battery. I'm not even sure the supplier would supply individual cells but no harm in asking.

 

[flecc]

The off load voltage of a dud cell will often be the same or nearly the same as a good one. It's only under load that the voltage instantly collapses, making it difficult to measure the cells since you'd have to have the meter connected cell by cell while you apply the load to get each reading. In the days of the old fashioned car batteries with top-up caps and exposed cell terminations, we were always able to apply a heavy discharge cell tester to instantly identify dud cells. Today's batteries don't make life so easy.

 

[morphix]

The off load voltage of a dud cell will often be the same or nearly the same as a good one. It's only under load that the voltage instantly collapses.

Thanks flecc that's a useful starting point. As this battery is destined for scrap there's nothing to lose by testing each cell under load until I can isolate it, depending on whether the vendor can supply a replacement cell.

Glad I didn't snip out the BMS now! Always best to check on the forum if unsure

 

[morphix]

I have just stripped the battery down and can see the 12 individual cells and how to separate them. It certainly looks feasible to isolate and replace the bad cell. Won't know if it's worthwhile though proceeding with the next stage until I hear back from the vendor. I've just e-mailed him...language barrier is a problem so I hope he understands!

 

[morphix]

The off load voltage of a dud cell will often be the same or nearly the same as a good one. It's only under load that the voltage instantly collapses, making it difficult to measure the cells since you'd have to have the meter connected cell by cell while you apply the load to get each reading. In the days of the old fashioned car batteries with top-up caps and exposed cell terminations, we were always able to apply a heavy discharge cell tester to instantly identify dud cells. Today's batteries don't make life so easy.

Flecc do you think I've just been unlucky with one of the cells going, or do you think 6-7 months life-span on a small 5Ah battery is about what you would expect under daily usage? I must admit i have been fairly punishing with it, riding on full power all the time and over the last few months on throttle-only a lot (even up steep hills) due to a knee injury.

I'm just wondering if it's economically cost effective to keep spending £160 every 12 months, and whether it might be better to get a bigger 10Ah batter that will last 2-3 years and cost £300-£400?

 

[banbury frank]

Hi

£150 delivered with 10 Amp continuous BMS Including charger 2.5 amp

36V 10Ah Lithium Ion Electric Bicycle Battery Pack - BMSBATTERY

Or 15 AH with 20 amp continuous bms £200 delivered including charger 4 amp

36V 15Ah Li-Ion Shrink Tube EBike Battery Pack - BMSBATTERY

Allow 2 / 6 weeks for delivery and pay with paypal to protect your Money

Frank

 

[morphix]

Hi

£150 delivered with 10 Amp continuous BMS Including charger 2.5 amp

36V 10Ah Lithium Ion Electric Bicycle Battery Pack - BMSBATTERY

Or 15 AH with 20 amp continuous bms £200 delivered including charger 4 amp

36V 15Ah Li-Ion Shrink Tube EBike Battery Pack - BMSBATTERY

Allow 2 / 6 weeks for delivery and pay with paypal to protect your Money

Frank

Thanks Frank, might give them ago I think and go for 10Ah this time.

Is that size for the 10Ah battery right?? 75mm X 140mm X 160mm seems rather small.

 

[flecc]

do you think 6-7 months life-span on a small 5Ah battery is about what you would expect under daily usage? I must admit i have been fairly punishing with it, riding on full power all the time and over the last few months on throttle-only a lot (even up steep hills) due to a knee injury.

I'd normally expect much longer even with harsh usage, but the discharge rate on lithium iron phosphate batteries does determine their life. Light discharge usage on current good batteries from the likes of Li Ping will give around 3 years, but with double that discharge rate the life can drop to around 1 year, these lives according to Ping's own figures previously published. So it is conceivable that very harsh usage could shorten that further, but I'd still be inclined to say a dud cell at a few months would have been a poor one at the outset.

 

[amigafan2003]

Thanks Frank, might give them ago I think and go for 10Ah this time.

Is that size for the 10Ah battery right?? 75mm X 140mm X 160mm seems rather small.

Yup - it's NiCoMn lipo - not lipo as in RC model lipo though.

 

[morphix]

Yup - it's NiCoMn lipo - not lipo as in RC model lipo though.

Interesting, that size might be ideal replacement then for my LiFePo4.. I'm not familiar with that type of battery, does it have any benefits or draw backs over LiFePo4?

 

[banbury frank]

Hi Lifepo4 is suppose to have 2,000 charge discharge Cycles the Li-ion 1,000

But lower cost More capacity for weight and size lower C rating

Frank

 

[flecc]

Interesting, that size might be ideal replacement then for my LiFePo4.. I'm not familiar with that type of battery, does it have any benefits or draw backs over LiFePo4?

It's what you might have seen me refer to as a compound cathode li-ion battery. Originally li-ion had cobalt based cathodes but they were a fire risk if the manufacturing standards were less than very good. So manganese became the preferred and safer replacement for cobalt, but didn't perform as well. To overcome that, compound cathodes reintroducing some cobalt with the manganese in a safe manner, together with other metals like nickel are now the preferred option for the best batteries. These don't have the longer life potential of LiFePO4, though many claim 800 charges instead of the former 500 charges, but they are a bit better at delivering the high currents our motors need.

 

[morphix]

Thanks for the info flecc and for all the help guys, feel much placed to make a decision now and I need to make one soon! I'm a bit nervous running only with my old battery which could give out at any time...can't imagine a single day without my bike ;-)

 

[Scottyf]

Still 7 to 8 months of use out of an a123 battery is pretty good.
Did you balance it or have a Bms?

I'm currently using lipo in a backpack and run 44v 5ah or 44v 10ah depending on the range.
I enjoy knowing what each cell is doing and the flexibility lipo gives. They also have a mental discharge rate should I want to go mad with amps.

The Bms batteries will be good so long as you don't require to many amps. 10ah give your current usage should give you loads of range.

What size controller do you have?

 

[morphix]

Following up this thread for those with an interest in my eBay LiFePo4 battery failure...

I did manage to the contact the seller and after requesting photos of the battery he told me that he can no longer supply suitable individual replacement cells as he's changed supply.

So it looks like I won't be repairing this battery now. I will however keep the battery and maybe if anyone else on the forum who bought the same battery type experiences a cell failure I can supply them with replacement cells.