1 Group has more voltage sag

[rudivb]

Hello, this is a Chamrider 52v37.7ah 14s13p battery with 18650 Samsung cells running a BBSHD.
I noticed in the Bluetooth BMS that group 1 at rest often has 0.01V difference from the other groups.
And as you can see in the video, group 1 has significantly more voltage sag than the other groups.
Is there any reason for concern? The battery is about 1 year and 4 months old, approximately 20000km.
I don't have any issues with the battery and everything works fine.

Edit: to clarify for.those who don't watch the video, under max load 1600 - 1800watt(33a) the voltage difference is close to 0.2v or 200 millivolts.

Thank you

(pictures are videos, click on them to play on imgur)

https://imgur.com/CqwjH5r

https://imgur.com/uLI8rEx

 

[DogDipstick]

Not enough to promote a great concern. Do not worry. Delta V across the cell banks is not that bad.

When the delta cell to cell reaches 100mV or more, then concern. This 5 incongruity might just be the BMS voltage reading error percent itself...

 

[Hillhater]

Is there any reason for concern?

Hmm ?, it certainly seems that you may have one or more bad cells in that group, which can only get worse !
What warranty came with the battery ?..at only 16 months you should not be seeing this type of issue, ..BUT,..20,000 km must mean how many charge cycles ?.. that may invalidate any warranty.
Either way, i suggest you talk to the supplier, Chamrider, who appear to be a established commercial business who would not want any cusromer issues with their products, and ought to be able to advise you .

 

[Hillhater]

When the delta cell to cell reaches 100mV or more, then concern

There is 200mv under load !

 

[rudivb]

Hmm ?, it certainly seems that you may have one or more bad cells in that group, which can only get worse !
What warranty came with the battery ?..at only 16 months you should not be seeing this type of issue, ..BUT,..20,000 km must mean how many charge cycles ?.. that may invalidate any warranty.
Either way, i suggest you talk to the supplier, Chamrider, who appear to be a established commercial business who would not want any cusromer issues with their products, and ought to be able to advise you .

Hi thanks a lot.
I guess the warranty is 1 year, but I could still contact them. I had a failed BMS about 6 months ago and replaced it with a Bluetooth BMS so I don't know what the complete cycle count would be.

What would the next plan of attack be?
Open the pack and I will have to dismantle the individual cells to test them right?
I was hoping not to, as I use this battery everyday for work and I don't have the experience or equipment to weld the nickel back, although I don't mind learning a new skill set.

 

[DogDipstick]

There is 200mv under load !

Yes thankyou for the correction. The static images told me different but I have watched the vid now. 200mV + under load near the bottom end of the SOC.

 

[eMark]

More than 100mV (general rule of thumb) is when there is a variance among the 14 groups of at least 100mV or greater when the battery is at rest. Currently the variance ranges from only 13mV to 17mV (delta), is due only because of Group 1. Being you are cycling the battery every day it's ok as long as you keep monitoring Group 1 should the delta variance increase from say 17mV to 25mV, then 35mV, then 50mV, then 100mV over the coming weeks/months. Until then try not to discharge the battery uner full load (33amp rate) any more than needed to prolong cycle life longevity and delay any future cell replacement in Group 1.

If possible let your battery sit for 3-7 days after a discharge cycle to see how much (if any) the delta voltage variance in Group 1 increases further. This would indicate at least one weak/bad cell in Group 1. If there is no increase in delta voltage variance (e.g. 17mV to 30mV) after a few days then no reason to tear into the battery at this time to test individual cells in Group 1. As long as you are charging every day or every other day you may not notice any inrease change in the delta voltage variance of Group 1.

200mV lag (under load) from just Grop 1 is no major concern at this time with your 13p battery as the discharge bounce back delta variance at rest is still only 0.017V (17mV) for Group 1 and the health of the other 13 parallel groups is AOK.

 

[rudivb]

More than 100mV (general rule of thumb) is when there is a variance among the 14 groups of at least 100mV or greater when the battery is at rest. Currently the variance ranges from only 13mV to 17mV (delta), is due only because of Group 1. Being you are cycling the battery every day it's ok as long as you keep monitoring Group 1 should the delta variance increase from say 17mV to 25mV, then 35mV, then 50mV, then 100mV over the coming weeks/months. Until then try not to discharge the battery uner full load (33amp rate) any more than needed to prolong cycle life longevity and delay any future cell replacement in Group 1.

If possible let your battery sit for 3-7 days after a discharge cycle to see how much (if any) the delta voltage variance in Group 1 increases further. This would indicate at least one weak/bad cell in Group 1. If there is no increase in delta voltage variance (e.g. 17mV to 30mV) after a few days then no reason to tear into the battery at this time to test individual cells in Group 1. As long as you are charging every day or every other day you may not notice any inrease change in the delta voltage variance of Group 1.

200mV lag (under load) from just Grop 1 is no major concern at this time with your 13p battery as the discharge bounce back delta variance at rest is still only 0.017V (17mV) for Group 1 and the health of the other 13 parallel groups is AOK.

Hi, thank you so much for your informed response!

Yes the variance at rest is indeed very low, I am looking at the voltages right now and currently the difference is only 4mV.
I do ride everyday (delivery driver), it's very rare for me to completely drain the battery, when I tour with the bike I get in the lower voltages, but day to day, I am usually hovering between 30% - 40% and 80% - 90% to preserve battery health. I also only rarely use the full 33amp when I need it, usually I ride with around 10 - 15 amp.

I have a second battery I could use for a while and let this one sit, how far should I discharge the battery first? It is currently at 53.4V. Also wouldn't the BMS keep the groups in balance even in those 3 - 7 days?

Thanks again for your response.

 

[eMark]

I have a second battery I could use for a while and let this one sit, how far should I discharge the battery first? It is currently at 53.4V.

At 53.4V approximate average of 3.814V per parallel group except Group 1 at say 3.800 (14mV delta variance). Let it sit for 2-3 days to see if there is any increase (if any) in delta variance. Depending on amount of increase over 2-3 days (if any) suggests there is a weak/faulty cell or possibly a BMS fault across Group 1.

Also wouldn't the BMS keep the groups in balance even in those 3 - 7 days?

Yes, but only if the BMS happens to be a more expensive "active equalizing" BMS" instead of a passive BMS which is most likely the kind of BMS in your Chamrider 52v37.7ah 14s13p with Samsung cells (30Q?). Being you charge the battery every day or every other day you may only need to let it sit for 2-3 days at most to see if there is an increase in Group 1 delta variance. If so suggests the fault is either with at least one weak/bad cell in Group 1 or a faulty BMS Group 1 circuit.

20,000 km is 12.427 miles. How many km do you travel between each charge/discharge cycle? How long have you noticed a 14mV to 17mV variance at rest on Group 1. Any way if no significant increase when monitoring delta variance of Group 1 over the coming weeks then it may just be a glitch to live with.

Overall your 14s13p battery is in fine condition due in large measure to your non-abusive use ...

200mV lag/sag (Group 1) under full load does suggest their may be at least one weak/bad cell in Group 1, but until the problem gets worse (say 500mV) just keep doing what you're doing.

 

[rudivb]

Yes, but only if the BMS happens to be a more expensive "active equalizing" BMS" instead of a passive BMS which is most likely the kind of BMS in your Chamrider 52v37.7ah 14s13p with Samsung cells (30Q?).

The BMS is not the standard BMS, I upgraded the BMS after a BMS failure, I think it does actively balance the cells. This is the BMS I have now: https://a.aliexpress.com/_EIuDAQj
(50A continues 150A peak)

Edit: actually there is a feature in the app for balancing, so I toggled that OFF
That will do the job!

20,000 km is 12.427 miles. How many km do you travel between each charge/discharge cycle? How long have you noticed a 14mV to 17mV variance at rest on Group 1. Any way if no significant increase when monitoring delta variance of Group 1 over the coming weeks then it may just be a glitch to live with.

It really varies from day to day, between 50km(31 miles) and 120+km(75+ miles). When I was touring last summer (going a lot slower), I might do 200km(124 miles) - 250km(155 miles), I will be touring again this summer.

I only recently noticed the variance, but it could've been there longer as I didn't really monitor it.

 

[eMark]

Edit: to clarify for.those who don't watch the video, under max load 1600 - 1800watt(33a) the voltage difference is close to 0.2v or 200 millivolts.

In your first post you said you have a 14s13p battery ... yet both of your above graphs indicate there are 14 parallel groups.

A 5% additional load sag on Group 1 at 4.0V (under max load) is only 200mV which is acceptable. In comparison what is the average sag on the other 13 parallel groups when "under max load of "1600 - 1800watt (33a)" So with your 14s14p at max load there is an additional 200mV sag on parallel Group 1 ... is that correct? In comparison what is the sag on each of the other 13 parallel Groups when battery is at max load (33a) at say 51.1 volts (2nd Chart) ?

The culprit could be a poor spotweld connection on Grpup 1 causing higher internal resistance leading to an additional sag of 200mV on Group 1. Eventually this could shorten the life of one or more cells in Group 1 ... causing the BMS to shut down the battery.

 

[rudivb]

In your first post you said you have a 14s13p battery ... yet both of your above graphs indicate there are 14 parallel groups.

Ah yes, I see now, my mistake, sorry I think that must be 13s14p, I'm pretty sure I counted 13 cells per group.

A 5% additional load sag on Group 1 at 4.0V (under max load) is only 200mV which is acceptable. In comparison what is the average sag on the other 13 parallel groups when "under max load of "1600 - 1800watt (33a)" So with your 14s14p at max load there is an additional 200mV sag on parallel Group 1 ... is that correct? In comparison what is the sag on each of the other 13 parallel Groups when battery is at max load (33a) at say 51.1 volts (2nd Chart) ?

Looking at the videos I made, under max load, the other cells (roughly) drop about 100mV - 120mV, while group 1 would drop about 300mV, which results in a 200mV difference.
(At 51V, group 1 would drop about 250mV)
(At 57V, group1 drops 300mV max)



I've let the battery sit now for 48 hours and there has been 0mV drain/difference.

I've done a quick visual inspection with my amateur eyes, I couldn't find anything special.

Also under a charge of 4a, group 1 becomes the high group with 8mV difference, which indicates higher internal resistance?

 

[eMark]

Ah yes, I see now, my mistake, sorry I think that must be 13s14p, I'm pretty sure I counted 13 cells per group.

You are correct (13p). What can be confusing to some is the 14 readouts in those charts. Each of the 14 readouts is the average combined parallel voltage (13p) of each of the individual series strings (14s). Each of the 182 cells is actually serving as both a cell in a series string as well as a cell in a parallel group. Optimum bus bar efficiency for keeping all the cells balanced assuming all cells are equal quality A-grade, no faulty spot welds, no faulty solder connections to sensing wires and no faulty BMS circuit.

I've let the battery sit now for 48 hours and there has been 0mV drain/difference.

There still could be at least one cell in that parallel group that's not up to par with the other cells during discharge.

I've done a quick visual inspection with my amateur eyes, I couldn't find anything special.
Also under a charge of 4a, group 1 becomes the high group with 8mV difference, which indicates higher internal resistance?

Now if only a so-called "SMART" BMS could tell you exactly where the problem point is located .

Have you tried placing your voltmeter probes on the neg & pos end of each of the 13 cells in group 1 to check for any noticeable disparity (even though all 13 cells in parallel). Depending on the disparity it may be time to replace at least 1 or more of the cells in group 1. Any noticeable disparity suggests at least one faulty cell in group 1 and over time it will become more noticeable; especially with frequent MAX load discharge riding.

As is (no further discharge at rest) just keep an eye on monitoring and have a backup battery when touring ... should the BMS shut down the battery due to MAX load on battery resulting in group 1 low voltage.

Would suspect to see some disparity among the 14 strings of 13p on that AliExpress 14s13p battery after 20,000km. It's anyone's guess how long you can continue to get satisfactory performance (as is) from that battery with occasional max load discharge riding before the BMS shuts down the battery (due to low voltage protection cut-off on group 1).

 

[eMark]

You are correct (13p). What can be confusing to some is the 14 readouts in those charts. Each of the 14 readouts is the average combined parallel voltage (13p) of each of the individual series strings (14s). Each of the 182 cells is actually serving as both a cell in a series string as well as a cell in a parallel group. Optimum bus bar efficiency for keeping all the cells balanced assuming all cells are equal quality A-grade, no faulty spot welds, no faulty solder connections to sensing wires and no faulty BMS circuit.

https://endless-sphere.com/sphere/attachments/_mg_6859-jpg.330090/

13s5p ...

Each cell is serving as both a cell in each 13s series string as well as a cell in each 5p parallel group. Being your 14s battery is 13p it will serve you better for many more cycles than if it were only 5p.. With only an additional 250mV drain on group 1 remains the same (with additional discharge cycles) you should be OK ... unless it progressively get worse. If possible limit your max discharge time of use to seconds instead of minutes.

 

[rudivb]

https://endless-sphere.com/sphere/attachments/_mg_6859-jpg.330090/

13s5p ...

Each cell is serving as both a cell in each 13s series string as well as a cell in each 5p parallel group. Being your 14s battery is 13p it will serve you better for many more cycles than if it were only 5p.. With only an additional 250mV drain on group 1 remains the same (with additional discharge cycles) you should be OK ... unless it progressively get worse. If possible limit your max discharge time of use to seconds instead of minutes.

Hi, thanks again for all the information you have provided, I have learned a lot.
I did open the battery once more and tried to check individual cells with a voltmeter, but it couldn't measure a difference, so I don't know which are the bad cell(s). I also redid some of the wiring of group 1, but the issue remains.

I am gonna leave it by this and just accept that we are dealing with a battery that already has a decent amount of mileage and it is to be expected. I will keep monitoring the battery closely (that's why I love having a bluetooth BMS) and will rarely use max discharge. When I am touring I go a lot slower anyway (to extend range), I might only use about 200W unless going up a hill.