Calling all Ping pack surgeons near Michigan

[ktrazz]

Hello ES'ers,

Long time lurker, and now that misfortune has found me, first time poster. I hate to come crying without even introducing myself. Hi, I'm ktrazz, 38 year-old, unemployed student from Ann Arbor, Michigan. I've been prowling the Sphere for about a year and a half. I own a couple of department store bikes (with steel forks), two WE BD36 kits, an ecrazyman 48V controller, some ampedbikes torque arms, a bunch of Anderson connectors, and a Ping 48V20A battery that I bought last summer. As you can see, I've made tons of newbie decisions based upon your experiences. I feel like I know many of you and I hold you in high regard (Hi, dogman).

My dilemma: I'm extremely sensitive to colder temperatures, so no riding for me over the winter. In my ignorance, when storing my bike, I decided to fully charge my Ping and store it on a shelf in the basement inside it's modified, steel toolbox... unattended... with no visual monitoring. Come riding season, I pull out my pack and panic at the sight of bulging prismatic cells.



I should have come to you then, but between dealing with a dramatic lifestyle change, losing my camera, and the sickness in my gut, I didn't get to it until August. After following Ping's instructions to the letter, he had me take the following readings:

View attachment 3

He said that a cell group is dead and recommended repair. After being reassured by my electronically inclined friend that we could do the repair, I had Ping send me a new group. When we finally got around to removing duct tape, this is what we found:

View attachment 2


My friend freaked and refuses to touch it. Aside from the leakage, he says that there are more cells bulging than I have replacements. So Ping says to send it back. When I found out shipping prices, I headed back here to find out how you guys were shipping them. Then I found out about lithium restrictions, hazmat, fines and possible jail time. That's when I freaked.

So here I am, looking for experienced Ping (or prismatic) pack surgeons, regional to Michigan, that I can spend a weekend driving to and fro for repair. nomad85? StevenR? Even ianmcnally2? Anyone or am I SOL?

Thanks.

ktrazz

 

[dnmun]

you gotta finish unwrapping it for anybody to estimate, but it looks like toast. they are all gone it appears.

did you do any charge/discharge cycles? through a load? if needed do you know how to do stuff like this or do you expect someone else will do it?

 

[floatingdog]

What went wrong? Was the temperature too cold? Should they not have been stored fully charged?

 

[TylerDurden]

Eco-Wheelz is in Plymouth, about 30min from A2. They might be able to help.

 

[ktrazz]

What went wrong? Was the temperature too cold? Should they not have been stored fully charged?

From what I'm reading, the proper way to store them is in a partially discharged state with weekly (bi-weekly?) regulation. Colder temps should be preferable. The basement was definitely warmer than the shed.

Eco-Wheelz is in Plymouth, about 30min from A2. They might be able to help.

Thanks Tyler. I'll pursue that avenue.

you gotta finish unwrapping it for anybody to estimate, but it looks like toast. they are all gone it appears.

did you do any charge/discharge cycles? through a load? if needed do you know how to do stuff like this or do you expect someone else will do it?

If they're all gone, I won't even sweat it and chalk it up to an expensive lesson learned. The BMS wouldn't accept a charge at all, so Ping had me charge through the discharge leads for strictly 30 minute intervals to get the BMS to kick in (which it didn't) before I took those readings. I'm a complete electronics neophyte. So when you talk about hooking up light bulbs and wall warts, I stare at the screen nodding while the lovely static plays in my head. At this point, I suppose it's probably not even worth parting out. Thanks for your help though.

ktrazz

 

[dnmun]

i suspect the BMS drained it while sitting.

can you just list the voltages you measured? without the XLS?

i suspect it will be a lot more than 3 pouches to replace.

 

[ktrazz]

Sorry about that:

Test 1
Balance Wires Volt
#1 #2 0.08
#2 #3 3.31
#3 #4 3.34
#4 #5 3.67
#5 #6 3.67
#6 #7 3.59
#7 #8 3.56
#8 #9 3.57
#9 #10 3.67
#10 #11 3.66
#11 #12 3.41
#12 #13 3.59
#13 #14 3.36
#14 #15 3.41
#15 #16 3.5
#16 B- 3.34

 

[swbluto]

Very interesting. It appears one cell group is completely dead (I'm going to guess it's the bulging one - Mine have similarly bulged when they were drained to 0V) while the other ones are holding a voltage which implies they're still functional. Makes me wonder then, how did they go about leaking like that? Their internal resistance might have risen substantially, though, due to leakage/whatever, so "functional" could be very relative. Might of also had a significant capacity decline as a year of high temperature and 100% charge typically results in a 60-70% decrease in capacity for regular lithium cobalt. The data on LiFePO4 is sparse, but I'd suspect a similar dynamic.

 

[ktrazz]

By high temperature, does the 63 degrees F. that I keep the house at fall in that bracket?

 

[swbluto]

By high temperature, does the 63 degrees F. that I keep the house at fall in that bracket?

No, haha. A "high temperature" would be like 104+ F like you might expect in the garage during the summer time.

I'm unaware of the storage conditions to make an accurate estimate on the impact, so that's something you'll have to evaluate if you want evaluation.

For regular lithium cobalt, 100% charge voltage during storage would result in a capacity drop to 80% in one year. A similar figure for LiFePO4 is unknown, but the dynamic would probably be similar.

 

[snowranger]

Why are you keeping the battery upside down? There is almost no protection on the top side. If you rode around like that, the PCBs connectors could have easily damaged the pouches or the tabs would flex until they broke. I'd just replace those bulging cells, and see what happens. Luckily, they're the outermost ones.

Maybe it isn't coincidence but all the bad cells on mine were on the front of the battery where the negative and positive terminals were.

 

[swbluto]

You know, it's interesting that only the first cell drained completely. I'm willing to bet that was the "ground cell"(Is there a negative/black power wire connected to that group?), and some BMS circuitry was feeding on that particular cell as a 2-3.3V supply.

 

[dnmun]

on the signalab v1 BMS the circuit current for the active devices comes off the top of the 4th cell. not sure at what point the leakage out of the battery through the circuit slows down, but maybe once the first cells reached an infinite open there was no more current loss, and then the rest of the pack has been sitting insulated by the burned out first pouches.

order three and fix it like every body else hadda. god's punishment. it's better to ride in the cold.

snowranger just did one and GCinDC has to do one yet. i have to replace #14 on another one for brandon, you got the end pouches that's easiest, i just did one, like greg and snowranger. there may be a battery shop in your town there who will do it too once you buy some replacements from ping.

 

[dogman dan]

Ouch, nothing to lose by trying to just replace those cells in the worst group now. Well, lots of time and maybe a costly soldering iron to buy.

Makes me glad my climate makes 12 month riding tolerable. Clearly, leaving a bms on a pack unatended for long is the the likely culprit.

 

[swbluto]

on the signalab v1 BMS the circuit current for the active devices comes off the top of the 4th cell. not sure at what point the leakage out of the battery through the circuit slows down

There must have been a LOT of leakage slow down for the rest to remain above 3.3 volts. The resting voltage for a cell at 50% is like 3.28 volts. Either that, or it could have been that that particular cell group had been unbalanced and so it was stored at a lower charge at the beginning of storage. I'd imagine not leaving it on the charger for long periods of time (Which is kind of a common habit) and higher than typical current consumption for the ground cell (as all current routes through ground) might have resulted in a lower than average initial charge for the ground cell group.

It does seem likely that the first few cells were tapped off for the active devices. The fact that a gate drive was needed for the mosfets implies more than the one cell would've been tapped.

 

[ktrazz]

Why are you keeping the battery upside down? There is almost no protection on the top side. If you rode around like that, the PCBs connectors could have easily damaged the pouches or the tabs would flex until they broke.

It just got flipped that way during tape removal. Your witnessing the only inversion.

You know, it's interesting that only the first cell drained completely. I'm willing to bet that was the "ground cell"(Is there a negative/black power wire connected to that group?), and some BMS circuitry was feeding on that particular cell as a 2-3.3V supply.

I'll check that when it gets opened.

order three and fix it like every body else hadda. god's punishment. it's better to ride in the cold.

snowranger just did one and GCinDC has to do one yet. i have to replace #14 on another one for brandon, you got the end pouches that's easiest, i just did one, like greg and snowranger. there may be a battery shop in your town there who will do it too once you buy some replacements from ping.

Ping sent five cells. Should I order more before surgery?

I greatly appreciate the guru-speak. I'll see if I can find some local help and update you accordingly.

 

[dnmun]

when i say the circuit current for the active devices comes off the top of the 4th cell it is because i mean it. not because i imagine it. you can look on the BMS and see it yourself where it runs under the low voltage cutoff optos on the outside row. it goes to a surface mount resistor on the back side which then doles out the current to the optos and the other active parts through a 1kohm resistor. so at full charge, 4 cells has 3.6x4=14.4V and when it is down close to the point where they start failing, around 8V. you can see them yourself since you have a ping BMS that came with your pack.

the output from the low voltage optos goes to pin 13 on the 14 pin package, for the current path from the 4th cell, there is a 102, 1kohm resistor on the backside in the current line before it splits and carries current to the other end of the optos, and also to the 14 pin package for running the cmos comparator circuits that end up driving the gates of the output FETs and that output is from pin 3 of the 4011c and it runs through a 103, 10k ohm gate resistor which is surface mounted on the back side then comes out at that through hole on the right of the 4011c and goes up to the output FET gates. the current that goes to the gate of the charging FET also comes off that lead that runs to the other end of the optos too, through all the high voltage cutoff optos in that second row, and you can see where the last output of the optos runs right up to the gate of the charging FET. also each of the comparators takes current from that cell that it monitors and each cell also drives the leds in the optos at each level too, both LVC and high voltage cutoff opto leds are on continuously while there is a connection between the cell and the BMS through the sense wires. so a lotta current can get used up over time.

you really need to evaluate each of the cells in the first 4 rows to see how much damage was done by drawing them down like that. but since they are holding voltage now, i would just replace as many as you can afford, then a little more and then you don't have to pay shipping twice, and all of them that have swollen are likely suspect, but they may be ok, i have never had a ping so can't say.

by evaluate, i mean see if they can discharge the 4Ah each they are rated at, they may conk out at 2 Ah and the other cells in your pack could go to the full 4Ah and so you would effectively lose most of the capacity of the battery and it would cut out early and be harder to balance, so that would be a useful test before ordering. too bad you are over there since i wanna order from ping too so we coulda split shipping.

snowranger had a buncha good pictures, jinba has a picture of the little pcb connectors all cleaned up.