Possible bad Battery

I finally got my new ebike kit up and running, and I've had non stop problems with the battery. When I first got the kit, I plugged the battery into the charger I had left over from Ping Battery. This is when I first noticed problems. The charger kept cycling on and off, but eventually brought it up to a full charge.

Not only that, the battery was giving me very inconsistent voltage readings. When I originally took a reading. It read 45.7 V. I found this odd, as it should of been charged up. I checked it later on and this time it read a little over 53 Volts. I figured, maybe I got a wrong reading. Shrugged it off. Checked again It was @ 45.7V.

I tried with a different meter and I get 52v. Tried my good meter again, and back to 45.7v. I'm like what is going on here. Why is this battery all over the place.

So I put it on the bike and use the bike for a little bit. I got MAYBE 8km out of it, before the BMS tripped @ 48v. Charged it again overnight. Ran it again, and it seemed fine. I had to take it to the bike shop for a wheel fix, as I got loose spokes and such.

Let it charge once again overnight, and when I got the bike back. I went to take it out for a test ride, and soon as I apply and sort of load. The BMS on battery will trip. I can have the bike upside down, and throttle on auto throttle and the motor will keep going for as long as I let it...

It seems to be only when a load is applied, and I draw a large amount of power at once from the battery, that the BMS trips. I'm leaning towards a bad battery, or bad BMS at this point. I've never seen a battery behave in this way before.

More info what battery do you have a link ? It this a diy ? Where you get the motor controller ?

Black Pie 5 Vector wheel motor with a built in Sinewave Controller. I'm using a 48V15AH LiFePO4 Battery. This a DIY Kit, installed on a regular Mountain bike.

I've also used a spare wire harness and throttle, wired up directly. I cut off the plug ends, and wired them up directly that way. Same thing happened. Soon as a load is applied, the battery cuts out.

The battery cycling on and off was probably the bms balancing the cells at the top of the charge

EbikeAus said:

The battery cycling on and off was probably the bms balancing the cells at the top of the charge

Click to expand...

Yes, that typically happens. However, it doesn't explain the erratic voltage reading. Nor the fact that it died so quickly.

Is this the same battery as your previous problems?

https://endless-sphere.com/forums/viewtopic.php?f=14&t=94234

https://endless-sphere.com/forums/viewtopic.php?f=14&t=94227

https://endless-sphere.com/forums/viewtopic.php?f=3&t=93456

https://endless-sphere.com/forums/viewtopic.php?f=14&t=93042

No, that battery along with bike was stolen. That pack had a weak cell, that was dragging pack down.

This is a new kit, but exact same as before. Battery is w Golden motor battery.

The owner of golden motor Canada has been extremely helpful and patient with me so far. However neither one of us know what's really going on yet so I'm just looking for possible feedback, to help figure out this problem.

Since the meters you're checking with give different readings, and troubleshooting requires knowing the readings you get are correct, then:

First thing I'd check is the batteries in your multimeters. If the batteries are low, they can give very wrong readings.

If the batteries in both meters are good, then you should check each meter with a second known-voltage source to see how different they read it.

If they read differently on a second voltage source, by the same ratio they are different when measuring your battery pack, then something is wrong with one of the meters, so the one that reads wrong on the second source should no longer be used to do any troubleshooting with on this battery pack, to eliminate possibility of confusion.

If they both read the same on a second voltage source, then there is something different about the input stage of each meter that affects the reading of the battery.

If the BMS has shutdown it's output (or input) (for whatever reason), it will be relatively high-impedance, so it will still present a voltage on it, but if the meter reading it has relatively low impedance, it will drain down that voltage and read lower than a meter with high impedance.

Similarly, it will drop in voltage (but not necessarily to zero) when connected to a capacitive load like a controller, but may not have enough voltage output to activate the controller. Even if it does, it would shutoff as soon as you blip the throttle, once a load is applied.



Things that will cause the BMS to shut off it's output would be a cell that's too low in voltage.

That can be a static issue, where it simply is just too low, and either won't or hasn't charged up enough to go above LVC.

Or it can be a probelm only under load, and usually only under a high enough load, which changes with it's state of charge.

In either case, it usually means a bad cell or cell group, even in a new battery.

When it is not a bad cell or cell group, then it can be a bad connection between the BMS and the cell group (bad solder joint, bad crimp, broken wire, etc). That keeps teh BMS from correctly detecting the cell group voltage, and what happens at that point depends on the BMS design and the actual problem/voltage reading.


Another problem that can happen is that the load is too high a current for the BMS itself, and it is shutting down to protect itself and the cells.

Some BMSs are programmable, so it might just be set wrong. If it's factory-programmed, it might not be user-alterable, but a growing number support user-settings via bluetooth or other methods.

Some are hardware only, and those aren't usually settable because it requires changing other things (like FETs, maybe shunts, etc)--they're just built for a specific current rating and usage.


If a cell / group is defective, it will also drop in voltage under laod, and trip the BMS.


But I'd test your voltmeters first. Then you'll at least know which readings are real, and be able to trust what you see, and what you post will then let us more accurately help you find the problem.


Either way, once you know your voltmeter is giving you a correct reading, you can use it to measure each cell group, and then measure them under load (not practical while riding, but possible by using a heater or similar as a load, or by putting bike upside down and using your mechanical brake on that wheel to put a load on the motor), and then measure them while charging right after you've drained them enough to trip the BMS on a ride.

The three lists of voltage you will get from doing this will show you which cell group(s) are a problem.

If no groups are a problem, then that leaves the BMS itself.




BTW: you said that when you first got the new kit, you hooked up the battery to your old Ping charger. Why not use the one that came with the new battery, which presumably would be a better match for it?

If something is wrong with the old charger, it might be causing (or have caused) the issues you're seeing. (using the new charger may not make any difference depending on what potentially could be wrong with the old one).

I was able to rule out my meters. I checked quite a few times with both. I would get 45v with my one meter, simply turn off the meter and back on again, and this time I would get 55v for example. I tried this with my other meter and I had the same pattern. Basically no matter what meter I used, I kept seeing the same 10v jump in voltage, which is not normal. I did keep a very detailed record over a couple weeks of the voltage readings from the battery if that would help any. Then again, with my exp. With a week cell, the reading will look fine until you start using the battery.

I also noticed that when I connected the meter to the battery, via clips and then turn it on. I would very briefly see a lower than normal voltage, before it showed a normal voltage. I've attached a picture to show this. I also have a small vid clip to show this behavior.



Link for Video: https://www.dropbox.com/s/ug7qjaoc5j07qt6/20181204_173138.mp4?dl=0

Link for Pictures of the Voltage drop off over time: https://www.dropbox.com/sh/667e19uzcrabvqi/AACuZDkmKIfxQ9yycwttK8gta?dl=0

The very first charge was with old charger, and each one after, was with new one that got mailed out to me.

As for a bad connection. I can rule out the wire harness that connects to the controller. The only thing that could be a issue is the two wires that come out of the controller. I did have the bike working for awile, despite the issues with the battery cutting off too early, and the strange voltage readings. The last time I used it, I went about 4.25km. Removed the battery, charged it up and then removed it from charger when it was.

I then put on a twist throttle to replace the thumb one. I by mistake plugged in the cable for the throttle, into the 2 button switch which has a horn and cruise control as I call it. The the bike/kit would not turn on, until I had properly connected the wires in the correct order. I did manage to push back 1/5 pins part way down, but it seemed to still make a connection.

I also tried a different throttle and a spare wiring harness, wired directly to the wire harness. This was to rule out the pin being pushed down, was causing problems. Despite this, the battery would still cut out under load.

BTW: you said that when you first got the new kit, you hooked up the battery to your old Ping charger. Why not use the one that came with the new battery, which presumably would be a better match for it?

Click to expand...

Because that charger was meant for a LiFePO4. It's rated for the same voltage, it just charges faster. This is why the kit did not come with a charger originally. I tried to save money, but not adding it on because I already had one that I used for years without problems.

Would like a link for your battery as the comtroller could be 25amps or more and bms
Cutoff at 20amps.

999zip999 said:

Would like a link for your battery as the comtroller could be 25amps or more and bms
Cutoff at 20amps.

Click to expand...

https://www.goldenmotor.ca/products/48V15AH-LiFePO4-Battery-in-FREE-Eco-pack-Bag.html


48V15AH Battery LiFePO4
Built in BMS
C-rating 3C (45A)
Weight 8.5 kgs
Life cycles >1500 Cycles

Controller for Motor:

Vector Controller with Sine Wave Technology and Bluetooth
24v/36v/48v Multi voltage
Max 25a Continuous power
Upgraded Mosfet
Regenerative Braking
Easily replaceable by plug
PC programmable
Forward/Reverse

Link to previous battery: http://www.pingbattery.com/servlet/the-12/48V-15AH-V2.5-LiFePO4/Detail

This ran fine for years, till I got a weak cell on the exact same motor, controller etc.

adamsavage79 said:

As for a bad connection. I can rule out the wire harness that connects to the controller.

Click to expand...

It's not the external connections I suggested as a potential problem.

It's the internal connections inside the battery, from the cell groups to the BMS.

If one of those connections is poor or broken, then depending on the BMS design it could cause various problems.

That can include being unable to balance a cell group, to drain it down from being too high in voltage, and the overvoltage would get worse over time, eventually possibly reaching a point where it could damage the cell group.

Or just shutting down becuse it can't read the voltage.

That's one reason that checking the cell voltages with a known-working voltmeter under all of the conditions I listed before is important, in troubleshooting this type of problem.

If you don't check the cell voltages, you won't know if one of the groups is causing a problem unless you just replace them and the problem goes away.

I'm not able to check, individual cells without voiding the warranty on the battery. Soon as I take the blue wrap off I void the warranty. So I'm just going to send the battery back and have him load test it. Then guess go from there. I'm 95% convinced it's a battery problem. It's been acting strange from day one.

If all you're trying to find out is whether or not it is a battery problem, you can simply use an external "dummy load" like a heater, to drain it from full until it shuts off. Using a wattmeter between it and the load will show you whether it is providing the rated capacity, and at what current range, and what voltage range.

It has to be a battery problem if the battery shuts off well before it should, and is not being used at higher currents than it is designed for.

If the dummy load test with wattmeter shows the correct amount of capacity, and it's current load is similar to that of the bike, then the problem is more likely to be outside the battery.

Even with a pair of meters it is possible both power sources(the batteries) are low on voltage and need replacing, I would check them on a well known power source first so would test on the battery charger outlet for consistency or similar item before considering internal battery issue. Using your charger output a constant 54.6 v would indicate the meters power source is fine any fluctuation points to the meters battery.

Ping sets his charger to 58.6 or a little more. Just email Ping he will help. Or is just your old battery a Ping. Your sentence was a little confusing.

999zip999 said:

Ping sets his charger to 58.6 or a little more. Just email Ping he will help. Or is just your old battery a Ping. Your sentence was a little confusing.

Click to expand...

My ping battery along with the entire ebike kit was stolen. This is another kit, with a new battery. The ping charger is 59.6v, the charger that comes with new kit is 58.4v.

ok reread post. 58.4 is fine. Have you contract golden motors ?

999zip999 said:

ok reread post. 58.4 is fine. Have you contract golden motors ?

Click to expand...

Yes, we are working on a possible solution.

So it turns it out, there was some issues with battery. Still don't know the problem was, but I was sent a new battery. The battery was load tested before being sent, and it passed it no problem. I used the bike today and everything worked fine.

Great how hard was it to ship and how much ?

Well I basically put it in a storage box, with original foam packing that was around battery and used some peanuts, as well. Taped up box really well, and shipped it out. They never asked what was in it, so I never told. Cost about $35 to mail, out express post.