Test cell V's in Ping LiFePO battery w/out dissecting it?

[wayover13]

I got a 36V 20Ah Ping LiFePO battery 3 or 4 years ago and it's been working pretty well. I actually had to take the bike out of service last year due to a maintenance issue. Well, it took me a whole year to get that maintenance issue addressed and I'm using the battery again this year. I should mention that, not realizing it was going to take me so long to attend to the maintenance issue mentioned, the battery actually sat for a full year without having been discharged the recommended amount (20%-80%, as I've been told: it was actually close to 100% when I took it out of service).

Fast forward to this year and I've resolved the maintenance issue and put the bike/battery back into use. Voltage on a full charge has been running a little lower than I saw in past years, with a reading at full charge of only 40.9 volts. While on a recent e-biking trip, my (greenbikekits) charger flaked out on me. Having read a review of the Cycle Satiator (CS, hereafter), I decided to spring for it.

Attempting to charge with the appropriate profile using the CS I get an error message toward the end of the charge cycle (Pwr fault). I consulted with Grin and they suggested creating a new, low power charging profile and trying that. No error message ensues when using that profile and I do get the expected "charge complete" message. But I note that the fully charged voltage still only gets up to 40.9 (43.2 is the voltage the preset LiFePO profiles are supposed to raise it to). In addition, I note that only 1 of the 12 LED's on the BMS lights up at charge completion whereas, previous to this year, I believe all were lighting up at the end of a charge cycle.

The guy at Grin suggests the battery cells have become unbalanced--a suggestion that, despite my rather sketchy understanding of electronics, makes sense to me. So he suggests testing individual cell voltages to determine how to proceed--another sensible suggestion, in my somewhat uninformed opinion. So I've begun looking into how the testing of individual cells in this battery might be accomplished.

I'm not real conversant in the anatomy of this battery, but had guessed I might find some point(s) on the BMS circuit board where I might be able to get individual cell readings. But my tests so far register only voltages in the 30-40V range. So I'm so far not finding any point that is easily accessible where I might test individual cell voltages. A friend who is more conversant in electronics is of the opinion that I'm going to need to start dissecting the battery to get those sorts of readings--meaning, I'm going to have to start cutting off shrink wrap, which I'd rather not do if not absolutely necessary.

So can anyone verify whether or not I should be able to get individual cell voltage readings from my Ping battery without starting to disassemble it? I sure hope so. Input will be appreciated.

 

[cell_man]

There will be a multi pin connector going to the battery bms. Get to that and measure the DC voltage on the adjacent pins and you will have the voltage for each series cell group in the pack.

 

[Gregory]

Try a search for BMS balance wires. Or measure cell voltages.

Measure between 1 & 2, 2 & 3, 3 & 4 etc.
Using a multimeter, maybe set to 20VDC range.
Carefully so you don't short anything out.
Write down each result.

They look like this.

 

[wayover13]

Thanks for the great tips and even photos! I was suspecting that might be where I could check, but was uncertain where to poke the probes. Using those tips, I was able to get voltages that appear to be close to what individual cells should be (all in the 3.31-3.32 volts range). At least for 11 cells: how would I get the 12th cell's reading? Judging from those readings, the battery looks to be balanced, though the cells do read a bit low for a battery that should still be charged to above 90% or so.

So I now have data to report to Grin. Meantime, any further input on these matters, e.g., what to make of the voltage readings I'm seeing and how to meter the 12th cell?

 

[cell_man]

Try going to the main positive or negative of the pack. Some BMS do not have a balance wire for every voltage point in the pack and might take that voltage from a large cable feeding into the pack. The pack positive, generally comes straight from the positive of the pack, it doesn't normally go through the bms.

Most bms have a balance plug with the number of balance wires equal to the number of series cells plus 1. If that is not the case, then look to take that last signal from a large cable.

Lifepo4 cells have a very flat charge/discharge curve. The voltage will only increase significantly, when the pack is near to fully charged.

 

[Gregory]

Ah yes, from memory the Signalab BMS uses the battery negative power wire. Most other BMS I have use 13 sense wires.

I THINK it is between wire 12 (which would be the black sense wire in those pics) wires and B-.......


BUT, do search for it first, because I'm not sure.



Edit: beaten by Paul.
Here's a pic, see how the last cell doesn't have a -ve sense wire, the BMS uses the B-, so thats what you need to probe.



P.S. if going to the main POS wire you need to switch your multimeter range to allow for >20V , unless you have a auto ranging DMM.

 

[wayover13]

Thanks so much for the additional tips and graphics, Gregory and cell_man. At this point though I'm having a bit of trouble--due to my own rudimentary understanding of electronics--comprehending what I should be doing to get a voltage reading from the 12th cell. Let me describe in my own words/understanding what I think you might be suggesting, to see if I'm understanding correctly.

So there's what I would call a header soldered onto my BMS, one that looks a lot like the ones circled in the photos posted by Gregory. It's from exposed metal parts of the back of that header that I've been getting my individual cell readings. The header is actually a sort of receptacle, by which I mean it serves as a plug into which a cable that comes out of one end of the battery inserts. That cable, incidentally, is comprised of 12 small individual wires.

The insulation on the individual wires that make up that cable are all white except for 2: there is one with red insulation and one with black insulation. When measuring the cell voltages by touching adjacent exposed metal leads on the back of the header, I started at the end with the red wire (on my right in the orientation in which I was working on the battery) and finished at the end with the black wire (on my left).

This battery also has, of course, two large wires coming out out it, and it is onto those wires that I attached the Anderson connector which allows me to plug the battery's power to the controller/motor on my bike. Those two wires also have colored insulation on them: one has black insulation, the other red. The back wire actually comes straight from the BMS, having been soldered to a point on the end of the board. What I'm gathering from what you've both said is that I need to put one of my volt meter's probes on one of those large wires, and the other probe on one of the exposed metal leads on the back of the header I've described--correct?

If so, let's say the last individual voltage reading I took was between the black-insulated wire at the left end of the header and the white-insulated wire right next to it. So, to get the 12th cell's voltage reading I would then put one probe on the lead where the black-insulated wire plugs into the header, and the other probe onto one of the large wires coming out of the battery pack? If so, which of the large wires? Red or black?

I hope I'm understanding more or less the directives. I don't feel comfortable proceeding without getting some confirmation about this.

PS I guess one of the main things I'm not understanding very well is what "B-" as used by Gregory means. Would that be "B" for black, and "-" for negative?

 

[wayover13]

I consulted with my more electronics-savvy friend about this. He said the right way to do it would be to touch the red probe of the volt meter to the contact on the back of the header to which the small black wire leads, and the black probe of the volt meter to the large black wire soldered to the end of the BMS board. When I did that, I got the same reading as I'd gotten for almost all the other cells--3.32 volts. So it looks like, in this slightly discharged state, the battery is balanced.

I heard back from Grin today too. They said 3.3 volts is a normal level to see on LiFePO cells that have been off the charger for a bit. So it seems voltage on these cells is not reading a bit low, as I'd assumed. The next step will be to give the battery a full charge, then check cell voltages while it's still attached to the charger. I'm to pay special attention to cells represented by lit LED's on the BMS (I've seen only 2 lighting lately when the battery gets fully charged).

Perhaps more to come.

 

[999zip999]

You can keep it on the ping charger for 24hr. This will let the slow balancing circuit balance the battery over time. Time you can use 1/3 1/2 discharge and leave on charger. Yes longer than that. Yes longer. Balance.
Cell monitoring is best. Balance.

 

[wayover13]

I'm not completely understanding your remarks, 999zip999, though the part about leaving the battery hooked to the charger for 24 hours (and maybe more?) is more or less clear.

So I'm learning some things about battery balancing. I left the battery hooked to the charger yesterday for 14 hours while awaiting a response from Grin. But so far they didn't answer. Meantime, more and more LED's on the BMS were coming on as the time passed. Eventually, all but one were lighted. I did not intnetionally turn off/disconnect the charger after 14 hours, by the way: it was hooked to a power strip that got inadvertently switched off. I noted, not long before that, that the top of the BMS where the LED's are located was getting pretty warm to the touch. I also measured total voltage not long before the shut-off--the pack was at 43 volts, very close to what is apparently the healthy voltage for this pack. I noted this morning that the LED's were flashing in what seemed a somewhat random fashion, and assumed that they'd probably been doing that over the course of the whole night.

This morning I decided to resume charging and check voltage on individual cells while doing so. The good news is that the "charge complete" message popped up pretty quickly and that the 11 LED's came on again. Voltage on all cells respresented by lit LED's looked like pretty much where it should be: right around 3.60, give or take. The cell represented by the unlighted LED was reading at around 3.37.

I went ahead and left it on for a few more hours and, lo and behold, the final, 12th LED began to light up. Checking the voltage on the corresponding cell showed that its voltage had risen to 3.57. So maybe it, too, is getting back toward full health?

I guess I won't hear back from Grin until Monday. Meantime, I decided to do a bit of reading on battery balancing for multi-cell lithium batteries and ran across this wikipedia article, which, though quite brief, was informative: https://en.wikipedia.org/wiki/Battery_balancing. I gather from that article that the heat I was feeling over the LED's may well have been a result of "passive balancing" where "energy is drawn from the most charged cell and is wasted as heat, usually through resistors." Sound right? As to the flashing LED's, once the charger is turned off, could that be the BMS causing individual cells to exchange voltage so as to bring them all to the same voltage level?

 

[999zip999]

Sorry a battery needs all the cells to work as a team. And do so they must balance. If not it's only as strong as the weakest cell. The balance circuit for a ping is very slow and would need a lot of time on a charger to give bms to work its magic. If left alone it will bleed down the first three cells.
It is best to write the voltage down on paper like.
1. 3.45v
2. 2.89v
3. 3.31v

12. Xxx volt

 

[Rassy]

Wayover, it looks like you are on top of your problem now. In my opinion it's very important to not only charge a Ping battery regularly, but to also make sure all the LEDs light up. I've given Ping batteries to a couple of relatives that didn't understand this and came close to damaging their batteries.

I didn't notice whether you stated what version Ping battery/BMS you have, but this thread may be of interest to you:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=83003

 

[wayover13]

Wayover, it looks like you are on top of your problem now.

Appearances can be deceiving . If "on top" you mean I am pretty close to having, or maybe even already have, all my cells balanced, you may be right. I think I'll try one more charging cycle using a standard LiFePO profile to test that. But the reason I don't feel on top of much of anything is because I'm not at all certain how my battery got imbalanced to begin with.

I suppose storing it at full charge, then not really touching it again for over a year could have played a role. I don't have enough knowledge in this area to even begin to say whether that could have played a role. Then, there's the wrinkle with a potentailly defective BMS that you've now thrown in. Thanks for pointing that out, btw, but I am unable to say for cetain what version my BMS is. And, as I said toward the beginning of this thread, I'd rather not have to dissect the battery pack to find that out (if the version number is printed on the BMS circuit board, it's completely obscured by heat-shrink, some or all of which would need to be cut away to view it)

Then there's the question, the answer to which I'm not at all certain about, as to whether battery balancing is just a regular maintenance routine e-bikers go through. And, finally, there's the question about how long one is to leave the battery connected to the charger, and if there's any possibility of doing that for too long: can the battery and/or charger be damaged by leaving them connected for too long? I see a whole lot more questions than answers here so far.

In my opinion it's very important to not only charge a Ping battery regularly, but to also make sure all the LEDs light up. I've given Ping batteries to a couple of relatives that didn't understand this and came close to damaging their batteries.

Should the charger be disconnected as soon as the last LED lights? If not, how long after it lights should the charger be disconnected?

I didn't notice whether you stated what version Ping battery/BMS you have, but this thread may be of interest to you:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=83003

Thanks, I've taken a cursory look but will be studying it more and thinking about how I might act on that information.

 

[Rassy]

Even though you have some doubts I still think you are pretty much up to speed. To answer a few of your questions, yes, once all the LEDs have lit up the battery is fully charged, and the actual balancing will be complete when all the LEDs have turned off. Other than the visual effect, that is the main job of the LEDs, to drain each cell group down to the same "exact" voltage.

It will not hurt your battery to leave the charger on for longer periods, such as overnight, but it's not recommended to leave it on for many days.

For that winter storage you are talking about, there are two safe ways that I know about. You can either put the charger on every week or two to keep the battery fully charged (not important) but also fully balanced (important), or you can unplug the BMS by disconnecting it from the battery by unplugging that little 12 pin plug where you tested the voltages. My 48 volt batteries have two places where that can be done, and Ping shipped them with the plugs remote from the BMS by several inches undone.

The V5.0 BMS is over 5 inches long and is all black on the top with just the LEDs showing. The V2.5 BMS is under 5 inches long and has lots of connection points showing along with the LEDs. This size is for a 48V BMS which has 16 LEDs, so the 36V BMS is probably shorter. Either version of BMS will draw down either some (V2.5) or all (V5.0) cells over time unless you unplug it as mentioned above. You already know that your battery ended up unbalanced after a long period of non use, and I think you are lucky it charged up and balanced as quickly as it did. I really don't like plugging in and unplugging those little, fairly delicate, balance wire plugs so I prefer to just plug in the charger every week or so. Some worry that keeping the battery charged up this way may shorten its life, but there is a discussion some place that indicates that isn't true for LiFeP04, only for other lithium chemistries, and in my experience I have batteries as much as 5 years old and others over 3 years old that are still performing good.

 

[wayover13]

Thanks for the further input and explanations, Rassy. I have another, perhaps daft question, regarding the lighted LED's we're discussing. They're all on now, and have been since I connected the battery last night and got the "charge complete" message from the CS. So when you say "the actual balancing will be complete when all the LEDs have turned off," do you mean those LED's should turn off while the battery is still connected to the powered-on charger? Or do you mean after they all turn off subsequent to disconnecting the battery from the charger?

I'm wondering because I've noted these last few times I've charged the battery that the LED's remain lighted, doing a little light show as they apparently randomly flash, after I've diconnected the battery from the charger. I'd been assuming this means that the battery is sort of balancing itself. Last time the last of those LED's did not stop flashing for about 24 hours. Is that normal?

My BMS is 7 inches long and the circuit board is green--at least the part of it I can see. There are two rows of little black electronic thingies soldered to the board below/adjacent to the LED's. I can see the characters j260119 printed on the circuit board right where the non-transparent heat shrink starts. See the attached photo for further reference:

"you can unplug the BMS by disconnecting it from the battery by unplugging that little 12 pin plug where you tested the voltages. My 48 volt batteries have two places where that can be done, and Ping shipped them with the plugs remote from the BMS by several inches undone." Yeah, mine has that remote detachment point as well. I'll be using that one. I did ask about this winter storage issue a couple years ago here and had understood that, for longer-term sotrage, disconnecting the BMS was not needed. Recent experience dictates disconnecting the BMS may be a better option in my case.

PS That image I embedded is not showing up in my browser: anyone else not seeing it either? Just in case, the url is . . . nevermind. Think I fixed it.

 

[Rassy]

The LEDs will stay on as long as the charger is connected. Then, when you disconnect it is when they will go off, and the dance they do is normal. It seems the longer the charger has been connected, the longer that little dance takes place, I think just draining off excess surface charge that has accumulated, and what you described sounds normal.

From your picture, I am pretty sure you have a V2.5 BMS which draws its power from just four of the sub groups, and consequently will cause an imbalance when left connected to the battery. Negligible for a few days, but noticeable after a few weeks.

 

[999zip999]

It got unbalance because cells 1 ,2 & 3 are the cells that power the bms. Yes just 3 cells so left alone it will become unbalance due to the bms. How many led lights turn on now ?
Oh your picture just came up I count 12 leds. Good

 

[wayover13]

It got unbalance because cells 1 ,2 & 3 are the cells that power the bms. Yes just 3 cells so left alone it will become unbalance due to the bms. How many led lights turn on now ?
Oh your picture just came up I count 12 leds. Good

Yes, I finally got all LED's to come on by leaving the charger connected for about 14 hours. After the latest charging cycle, fresh off the charger, all cells tested at between 3.56 and 3.60 volts. So it looks like the maximum capacity of this battery is going to be 42.7 volts or so. I still get the "Pwr fault" error when trying to charge the battery with the LiFePO profiles that came pre-programmed in the CS. Maybe that's because those profiles are configured to raise voltage to 43.2? I was hoping Grin might help clarify some of this. But it's already been 4 or 5 days since I wrote to them, so I'm beginning to think they won't answer. It would be nice to know whether I will need to create a new profile for the CS, where fully-charged voltage matches the maximum readings I'm getting from my volt meter (i.e., 42.8ish). Doing something like that seems the sensible resolution to me, but I'm far from having any sort of authoritative knowledge in this area.

PS I spoke too soon regarding not having heard back from Grin. When I checked my e-mail after having written this post, there was, indeed, a response from them there. True, they didn't answer one of my pointed questions, but they did respond to parts of the e-mail I wrote to them 4 or 5 days ago.

 

[999zip999]

I Gage light for cells or over the man perspective cells they will last 2,000 Cycles if the demand an of amps is reasonable for the cell use lifepo4 no cell for average demand. Average.

 

[wayover13]

Sorry, but I really cannot understand what you are trying to say in your latest response, 999zip999. Something(s?) crucial to deciphering your meaning seems to be missing there.

I'm trying to resolve discussion in this thread regarding battery balancing into some prescriptive advice I can follow in the future. As I'm currently understanding the matter, here are my options:

1. Stick with my likely-defective BMS and, when the battery will be out of commission for an "extended period"
a. top off the battery's charge every month or so to keep it balanced, or
b. disconnect the BMS from the battery during that period and forego the charging

2. Replace the likely-defective BMS and follow options a. or b. above, just as I would have done with the likely-defective BMS.

Have I successfully distilled salient prescriptive advice on battery balanciing from input offered thus far?

What remains to be determined is what, exactly, constitutes an extended period? A certain number of days of disuse? A certain number of weeks? What is the threshold?

The same question applies to leaving the battery connected, after the "charge complete" message has appeared, to the powered-on charger: how long is too long? Grin did not answer that question, though I asked them pointedly. According to Rassy, it's not a good idea to leave the battery connected for "days." So, is 24 hours the threshold, i.e., the point after which it becomes harmful for battery and/or charger to leave the two connected? Those are the outstanding questions I still have.

A sort of final data point I can offer regaridng this battery is that it now seems to settle down, once it's been through a charging cycle and all the LED's have stopped flashing, at about 42.4 volts, according to my volt meter. Individual cells register, in this state, at between 3.51 and 3.56 volts. I will be asking Grin in my next note whether I should go ahead and create a new charging profile that brings the battery up to 42.8 volts or so, which is the maximum voltage I've been able to read from it after a few recent charge cycles (I'm assuming the "Pwr fault" error message I get when trying to charge the battery using any of the pre-programmed LiFePO profiles results from the fact that the battery cannot reach the 43.2 volt level to which those profiles are programmed to raise the battery)

 

[999zip999]

Wow I read my message it's all messed up. It's not that problem of the BMS it is the problem of the design of the BMS it drains 3 cells to power is Balancing Act. The reason is they expect you to use it as a weekly or daily Rider the battery was meant to be used daily and not stored. Use it or lose it

 

[999zip999]

Funny I read my post before I post it and it all looks good and I'll try to send it it sometimes not often goes the Bibbblleeo

 

[Rassy]

My cat just stepped on my keyboard when I was reading your last post and the reply screen popped up, so I guess I need to write a reply.

Everything I've said on this thread is based on personal experience along with bits and pieces I've read from others over the last 5 years using Ping batteries. All but one of these nine batteries is still performing properly, including two that were abused by relatives that took extra effort to recover. The one not currently working went to a friend who subsequently abused it, including some sort of shorting incident. He returned it to me to see if I could fix it, but I was unable to get that battery to hold a full balanced charge.

So first, I agree with zip that the BMS is working properly, it's just a poor design. Even if you replaced it with a V5.0 BMS it would still run the battery down over a period of time.

Which method you choose to use when storing your battery is a personal choice. I prefer to charge my unused batteries at least once a week, although two years ago I took a 3 week trip and did not notice any ill effects on the batteries once I got home and gave them time to charge fully (i.e., to have all the LEDs turn on).

Then you threw my a curve. Apparently you purchased a special charger with custom profiles and it doesn't do what you expect. All I can say is you apparently have a charger that will charge the battery until all LEDs turn on. Once the LEDs all turn on there is no point leaving the charger hooked up because the LEDs will drain every cell group to "approximately" the same voltage, and when you start using the battery it will drop to an even lower operating voltage. In other words, nothing can be gained by trying to force a higher voltage into the battery once all the LEDs have come on, and in fact the BMS will signal to the charger to shut down when any one cell group reaches a predetermined max voltage regardless of the total battery voltage at that point. This is the point where the charger goes into trickle mode which is basically just enough current to keep the LEDs lit and also to slowly keep charging any cell group that hasn't yet reached the voltage that lights the LED.

I have no idea if the battery will be harmed by leaving it on the charger once it is fully charged, but some experts have indicated that many battery chemistries (not SLA) last longer if kept at something like 80% charge. Apparently this does not apply to LiFeP04, but the specs for these batteries are for the operating voltages and not for the higher charging/balancing voltages that don't increase the total AH you're going to be able to pull from the battery. I.e., within seconds of hitting the throttle your battery will drop to its operating voltage regardless of whether it is 1 or 2 volts higher initially.

 

[wayover13]

Thanks for your response, Rassy. Do you ever have misgivings about your storage strategy, given that storing these batteries at full charge is said to cause them to degrade? (see the recent thread "Is mantaining a cell at 3.6V overcharging?," where I specifically asked about advisability of long-term storage of LiFePO at full charge) It seems like there should be some sort of device made for storage scenarios wherein, if the battery is expected to sit idle for some longer period, it could be periodically topped off, then immediately drained by about 20%. That would be a way of sort of artificially keeping the battery in use since, as we seem to be concluding in this thread, these types of batteries are not really designed for storage. At least based on my current understanding of the issues a device of this sort seems needed. Is there anything like that out there?

As to the "curve" I threw, that was actually thrown at the very beginning of the thread--in the very first post. I alluded to the fact that it was the failure of my charger and its replacement with the Cycle Satiator (CS) that initially clued me in to the fact that my battery was having issues: I was getting the "Pwr fault" error message toward the end of the cycle when trying to charge it with any of the pre-programmed LiFePO profiles on the CS. Of course the longer the thread the less likely we all are to read most or all of it, but I do want to point out that the CS was one of the premises of the ensuing discussion.

On an unrelated note, I took a look at the delta trike referenced in your sig. It caught my attention because that is the sort of assist I was initially thinking of putting on my delta tandem trike. Among the considerations that made me opt for the simpler solution of just using a hub motor was the fact that none of the mid-drive mounting kits looked like they could be modified to accommodate the delta trike frame/drivetrain: I decided, were I to go that route, I'd have to design and build my own mounting kit. Did you create a thread documenting how you adapted the mid-drive to that bike's frame/drivetrain? Or perhaps post some information on the topic elsewhere? If so, please point me to it; I'd like to try and learn from your project in case I ever decide to convert my delta tandem trike to a mid-drive.

 

[Rassy]

Yeah, I read about the CS in the first post, but ignored it since I know nothing about it, and instead directed my responses towards the unbalanced battery and long term storage. As far as the CS is concerned, it seems it successfully fully charges your battery (all LEDs turn on) and other than giving you the ability to adjust the rate (amps) of charging there is nothing else needing attention. At various points you mentioned 42.4, 42.7, 42.8, and 43.2 volts, all of which are above the operating voltage of the battery and hence of no individual importance (IMHO). I.e., once all the LEDs turn on the chargers job is done and the LEDs will proceed to drain all cell groups down to the same voltage once the charger is disconnected. The voltage will continue to drop a little, partially because the BMS is using some juice, but it is when you start using the battery that it quickly drops to its operating voltage.

I have also seen where some don't think that 80% storage level applies to LiFeP04. Personally, I like to keep my batteries balanced and ready to use. If the life span of the battery drops a little because of this attitude it just gives me the excuse to try a newer, lighter, improved battery chemistry. But so far, the Ping batteries are still my favorite and when I did try a couple of different batteries I missed having the LEDs for the feedback, confidence, and turnkey simplicity they provide. I normally include a Watts Up meter on my conversions since the voltage of a LiFeP04 is not an adequate indicator of the State of Charge.

As far as my EZ3 delta trike, it was a quick, straight forward, and easy install of the BBS02. However, on a tandem bike or trike where the front crank has the chain wheel on the left side the crank drives such as the BBS02 are not compatible. I have seen this discussed on the forum but never had any interest in it and did not follow the threads.