Sudden loss of power

[luvxu]

Hi All,

My bike had been running well for over 400 miles until last week where it suddenly loss power and can't climb any slight hill. I noticed that every time I tried to climb a hill, the CA would show that the voltage quickly drop from 51V down to 35V and backup to 51V on flat land.

My bike is set up with the following:
1. Golden Motor Magic pie 2
2. Lyen controller
3. 48V 20A (panasonic 18650 lifepo4)

The curious thing is that if I reprogrammed the Lyen controller to board type EB318 instead of EB312 then it doesn't lose power on hill climbing anymore. However, the bike run a lot slower, the top speed would become 20mph instead of 30mph.

any help to diagnos this problem is much appreciated.

 

[John in CR]

It sounds like a bad phase wire connection that heats up at higher currents. It ran better at the different programming, because the controller expected a lower resistance shunt. That lowered the current limits because the shunt was actually unchanged causing the controller to believe current was higher than actual.

 

[skeetab5780]

if the voltage is sagging that bad on the input side wouldnt that usually mean the battery has an issue? do you have another battery pack you can borrow and try to power it with?
If you dont maybe you can check the voltage of each cell group while under a load to assure that its working properly.

 

[luvxu]

I wonder how to check for the battery cell voltage under load? One day I woke up to find out that the battery has 0V. I then checked each cell and found one cell that was shorted. I replaced that cell and the voltage went back up to 52V. Do you think that other cells may have a problem? I still don't know why one of the cell died like that. It is possible that other cells were weaken some how?

 

[dnmun]

put the battery on the charger and measure the cell voltage of each cell to .01V and post them up here on your thread.

 

[skeetab5780]

if that is a 48v 20ah pack made of 18650 cells you cant just replace one cell, it probable has a bunch of paralleled strings of 10 or more cells wired in series. If you replaced one group of cells with new ones, they could be conflicting with the older ones.

pics and cell level voltages will help us diagnose.

 

[luvxu]

I finally got the time to disassemble the battery. Include is a picture of it.

When I don't have any load, the battery read 49.93V. The 12 cells read:
1. 3.97V
2. 3.84V
3. 3.97V
4. 4.10V
5. 4.17V
6. 4.16V
7. 4.17V
8. 4.17V
9. 4.17V
10. 4.16V
11. 3.99V
12. 3.99V

I noticed that when I put a load on the bike, then the battery voltage drop down to 29.95V. The 12 cells read
1. -15.95V
2. to 12. read between 3.80V-4.10V

I just don't understand why there is such a big negative reading on the first cell. This is why my voltage dropped from 49.93V (no load) to 29.95V (with load).

Any help to explain the problem would very much appreciated.

 

[dnmun]

the BMS should have turned off the battery originally when you left it to discharge to 0V. your pack is very unbalanced for lipo so i doubt if the BMS is actually balancing the cells or you have the wrong charging voltage to force the BMS to balance the cells.

the reason the first cell goes to negative volts is because it is ruined beyond all hope and if all of the cells had previously drained to 0V then they are also dead.

what kind of BMS do you use on this and can you show a picture of it?

 

[eva-michael]

Your motor should be rated 500W and 1000W peak. That's totally not high. After cells confirmation, you should check the connections of the BMS and then see if this bms has been fault.
Climbing is always easy to cause incompetent BMS get broken as it does not capabible enough to load the current during climbing. Mosfet may over heat and get demage.

 

[luvxu]

I replaced the -15v cell and everything is back to normal now.

I think someone is going to yell at me knowing that I had not been using any BMS for the whole time. In the past I have had some battery packs with built in BMS and it always cut out on my motor when I need it most so I built this these new packs for the magic pie bike without any BMS. I figure that if I set my Meanwel bull charger voltage to under 52V and use a cheap homedepot timer (http://www.homedepot.com/p/Defiant-15-Amp-24-Hour-Plug-In-Mechanical-Big-Button-Timer-49810DI/203678109#.UfJufI3vu1g) which set to about 3 hours then the battery is protected from over charging. On the other hand, to protect the battery from over discharging, I set the cycle analyst min voltage to 40V.

The trick worked well for me for over half a year until one day I blew my Meanwel charger through connection carelessness. Then, I replaced the charger with similar one but this time I had the voltage set to 57V instead of 52V (I was using a small analog voltmeter and wasn't able to tell the voltage exactly). So, after a night of charging the next morning I found that the pack voltage dropped to 0V. I had to unpack the battery and replaced the cells that has 0V. The pack voltage went back to to 51V after charging but I start to notice this sudden loss of power after a short ride or whenever I had to climb up hill.

Now you have the complete history.

 

[dnmun]

that is what a BMS does. prevents you from damaging the pack. you most likely damaged the other cells too. no way to know how high they got when you charged it and the high cells don't usually die until you go over 4.2V.