Vpower / CamyCC Battery LifePo4 48v20ah rebuild

T

tstephens

10 mW
Joined
Nov 6, 2013
Messages
24
Location
San Diego, CA
EDIT: I gave up on this repair, but wanted to add additional info on the BMS pin out in case someone else needs it.

I picked up my first eBike at a local auction this week. I got it for a steal, but the downside is the battery pack will not charge over 40.8V. The pack is one of those duct tape Vpower CamyCC LifePo4 48v20ah batteries. I opened the pack up and found a loose battery tab on one half of the pack, there may be others. I'm still working to disassemble the other half of the pack. There are miles of duct tape wrapping this beast.

I charged the battery with BMS for 4 hours and noted that the voltage was only 40.8V with the BMS disconnected. Here are the battery voltages as measured from the sense wires on the battery with the BMS disconnected.

Here is a picture of the BMS pins. (Picture was taken from the manual on: http://www.batteryspace.com/prod-specs/6128_2.pdf, which appears to be a similar BMS)


1 = B- & B1 (Place - probe on B-, + probe on B1)
2 = B1 & B2 (Place - probe on B1, + probe on B2)
3 = B2 & B3
:
15 = B14 & B15
16 = B15 & B+


1: 3.245V
2: 3.236V
3: 267mV
4: 3.258V
5: 3.267V
6: 1.878V
7: 3.255V
8: 2.998V
9: 320mV
10: 3.260V
11: 3.054V
12: 3.234V
13: 3.227V
14: 3.229V
15: 334mV
16: 3.282V

I'm a newbie with a computer engineering background and can do basic electronics repair.

If possible, I would like to replace the dead/dying strings (3,6,9,15). Does this seem feasible? Is it simply a matter of cutting the strings out and replacing them with new and re-balancing?
 
It doesn't look good. I have repaired two of these packs I can tell you its a pain.
If you want an education on large series parallel batteries, It may be fun to mess around with.
Recheck and resolder any broken tabs you find but as it stands my advice would be to make a smaller pack out of the best strings of batteries, only use the ones that are currently holding over 3.2v. I would not use #3,6,8,9,11,15 the rest are pretty close so it may be possible to restore them at a reduced capacity if your lucky you could make a 10 ah pack out of this. Then charge them for 24 hrs so they will balance do a shallow discharge and repeat. Instead of buying new batteries to replace the bad ones put that money towards a decent battery pack this will save you money in the long run.

You may be able to find good cells in string # 8 and 11 if you have the right tools but its a huge amount of time and not really worth it imo.


Mushy
 
I feel your pain on the disassembly process. I just tore my 15ah vpower apart and it was a huge pain.
All my cells were ok but I did exactly as suggested above and broke it down to more manageable configurations and then built it back up.

How long did you leave it on charge and balance?
 
Thanks for the replies. I'm thinking breaking the pack down into a usable Ahr size is the best approach, limiting my investment in time and $$.

This battery is of unknown condition, I assume it has been sitting discharged for quite some time. One of the end cells was dented/crushed at the top.

I had the pack hooked up to the charger for 4hrs. I took bulk measurements at 1hr, 2hr, 3hr and 4hr. I stopped charging at 4hrs, because the pack voltage was peaked at 40v with little to no change over the previous hour. I was afraid that this battery of unknown condition may be damaged and could cause an accident while charging.

Is it safe to continue charging, or should I just jump in and start cutting?

Right now the only charger I have is the bulk charger that feeds the BMS. Do I need to acquire another smaller charger to individually charge the strings to verify the health?
 
I would not charge it. One of the groups is jumping up to full voltage very fast tripping the bms. I had cells that would jump to full voltage very fast and trip the bms then drop off immediately after to 1 volt. any energy that goes into these cell just turns into heat.

You can use a volt meter and the bulk charger to reconfigure the pack if the bms is working properly.
You will need 16 groups of batteries figure out how many good cells you have and divide them evenly.
label each bms wire take pictures of where they are draw it out if it helps.

I reused the tabbing material from the crap groups to solder the pack back up. Only remove the tabbing where you need to.
 
post up a picture of the battery now that you have it open. can you verify the voltage measurements of .3V on those three cells or is that because the sense wire is broken?

we have parts for them but the BMS could be damaged by the previous owner and caused the 3 channels to die, but you need to use a watt meter to measure the current so order one from ebay.

http://www.ebay.com/sch/i.html?_trksid=p2050601.m570.l1313.TR0.TRC0.X60V+100A+wattmeter&_nkw=60V+100A+wattmeter&_sacat=0&_from=R40

you need to measure the cell voltages while it is charging and you should use the BMS to charge it if you don't know how to keep it from overcharging. if you charged it without the BMS already that could be why the 3 channels are now dead.
 
I managed to get the battery fully unwrapped tonight. My original voltage readings were verified on the cells themselves, so this does not appear to be a faulty sense wire.

Here is picture of the battery unwrapped and filleted open.
battery unwrapped 2.jpg
battery unwrapped.jpg

I started to take voltage measurements when charging, then noticed that the battery cable connector was not always making a connection. I tugged on the back of the connector and both wires came out! I disassembled the connector and re-crimped the connections and for good measure filled with soldier. I seemed like the connectors were a bit undersized for the size of the wire.

negative.jpg
positive.jpg

Picture of the BMS, no visual faults on the board
bms.jpg

After refitting the battery connector, I started up the charger with the BMS attached ( I have always charged with the BMS attached ). I took some voltage readings and for the most part they were the same, except for the <1V cells.

3 - 360mV
6 - 1.943V
9 - 467mV
15 - 482mV

I was expecting to see higher voltages when charging, is that not the case? The charger is putting out 59.2V

I ordered one of those eBay watt meters. How do I go about conducting the current test?
 
Noe that is a sizable battery.
Make sure to remove all cells below 1.5V and recycle them. These cells might look OK, but actually they can start a very nasty fire one day due to internal short. These internal shorts develop due to too low voltage.
 
Don't start cutting it up yet. Though unlikely, it is still possible that the pack is now just so out of balance it cannot possibly get balanced by the slow bms that is designed only to handle slight inbalance.

Try to charge the lowest strings up to closer to 3.5v manually, like with a low voltage power supply like a phone charger or other 5v wall wart. Then put it back on the bms and charger, and see if it will balance, hold some of that charge overnight, etc. It may still have such low capacity that it cannot run the bike at any speed without exceeding the max discharge rate.

But if that fails, then..

What I would do, is get something else to run the bike. Why? Because these packs are so notoriously bad. In general, lots of the problems come from mechanical faults at the tabs, or wires to the bms. But yours seems to have truly dead cells randomly scattered through the pack.

To me, that's a big red flag that the remaining strings don't have a ton of lifespan left in them. Not worth it to add any new cells, unless they are other used cells obtained close to free. And then, again, it might be running at 4c when you use full throttle. This type of cell needs to remain below 2c discharge rate.

But you will have some cells to play with now, for other purposes. A large 12v pack can power all kinds of portable stuff, like a boom box for example. Lighting for the bike is another obvious use. A 12v lighting pack on the bike will also charge you phone, etc, with a cigarette lighter port.
 
i had the same disconnected crimps on the connector of the Vpower pack i worked on too. good to know it wasn't just my battery they did it too, they do it to all of them. you don't wanna use that kinda connector anyway.

to identify what kinda shape the battery is in, start by charging up the low cells with a single cell charger. you don't have to charge through the BMS for that since it is only for one cell at a time. most of us use the cell phone charger wall warts but you can buy a lifepo4 single cell charger.

the cell phone wall warts put out 5V so they will destroy the battery cell if you leave it charging unmonitored. you can find them in thrift shops or in the dumpster is where i find a lot of them. cut the plug off the end and solder alligator clips on the wires. make sure to label each alligator clip so you always know which is positive and which is negative.

then clip the cell phone charger onto the ends of the cells, on those parallel strips on each end of the cell. make sure to verify it is the correct one to charge and that you have the correct polarity before proceeding. you have to monitor the cell voltage constantly because it will go from 3.4V to 4V in a few minutes when it gets full. over 4V is very very bad. if you charge through the watt meter you can measure the current so you will have an idea of how much charge it is taking. all of the low ones should take up to 20-21Ah of charge through the meter.

then charge the low cells up to 3.65V and stop charging then, disconnect the clips and go to the next one, until all 4 have been charged. then reconnect the BMS and charge the pack up to full charge. then you can use the watt meter to see how much charge the low cells still can store by discharging the battery into a dummy load like an electric oil radiator type space heater.

that is when you will finally know enuff to make some decisions. until then do not cut it apart. i have a buncha those cells and i think AW amy have some too so if you have some dead rows we can provide parts.
 
Amberwolf - Your http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=22750&start=0 thread was my inspiration for even considering tackling this project.

Right now I have a cell phone charger (5v 800mA) hooked up to string 2 to see if I can get it to charge up. The charger leads are being held to the battery using rare earth magnets to hold it down. The charging is a very slow process, like watching paint dry. Right now I have a DMM hooked up with a webcam pointing at it. Currently at 3.31v (started at 3.26v) That way I can monitor it while working on the computer. How closely do I have to monitor the charge?

Next week I will pick through my works scrap pile to get some extra cell phone chargers, so I can do this operation in parallel.

Should I even attempt to charge the strings that are < 1V? I assume those are a lost cause until the dead cells are removed.
 
do not charge the cells that are already above 2V. you only need to charge the cells that are down below 2V and when they get charged up then you can use the charger to finish charging them. you should be using alligator clips and measure the current through your ammeter function on you voltmeter. use another voltmeter to monitor the voltage.
 
Thanks for the clarification.

I hooked everything up with alligator clips and moved the charging/monitoring setup to string 6. The voltage is now 3.07v @ .88A (1.878v prior to charging).

What should I expect to see with respect to the current measurements as I approach 3.4v?
 
your wall wart will push the full 800mA up to about 3.65V where it should slow down below 40-100mA as the cell reaches full charge. but if you can just get it up to 3.4-3.5V then go charge the other low ones.

once you have them all around 3.4-3.5V then you can use the charger with the BMS and the BMS will balance the pack then. it cannot balance the pack now because it is so far outa balance to begin with.
 
dnmun said:
i have a buncha those cells and i think AW amy have some too so if you have some dead rows we can provide parts.
Click to expand...
I wish I did, but the looters stole that pack along wiht some other stuff in early October. :(
http://www.endless-sphere.com/forums/viewtopic.php?f=1&t=49550&p=808723#p808723
 
To get the battery closer to all the same state of charge, you can also pull down 3.5v charged cell groups so you don't start out with some cells at 3v and some at 3.5v. Use a turn signal light bulb to, again very slowly, pull down the voltage on the higher strings. Anything can be your load, but a turn signal bulb is cheap.

Then put the whole thing back on the charger and bms, with all cells at a closer to the same voltage. The bms will do the rest, without so many max charge and pull back down cycles on the highest cells.

With luck, once you get all cell groups charged, none lose that much overnight. If so, you have a chance of it being still usable, even if capacity is low compared to the nominal capacity.
 
yep, the trick is get the pack to charge up all channels initially and then see how fast the voltage drops on these impacted cells. we still have no idea about the condition of the BMS itself. there is the possibility that those channels are low because the BMS has shorted shunt transistors which drained the charge from the cells. but you measured a few millivolts instead of 0V so it is not clear if the BMS is damaged. but that can be replaced too.

if none of this works and you have dead channels you can also go down to 12S or 14S too with the cells you have now.
 
Here is an update on the battery pack.

Strings 3, 9 and 15 will not hold a charge. The voltage drops off very quickly after removing the charger. I'm guessing there is an internal short in these strings.

The rest of the strings are all at approximately 3.25v and are holding the partial charge just fine. I will fully charge each of these strings when my single cell chargers arrive next week. Manually charging with the cell phone charger requires too much vigilance.

How should I proceed with 3,9,15? Should I cut the parallel connections and try to locate the bad cell(s) via the binary search approach?
 
I'd be thinking in terms of building it into a 12s pack myself at this point. Now that you confirmed bad cells in those three strings, I'd cut out and toss those strings. If, you can pry it apart without damaging the adjacent good cells.

Cool that you seem to have enough good cells to do that much. I expected worse to be honest.
 
if the BMS is connected when the voltage drops then you have to verify that the BMS is not shorted on those channels so put the battery on the charger and measure the voltage across the shunt resistor on all channels and make a list.
 
The bms is not connected at this point. So the fault is definitely in the battery. Is there a way to test the bms without having all of strings functional/present?

If I cut out the bad strings and make this pack 12s I assume I add another new 4s pack in series to have a functional 16s pack. What are the downsides of this approach?

My other thought is to reconfigure the pack into a lower AHr configuration that will fit in the bike triangle. The current configuration is not ideal with all of the weight on a rack above the rear hub wheel.
 
I cut out the bad strings. Now I am charging up the good strings. This is taking forever without a single cell lifepo4 charger with hvc due to the manual monitoring and my paranoia of overcharging. I have a charger on order, just waiting for delivery.

In parallel I'm modeling different pack layouts trying to see how to fit the remaining good cells in a triangle bag.

Does anyone have recommendations on how to solder the remaining cells together? Is copper braid the way to go? What size of braid do I need for this pack?
 
yep, that is what i was gonna do. but when i was taking the pack apart a lot of the spot welds pulled off the case(negative end) and i decided to cut it into sections that were at least equal to or a little more than the minimum series i could find which was not a bad weld. so now i have 17 strings(8 in parallel) of 8P now. so i could build a 17S or even 18S with a little bit of soldering. but i decided to not build it into a pack because of the design. i even worry about having them on the table but they are mostly discharged. i do have an 18S BMS though.
 
You must log in or register to reply here.
Back
Top