Chris's Build Thread

They huge! But they pretty! :D

This is still considering an ATS. You are Automatically Transfer Switching from one source to another. The one thing we (I) was missing, was the direction in which you were sourcing. I think you were going from Mains to Battery and back to Mains if there was a problem on the line. Didn't realize you were going from Battery to Mains when the batteries get low. Still ATS, just using 1/2 of what other use it for ;)
 
So I got some batteries today...

image_agdtlw.jpg


This is my prototype pack:

image_rapisc.jpg


image_dnkaah.jpg


I made a pack that had about 7 cells in it, But as I was tightening the nuts down I noticed smoke. So I frantically took it apart and found this:

image_hrfpsy.jpg


I have no idea what caused this. I am not sure on what the washer shorted out on? Thoughts? Maybe I had the cell in the packbackwards? I was in such a hurry to break down the pack I didn't check for anything weird.
 
looks like a puncture from a washer
 
I talked to someone familiar with these cells. There is a clear insulating layer on the outside of the cells, and the pouch is made of metal. So the negative and positive on the pack shorted briefly. Back to the drawing board
 
I am still working on a way to connect these batteries in packs. I made a mistake of thinking washers would work. Technically, I still think they would if I make sure the 'rounded' side of the washer to contact with the battery, since then it wouldn't cut into the insulating layer. However practically it will take something like 24000 washers in all for 1050 cells to make my bank. Currently I am thinking about using 3/8" hard copper pipe cut to height between cells. That has a decent wall thickness with a 1/4" inner diameter. Its fairly cost efficient as well. Any else have any ideas?

I am trying to not lose hope here on these batteries, and I really hope I didn't make a mistake with these batteries, as SLA batteries are starting to look real good right now in terms of complication of implementation.
 
You could go Nut, cell, Nut, Nut, cell, Nut etc.
Might be easier and lighter than all the washers.
You could more easily control the cell contact.
 
Bubba said:
You could go Nut, cell, Nut, Nut, cell, Nut etc.
Might be easier and lighter than all the washers.
You could more easily control the cell contact.
Click to expand...

I was thinking about this too, just might take a long time plus the issue of having appropriate spacing between the cell terminals/cells themselves.

Sean said:
How about plain or threaded stand offs ....
Click to expand...

The problem I found with those is getting the correct height. I need 11/32" tall with at least a 3/8" hex width.

I am still poking around at both options though.

I had an idea of havingbus bars made for the positive and negative side, each terminal can then be bent and screwed into it using M6 machines screws.Then use the Acrylic on the ends with the threadedrods for compression. The rodswould go down the short sides instead of through the terminals obviously.


I whipped this up in CAD:

image_cnkxmq.jpg
 
The professional ones I have use a Circuit board and the cells (much like the board your thinking of) are strapped applying pressure to them.
 
The problem looks like you need to extend the exposed tab part so you don't risk pinching the pouch outer layer.
Maybe use some thin copper sheet eg 30mm x 50mm, fold in half (on middle of 50mm) + punch/drill hole for rod.
Then solder copper part to cell tab.

I have some similar pouch cells & they are assembled with tabs through PCB slots, folded 90 deg & soldered flat. 4 cells to a PCB.
 
More work on the bus bar design.


image_njtnmg.jpg


image_yrwzfe.jpg


Redpacket said:
The problem looks like you need to extend the exposed tab part so you don't risk pinching the pouch outer layer.
Maybe use some thin copper sheet eg 30mm x 50mm, fold in half (on middle of 50mm) + punch/drill hole for rod.
Then solder copper part to cell tab.

I have some similar pouch cells & they are assembled with tabs through PCB slots, folded 90 deg & soldered flat. 4 cells to a PCB.
Click to expand...


Thanks for the input. I think I am abandoning the threaded rod idea. I will either get the bus bars I am designing made or use stand-offs that are small enough not to engage with the pouch. Only thing that might save the threaded rod idea is some spacers as Sean posted. I just haven't looked around for that yet.
 
So I have ordered some stand-offs and spacers, enough for one pack to test that.

I also have an updated rendering. I have quote requests out for the bus bars, so we shall see what they come back with.


image_kpwrkm.jpg
 
Mocked up a version with 11/32" (8.73125mm) aluminum spacers. There will be spacers between the acrylic and terminals as well of course.


image_lskioh.jpg

image_nouwvx.jpg
 
I made a mistake in my drawing and the cells were 8.91mm causing me to choose the wrong spacer size, 8.74mm. So after fixing that (cells are average 8.45mm), I ordered 100 7.94mm and that should do it. I might need shims between the cells and the acrylic to make sure the cells are properly compressed. But the idea seems to work.

image_mubfoa.jpg
 
Lookin good
 
What rate of charge/discharge will you be wanting to put through the assembled packs ?

You might need to make use of much more contact area available on the tab if you are expecting high amps, you will also likely need clamping plates with far greater rigidity when you start charge/discharging them, as even at low rates the acrylic will deform at the midpoint.
 
Sean said:
What rate of charge/discharge will you be wanting to put through the assembled packs ?

You might need to make use of much more contact area available on the tab if you are expecting high amps, you will also likely need clamping plates with far greater rigidity when you start charge/discharging them, as even at low rates the acrylic will deform at the midpoint.
Click to expand...

80A charge, 138A discharge max to the entire bank. Each series string will be 14s25p. Three of these strings in parallel making 14s75p in total.

So:
-Charging-
Each string will see ~27amps, and thusly each pack (roughly)
Each cell will see ~1.08amps
This looks safe with this configuration

-Discharging-
Each string will see ~46amps, and thusly each pack (roughly)
Each cell will see ~1.84amps
This looks safe as well.

Correct me of my math is off, but this all looks good to me. I think I will do some heat tests with a camera to check in my prototype on Monday.

Regarding the acrylic, it is 0.25" thick. At 8.75" x 6" it would take a great deal of force to warp these much if at all, specially at the charge/discharge rates I am looking at. Since many packs are held together with tape and do fine, I figured 0.25" acrylic with 0.25" threaded rod should be plenty.
 
Its been a while since I posted. Progress is being made, 28 of the 42 packs are built. Permit paperwork is submitted to the AHJ. I haven't decided where the longmons are going to be mounted, but definitely not where they are in the picture.


image_swfpfp.jpg

image_hgbxfn.jpg

image_szogqc.jpg
 
What's youroverall capacitygoing to be ?

I have 114kWh of Leafs cells for a home project sat waiting.


image_bfmjxo.jpg
 
Sean said:
What's youroverall capacitygoing to be ?

I have 114kWh of Leafs cells for a home project sat waiting.


image_bfmjxo.jpg
Click to expand...

Only ?? :D



(this is just not fair....)
 
You must log in or register to reply here.
Back
Top