Battery Issue

Joined
Apr 18, 2022
Messages
10
I created a 7S22P 18650 battery pack. For some reason, the first and last series have slightly higher resistance and the positive and negative ends of the pack get hot during high (20+A) current draw. What would be causing this and how do I fix? I did connect the balance charger wires incorrectly initially which caused a spark but I guess that wouldn't cause this?

I used .15mm thick x 8mm wide nickel strips throughout.

Thanks
 

Attachments

  • 20220417_120859.jpg
    20220417_120859.jpg
    1.5 MB · Views: 691
First and last of the series are getting hot and the wires you used are not?
I would say you need a more even flow, meaning at 22P, I would put 3 or 4 smaller wires every ~5 cans to one big wire on both the first and last series.

Are your tabs genuine metal or some tinned knockoff.

FOUR

1
2
3
4 -----
5
6
7
8
9 -----
10
11
12
13
14 -----
15
16
17
18
19 -----
20
21
22


FIVE

1
2
3 -----
4
5
6
7
8 -----
9
10
11
12 -----
13
14
15
16 -----
17
18
19
20 -----
21
22

THREE MIGHT DUE TOO :thumb:
Gauge of wire used on the branch from positive or negative?
Depends I guess, 3x12awg to 8awg, be hard to solder three 12awg to one 8awg. 4 or 5 branches even harder, need like a 100w iron.

You could also try soldering some solid core wire from Home Depot, house wiring, 10awg solid core from cable of 3 or 4 wires. Be not the first thing to do, I would do the branches first.
Solder one continuous 10awg solid core wire on the first and last in the series, would be plan B. plan C, every Parallel. Strip the 10awg of its insulation, pre-tin the wire, pre-tin the tabs where you will solder.

Also, be sure to use good quality, new wire, make 100% sure your solder connections are good.
Inspect all the tab welds. And maybe test some spare or scrap tabs to see if they are the right ones, I remember reading someone buying tabs and they werent the right tabs, I believe he put the tab in some cola for acid, and the tabs were tinned steel or some b.s.

Finally something that is an option is to double up the tabs, whether it be s and p or just one. In that regard I would do the first and last only first, and do test runs, work that battery man. So if you can weld tabs, that might be the first option rather then solder solid core 10awg wire onto the tabs.
 
Thanks, I just saw your reply after I had soldered 12 gauge solid core copper wire across the pack positive, which dropped the resistance from 3 milli ohms on that series to around 1.5 (around the same as the others). The only other wire I had available was 8 gauge and that seems a little too big.

Maybe this will be enough if I do the same to the negative end? Afterwards I'll test it with ~40 amps to see if it still gets hot at the ends. I also ordered some nickel plated copper strips that I might use on the other series'.

The nickel strips I used didn't spark when I took a Dremel to it so I guess it's pure. I'm also letting a piece I scuffed up sit in saltwater overnight.
 

Attachments

  • 20220421_072459.jpg
    20220421_072459.jpg
    1.2 MB · Views: 581
Also noticed that the 1st and last battery in those end parallel groups were farther off with their voltages vs the other sets after charging. So like you said, uneven flow.
 
Batteryguy85 said:
Also noticed that the 1st and last battery in those end parallel groups were farther off with their voltages vs the other sets after charging. So like you said, uneven flow.
The outer negative and positive parallel line (1S&7S) ? How far off were the voltages ?
What amp rating (0.5C, 1C, 2C) during charging ? Did you charge to 90% or FULL ?

In addition to calab's post what about soldering 14awg solid copper coming off each end (last three cell busbars) of the two outer 22P lines (1S&7S) then joining these two short equal lengths of 14awg to 12awg copper ?

What brand, capacity and MCD rating are those 18650 cells ?
Are they new/fresh cells ... unused, but in storage at least 4-5 yrs ... salvaged cells that you tested--sorted as best as possible before assembly ?

Is it for say a teens' custom-made 24V Skateboard or GoKart cruising at MCD most the time ?

Sounds like a FUN Project :thumb:
 
The outer negative and positive parallel line (1S&7S) ? How far off ?
What amp rating (0.5C, 1C, 2C) during charge ? Did you charge to 90% or FULL ?
Yes, 1S and 7S. 0.4C (30 amps) charging to target 4.10v Li io setting on iCharger X12. I guess i should stop at 90% instead. First cell in the row ends up at like 3.80v while last cell is at 3.71. Then it takes a while for them to equalize afterwards. The cells in the other groups are closer in voltage immediately after charging.

They're all new Molicel M35A 18650 3500mAh 10A and were not in storage.

In addition to calab's post what about soldering 14awg solid copper coming off each end (last three cell busbars) of the two outer 22P lines (1S&7S) then joining these two short equal lengths of 14awg to 12awg copper ?

You mean soldering copper on the other side of the pack positive and negative? I.e. opposite side of where I soldered it in picture above?

It's currently intended as a battery backup for my aquariums and other devices in a power outage (via 2000W inverter)
 
All the series have around the same resistance now after adding the 12 gauge copper across the pack positive and negative ends.
 

Attachments

  • 20220421_143317.jpg
    20220421_143317.jpg
    1.2 MB · Views: 558
Now at ~45 amps and after 20 minutes, the output 10 gauge silicone wire is 108 degrees F and the batteries closest to the silicone wire are 95 degrees F
 

Attachments

  • 20220421_145548.jpg
    20220421_145548.jpg
    2 MB · Views: 556
Batteryguy85 said:
Yes, 1S and 7S. 0.4C (30 amps) charging to target 4.10v Li io setting on iCharger X12. I guess i should stop at 90% instead.
Isn't 4.10v 90% ?
Why not at 0.5C (1.75 amp charge rate with your iCharger X12) instead of 0.4C ?
Batteryguy85 said:
Also noticed that the 1st and last battery in those end parallel groups were farther off with their voltages vs the other sets after charging. So like you said, uneven flow.
So the imblance is only at each end of rows/groups 1S22P and 7S22P (e.g. 3.80v and 3.71v) with the other 150 cells (154-4) within a few millivolts (give or take) of 4.10v ?
Batteryguy85 said:
First cell in the row ends up at like 3.80v while last cell is at 3.71. Then it takes a while for them to equalize afterwards. The cells in the other groups are closer in voltage immediately after charging.
So were the other 110 cells (including the cells at each end) in the other five 22P rows/groups within a few millvolts of 4.10v. So the end cell imbalance (3.80v and 3.71v) is just with the negative row/group 1S22P and positive row/group 7S22P or moreso with the negative 1S22P ?

By "row" are you referring to the first cell at the end of just 1S22P at 3.80v and the last cell on the other end of 1S22P at 3.71v OR also the first cell and last cell at each end of positive 7S22P or a noticeable cell imbalance at each end of all seven 22P rows/groups (14 cells) ?
Batteryguy85 said:
They're all new Molicel M35A 18650 3500mAh 10A and were not in storage.
Wouldn't 0.5C be a 1.75amp (3.5Ah ÷ 2) charge rate ?
Batteryguy85 said:
You mean soldering copper on the other side of the pack positive and negative? I.e. opposite side of where I soldered it in picture above?
Yes, because it's and odd series (7S) so the neg & pos are on opposite sides of the battery. Have edited my previous suggestion to see if others think it makes any sense with improving cell balancing during both charging and discharging ...
In addition to calab's post what about soldering 14awg solid copper coming off each end (say last three cell busbars) of the two outer 22P groups (1S neg & 7S pos) then joining each of the 2 equal 14awg lengths (2-neg and 2-pos). 2-14awg from each end of 1S22P and 2-14awg from each end of 7S22P. Then joining the two 14awg from neg 1S to 12awg and two 14awg from pos 7S to 12awg copper.
Batteryguy85 said:
It's currently intended as a battery backup for my aquariums and other devices in a power outage (via 2000W inverter)
Hope the above isn't too confusing and will be of some help for other suggestions for keeping all the cells in your 7S22P battery balanced within at least 25-30 millivolts of each other in that all 154 are new/fresh Molicel M35A 18650 3500mAh 10A cells.

What BMS and fuse protection ?
 
Isn't 4.10v 90% ?
Why not at 0.5C (1.75 amp charge rate with your iCharger X12) instead of 0.4C ?

Yeah I guess it is 90% huh. Icharger will only charge at 30A max.

So the imblance is only at each end of rows/groups 1S22P and 7S22P (e.g. 3.80v and 3.71v)
The others were ending up at like 3.9v after resting. I guess the charger is cutting off before they can balance out?

So were the other 132 cells in the other six 22P-groups within a few millvolts of 4.10. Was there also a similar imbalance (3.80v and 3.71v) with both the negative row/group 1S22P and positive row/group 7S22P or just the negative 1S22P ?

The others ended up at around 3.9-3.95
By "row" are you referring to the first cell at the end of just 1S and the last cell on the other end of 1S22P at 3.71v OR also the first cell and last cell at each end of positive 7S22P or the cell at each end of all seven 22P rows/groups (14 cells)?
Cell difference the beginning vs end of 1S AND separately at the beginning vs end of 7S

Wouldn't 0.5C be a 1.75amp (3.5Ah ÷ 2) charge rate?
Would be (3.5Ah x 22) ÷ 2 = 38.5A right? Charger can't do over 30A.

What BMS and fuse protection ?
No BMS yet, just balance charger. What high quality & economical BMS would you recommend? Are these Daly Bluetooth ones any good?
https://www.ebay.com/itm/124907475081

No fuse protection. Haven't figured out how to do that. Any ideas/tutorials?
 
Batteryguy85 said:
0.4C (30 amps) charging to target 4.10v Li io setting on iCharger X12 ... iCharger will only charge at 30A max.
The maximum recommended charge current (datasheet) for 18650 Molicel M35A is 1.7A. You're not charging some RC Lipo with a 50C discharge rating. If it's capacity is 5000mAh, or 5 Amps, that means that LiPo battery can be safely charged at a maximum of 15 Amps!), but that RC mindset doesn't apply to 1865/2170 cells. At least not the last time i checked :wink:

That iChargerX12 is just what DIY builders of 10S-12S packs can make good use of ... overall good reviews :thumb:
https://www.progressiverc.com/products/icharger-x12?pr_prod_strat=description&pr_rec_id=4f458320c&pr_rec_pid=5201078222983&pr_ref_pid=5201078124679&pr_seq=uniform
https://www.amazon.com/iCharger-1100W-Balance-Battery-Charger/dp/B08MS9N22X

Batteryguy85 said:
I charged again and now they're within 3mv
Couldn't ask for any less :thumb:
What's your actual amp charge ? ____ ... did you figure 0.4C x 22 = 30.8A charge rate?
 
The maximum recommended charge current (datasheet) for 18650 Molicel M35A is 1.7A. You're not charging some RC Lipo with a 50C discharge rating. If it's capacity is 5000mAh, or 5 Amps, that means that LiPo battery can be safely charged at a maximum of 15 Amps!), but that RC mindset doesn't apply to 1865/2170 cells. At least not the last time i checked :wink:

The capacity of the pack is 77 Ah because there's 22 batteries in parallel. Doesn't that mean I can charge the pack at 77/2 = 38.5A? That's only 1.75A per battery. That's also how I'm able to safely discharge the pack at over the max individual cell rating of 10A...

What's your actual amp charge ? ____ ... did you figure 0.4C x 22 = 30.8A charge rate?
Yes, but 30A is the max the icharger can output.
 
Batteryguy85 said:
The capacity of the pack is 77 Ah because there's 22 batteries in parallel. Doesn't that mean I can charge the pack at 77/2 = 38.5A? That's only 1.75A per battery. That's also how I'm able to safely discharge the pack at over the max individual cell rating of 10A...
Datasheet says maximum charge current is 1.70A so why do you say "That's only 1.75A per battery". Charging to 4.10V and discharging to say 3.00V ... so figure cell capacity of 3.25A (when new) and 80% cell capacity (2.60A) as pack ages.
Batteryguy85 said:
Yes, but 30A is the max the icharger can output.
Depends on how FAST you need to charge your 7S22P pack and how many c/d cycles you hope to get out of your DIY build. Some may agree with your charge rate of 30A; while others might consider a 10A rate tops.

You've probably already read this ES thread ...
https://endless-sphere.com/forums/viewtopic.php?t=74339

Bottomline Concern: Your USE depending on how FAST you need to charge; Your amp rate of discharge; How many c/d cycles you hope to achieve at your c/d rates.
 
eMark said:
Batteryguy85 said:
The capacity of the pack is 77 Ah because there's 22 batteries in parallel. Doesn't that mean I can charge the pack at 77/2 = 38.5A? That's only 1.75A per battery. That's also how I'm able to safely discharge the pack at over the max individual cell rating of 10A...
Datasheet says maximum charge current is 1.70A so why do you say "That's only 1.75A per battery". Charging to 4.10V and discharging to say 3.00V ... so figure cell capacity of 3.25A (when new) and 80% cell capacity (2.60A) as pack ages.
Batteryguy85 said:
Yes, but 30A is the max the icharger can output.
You've probably already read this ES thread ...
https://endless-sphere.com/forums/viewtopic.php?t=74339

Depends on how FAST you need to charge your 7S22P pack and how many c/d cycles you hope to get out of your DIY build. Some may agree with your charge rate of 30A; while others might consider a 10A rate tops.

Bottomline Concern: Your USE depending on how FAST you need to charge; Your amp rate of discharge; How many c/d cycles you hope to achieve at your c/d rates.
1.75 and 1.70 are close enough but I'm not charging at 38.5A anyway. I'll probably even drop down to 15-20A since charging speed isn't much of a concern for me
 
m3vuv said:
why all the quotesmakes threads impossible to keep track of!

You can't scan the text and find the part being replied to?

How did you do in school?
 
Back
Top