Molicel P42A 20s8p (Part 2)

DIY Builds Battery Builds
1

Part 1 - Molicel P42A 20s8p - #2 by Nacho

I’m planning on using .15mm x 10mm nickel strips.

The Molicel P42A is rated for 45A.

According to this graph, a .15mm x 10mm is able to handle 20A. I would have to triple stack it.

4295919F-4356-4A2F-BD96-0A0B60A78210
4295919F-4356-4A2F-BD96-0A0B60A78210588×932 217 KB

But then there’s some graphs that show 7A, such as the one below, which would mean I would have to stack 7 of it. I can’t really get my head around it.

image
image1024×547 73.8 KB

I would try a copper nickel sandwich but I can’t seem to find an ampacity chart for copper.

I found this

It’s a .15 x 10mm and according to this chart, it can do 21A which would mean I would have to triple stack it.

image
image1152×771 188 KB

Look at the .15mm x 10mm of the nickel on that chart. It says 6A which is close to the 7A on the second chart. The first chart is made by esk8 though which is a better resource, I would assume, so I don’t really know who to believe at this point.

If someone can point me in the right direction. Thank you.

2

Just a side note…that 45A rating is for non-continuous use as long as the cell doesn’t get too hot. And operating at anywhere near that level results in huge voltage sag (same as for any cell).

3 Likes
3

Just use a single stack. If this is for esk8 you’ll be fiiiine

1 Like
4

This seems super ambitious for one’s first battery.

Your 8 parallel cells do not require thick conductors joining them, the 20 series connections do.

I see most high amp battery packs use doubled or tripled 12awg silicone wire or tinned copper braid as series connects.

I’ve also read that the cans of P42a are more difficult to spot weld.

I’ve only built one Esk8 battery so far.
I made plenty of mistakes
I learned tons in the process.

I’m glad I decided on a small pack of inexpensive cells to learn with, and I actually made two 3s1p batteries first, one with 0.15 nickel, the second with 0.15 copper/0.10 nickel plated steel sandwich, even though these packs will never see more than 4 amps, and only briefly.

I thought I had the copper nickle method down, and tried it on my 10s2p battery with infinite slot method, and then found welds which appeared, to the eye, to be wonderful, were indeed totally and cometely inadequate.

More practice with my welder maxed out, could not weld 0.15 copper nickle sandwich on my new cells, but seemed more than awesome, on my practice cells.

I had to step down to 0.1 copper and 0.1 nickel plated steel to get proper welds on the new cells, and this battery will never see 30+ amps and will only approach that for a second or three.

A single layer of 0.15x10mm would have been far easier, and more than adequate.

But yea, I wish it were 24s 12p battery with P50b cells, welded with 0.3mm copper feed, 2 awg series interconnects, 4 2000 watt motors, and smoke this guy in a drag race.
717YGkfr-FL.AC_UF1000,1000_QL80

I get the desire for a huge battery with massive interconnects.

I remember searching for nickel and copper ampacity charts, and finding those one you linked, and look back at how much less I understood then, than I do now, having built only one (untested) battery so far, and how much more I still need to learn.

I am very glad I started with a modest size and capability Battery with 40$ worth of cells.

20s8p … 160 P42a cells at 4.60$ each…760$

I’d be dang sure of my skills before taking on such a battery, and one can only learn so much by reading, watching researching.

Experience teaches far better.

3 Likes
5

I see. Thank you.

There will be 4 connections between each series so that’s 360/4 = 90A for each connection.

You mentioned 12 AWG Silicon Wire. I think that would be great for the 20s connections. Thank you for that idea. I’ll also look into the tinned copper braids that you mentioned.

If you have any more ideas you want to throw my way, I’d be incredibly grateful.

Thank you.

6

You do not need to build the pack to the spec of the cells… just to the spec of what your drive system will be pulling. Stacking nickel ain’t the way.

3 Likes
7

I have 2 5000W motors. Each motor will have a 70A controller.

That would be 140A peak from the battery. Potentially more if the controllers are shunted (Still thinking about if I should do this or not).

The battery pack isn’t going to be bent in anyway. so if I stack the nickels, they’ll lay flat anyway. It will just be thicker. I’m trying to find alternatives to nickel strips since I don’t really want to stack nickel either. I don’t want to solder wires from each cell.

I could go the copper nickel sandwich route. I’m still thinking about this.

I’m heard of people putting blobs of solder under the copper strips and using a spot welder to poof the solder and the copper together. I’m not too sure about this TBH.

Copper nickel sandwich might be the route. If I could just find out the ampacity of copper.

8

140A peak ≠ continuous.

Don’t build your pack to the peak spec. Building a complex and dangerous electrical component like a battery by just adding up peak specs is a sign that more research needs to be done.

3 Likes
9

This depends on the acceptable temp rise. Check out the ratings for copper wire used in residential construction and see if the allowed temp rise is a decent match for your requirements.

If yes then you can easily use the circular cross-sectional area of different gauge wire to convert it to the rectangular cross-section of strip and go from there.