As Ben has mentioned go over the BMS with a mag glass or takes clear pics and use your pc to blow them up to look for any damage how ever minor.

 

The charger has been ruled out so only the battery/BMS & wiring can be the issue.

Not very technical I'm afraid but usually faults stand out to see.

Start with the battery charge port and try continuation from charge pins to BMS solder joint on the Negative and to the battery + on the positive (though unlikely as early in the thread you registered 36.8v on the charge pins).

All BMS wiring is negative except the sense wires which are all positive bar one.

 

My thoughts are BMS, sense wire connections to battery groups there isn't much else to go wrong (we know sense wires are ok as you have got ten good cell voltage readings).

Continuity check either side of the charge fuse to rule out a break down on that line.

 

Once every thing has been exhausted check wise then we can only assume the BMS has a fault.

Edited May 19, 20205 yr by Nealh

Curioser and curioser..

 

I rechecked the output voltage of the battery, which was 37.2V last week, to find it has now dropped to 29.9V. I then rechecked the voltages of the original cells, which were previously all 3.70 or 3.71V, to find that they are now all between 2.97 and 3.01V. These seem quite significant decreases, but the battery has not been connected to anything in the interim.

 

I then placed the black probe on B- and the red probe consecutively on B1 to B10, which I understand should produce increments of 4.2V in a healthy battery, and obtained the following voltages;

 

B1 2.97V

B2 6.97V

B3 10.91V

B4 13.95V

B5 17.87V

B6 19.60V

B7 19.60V

B8 29.60V

B9 29.60V

B10 33.30V

 

These results suggest some very significant imbalances between cells, unless there is some reason why this is not a valid method of measurement?

 

I then rechecked all the cells using the method Nealh recommended and found that they were still all between 2.97 and 3.01V, which suggests good balance.

 

They can't both be right, so why are they producing different results, which I've now checked and confirmed several times?

Ok, we might be getting somewhere now as we were basing issues around all previous voltage readings. There are some big differences from one week to another either you messed up somewhere or the meter is giving you incorrect readings, replace the meter PP3 battery with a new one and take the readings again. A meter battery that is low can give erroneous readings.

Now we have at this time new lower voltages to go by then they now make more sense as to why the charger isn't charging, because the BMS isn't switching on due to all cell groups being way too low voltage wise. The BMS won't switch on unless the voltages are above 3.2 -3.3v on each cell group.

Curioser and curioser..

 

I rechecked the output voltage of the battery, which was 37.2V last week, to find it has now dropped to 29.9V. I then rechecked the voltages of the original cells, which were previously all 3.70 or 3.71V, to find that they are now all between 2.97 and 3.01V. These seem quite significant decreases, but the battery has not been connected to anything in the interim.

 

I then placed the black probe on B- and the red probe consecutively on B1 to B10, which I understand should produce increments of 4.2V in a healthy battery, and obtained the following voltages;

 

B1 2.97V

B2 6.97V

B3 10.91V

B4 13.95V

B5 17.87V

B6 19.60V

B7 19.60V

B8 29.60V

B9 29.60V

B10 33.30V

 

 

Elaine going by those new readings your battery is dead/a goner.

These are extrapolated to individual cell readings from the above something isn't correct esp B8 10v isn't possible.

Recheck all again with a new PP3 battery in the meter please.

B1 2.97v

B2 4.0v

B3 3.94v

B4 3.04v

B5 3.92v

B6 1.73v

B7 0v

B8 10v ?

B9 0v

B10 3.7v

Elaine going by those new readings your battery is dead/a goner.

These are extrapolated to individual cell readings from the above something isn't correct esp B8 10v isn't possible.

Recheck all again with a new PP3 battery in the meter please.

B1 2.97v

B2 4.0v

B3 3.94v

B4 3.04v

B5 3.92v

B6 1.73v

B7 0v

B8 10v ?

B9 0v

B10 3.7v

 

 

OK, I'll replace the meter battery and report back. Could a fault with the meter explain why the two methods of measurement gave such different results, which were consistent as I measured them several times?

Certainly yes if the meter battery is tired, voltage readings can be all over the place.

 

Following the second batch of voltages I certainly now believe your ebike battery has a low cell issue and why the BMS won't allow both chargers to work.

 

If the meter battery is old or failing and hasn't been used for a while, on the first use energy may have been stored up and then on the second use it wained and why you have got widely different results. I can't believe the BMS has somehow failed over such a short period of time.

See what results occur with a new battery installed.

These are the readings with new batteries in the meter. They were taken from the pins in the plug after first disconnecting the plug from the BMS. For each reading I held the probes in place until the meter gave a steady reading, and where the values differed from those I had obtained previously I took extra care to make sure I had a good contact.

 

Using the method you first suggested, i.e. measuring across B-/B1, B1/B2, B2/B3 etc. all ten cells gave readings of 2.97 or 2.98.

 

Measuring from B1 to B10 using B- as the neutral gave the following five sets of readings;

B1 2.98 2.98 2.98 2.98 2.97

B2 7.00 7.00 6.98 6.99 6.69

B3 10.70 10.99 10.69 10.96 10.40

B4 14.42 13.98 14.41 14.67 14.11

B5 18.13 17.94 17.90 18.38 17.82

B6 19.90 19.90 19.90 19.90 19.90

B7 19.90 19.90 19.90 19.90 19.90 B8 19.90 19.90 19.90 19.90 19.90 B9 29.90 29.90 30.00 27.30 29.90 B10 29.90 33.50 33.60 30.0 29.90 I appreciate that the 10V jump from B8 to B9 is impossible, but that's what I got, which is why I took multiple readings. Looks like I need a new battery, but I still don't understand why the method of measuring across the pins in pairs gives near identical readings for all ten cells?

Due to the lack of use all of the cell groups have equally drained so whilst connected on the bike a residual small current has done this. Now we have got to the bottom of things and we have 10 pretty well balanced low voltage readings they are recoverable at above 2.8v each.

 

All your battery needs is for all 10 cell groups to be manually raised to above 3.3v each then the BMS will allow charging.

 

You need an old phone charger it will be 5v rated but that doesn't matter, chop off the charge plug .

Bare the two wires and solder on a pin to each wire using one from a computer pin strip, insulate each end with a bit of Red and Black heat shrink or tape leaving 5mm of pin showing make sure polarity is checked and correct.

 

This my one.

Edited May 20, 20205 yr by Nealh

Using the modified 5v charger you can now charge each cell group up a little one at time, you will need your meter to check on process and must not leave the battery unattended. Each cell group if viable will take 40 - 60 mins approx. to raise above 3.3 v or more to be sure 3.4/3.5v will be better. Time the first cell group and then use that as a basis for each cell group to 3.5v.

Insert the pins with correct polarity in to the sense wire connector as you did earlier B-/B1, B1/B2, B2/B3 etc,etc until all are about 3.5v, don't rush the process and if it takes a day or two doesn't matter then reconnect BMS and try the charger. It should click and start the charge with the Red light showing and turn Green when finished, if all looks quite good put the battery back together and use. Balance might be a bit out so may take a few discharge/charge cycles to get right, for the first one or two charges try leaving the charger connected for a 3 - 4 hrs as the BMS should bleed down any higher cells then raise lower cells to balance all out. The balance process is slow and typically only about 50 ma current is sued hence the 3 - 4hrs balancing needed.

Edited May 20, 20205 yr by Nealh

I have gone over the last few posts and something is a miss with the measurements, the B1- to B1 , B2 , B3 etc,etc, measurements should go up fairly equally by 2.97 - 3v all the way to 29.9v or so, as indicated by the B-/B1, B/B2, B2/B3 etc, etc measurements. Both sets of results should correlate and back each other up.

 

Do the B-/B1, B1/B2, B2/B3 etc,etc measurements again and write each one down as you see it and post the ten measurements /readings.

Edited May 20, 20205 yr by Nealh

I have gone over the last few posts and something is a miss with the measurements, the B1- to B1 , B2 , B3 etc,etc, measurements should go up fairly equally by 2.97 - 3v all the way to 29.9v or so, as indicated by the B-/B1, B/B2, B2/B3 etc, etc measurements. Both sets of results should correlate and back each other up.

 

Do the B-/B1, B1/B2, B2/B3 etc,etc measurements again and write each one down as you see it and post the ten measurements /readings.

 

I repeated the B-/B1, B1/B2, B2/B3 measurements several times with the new batteries in the meter, and got very consistent results with all ten cell groups reading either 2.97 or 2.98.

 

Measuring from B1 to B10 using B- as the neutral gave noticeably more erratic results, as above, with three consecutive cell groups showing 19.90v, then jumping by 10V to 29.9.

 

Since I've now loaded the meter with brand new Duracells I don't understand why the two different methods of measurement are providing such conflicting results. How could this be?

I repeated the B-/B1, B1/B2, B2/B3 measurements several times with the new batteries in the meter, and got very consistent results with all ten cell groups reading either 2.97 or 2.98.

 

Measuring from B1 to B10 using B- as the neutral gave noticeably more erratic results, as above, with three consecutive cell groups showing 19.90v, then jumping by 10V to 29.9.

 

Since I've now loaded the meter with brand new Duracells I don't understand why the two different methods of measurement are providing such conflicting results. How could this be?

 

A thought, with your meter did you have voltage set at 20v or 200v for the B- measuring .

A thought, with your meter did you have voltage set at 20v or 200v for the B- measuring .

 

My meter doesn't have selectable voltage ranges, it scales automatically.

 

Something doesn't add up though. As I understand it the two methods are essentially measuring the same thing but in different ways, and reading along the cells from B- should show a cumulative addition of the individual voltages measured between them, surely. I don't understand how the same meter measuring the same cells using the two methods produces results which don't correlate, and it's an important contradiction, because one set of results says charge the individual cells and it'll be OK, whilst the other says buy a new battery...Very odd

I have a similar meter which auto registers voltage and has no issue with both methods correlating. The important factor is the first and usual method of individual cell group measuring confirms all are near balanced, I would tread carefully and with the nice weather raise voltages and charge outdoors.

OK, but there is also the mystery of the voltage drop over this week. Individual cell groups measured 3.70 last week, but still wouldn't charge.

 

 

I have a similar meter which auto registers voltage and has no issue with both methods correlating. The important factor is the first and usual method of individual cell group measuring confirms all are near balanced, I would tread carefully and with the nice weather raise voltages and charge outdoors.

We don't know if they were false readings or not ?

Your experience certainly isn't what I have seen before.

 

If you think the pack is scrap and plan on replacing it, then no loss by raising cell voltage and charging outside away from the dwelling for safety reasons alone.

If all goes well then can only assume an anomaly with meter reading or BMS at fault.

If a cell group has an issue one or two may be different but all appear equal under the first voltage measuring practice.

Edited May 21, 20205 yr by Nealh

We don't know if they were false readings or not ?

 

Well although the battery output and individual cell voltages declined significantly over the week the output of the battery charger remained exactly the same, suggesting that the meter was not the variable. Moreover, the new readings were obtained before changing the meter batteries, which didn't have any effect on voltage readings.

 

If you think the pack is scrap and plan on replacing it, then no loss by raising cell voltage and charging outside away from the dwelling for safety reasons alone.

If all goes well then can only assume an anomaly with meter reading or BMS at fault.

If a cell group has an issue one or two may be different but all appear equal under the first voltage measuring practice.

 

Fair point, nothing to lose as long as I avoid burning the house down.

 

I'll do as you suggest and let you know what happens....

I soldered two pins to a 5V phone charger, checked the polarity and connected to B-/B1. The voltage of the cell was 2.98V when I started, and after charging for 2.5 hours the voltage was 2.97. I checked the output of the phone charger at the pins before and after the session and confirmed that it was delivering 5V at the pins.

 

This battery continues to defy logic....

 

Thank you for all your help, it really is appreciated, but I think this is now a job for re-celling. Maybe the inability of this battery to respond is down to the fact that it is 8 years old, and so just a little too far over the hill.

Eight years old is certainly going some, I think you've had your money out of it.

Sounds like another job for Jimmy. He did mine with Boston Power swing cells - think it was in the £250 region.

Yes scrappy time.

Thanks again, and yes, Jimmy will be fitting the Boston Power Swings....