Any advice about my pack design

[atkforever]

Hello Guys, I am designing my 3rd LI-ION pack and this time i'm unsing an Hailong like DIY kit (91cells).

My pack is 16S5P, and max amp will be limited to 40A.

I red line is the nickel strip for the serie connection of the last 2 groups, and green line nickel strip of the parallel connection of cells of the 15th cells pack. Yellow cells are then the positive bus (B16).

I see 2 issues :

The 3 red strips will handle a lot of stress, more than 13amps each (I'll double or trippe them ), but the main issue is that the number of connection is odd so one cell will "receive" twice the current, it's the cell pointed by the purple arrow. What is clearly the risk here ?

The second issue is that current will flow through green nickel strip line, and there might be a "bottleneck" on the section pointed by the orange arrow, about 20amp going thru this section, am I right or do I miss something?


My pack design is probably not the best, If u have any clue on how to put the cells, feel free to tell me

 

[atkforever]

I am about to solder everything soon so if I'm doing crazy mistake please tell me

 

[serious_sam]

I am about to solder everything soon so if I'm doing crazy mistake please tell me

If you're going to solder, use copper, not nickel.

 

[atkforever]

Yes sorry, about to spot weld.

 

[eMark]

Yes sorry, about to spot weld.

Isn't there almost as much [orange arrowhead bottleneck] concern with p-14 (gray) connecting to p-15 (white) in comparison to the interconnections between the first twelve p-groups ?

Spot-weld another longer bus bar strip between the p-15 white cell (closest to the orange arrowhead) to the top white cell (next to the p-16 yellow cell). There isn't as much increase as one might think in mΩ resistance when doubling the length of a nickel-plated copper strip bus bar.

One of my many experimental tests with my VRUZEND kit was to determine the difference in mΩ resistance between nickel-plated copper bus bars and stainless steel bus bars. The mΩ measurement was the distance from one outer connection hole to the other (see attached photo).

• a nickel-plated copper bus bar measured 2.4mΩ ... doubling length only 2.6mΩ

• a stainless steel bus bar measured 5.2mΩ ... doubling length was 11.7mΩ

So, its aok to spot-weld a longer strip between those two white (p-15) cells. With several bus bar spot-welds on can end of that p-15 cell it's better to spot-weld in two or three places closer to side of can than all on or near center. Some believe that it's always better to spot-weld off-center on can end (closer to side of can) than on the center of can.

Bottomline: Doubling length of nickel-plated bus bar strip doesn't come anywhere close to doubling the mΩ resistance. I removed some of the nickel plating to expose the copper in the photo. You can see the exposed copper in the bottom bus bar (close to threaded stud connection).

 

[atkforever]

So, its aok to spot-weld a longer strip between those two white (p-15) cells. With several bus bar spot-welds on can end of that p-15 cell it's better to spot-weld in two or three places closer to side of can than all on or near center. Some believe that it's always better to spot-weld off-center on can end (closer to side of can) than on the center of can.

Thank you for your accurate answer, that will help for sure.

What do you think about the fact that only two cells of the P16 are "connected" to P15 ? Will these cells wear out faster ?

 

[eMark]

Thank you for your accurate answer, that will help for sure.

How accurate is questionable?? Any doubling of resistance when doubling the connection distance can be mitigated by increasing size of bus bar strip to equalize overall pack demand (e.g the connections between p-15 to p-16).

What do you think about the fact that only two cells of the P16 are "connected" to P15? Will these cells wear out faster?

Shouldn't as long as balanced with appropriately sized bus bar connections between just two cells of P15 and P16. There should be a way you can rearrange P11-P16 to allow for three cell connection between P15 to P16 instead of just two cells. Will do some figuring to see what's possible ... be back tomorrow.

 

[eMark]

This arrangement allows for three series connections between P10-P16. What about just three series connections (1S,3S,5S) between P1-P9 instead of all five.

Made good use of that one empty cell space in bottom right corner
https://scontent-cdt1-1.xx.fbcdn.net/v/t1.15752-9/151937771_127728829232837_4278769972271085050_n.png?_nc_cat=106&ccb=3&_nc_sid=ae9488&_nc_ohc=MRnCICE107YAX-07FsK&_nc_ht=scontent-cdt1-1.xx&oh=e9cbd6a75a05360453be4e9ca93b76eb&oe=60515E39

 

[atkforever]

This arrangement allows for three series connections between P10-P16. What about just three series connections (1S,3S,5S) between P1-P9 instead of all five.

Thank you so much.

Regarding the series connections you are right; since i'm using 0.15 pure nickel I'll allow 7amp max through one strip. In this configuration (3 connections) I'll have to stack the strips (2x); indeed this is a good option.

What configuration gives the lowest resistance ? 5x 0.15 strip or 3x (doubled) 0.15 strip ? I'd say 3x doubled strip.