48v 15Ah LiFePO4 battery testing... Is my BMS dead???

[Sacman]

I started this topic because discussion about my BMS started to go off into a tangent direction from the original topic "Is Jimmywu66 & EastUnitedMusic the same person?" ( http://endless-sphere.com/forums/viewtopic.php?f=14&t=6034 ) I started this topic in order to preserve the original discussion about possible "rouge" battery vendors on eBay in the original thread.

The BMS is part of a package deal (48v LiFePO4 battery, charger, and BMS) that I bought from a battery vendor on eBay called EastUnitedMusic (Dan).

Here's what the BMS looks like with the black heatshrink cover removed.

It is a 16-channel BMS. Here's the view from the front.

... and here's a view from the back.

Sooo.... I tried charging the batteries he sent me 3 weeks ago and it looks like the BMS is shot and doesn't work. No signs of voltage increase on my volt meter even after 20 minutes on the charger. DAMN!!! So I unhooked the BMS and connected the charger directly to the red and black wires of the battery...and it looked like it started charging. I noticed the built-in fan in the charger kicked into hi gear. Thirty minutes later I try the battery on my ebike and went for a short but successful 3 mile ride. Well looks like it worked so I'm fairly relieved. :wink:

Lifepo4 don't act like sla's. It will take a long time to show any voltage change charging. Not 20 minuites. Charge for 12 hours through the normal charging wires before you decide it doesn't work. The full capacity of the pack will take about a half dozen cycles or more to develop.

That very night I plugged the BMS back onto the battery and tried charging the battery with it on again. But this time I left it on ALL NIGHT (probably about 10 hours). It still doesn't charge. My volt meter showed the battery started at 49.2 volts at night and it still showed that in the morning. The 2 lights on the charger stayed 1-green and 1-red the whole time (indicating charging complete). And the charger's fan speed stayed at low rpm's. I dunno... it's really looking like the BMS doesn't work.

Okay so just to verify... I unplugged and removed the BMS again and directly connected the charger back to the battery leads. Right away I could see and hear the charger's fan speed jump into higher rpm's and the 2 lights on the charger both turned red (indicating charging). My volt meter showed the battery's voltage started to slowly increase and in 5 minutes was showing 50.6 volts.

I left it to continue to charge and came back to check on it 5 hours later. Looked like it was done charging. The charger fan was back down to low rpm's and the charger's 2 lights showed 1-green and 1-red again (charge complete). The battery voltage displaying on the volt meter was showing 55.4 volts.

I wrote an Dan (aka EastUnitedMusic, aka Mr. Lau) and told him about the BMS appearing to be bad and although I was glad to see the battery works but I was still concerned that I didn't have a working BMS to protect the battery. I asked him if he could help me fix the BMS. I wasn't really expecting an answer from him since he ignored my many emails in the past but Wow!... he actually replied with an email to me later on that night.

In his email he told me to remove the BMS when charging the battery and plug it back in when I use the battery on my ebike. Huh? ... well that's a "screwy answer". Sure the LVC on the BMS will prevent excessively deep dischages on the battery, but with no BMS during charging there's no protection from overcharging and no cell balancing.

So what do you guys think? Is this BMS dead? Should I just order another one from Ping?

 

[Link]

I think it's dead. Mine seemed to have failed in exactly the same way.

Do any resistors on the board get abnormally hot, perchance?

 

[Sacman]

As for the review on battery so far...

The 48v 15Ah battery drives my c-lyte 408 motor on the back wheel, and I'm using a C-lyte Journey 20A controller. The bike's top speed on level ground is around 24mph and when climing hills around 5% grade it slows down to about 17mph. It is a very significant jump in performance from my previous battery setup (36v 10Ah NiMh pack).

Except for the problem with the BMS the battery seems to be performing pretty well. I'm actually fairly happy with this setup for now. This morning I just came back from a 26 mile bike ride and over half of that distance was over rolling hills with about 5% grade. I'm normally pedalling most of the time too. The last 2 miles was on mostly grades of 10%. Yesterday I did a 21 mile bike ride over similar terrain but with stronger gusting headwinds.

The 3A charger seems to be doing a decent job too. I've done about 9 charge/discharge cycles now and batteries seem to get to full charge in about 4 hours and shows a final voltage of 54.4 volts. Does anyone know if this voltage sounds about right? I'm a little concerned that this voltage might be a little on the "low side". Reason why is because I contacted Ping a week ago to try to buy one of his 48v LiFePO4 chargers and he said his chargers actually put out 61 volts. What should the correct voltage on these chargers be? Does anyone here know?

 

[Sacman]

I think it's dead. Mine seemed to have failed in exactly the same way.

Do any resistors on the board get abnormally hot, perchance?

Nope... the BMS stays pretty cool to the touch. Which is another thing that makes me believe it's dead... I've read others who say thier BMS get's hot to the touch when charging, and so I was expecting mine to get hot too.

 

[dogman dan]

yep, I'd say your bms doesn't work for charging. Bummer. I just thought maybe you had the idea a 30 minuite charge was possible. So no warranty on your bms? I'd say you are officially ripped off then. If your pack was balanced and broken in, I'd say no problem to charge with no bms, but I think you may be flirting with cell damage if you never get an initial balancing. the damage will occur discharging. Charging lifepo4 is pretty forgiving, and very tolerant of overcharging. The actual voltage of your pack will be a lot more than 48, for my 36v it's about 44.5. If you want to undercharge a bit, use a regular cheap scooter charger. It will shut off a volt or two less than the lifepo4 one. The charging cycles are the same.

 

[dnmun]

send him that pair of pictures. there is an active device in the upper left corner which was hit and damaged, you can see how it is displaced, maybe the leads have broken inside the package or torn out of the solder itself, and the red wire with the bare end should be attached to the PCB somewhere, it is the charging lead to the BMS, i suspect he will send you another BMS immediately, but i think there is enuff smarts around here that someone can tell where it should still be attached if he doesn't, too bad they scratched off the part numbers, but i think you have enuff in those pictures to get them moving and they could even allow you to keep the old BMS too, if you are lucky. ask them if you can keep it to try to repair it, but deliver another. BOL

did you get 3 different packs with a BMS for each pack? why not one 15Ah pack instead or do they only have 5Ah cells? prismatic cells?

don't worry too much about over charging, if you keep it to less than 3.7-3.8V/cell, but you will have a new BMS soon i think.

ps: i have identical tile in my basement sunroom, but i'm gonna use white grout. thanks for the pictures, dennis

 

[Rassy]

dnmun wrote:

and the red wire with the bare end should be attached to the PCB somewhere, it is the charging lead to the BMS

I belive the red wire with the bare end should just be connected to the main positive wire from the battery. It just completes the circuit to the battery for charging purposes.

 

[leamcorp]

I think it's dead. Mine seemed to have failed in exactly the same way.

Do any resistors on the board get abnormally hot, perchance?

Nope... the BMS stays pretty cool to the touch. Which is another thing that makes me believe it's dead... I've read others who say thier BMS get's hot to the touch when charging, and so I was expecting mine to get hot too.

Yep I think its dead too. Mine get super hot ... to a point where its will melt anything it touch. I now have a metal casing for the BMS. This doesn't seem to affect the operation of the battery. Also, I'm thinking that individual cells used are protected.

The final charge on mine is 55.4v

I've just ordered a 48v BMS from Ping. I'll let you know if that worked.

 

[Sacman]

dnmun wrote:

and the red wire with the bare end should be attached to the PCB somewhere, it is the charging lead to the BMS

I belive the red wire with the bare end should just be connected to the main positive wire from the battery. It just completes the circuit to the battery for charging purposes.

Yes, that's the way I connect the red wire (to the main positive wire on the battery). I was told that was the correct way to hook it up.

 

[Sacman]

send him that pair of pictures. there is an active device in the upper left corner which was hit and damaged, you can see how it is displaced, maybe the leads have broken inside the package or torn out of the solder itself, and the red wire with the bare end should be attached to the PCB somewhere, it is the charging lead to the BMS, i suspect he will send you another BMS immediately, but i think there is enuff smarts around here that someone can tell where it should still be attached if he doesn't, too bad they scratched off the part numbers, but i think you have enuff in those pictures to get them moving and they could even allow you to keep the old BMS too, if you are lucky. ask them if you can keep it to try to repair it, but deliver another. BOL
thanks for the pictures, dennis

Hmmm... you're right... that little device in the upper left corner does look like it was hit or damaged. I'll send him the pictures and see what he says. It's worth a try.

 

[Sacman]

did you get 3 different packs with a BMS for each pack? why not one 15Ah pack instead or do they only have 5Ah cells? prismatic cells?

dennis

And yes you're also right that I got three 5ah battery packs each with it's own BMS (instead of ordering one big 15ah pack). I ordered the three 5ah packs to give me flexibility to mount them on different locations of my ebike to distribute the weight. It was his mistake when he had my 3 battery packs made with 3 separate BMS's he told me he lost money on the deal because if it. It's the only thing that makes me feel better about these BMS's not working. Maybe I can try different things or canibalize parts from one to get the others working.

 

[Sacman]

don't worry too much about over charging, if you keep it to less than 3.7-3.8V/cell, but you will have a new BMS soon i think.

Hmmm... I forgot where I saw it, but somewhere I thought I read that the nominal voltage on these cells was 3.2 Volts. Now I got a 16 channel BMS. So my battery pack voltage should be reading (3.2) x (16) = 51.2 Volts when fully charged right? Well after leaving the charger on the battery for 12 hours my volt meter holds a reading of 54.4 Volts ... which translates to (54.4)/(16) = 3.4 Volts per cell. So I guess I don't have to worry about overcharging ... does this sound right?

 

[dnmun]

i agree the red wire should be attached to the battery plus terminal, since the B- is what gets switched by the fets across the end to disconnect the battery. don't scrap anything yet, you should be able to get parts, but i think he will send you a new BMS. that package part in the corner could be all your problem. this is for richard or gary to decipher i think.

 

[recumbent]

leaving the charger on the battery for 12 hours my volt meter holds a reading of 54.4 Volts ... which translates to (54.4)/(16) = 3.4 Volts per cell. So I guess I don't have to worry about overcharging ... does this sound right?

My LIFEPO4 bats top out at around 3.6 volts each, and settle down to about 3.4 volts when the BMS tries to balance the pack after a few hours on the charger.
These are different batteries than yours, but just to let you know that charging up to 3.6 is just fine. They never heat-up either. My BMS does not balance properly like many people, we are just going to buy Garys & Richards BMS when they get the parts sorted. Then we will have a good BMS that we can all fix together on this forum, if need be

 

[Sacman]

leaving the charger on the battery for 12 hours my volt meter holds a reading of 54.4 Volts ... which translates to (54.4)/(16) = 3.4 Volts per cell. So I guess I don't have to worry about overcharging ... does this sound right?

My LIFEPO4 bats top out at around 3.6 volts each, and settle down to about 3.4 volts when the BMS tries to balance the pack after a few hours on the charger.
These are different batteries than yours, but just to let you know that charging up to 3.6 is just fine. They never heat-up either. My BMS does not balance properly like many people, we are just going to buy Garys & Richards BMS when they get the parts sorted. Then we will have a good BMS that we can all fix together on this forum, if need be

Thanks, that helps me understand and feel better that a little overcharging is okay.

If you do get a new BMS won't you have to take your battery pack apart and solder wires to each cell?

 

[swbluto]

That BMS looks *exactly* like the one used for my Ping 1.0 48V 10AH battery, resistor for resistor, amp-for-amp,mosfet-for-mosfet. Except the chip-identification markings on mine are still intact.

*wonders why they're blurred on yours*

 

[swbluto]

don't worry too much about over charging, if you keep it to less than 3.7-3.8V/cell, but you will have a new BMS soon i think.

Hmmm... I forgot where I saw it, but somewhere I thought I read that the nominal voltage on these cells was 3.2 Volts. Now I got a 16 channel BMS. So my battery pack voltage should be reading (3.2) x (16) = 51.2 Volts when fully charged right? Well after leaving the charger on the battery for 12 hours my volt meter holds a reading of 54.4 Volts ... which translates to (54.4)/(16) = 3.4 Volts per cell. So I guess I don't have to worry about overcharging ... does this sound right?

My ping charger stops doing the bulk of its charging when the cells reach 3.4 volts. Past that, they seem to slowly float to 3.8 volts and then the charger just stops charging past that. If you did what I did, and just constantly plug and unplug the charger, you can get the cell voltages to 3.9 but they never went past that. The 3.4-3.7 resting voltage range seems to be a "superficial" voltage so-to-say, as it pretty much disappears past some rather nominal discharge(like less than .5 ahs).

It seems my 48V battery is 60V hot off the charger, usually, so that's about 3.75 volts per cell.

 

[Sacman]

Thanks for sharing that info. I've just been studying the discharge curves for these LiFePO4 battery cells and your explanation just clarifys and confirms my understanding now. A major desireable benefit of thes LiFePO4 is their ability to hold a relatively constant voltage (discharge plateau) over their discharge cycle. The specs on these cells show that the discharge plateau occurs between 3.2 and 3.3volts. This roughly translates to about 51 to 53 volts. So any stored charge higher than 53 volts is gonna get used up quickly (just like you explained).

 

[dnmun]

after your comments on the chinamusic thread, i wondered why you never attempted to get a replacement BMS. i now realize you actually have 2 others on the other packs. do you swap the BMS from one of the batteries that will charge and balance and use it on the battery without the BMS? did you ever try to bridge source to drain on the charging fet on that bad BMS with the damaged opto and see if your charger would then turn on?

also, what is with the avatar? is it some sort of double exposure you did of yourself or did you toss some marbles in front of the camera? are you a magician or is it just an ego image?

did you ever actually measure the charge current as the voltage reaches 3.4 up to 3.8V? how rapidly does it drop off in that range and can you tell where the charging current drops to 100ma or less? that could be the final shunt current, maybe pile up the cumulative resistance of the shunts and divide that into the final voltage to determine where it won't accept charge anymore.

just curious, do all the batteries have the same profiles now? or do they show some variation in standby voltage? can you tell how balanced the cells are? does the BMS really balance the cells or does it have the same variations in voltage as the battery without the BMS?

and you never had the problem so many others had with the overheating output fets either. is that because the current is so much lower from the 5Ah packs? seems most likely, just wondered if your fets got hot. sure wish there was knowledge of which fets would work as replacement for people like mr R2 and leamcomp. i could not tell what fets were there from mr R2's picture, sure wish i knew more electronics stuff.

 

[Sacman]

Okay I'll try to answer as many of your questions as I can.

after your comments on the chinamusic thread, i wondered why you never attempted to get a replacement BMS. i now realize you actually have 2 others on the other packs. do you swap the BMS from one of the batteries that will charge and balance and use it on the battery without the BMS? did you ever try to bridge source to drain on the charging fet on that bad BMS with the damaged opto and see if your charger would then turn on?

I simply have been busy fixing other, more important problems with my ebike. Have you seen my other recent posts? I felt that those problems required more immediate attention than getting my BMS's to work since I've been using my batteries fine without the BMS so far.
1) I kept getting lots of flat tires about 2 months ago. I went thru the whole hassle of trying various thicker tire tubes and puncture resistant tire liners and even Slime tubes. It turned out that the types of flats I was getting were pinch flats from running too thin road-like tires and rolling over small potholes in the streets. I didn't realize it until I changed up to wider tires. This problem took me awhile to figure out it got as bad as me getting 2 flats in one day.
2) The cheap galvanized spokes on my Crystalyte rear wheel (with motor) kept breaking and I finally got the whole wheel re-laced with high quality GT brand stainless steel spokes.
3) My half-twist throttle is stuck on WOT. I've traced that problem with something wrong in controller ( not in the throttle itself).

also, what is with the avatar? is it some sort of double exposure you did of yourself or did you toss some marbles in front of the camera? are you a magician or is it just an ego image?

I'm a fan of the movie The Matrix. My avatar is not a picture of me... it's a picture of the main character of the movie (his name is Neo). It's a picture of Neo using his mind to stop a bunch of bullets that were fired at him. The main storyline of the movie shows how Neo slowly discovers that he has God-like abilities and powers that he uses to eventually to SAVE THE WORLD AND THE HUMAN RACE. Near the end of the movIe one of the abilities he learns he can do is to stop bullets. I thought it was apropriate... as members of this forum we all learn new abilities to help SAVE OUR PLANET.

did you ever actually measure the charge current as the voltage reaches 3.4 up to 3.8V? how rapidly does it drop off in that range and can you tell where the charging current drops to 100ma or less? that could be the final shunt current, maybe pile up the cumulative resistance of the shunts and divide that into the final voltage to determine where it won't accept charge anymore.

Sorry I don't have an ammeter. I only have a simple volt meter and that's what I've been using to check the battery and BMS.

just curious, do all the batteries have the same profiles now? or do they show some variation in standby voltage? can you tell how balanced the cells are? does the BMS really balance the cells or does it have the same variations in voltage as the battery without the BMS?

I've measured all the cell voltages (3 x 16 = 48 total) of each of the 3 battery packs about 4 times now during the 2 months I've had them. I take my measurements at the connector plug of the sense wires. 46 of the 48 cells read very close to 3.4V at final full charge. Cell 47 reads 3.6V and Cell 48 reads 3.7V. Resting voltages of all the cells (after an hour) range from 3.22V to 3.28V. To me that looks like a pretty darn good balanced pack of cells (considering I'm not using a BMS). That's also why I didn't put to much priority on fixing the BMSs.

and you never had the problem so many others had with the overheating output fets either. is that because the current is so much lower from the 5Ah packs? seems most likely, just wondered if your fets got hot. sure wish there was knowledge of which fets would work as replacement for people like mr R2 and leamcomp. i could not tell what fets were there from mr R2's picture, sure wish i knew more electronics stuff.

When Charging...
Yeah that's interesting and possibly true...each BMS is actually handling 1/3 the current ... makes sense that it could be why the FETs are not getting overheated... but only for discharging. I only have a 3 amp charger which means the BMS is only seeing 1 amp during charging. I'm very sure the FETs and the rest of the BMS can handle more than 1 amp of current and shouldn't get hot.
But when I had the BMS's hooked up to my batts the charger never turned on... so the FETs never had a reason to get hot ... at least for charging.
When charging without the BMS the batts get warm but only very slightly. The rise in temperature is so small that you might miss it if you took your hand away from the batts too fast.

When Discharging...
When I had the BMS's hooked up to my batts I only had 6 miles of range then the LVC cut the off. I remember only 1 battery pack was actually pretty hot to the touch. The other 2 battery packs and 3 BMS's were not...they were cool.
I've been using my batts and riding around without the BMS for 2 months now. The battery packs stay pretty cool most of the time. The only time they get warm is when I've been riding up long steep hills. And even then ... they're just warm (not hot).

I'm thinking it has a lot to do with my riding habbits too. You see unlike most ebikers on this forum... I actually peddle moderately hard and keep my speed close to max speed (24Mph) most of the time...keeping the current draw from the batteries fairly low. I only have 16-gauge wire between my battery and controller. And that wire never get's hot.

 

[dnmun]

thanks a lot, i had heard of the movie, but never seen it. in that image you can see him projecting some mental force almost. concentrating behind the shades.

i learned a lot from following you guys problems. good thing you started these threads. sounds like the cells are well balanced, i wonder if the LVC is set to some bogus number too, that jimmy wu has no clue about, and that was why your pack cut out early. the open source BMS thread is all i know about them, i am now beginning to see how their idea of how the BMS should work involved eliminating the BMS control over output, and therefore the energy and voltage loss across the fets, and the individual cell level LVC to protect the individual cells.

you could bypass the output fets too, just like the shorted output on nopingbubba's pack where the melted solder shorted it out. jimmy wu really should have used better fets, these have such high resistance that they overheat and melt the solder they get so hot.

you should be able to use your voltmeter as an ammeter with a 10A scale, and all the charging current could go through the ammeter, and as it dropped you could shift the leads down into the lower ranges to see how the charging current goes. you need alligator clips on jumpers, connecting the ammeter in line from C-(the black charging lead) to battery negative terminal (or the lead coming off that charging fet).

do you have an idea how to help R2 or some of the others? do you think he may have some bad cells or that his battery just doesn't charge up enuff?