VPower / CammyCC 48V20Ah LiFePO4 Repair by Amberwolf

More later once I have time to troubleshoot it, but I suddenly had a problem tonight, just as I was about to leave work after closing time, with 80lbs of dog food on CrazyBike2.

Under a startup load (typically 60A+ for an instant), the battery would cut out as if it hit LVC, and require actually disconnecting the main power plug at the BMS from the battery itself, in order to reset it. It'd run fine just spinning the wheel, even with the bike weight on it, skidding on the ground, but if I tried to ride the bike, or have the load on the wheel enough to keep it from skidding/spinning, it'd cut out completely.

When I have hit LVC on it during the capacity testing I did, it only had to have the pack/BMS disconnected from the bike to reset it, IIRC (but i'd have to go back and check previous posts about that to be sure).

Unplugging the cell balance taps didnt' work, either, as I couldn't even get it to reset with them disconnected, while I was trying to do this to get home. It did work later, once I got home.

I started trying to measure at the balance taps for voltages, but my hands are so unsteady from being exhausted that I can't get good readings. Just typing all this is quite the effort, but I can't sleep due to worrying about the fate of ES
http://www.endless-sphere.com/forums/viewtopic.php?f=1&t=30607
after the buyout debacle began.

Since I would like to use the battery to ride with to work tomorrow, I put it on the charger and will leave it to run and hopefully balance some overnight, even though this will take away some evidence of what might be wrong. I'm sure I can fix that just by running a high load on the pack.

I suspect it's one of those strings that was running low during the discharge/capacity tests, or another runt cell coming to light after pushing the pack like I have been.


I PMd and emailed Cell_man to see what cost of new cells similar to what's in here might be, in case I need several (or more) to fix it (though mostly to repair another member's pack if it turns out to be cells on his similarly-built pack, once it arrives here and I can get to testing it). No response from CM yet, but I know he has some pretty busy times, sometimes.


More data/etc later.

Well, Cell_man doesnt' carry this type, which I kinda expected, but it was worth a shot.

After working today and then spending way more time than I'd like (or can spare) checking up on recent events here on ES, I've not yet had time to do cell-level checks, but the pack charges up to it's normal nominal voltages just fine, and it cycles thru balancing/re-topping-off, keeping the bleeding resistors in the BMS warm, though I haven't yet caught it actually restarting the charger, it must be doing so periodically.

Today, in practice for repairing RTLSHIP's very similar pack here:
http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=30864
I began by troubleshooting my own. Might as well make the mistakes on mine, instead of someone else's. :lol:

I've had it disconnected from the charger for over a day, after the full charge finished. Still hooked up to the BMS but that is unlikely to be causing the problem with it, based on previous problems.

Checked cell voltages at the balance connectors with the BMS disconnected, and found that cell group 6 down from the most positive end (exact opposite of the group in RTLSHIP's pack, interestingly enough) was down to 1.72V.

All the rest were between 3.38 and 3.47V, mostly closer to the latter.

It is not surprising that it was this group, as it is one of the ones that has always read a bit lower than the others. There are still two other groups that tend to read lower, though, that haven't failed yet. When I have time, I probably should do the binary search on those to eliminate the weak cell(s) in each of those groups. For now, just going to be this one.


I did the binary search for the problematic cell by cutting the tape back, sticking it to itelf by rolling it back away from the group, and then cutting the tabs first between the group's two halves of 9 cells each.

Measuring each half of the group showed rapidly rising voltage on one 9-cell group, and slowly dropping voltage on the other, making it obvious which one probalby had the bad cell(s).

I then cut tabs between a set of 4 cells out of the 9 remaining and the other 5 in that group with the dropping voltage, and the group of 4 measured still dropping (evne faster now, with less cells to hold it up), with the other 5 rising.

Dividing the 4 into two of 2 each, the one with rising voltage was left alone and the one still very rapidly dropping was then divided into the two remaining cells as separate entities, one of which began rising, and the other dropping like a stone--half a volt while I watched, over a few seconds!
View attachment 6

View attachment 5

Turned out to be the end cell, and thus very easy to remove.

All the other cells rapidly rose over the next 20 minutes or so to over 2V, the first largest group up to 2.2V. So probably all still good. I repaired the tab cuts by soldering used bits of solderwick across the gaps, using the Weller 80W fat-flat-tip iron so it only took a second or so on each one.
View attachment 4

View attachment 3

I waited to see what voltage it would level off at, and it stopped at 2.25V after 40-something minutes.

Dug out the little Sorenson supply, and set it to 3.65V and 250mA current limit. Hooked it up to the meter Bigmoose sent me so I could watch the voltage rise, and set the Sorenson's own meter on A so I could see if it started to fall off.


I took a couple short tiny bits of solid-type wire from a snippet of CAT5 cable leftover from something long ago, and placed those in the meter's banana plug holes along with the meter leads to hold them in,
View attachment 2
then inserted the other end of those wires into the balance plug holes for that cell group, and began recharging that group.

View attachment 1

After 30 minutes or so it's up to 2.83V and constantly rising. Assuming essentially zero Ah in the cells to start with, that leaves theoretically about 17-18AH to fill up the 17 cells in the group, so in theory around 3 days to recharge at the 250mA rate.

I will only recharge for overnight at the low rate, then I will up it to 500mA and see if anything bad happens. If it doesn't, after a couple of hours, I'll go to the full 1A or so this Sorenson can put out. That will bring the recharge time down to a bit over a day, total.

At that point, I should be able to test the pack on the bike with essentially full charge on Sunday, which is my next day "off". I'll probably test it on DayGlo Avenger since A) it doesn't pull as much power as CrazyBike2, and B) I can still pedal DGA home if the pack does turn out to have some other problem.

The cell group is now up to 3.50V (with the power supply on. It falls to 3.37V once disconnected). The Sorenson gets awfully hot, smells like hot solder, because it's already nearly 100F in the house, so I had to put a small fan on it to pull air thru it better than convection was doing.



Since the dogs could accidentally knock stuff off where this is, I used a fan that is already safe to use around them; it's an old "mac saver" from I think Kensington, basicall a power strip with a fan built in that sits in the handle location on top of the old Mac Plus types, from the mid-80s. Doesnt' move much air, but it is very quiet and it cools the Sorenson off enough to not smell like it's about to burn. :lol:

I turned up the current to half an amp now, and let it run while I dealt with other problems after getting home from my dogfood trip today
http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=15570&p=447022#p447022
http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=18671&p=447026#p447026
and it seems to be doing fine, so I went up to 1A for a while to see what happens. It's still running right now, an hour or so later, with no issues so far, and still at 3.50V.

It's nearly finished charging that cell group. If I turn off the single-cell charger (sorenson) and disconnect it, it sits at 3.34V steady, even after a couple of hours on the meter. It's now continuing the charge, at 500mA (I haven't had time to watch it for the 1A charge, so ti's just gonna have to take longer). I may not have time to test it on the bike or bench under load today, though.

It ifnished last night at 3.64V and zero noticeable current. Unplugged the sorenson and it dropped o 3.62V, left it until i woke for work this morning it eventually went to flickering between 3.50V and 3.51V, where it has stayed so far today.

Rest of the cells are just below that, so it seems that everything is charged ok. I'll have to test it when I feel a bit better; currently recovering from a little crash on DayGlo Avenger.
http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=15570&p=447763#p447763

Tomorrow is a day off, so after I get started on RTLSHIP's pack repair, then I will fix a problem with DGA's wiring probably caused by the crash; an intermittent cutout. After that, I'll test this pack on DGA, since at least I can pedal DGA around if I failed to fully repair this pack.

Hope you will recover soon, man.
Good to know you´re wearing a helmet.

greetings,

christian

Got RTLSHIP's pack diagnosed and the bad cell removed, recharging the group now and then just need to find a good cell to replace it with.
http://www.endless-sphere.com/forums/viewtopic.php?p=449132#p449132


Tested this pack today on DGA:
http://www.endless-sphere.com/forums/viewtopic.php?p=449498#p449498
After a bunch of yard cleanup, and processing a new arrival
http://www.endless-sphere.com/forums/viewtopic.php?f=3&t=16920&p=449421#p449421
and then more household chores, I got a chance to do a short two mile test run of DGA with the Vpower/CammyCC pack on there. First a quick check of all the cells via balance connectors showed they haven't changed since I took it off the single-cell charger (Sorenson) for the recently-fixed group, so that's good.

I didnt' have any problems with it, and I tried abusing it as much as the Fusin and 6FET could tolerate (which is nowhere near what CrazyBike2 puts it thru), including a grocery run hauling about 60lbs of stuff, mostly lashed down on the rear rack in a backpack, and some lashed into the top portion of the pod.

The battery pack (still in the big black box) takes up almost all the space in the pod, so I can't fit much else in there. I'd like to test the pack for my work commutes, too, but unless I take it out of the box, I can't, because otherwise there isn't room for my work uniform and leaving my lunch and cold water/ice/etc. in the insulated pod (I don't have anywhere else to keep it at work, since I can no longer take an insulated cooler or backpack in, and there isn't space in the little refrigerator for my stuff).

I *might* be able to put the box vertically and still have my stuff fit, but I don't think it will work. Have to test the theory; the pack may shift around in the box that way, while it doesn't if it's horizontal.

Anyway, the numbers from the run (both legs as one set), using the Turnigy Watt Meter:
55.6Vstart
53.44Vrest
50.93Vmin

1.063Ah
55.9Wh

1225.7Wpeak
23.73Ap

The Fusin got pretty warm, as I was running it as close to full throttle as I could, every chance I could, but I still pedalled because I don't want to cook it. I did let it take all the startups by itself, though, to see if the battery would be ok with that, and it is. But i do have the 6FET set to limit current at 19A, IIRC, so even the 23.73A peak current must not be very long.

The controller was kinda warm, but it is fully open and gets plenty of airflow at speed, so I'm sure it would be a lot hotter if it were enclosed and/or out of the airflow. I think it really needs more caps, not just the big ones but some little ones, too.


Anyhow, the battery appears to be behaving, as the cells are still just about all in balance with each other, except for that one low one at the end is almost 0.2V less than the others.

Do you know if vpower will sell just the bms ? I see they don't offer it on the web site.
I have this battery and I believe the bms is bad as the volt meter says the cells are full power, until you plug in the bms and then the power drops to zero.

I doubt it's the BMS, but it's always possible. You'd need to see what the voltage of each cell group is, both idle and under load, to see if a cell group has a problem. Seeing the pack voltage under load would tell you if it's the BMS or the pack; if there's lots of sag then it's probably the pack and the BMS is just doing what it's designed to to save your pack.

I also answered your other PM'd question in a PM reply, but I think you'd be better off either starting a thread for it or continuing the one you orignally had. (since you never replied to it after some other questions were asked, there's no way to know if even those problems were resolved).

aw if you get this working - I have another vPower - Cammy which I have double tapped to have 10S and 4S parts (seperate charge parts), no reason for the distribution... The point is I tried fairly hard to ressurect it and got maybe 80% of the cells back (still 12 S maybe 15AH in real life... it's is yours for the price of ground shipping (I won't airborne lithium uneless laptop battery suplu size).

Regards,
Mike
PS: Earnest Offer

mwkeefer said:

aw if you get this working - I have another vPower - Cammy which I have double tapped to have 10S and 4S parts (seperate charge parts), no reason for the distribution... The point is I tried fairly hard to ressurect it and got maybe 80% of the cells back (still 12 S maybe 15AH in real life... it's is yours for the price of ground shipping (I won't airborne lithium uneless laptop battery suplu size).

Regards,
Mike
PS: Earnest Offer

Click to expand...

Once I can save up for that, or earn enough doing repairs of things for people here on ES, I will take you up on that if it's still available by then. I don't know how long that will be, though. How much would ground for it be to Phoenix, AZ, 85051?



So far the repair seems to be behaving itself. After a couple of work commutes, plus a short around-the-block trailer test with Nana the St. Bernard in it, it is at the following stats from the CA. (it has not yet been recharged since the above tests, so you have to add together the duration based stuff, Ah, Wh, etc).

43m 41s trip time
10.09miles
23.7mph max
13.8mph avg

27.0Wh/mile (lots of pedalling)
5.390Ah
272.5Wh
58.46Amax

53.6Vstart
52.5Vrest
46.1Vmin

0% Regen
0Ah Regen
-11.9A peak Regen (must be only instantaneously momentary as there is no braking occuring)


It's now recharging via it's normal charger and the BMS, as a pack. I'd rather wait and see when it hits LVC but I need all the power available tomorrow to take Nana to work with me and back home afterward.

Wolf - your not well right? Frag that man, I'll ship it out to you at my own expense - enjoy, they are a heavy pain in the ass but as I recall I broke it at the 8S point (iCharger 208B+) but PM me before I ship it, lets see if I have some bulk charging kit you could use or some hub/frock motors or other RC motors (kinda clearing house)..... but the vPower is all yours brother along with the POS BMS which never worked and I suppose I can toss in a 2.5A charger too... I'll PM you back tommorow (after I get your mailing address in AZ) and I will let you know when to expect it... Need some Lipos? I may be able to oblige, need some whatever (long as it's legal in all states) I got ya if I can....

Regards,
Mike

mwkeefer said:

Wolf - your not well right? Frag that man, I'll ship it out to you at my own expense - enjoy, they are a heavy pain in the ass but as I recall I broke it at the 8S point (iCharger 208B+) but PM me before I ship it, lets see if I have some bulk charging kit you could use or some hub/frock motors or other RC motors (kinda clearing house)..... but the vPower is all yours brother along with the POS BMS which never worked and I suppose I can toss in a 2.5A charger too... I'll PM you back tommorow (after I get your mailing address in AZ) and I will let you know when to expect it... Need some Lipos? I may be able to oblige, need some whatever (long as it's legal in all states) I got ya if I can....

Regards,
Mike

Click to expand...

Well, I'm up for anything that's otherwise scrap; if I don't use it "now" I'd use it eventually. Or find someone else in need of it and rehome it, if I find it's not something I can use, but that would be after not using it for a really long time (whcih for me is usually counted in years, sometimes decades. :lol

As for not being well, I think much of it is just being so tired from stress, lack of sleep caused by stress (and the heat right now) and constantly changing work schedules, etc., and the gut problems from the stress. Well, that and apparently falling asleep and riding into bushes/gravel. My ankle/leg from the DeathRace is way better, and is now just a limp that finally sometimes is less on the right ankle than my original left-knee limp (which itself is a bit better these days).


I can pay at least shipping for stuff, but it will take time to gather up. I've had a couple of donations to the dog-stuff fund recently, that will help keep them taken care of and fed, but I have to catch up on bills and rent for this summer, as I've gotten behind from not having enough hours at work (partly due to my Death Race ankle/leg injury).

That said, I won't argue with anyone that doesn't want any money from me; I can't afford to. :lol:

Just as a note, all the groups appear to be staying relatively balanced. At about 6Ah used from yesterday and today's commute plus a store trip, I checked the balance taps and all groups read 3.29xV. I'm sure the meter's not even accurate to that 3rd decimal place. :lol:

amberwolf said:

I'm sure the meter's not even accurate to that 3rd decimal place. :lol:

Click to expand...

When you are just looking for a change in voltage , it doesn't have to be accurate... just repeatable (i.e. the difference between accuracy and resolution)

I have several HP3458A meters that resolve to 8.5 digits, absolute accuracy is around 6.5 digits. These are the best meters in the world and are basically at the theoretical limit of what silicon can do. Their design has not been improved upon in over 20 years...

For my purposes, I think even just two digit accuracy is probably ok; 3 is a bonus. :lol:

The best modern meter I have is a Fluke 77-III, but it's currently buried in a box in one of the back rooms with a bunch of other stuff I can't get to after the cleanup, until I get time to re-sort and move stuff I had to just box up and get inside out of the city's view.

The best (but least likely to be calibrated) meter I have access to ATM is an old HP nixie-display type (I forget which one). But seeing as how it's way more convenient to use the little handheld Centech types, I end up using one of them (the same one each time, wherever possible).

I've really not used this pack for the last couple weeks or so, since I started the testing of the Volgood pack repair, so it's been sitting on the bike but not being recharged for that time, and it has not had any groups going low, so I think I've found all the really bad cells for now. More will rpobably fail later.

amberwolf said:

More will rpobably fail later.

Click to expand...

And, as always, at the worst possible time... and in the worst possible place... Amazing how that's ALWAYS the case.

Probably when I am 10 miles from home, having just picked up a 500lb load of dog food. :lol: And it'll be late at night so I can't get hold of anyone to help out.

You need smaller dogs... or a Korean cookbook...

amberwolf said:

The best (but least likely to be calibrated) meter I have access to ATM is an old HP nixie-display type (I forget which one).

Click to expand...

:x
:Luke's jealous face:
:x

Nixie tube DMM is big pimpin. I would love to have that instead of my 7.5 digit ethernet logging Keithly for my office bench meter.

Want to check a voltage? Hold on a few minutes guys, my tubes are warming up, you don't want to rush things like a voltage reading.

It's not exactly precision equipment...


Neither is my scope:


But they all "work" well enough for my purposes. :lol:

amberwolf said:

It's not exactly precision equipment...


Neither is my scope:
View attachment 1

But they all "work" well enough for my purposes. :lol:

Click to expand...

mmm sexy... nothing like the oldies. =P