Battery Management System Tripping early

[Adamsavage79]

I think I have a bad or the very least problematic BMS. The problem I'm having, is that the BMS is tripping too early. It thinks the battery is too low, and cuts the power. However, when I unplug one of the BMS plugs, and plug it back in. This seems to reset it. When I check the voltage. I'm getting a reading of 51.72 Volts. This isn't generally low enough for the BMS to trip. Whenever the battery pack gets to this voltage, this is when the BMS is tripping. When I reset the BMS, I can typically go a couple hundred feet on the bike, before the BMS trips and cuts the power.

Is it a bad BMS, or have I reached the end of the lifespan for my pack ?

Thanks, John

Edit: This might be the Golden Motor kit.. Seems @ 50.9 volts, nothing will turn on. Seems like a rather high cut off voltage for a 48v pack.

[Bikemad]

Hi John,

Without knowing what battery pack you have, how old it is, cell count, chemistry, and how many charging cycles etc. it is difficult to say for sure, but I suspect it will be one or more of the cells beginning to weaken.

Was your battery fully charged before setting off on your ride, and if so, how far did you ride before it started cutting out?

Were there lots of hills or maintained high speed runs during the ride?

If your batterys is a LiFePO4 pack, it will have 16 cells (or 16 groups of paralleled cells) in series.
Each of these cells has a nominal voltage of 3.2V, therefore the battery is technically a 51.2V battery not a 48V battery.
The typical charging voltage for this battery would be 58.4V (16 cells @ 3.65V per cell) or 58.8 if it's a 14 cell LiMn battery.

If your battery is reading 51.72V in a "no load" state (that's an average of ~3.23V per cell if it's a LiFePO4 battery), then the cells are probably ~90% discharged, and there's a good chance that a single weak cell could easily trigger the low voltage cutoff on the battery's BMS when load is applied.

Without carefully monitoring the exact voltage of the battery under load, immediately prior to the BMS cutting out, it is difficult to say what the loaded voltage was, because all you can measure is the unloaded battery voltage, which will have risen significantly because the load that was applied to it has been removed.

Also, if the battery LEDs are not lighting up on the throttle, then I would say that the battery is not supplying sufficient voltage, and none of the settings in the controller could cause this to happen. 

Alan
 

[Adamsavage79]

I was getting 51.7 so I thought. Then I found out the voltage meter I had was defective. After I changed the battery in my other one, I then got a reading of 50.9. The Smart Display and throttle would basically flash at me, to indicate a low battery. This was with the bike not moving, and no throttle applied. The pack is a LiFePO4 pack, and they are flat cells that are grouped up. I did get a reading of 6.46 volts from basically all the groups I checked, expect one. There was one that was reading 6.06 Volts, or something close to that.

As for the ride. The first leg was 11.5 km, of which max throttle was applied about 60-70% of the time. I had a pretty big hill along the way, which I reduced the throttle back to help extend range. 2nd leg was 600 Meter. 3rd was 2.3km. The final leg was 16.2 km with a mixture of pedaling, costing and applying the throttle on and off quite a few times. There was also lots of stop and go, but I would pedal to get the bike going up to speed first, before using the throttle. There was only a couple times, I would start from a dead stop.

As for the age of the Battery. It's a little over 3 years of heavy use, so I now suspect the battery is near the end of it's life cycle. I was also reading, that when you apply a very high load to the battery, it shortens the life of the battery each time.

So with that, I'm thinking either Ping Battery again, or trying out a Golden Motor battery, which is located in Canada and would result in quick shipping, however it is about 3 pounds heavier vs the ping battery.

[Adamsavage79]

I ran the bike today till LVC, and ran it hard to shorten the distance needed. I got about 17 ish KM out of it, due to a headwind and a few going from a dead stop.

I did two different measurments. I did the method you guys asked me and the method Ping asked me to do. I did this with a new multimeter.

The 1st Method I did was, measuring the cells by putting the black prob on the "B-" and then uncliping the 16 wire BMS plug, and putting the red probe on the 1st pin. Then moved the red probe to 2nd pin and black probe on to the 1st pin, and moved from left to right, all the way to the 16th pin. This is what I got.

Pin 1: 3.25v
Pin 2: 3.27v
Pin 3: 3.27v
Pin 4: 3.26v
Pin 5: 2.90v
Pin 6: 3.26v
Pin 7: 3.27
Pin 8: 3.27
Pin 9: 3.27
Pin 10:3.26
Pin 11:3.26
Pin: 12: 3.26-3.27 (fluctuates)
Pin: 13: 3.27v
Pin: 14: 3.27v
Pin: 15: 3.26v
Pin: 16: 3.26v


The 2nd Method, what I did was, measuring the cells by putting the black prob on the "B-" and then uncliping the 16 wire BMS plug, and putting the red probe on the 1st pin, then keeping the black probe on B-, I moved left to right, till I reached the 16th pin on the wire. This was the results.

Pin 1: 3.25v
Pin 2: 6.53 - 6.54v (fluctuates)
Pin 3: 9.81v
Pin 4: 13.09v
Pin 5: 16.03v
Pin 6: 19.30v
Pin 7: 22.5v
Pin 8: 25.8v
Pin 9: 29.1v
Pin 10: 32.3v
Pin 11: 35.6v
Pin: 12: 38.9v
Pin: 13: 42.2v
Pin: 14: 45.4v
Pin: 15: 48.7v
Pin: 16: 52.0v

[Bikemad]

It looks like cell no 5 is at least 0.35V lower than all the other cells, and is causing your BMS to cut the power from the battery to prevent it from being further discharged.

I suggest that you fully charge the battery and then measure the cells again to check that they are all fully charged.

If cell 5 is at the same voltage as the others after the initial charge, I would say the cell is faulty and will need to be replaced.

However, if cell 5 is still substantially lower than the others after a full charge, I would leave the battery on charge for a few more hours to see if it improves.

If it does improve (but not 100%) leave it charging to see if the BMS can balance the cells to allow the low cell to reach the same voltage as all the other cells. If it is just an out of balance cell, it could take a long time to fully charge.

To prevent the other 15 (already fully charged) cells from being overcharged, the BMS uses resistors to bleed off any excess voltage. But as this discharge is usually only around 200mA, it usually limits the charging current to a similar level (or cycles it on and off) during the balancing phase at the end of the initial charging process.

If cell 5 is unable to reach the same voltage as the other cells, it is probably a faulty cell (although it is also possible that a faulty BMS could also prevent it from being correctly balanced).

Measuring the cells one by one is generally more accurate as they can all be measured on the same lower scale setting on your multimeter (e.g. 0-20VDC) which gives a higher precision reading (2 decimal places).



Using the 0-200VDC scale is generally a much lower precision readout to just 1 decimal place, which is not really accurate enough for cell voltages. 

Your results from measuring the cells the way Ping suggested illustrates this nicely:

Pin 1: 3.25v  3.25V
Pin 2: 6.53 - 6.54v (fluctuates)  3.28V
Pin 3: 9.81v  3.28V
Pin 4: 13.09v  3.28V
Pin 5: 16.03v  2.94V
Pin 6: 19.30v  3.27V
Pin 7: 22.5v  3.2V
Pin 8: 25.8v  3.3V
Pin 9: 29.1v  3.3V
Pin 10: 32.3v  3.2V
Pin 11: 35.6v  3.3V
Pin: 12: 38.9v  3.3V
Pin: 13: 42.2v  3.3V
Pin: 14: 45.4v  3.2V
Pin: 15: 48.7v  3.3V
Pin: 16: 52.0v  3.3V

The results in green are rounded to the nearest tenth of a Volt whereas the results in blue are within one hundredth of a Volt.

At least you now know why the battery was cutting out (the BMS main power cut-off being triggered by the low cell voltage).

Alan
 

[Adamsavage79]

I did leave the battery on the charger for about 15 hours, and took a reading after. All the cells where reading the same voltage. I didn't however, check just before I rode the bike, as the battery sat overnight. That being said, that same weak cell right after the charge was done, was reading a little bit lower of a voltage than the rest.

This will mark the 2nd time I've had cell 5 give a much lower reading, than the other cells, and I agree that this is why the LVC is turning on. If it turns out, I have a weak cell. I'm not sure it would be a good thing to replace the weak and old cell, with a brand cell. It would be much cheaper, but I wonder if it could cause problems down the road.

[Adamsavage79]

I let the battery stay on the charger for 23 hours roughly and then let it sit for a few hours. Then I took a reading and it's the same as before. All 16 cells are giving me equal readings of 3.60v. It's only when I start to use it, that the one cell is discharging faster than the others, it seems.

[Bikemad]

That same weak cell right after the charge was done, was reading a little bit lower of a voltage than the rest.

All 16 cells are giving me equal readings of 3.60v.

Looks like the BMS is working fine and has balanced the cells correctly, now you'll need to find a replacement cell if your range has not improved now that cell 5 is fully charged to the same level as the other cells.

If you replace the weak cell with a single brand new cell of the same capacity, your available battery capacity will be as good as the remaining  weakest cell (Probably cell 1, not cell 5 that you are replacing )

Alan
 

[Adamsavage79]

This would be my cheapest option. The health of the battery's themselves to be good, so I'm not sure if mixing old with new would be bad. Will talk to the Battery store Tuesday and see what they say.

[Adamsavage79]

It would seem I can't replace the cell, as I don't know enought about it or where to get a new one, aside from ping. Even then, the way PING made these, it is very tricky remove them and then somehow get the group of them to connect to the square. I attached a picture to show what my pack basically looks like.

I tried going to a battery store, that has some bold claims on it's website, only to be told they don't build batterys and they refuse to deal with Lion batterys, as they are too dangerous. This is from their website, under "Battery Builds"


"FOR NEARLY 25 YEARS TOTAL BATTERY HAS BEEN SPECIALIZING IN THE DESIGN AND ASSEMBLY OF SPECIALITY AND CUSTOM BATTERY PACKS.
No matter if your requirement is for a single custom assembly or you require production runs of many thousands of batteries we have the resources and industry contact to facilitate your requirements.

For any custom battery pack or specialty battery please contact us directly."


So it looks like I will be stuck buying a new pack. I'm thinking about going with Golden motor, but not sure. I would like know though, if I can get replacement cells should I get a weak cell. As for the cell that is weak, I know where it phyically is in the pack as I measured each cell before. I will have to mark it with a marker for future reference.

[Hastings]

What happened with your pack ? Found a cell -Sent to Ping or bought a new one

Almost every glue dissolves in brake cleaner ! Or is it Heat glue ?