How do I diagnose my battery (random shutdowns)

Hi ESFMs,

I'm having this recurring issue lately with my battery where it will just shut down.

The first time it happened, it was after a phase wire broke. After repairing the connection, not only did my battery not respond but I was also getting low voltage readings of 19v. It's a 48v pack.

After pulling the pack apart that night and discovering that I had no continuity because of a connector, I replaced the connector and started getting healthy readings of 50v again and it had been fine since.

Until today. I did everything I did last time. I opened the pack, checked for continuity on both the red and the black discharge wires. This time they were solid. Using my volt meter, I kept getting readings of 10v, 14v or sometimes 19v. Then somehow, I don't know how, it fixed itself and started giving healthy readings of 51v. I finished my frustrating journey without any problem but it did it later in the day.

What is happening and what do I look for? How do I troubleshoot this? I'm not sure, but it seems to fix itself when I move the red wire around.

I also checked on the bms voltage per series(the first time it happened). Though I sorta rushed it because I kept sparking them by accident and I didn't want to ruin any cells. They all had a reading of 3.7v (if memory serves me) how do you test those without accidentally crossing our touching 2 at the same time?

ClintBX said:

What is happening and what do I look for? How do I troubleshoot this? I'm not sure, but it seems to fix itself when I move the red wire around.

Click to expand...

Then that wire, or whatever it connects to, is probably broken inside and needs to be replaced or respliced.

Or moving it is causing something else to move, reconnecting or disconnecting it. (like a connection between the cells, or a cell and the BMS).

WHenever you have any sort of intermittent problem, then whatever you do that "fixes" it evn momentarily is almost certainly the cause. And almost always, intermittent problems are caused by connections or wires.


BTW, sometimes when working on things that are powered on (like with batteries) continuity tests may fail to show the correct reading, because of voltages across things (that should be connected but aren't, for instance).

But a voltage check across things will show connections. If you connect a meter across a wire that should have continuity, and it shows a voltage, then it has a broken connection somewhere between the meter probes (otherwise there'd be no voltage because it'd just be a wire).

You have to be careful after sparking them as the terminals sometimes don't quite connect right afterwards depending on the spark size.
And you either have to have really steady hands and skinny probes, or really thin probes to go into the actual holes. Also if your math is good, you can leave one probe at one end, and keep subtracting the voltage of the previous cells as you move the other probe down the line.

Totally a guess, but if all the groups are even, then maybe where the power wires are attached to the BMS is getting loose, which could explain the fluctuating voltage. I just had to fix two batteries where the solder blob attaching the positive wires came off the board.

amberwolf said:

BTW, sometimes when working on things that are powered on (like with batteries) continuity tests may fail to show the correct reading, because of voltages across things (that should be connected but aren't, for instance).

But a voltage check across things will show connections. If you connect a meter across a wire that should have continuity, and it shows a voltage, then it has a broken connection somewhere between the meter probes (otherwise there'd be no voltage because it'd just be a wire).

Click to expand...

Thanks for that tip. I'll try that next time I get issues.

So far, it hasn't happened since then. When it happens again, I plan to do a volt test right at the base of the discharge wires where the solder joins are.

I also suspect a probably loose spot weld on one of the cells. I noticed that my volt readings would change after flipping the battery around. I'm just speculating but it could be a loose tab that connects most of the time but under certain conditions, may be losing contact.

Voltron said:

You have to be careful after sparking them as the terminals sometimes don't quite connect right afterwards depending on the spark size.
And you either have to have really steady hands and skinny probes, or really thin probes to go into the actual holes. Also if your math is good, you can leave one probe at one end, and keep subtracting the voltage of the previous cells as you move the other probe down the line.

Totally a guess, but if all the groups are even, then maybe where the power wires are attached to the BMS is getting loose, which could explain the fluctuating voltage. I just had to fix two batteries where the solder blob attaching the positive wires came off the board.

Click to expand...

Thanks for that. I'm definitely using that when I need to.

Though, I think I need smaller probes because I was accidentally sparking them with just one probe.

Well, it happened again today. Numerous times. After having to stop every 20 meters to unplug and plug back in, I decided to look at the connectors. I found that one of the black negative connectors was in need of changing. So I changed that one but it persisted. I looked at the red connectors and changed some of those. I had to use butt splices to connect 2 different gauge wires because the connector in using were too thin.

And when I thought it had stopped. (maybe, it has ), I go to plug my battery for charging and the charger blows on me. I'm losing my mind right now.

I just went to test the charging port's voltage reading and it sparked. I only put one probe on the center pin. Unless I accidentally touched the side. It's one of those 3.8mm pin charging connectors. (I think that's the measurement. Not sure) how do I test that safely without sparking it?

I have a spare charger but I don't want to blow that too.

It's still shutting off randomly and I'm on my way home to open the pack and take a deep look.

I'm home now. I've got the pack open and can't see anything. The volt tests keep checking out as okay. But oddly enough,it was shutting off just waiting for the train when I plugged it in a couple of times.

It worked well enough for me to ride from the train station to home. 3kms I think.

I tested my other battery. A 48v 20ah and it didn't have any similar issues. So, that rules out the controller.

I tired to test the bms series and I'm getting mostly 3.8of the ones I was able to do. I keep sparking them I even managed to spark them with a probe on both ends.

The pack is still reading 49v. So, I don't think I did any damage but there is a char Mark where I sparked the end series on the bms.

I'm gonna keep looking but I'm totally lost. I can't peel back all the foam and paper because there are layers under layers and the connectors to that bms (with all the white writers running to the cells ) is glued in or something.

Here's a few shots of what I'm looking at.

Here are more shots of the bms. I don't see any loose solders or connections.
And the back of the charger. It bad.

I had tab welds break off, but my battery was breaking zip ties and fell out of its rear holding.

After troubleshooting like crazy, I gave up my search and put everything back together again. Even gave it a test spin and had no recurrence in the 5 km ride.

I'm at a loss. The only thing I can think of, is that I have a bad bms. Maybe it's a heat issue. The battery barely feels Warm when I ride but I did upgrade controller recently to a 35 amp Max. Before, I had a 25, and before that, I had a 30 Max. Then again, the first time I had this problem was right before putting in the new controller.

Any ideas or experience guys?

Oh! And by the way, my spare charger didn't blow thankfully. It's charging right now.

BMS failure is quite possible. And not sure from the pics, but pulling the big flat white JST plug for the balance wires makes it easier to probe them, even it it means peeling the black goo off.

Voltron said:

BMS failure is quite possible. And not sure from the pics, but pulling the big flat white JST plug for the balance wires makes it easier to probe them, even it it means peeling the black goo off.

Click to expand...

Yeah, that's left over foam you see there. It also has some hardy glue holding it in place.

Not quite sure how to here past the glue seal. Any ideas?

A little controlled blow dryer or heat gun use can soften things up sometimes.... also since you have the battery that far open now you could skip the balance wires and measure directly from the groups.

Voltron said:

A little controlled blow dryer or heat gun use can soften things up sometimes.... also since you have the battery that far open now you could skip the balance wires and measure directly from the groups.

Click to expand...

I did that. That also checked out as okay with the 49v reading.

How can you tell that it's a bad bms? Is it by process of elimination?

I mean, if everything is reading as okay, does that mean it must be the bms or... If it was the bms, then something would have read wrongly in my tests?

ClintBX said:

How can you tell that it's a bad bms? Is it by process of elimination?

I mean, if everything is reading as okay, does that mean it must be the bms or... If it was the bms, then something would have read wrongly in my tests?

Click to expand...

BMS are cheap, just slow to reciev3 from China. They ar3 chea0 enough that i order a spare with battery whenever i can. If you have one its easier to eliminate the issue.

https://annpower.aliexpress.com/store/group/BMS/623665_250784538.html?spm=2114.12010608.0.0.652d36d5Vr4ZSe

Well, it seems that my issues have abated for now. I started my day with no issues and after more than 20 kms of riding, so far so good.

I don't know what I did or if I even did anything for it to stop but I have a theory.

I think it happens when I charge it and it either doesn't quite fully charge or the charger has a bad connection and flickering red (I remember when my last charger blew, it caused the battery to shutdown for a few minutes ).

So I think that maybe there was something buggy with the charger.

When it happened yesterday, it was right after I took it in to a cafe to charge it. It registered as full earlier than usual. I unplugged it and plugged it back in and it was flashing green. It was shy of about half a volt from usual. That's when it started doing is random shutdowns again.

And the other day, when I had it happen on my way out, I noticed that my charger stopped short of a few decimals (54.4 instead of 54.6) I didn't think it was a big deal.

My best guess is you had the same ol problem I've had a few times,, a bad plug on a charger disconnects it, and the light on the charger goes green. Green means charged, or disconnected.

Your problem might have been inside the charger, I've seen that too, but the result is the same, you take off with a low battery, and viola,, it stops quick.

Confirm your battery is really full each charge somehow, with a real voltmeter. If this was your problem, the old charger never got it full, or disconnected somehow, the problem will go away now.

It could be a loose bms wire, or plug pin too, that would spoof the bms that a cell group was empty.

dogman dan said:

My best guess is you had the same ol problem I've had a few times,, a bad plug on a charger disconnects it, and the light on the charger goes green. Green means charged, or disconnected.

Your problem might have been inside the charger, I've seen that too, but the result is the same, you take off with a low battery, and viola,, it stops quick.

Confirm your battery is really full each charge somehow, with a real voltmeter. If this was your problem, the old charger never got it full, or disconnected somehow, the problem will go away now.

It could be a loose bms wire, or plug pin too, that would spoof the bms that a cell group was empty.

Click to expand...

I've been pretty much making a habit of checking after every charge with my trusty multimeter. I also have a throttle indicator (though less reliable but gives me Ball park figure when riding) and I have my watt meter mounted on my "dash" that I sometimes use.

But your right. Without opening the pack and testing each set, there's no way of knowing when there's an under charged one that could be tripping things.

That said, I still suspect that charger. I bought this charger twice. The first one, when it blew I put it down to the wires that were getting worse for wear. This second identical charger that I've only been using for a couple of months has me Suspicious of its quality.

Anyway, I'm now using a spare charger that came with the battery when I got it. It tied me over last time without any issues. Hopefully, it'll do until I get a new charger again.

I don't think anyone has mentioned downtiube cases, wire corrosion , and connection problems. The worst being bottle and shark batteries. Not like the case here, but very similar symptoms.

tomjasz said:

I don't think anyone has mentioned downtiube cases, wire corrosion , and connection problems. The worst being bottle and shark batteries. Not like the case here, but very similar symptoms.

Click to expand...

I ruled out connection problems when I was fighting with it a few days ago. I even changed old connectors incrementally to rule out sources. Though I did the phases last later that night after opening the pack and taking the snaps I provided earlier. But I can't be certain that the phase connectors could have caused it because I did 2 trouble free test rides. One with this battery and another with my smaller 20ah battery pack.

I didn't think about wire corrosion. Most of the wires are quite new but I have been using a couple of extension wires to bridge the length between battery and controller. Though, they're about the same age as the battery itself. 9 months.

As for downtube cases, I don't think that's an issue. I have a triangle pack in a triangular frame bag.

So far, touch wood. I haven't had any issues since I blew my last charger and have had to use my spare 3A charger. For some reason, it only charges my battery to 53v instead of the 54.6 it's supposed to do. So, I won't be getting any top balancing for a while until I get my hands on a new 4 or 5A charger.

Well, I must have jynxed myself with my last post. It happened again yesterday. Though, this time I think it was a badly crimped connector I changed recently that set it off.

At first I thought it was that wire because it was showing no continuity. This was the black ground extension wire I mentioned earlier. So I swapped it with a new 10 AWG wire I had set up earlier. Despite replacing the suspected culprit, I still got random shutdowns for the next 3 kms.

I later swapped the red extension wire with another 10 AWG copper wire and opened my pack to thoroughly measure my cells and test the grounds on the bms.

I found a way to measure the cells through that white connector on the bms without accidently sparking them. It involves electrical tape wrapped around the probes. The width of the tape is just wide enough to wrap the probes of my volt meter while leaving just the tip exposed to take a reading. There's a hack for ya.

Anyway, I got an even reading of 3.5v per cell and with a total pack volt reading of 46.2v. I know, there's a discrepency of 0.7v but my multimeter on had a one digit decimal placing. So, maybe they were something like 3.55v each.

I've changed connectors, I've uprdaded some wires and I'm using a different charger than when this problem first arose.

If this happens again, my next plan of attack will be to switch back to my 25 amp controller.

If that fails, I'll question my throttle. Seems unlikely but it could be creating a short that I'm not aware of.

Nice probe hack.... what kind of displays do you have, i.e. when it shuts off do all the displays or voltage LEDs go dark too, or does the power just cut out but leave power indicators lit?