Lifepo4 48v won't go beyond 42.2v

[hackspacenige]

After one ride of 30 miles at 30mph motor stopped and two bms components were fried and replaced.
25ah battery (£430 on ebay)
Any advice appreciated

 

[dmwahl]

More details would be helpful, such as the BMS model, which components were fried, and what all the cell voltages are when it stops charging at the least. You should also delete the duplicate threads.

 

[hackspacenige]

Thanks for reply. Bms came with battery which came from sunthing-28 ebay seller.
Wasn't certai I was measurinv correctly buytI was getting mostly 2.7v and a 1.8v. Components burnt out were larger square black ones with three legs.
Thanks again

 

[dnmun]

i will repeat what he asked. post up information we can use to help you. you only need one thread so delete the others.

which mosfets burned? post up a picture of the BMS. what is the cell voltages of each of the cells? list them here.

how did you burn up the mosfets? did they burn out during discharge or during charging?

did this happen the first time you used the battery and before it had a chance to balance? did you charge the pack before using it the first time? did it fail on first use?

 

[cam3087]

This is the first I've heard of a sunthing battery getting toast. Hope you can get it fixed cuz I get a feeling the eBay seller is just that, a seller not fixer good luck

 

[dogman dan]

At this point, I'd be trying to get the battery recharged without the bms.

Remove the bms, then monitor the voltage of each group of cells as it charges. Once one cell gets above 3.8v, you can put rig a 12v turn signal light bulb to discharge that cell a bit, before putting the whole pack back on the charger.

Tedious as hell, but you will find out immediately if you have any problems with the cells themselves. (like if one group won't hold any charge)

Later on, you can figure out if the bms can be repaired, or replace it with another from a different source if sunthing won't send you one.

Why did the fets on the bms melt? I don't know, but I suspect you hit the battery pretty hard. 30 mph should not have been much more than 20 amps continuous though, so no reason to think you "killed it" . Just not wise to test things to their limits right out of the box, if that was your first ride.

 

[hackspacenige]

]Hi - Have removed repeated threads - don't know how that happened.
I've measured the voltages - am getting:
2.68
2.68
2.64
2.59
2.81
2.81
2.77
1.84
3.19
3.05
3.2
3.09
2.85
2.13
2.41

It failed on discharge at the end of it's first long ride. I had charged it overnight beforehand and was getting 53ish volts.
Struggling to get a photo on the forum. I suppose I have to make photo a lower quality then put a link to flickr.

 

[dnmun]

set your camera to VGA and the pictures will fit here.

the cell at 1.84 is way low. the BMS should have cut out before it got that low. i expected the voltages would be higher if it is on the charger so i think your charger is dead so find a charger that works and put it on the battery and then measure again while it is charging.

ok, edit, forgot this is the one that has the dead mosfets, that is why it won't charge. your charging mosfets are burned open circuit so they won't allow current to flow.

if you have a voltmeter we can walk you through how to test the mosfets.

 

[hackspacenige]

thanks dnmun
that's really helpful - will have a look on ebay.
Now first I'll try with photo

 

[dnmun]

load the picture to your disk. in the dialog box under the posting section there is 'upload attachment', then clip 'choose file' and go find your picture on the disk, and 'open' it and after it is in the upload box then click 'add file'.

post your location in you personals, i have some parts and 16S BMSs too.

 

[hackspacenige]

http://www.flickr.com/photos/87981029@N02/with/9971415974/

Had to go the flickr way - couldn't find vga on cam.

Photos 2,3,4 show damaged BMS before I replaced mosfets
1,5,6,7 show what it looks like now

Thanks

 

[dnmun]

to test the mosfets with your voltmeter, use the diode tester function. the mosfet has a body diode between the source and drain. the mosfet legs are gate, drain, source from left to right.

those are not the same BMS.

to test the charging mosfets, the ones connected to C-, you put the red probe on the source leg and the black probe on the drain leg. measure the voltage from the diode tester and record it. then reverse the leads and measure the voltage, red on the drain and black on the source. it should read open circuit.

 

[hackspacenige]

right - sorry other photos are of the first bms sunthin-28 sent me with the battery. That one died whilst charging. The second one is the one he sent me as a replacement (a used one) When the second one burnt out on the ride (admittedly long ride ) I used unburnt mosfets from the old bms to replace burnt ones.

Recommended charger - any that do load test and lithium equivalent of disulphating par chance

 

[dnmun]

you don't need a charger. the BMS charging mosfet is burned open circuit. did you test the diodes you used as replacements? why did you burn up so many sets of mosfets? especially the charging mosfets?

 

[dogman dan]

At least he's not at 0v on any cells. That's good news. The battery itself should be recoverable. You don't need desulphinating, just charging and balancing.

I'd be putting the charger back on that pack with the bms completely removed, at least until one cell got charged up to 3.8v. At that point, you could use a 12v light bulb to discharge the high cell awhile, then put the charger back on.

Whenever you do get the bms fixed, it would be nice to install it onto a battery at least a little bit better balanced than what you have now. It would be nice to know all your cells are reasonably ok, by charging them, manually balancing the pack, and then seeing if any cell groups start discharging themselves a lot more than the others. With luck, you balance the pack, and a few days later still have 55-56v.

 

[dnmun]

i agree. the 1.84 is not fatal but you should try to charge it without the BMS to get it above 2.5-3V or so. the longer it stays below the 2V level the worse off it is. just don't let the high cells go above 3.9V.

you will need a new BMS. if you can test the mosfets on that one we could know something. but it sounds like they are dead. either open or shorted. if the output mosfets are shorted it would have allowed you to run the pack down below the LVC and that was why the cell got to 1.84. it might have been much lower. under load. so it could be bad.

 

[dogman dan]

That's what I was thinking too, get that low one higher at least, see if it can hold a charge. If a cell or two is toast, then that may change his decision on fix or replace the bms, and which bms if replace is the choice.

 

[hackspacenige]

Getting a replacement bms would be very welcome if you have got one bnmun. I'm based in Bristol, UK. I'm off to France til Thurs but if you or anyone has a bms that would work I'd be happy to buy it. Otherwise won't be able to tinker until friday. This sounds promising though folks. Thanks for all your advice.

 

[hackspacenige]

Back in action. Me not the battery. Have done the diode testing.
Getting
1
1
1
1
001
001
001
001

The latter 4 being the burnt/replaced ones. So would like certainly to buy a replacement bms if still on offer.
Meanwhile, a guy at my local hackspace has recommended I get a bench power supply to charge with variable curr&voltage&curr limit& states curr. ato enable me to charge indiv. cells. Any recommended models?

 

[dnmun]

i think BMSs from henry at Bestechpower is the best bet for you. the D167 is good to 35A continuous and 70A cut out. the D166 goes to 40A continuous and 80A cut out. it should cost about $48-50 for one and $12 shipping to UK. cheaper than i can send you one.

here is what they offer in 16S lifepo4, go back to the main page for their contact info and request a bid for the D167:

http://www.bestekpower.com/512v16spcmbmspcbforlifepo4batterypack/

 

[hackspacenige]

thanks dnmun
have managed to charge the 1.85v up to 2.6v and it's staying there days later. Have also ordered power supply and a BMS (or two - min. order) from Henry.
Very grateful for your help.

 

[dnmun]

henry told me he was at a trade conference so i put off my order until he gets back the 18th, i am ordering 4 of the D167 and the 4 more of the D126 then. but i posted up the other day asking if anyone wanted to join a shared shipping order and got no hits.

i have been able to help a few people this way. just cumbersome and i have to put up a lot of cash up front to make it work with henry. they are a manufacturer and wanna ship hundreds at a time. they really do seem to be the best source in china so i think they will be successful long term. but nobody buys a lipo pack with a BMS since no manufacturer wants the liabilities of selling a big lipo pack. hobby king does it because they sell to the RC guys who only use them outside on model planes and not being charged in a an apartment building with 100 residents who can be burned out when the balancing charger shorts out charging an ebike.

i was reading on the evchargernews.com network that the parking garage at the disney parking garage in downtown LA has now posted signs restricting people from charging because they are concerned with the risk of fire. they will disconnect and file a complaint if they catch you charging from the 120V outlets. i am sure it is because the insurance company for the garage has put that in the contract.

the insurance companies are getting wise and before long nobody will be able to take an ebike inside a building or an apartment building and charging batteries inside an apartment will become against the building rules.

 

[hackspacenige]

well done - so thank-you for talking him down to accept order of two.
What are you doig with them all then. You have alot of bikes/projects?

 

[dnmun]

i try to sell them to people to use. by keeping the shipping low and buying enuff that he gives me a better price i can send them 1st class mail to other people for my cost and save them 30% so it makes it easier to try using them. i am just trying to promote the use of BMSs on these lipo packs to prevent more fires from balancing chargers and bulk charging with no BMS.

the insurance industry has already taken notice of these fires and if we cannot stop it then nobody will be able to take their ebike inside a building anymore. plus if i can just prevent one more house or apartment building fire it will be worth the expense.

it is really petty change when i have money, but sometimes i have no money too, so i can relate to people thinking they don't wanna spend money on a BMS for their lipo because all the cool people don't use them either.

the reason nobody uses them is because the lipo comes from HK in these pouch only formats for RC apps. if they were built with a BMS from the factory then people would already be using them.

add a little complexity of wiring up a BMS, and it is easier to take the simple way out imo. but i don't consider them complex.

 

[hackspacenige]

Hi Dnmun
Just received two 16s d167's - had a thought - it is definately 16 and not 15 I needed is it?
I balanced the cells manually and with power terminals on 1-2, 2-3, 3-4, 4-5 .....and only charge 15 cells. Is there a way to charge the 16th if I have one?