My try on no solder/no weld 18650

[riba2233]

Hi! This is basically hybrid of two methods already known, but I was trying to get the most out of each one, and method that would be easier to work with. It works like this:

So, we got the dimpled copper strip that goes on the cells, then the plastic pieces, and the zip tie (not on picture) that goes into these slots, and compresses the parallel pack. On top of all that, we should add shrink wrap, and all cells should be glued. Latter we just connect parallel packs in series.

So what do you think, yea or nay?

 

[zener]

I would do it with nuts and bolts not zip ties. (than its possible to work with precise tightening torque)
And something stiffer than plastic pieces.

 

[riba2233]

Yup, that's also an option, just be careful not to short tabs with screw.

One could also use some kind of rubber ring instead of zip ties, but I wold be worried that they would loose compression over the time.

 

[Punx0r]

With three cells between each pair of zip ties there is a possibility the middle cell will have weak contact as the plastic board will be pulled into a slightly arched shape. The heat shrink is better because it provides a more uniform compression.

The construction of the Zero motorcycle packs is very interesting: rubber hoops on each cell and then covered in heat shrink.

 

[riba2233]

That's why I'm putting only two cells in a row, and I'm using both zip ties AND heat shrink.

Yes, I got inspiration for this one from zero battery pack and of course this one:

http://endless-sphere.com/forums/viewtopic.php?f=14&t=57810

 

[marty]

x
The construction of the Zero motorcycle packs is very interesting: rubber hoops on each cell and then covered in heat shrink.

Agree
Here is some Grounding Braid and Ribbon
http://www.mcmaster.com/#grounding-braid/=s8ftkv

 

[j3tch1u]

i am using something similar. i'd recommend using copper beryllium curved bus bars (0.6-0.8mm) otherwise you will have connection issues with the copper deforming over time. the BeCu acts as a spring to distribute strong, even pressure across the terminals (you won't need the plastic plates). in this case, i would cut the zip tie notches a little deeper to act as "fingers" and allow tolerance for the adjacent cells. be sure to use heat stabilized, uv resistant zip ties as they be susceptible to heating/cooling cycles. personally i don't think it's a good idea to have the cells densely packed together--the heat will diminish their life/performance.

Hi! This is basically hybrid of two methods already known, but I was trying to get the most out of each one, and method that would be easier to work with. It works like this:

So, we got the dimpled copper strip that goes on the cells, then the plastic pieces, and the zip tie (not on picture) that goes into these slots, and compresses the parallel pack. On top of all that, we should add shrink wrap, and all cells should be glued. Latter we just connect parallel packs in series.

So what do you think, yea or nay?

 

[riba2233]

That sounds great, can you post some pictures and link to copper beryllium bars?

Thanks for the great info!

 

[j3tch1u]

if you google it, you can find many places that sell it. here is a listing on ebay (not cheap though):
http://www.ebay.com/itm/1pcs-Copper-Beryllium-Alloy-BeCu-C17200-Plate-Sheet-0-8mm-200mm-200mm-E5-P-/181410043847?_trksid=p2054897.l5658

try to get C17500 or C17510 grade (high conductivity, medium strength)

That sounds great, can you post some pictures and link to copper beryllium bars?

Thanks for the great info!

 

[riba2233]

Thanks! Do you have any pictures or info on how to curve the bars?

I would really like to see the pictures

 

[j3tch1u]

we're doing it a little differently with custom cell holders made out of thermally conductive polymer (similar to the plastic used in LED bulbs). the busbars are bolted onto both sides of each 12p module and then connected in series using copper braided strap. there are many ways to bend the copper just use your imagination :lol: (and i don't mean your telekinetic powers).

Thanks! Do you have any pictures or info on how to curve the bars?

I would really like to see the pictures

 

[riba2233]

Nice! Thank you so much, I own you one

 

[j3tch1u]

all in the spirit of sharing my friend

Nice! Thank you so much, I own you one

 

[riba2233]

It really looks awesome. How much weight does the plastic add? And is it costly to print that much plastic?

 

[j3tch1u]

the 3d printed mockup is around 100g (0.1mm layer height, 20% fill). took about 16 hours to print and used about 15m filament. our production blocks will be compression molded (can't 3d print the thermal plastic as it only comes in pellets and requires high pressure to achieve good thermal conductivity ~ 5 W/mK). the cell holder acts as a heat sink which is essential when using some of the newer 30A discharge cells. cell life deteriorates at >50 deg-c.

It really looks awesome. How much weight does the plastic add? And is it costly to print that much plastic?

 

[savaoaknyc]

Really nice,
i was looking one year ago and stumbled on this one
http://www.coolpolymers.com/files/ds/Datasheet_d5506.pdf
It has thermal conductivity 10w/mk.
Anyway i'am not into this stuff.Don't have the knowledge in thermodynamics etc..

Any test data about the temperature at 3c discharge of this pack???

 

[j3tch1u]

not yet, the parts haven't been made yet. coolpolymer's sample prices were too expensive so i'm using a local supplier (in taiwan).

Really nice,
i was looking one year ago and stumbled on this one
http://www.coolpolymers.com/files/ds/Datasheet_d5506.pdf
It has thermal conductivity 10w/mk.
Anyway i'am not into this stuff.Don't have the knowledge in thermodynamics etc..

Any test data about the temperature at 3c discharge of this pack???

 

[spinningmagnets]

Many years ago I was a parts inspector in an aerospace company. One of the jobs we would do on occasion was to pressure test and cycle (roughly 1,000 cycles) sevral types of transducers that involved the flexing of a beryllium-copper disc with a strain-gauge attached to it.

What I'm getting at is that: beryllium-copper in the right alloy is a phenomenal "leaf spring" for these contacts. We've all handled copper wire that bends easily, and once bent...stays in the new shape. As long as the BeCu is not bent past its allowable spec, it will literally cycle millions of times and return to its original shape.

The way that Ben is using these, the contact strips will have a very small flexing to them as the cells get warm (expand) and then cool down.

This is a brilliant solution.

________________________________

edit: this style of contruction will likely not need any goop between the cell-ends and the bus bar contact points, but...a while back when I asked Luke if there was any conductive paste, where a "tiny" dab would ensure solid contact and help avoid oxidation over time, he recommended this: http://endless-sphere.com/forums/viewtopic.php?f=14&t=57810#p862253

 

[PaulD]

our production blocks will be compression molded (can't 3d print the thermal plastic as it only comes in pellets and requires high pressure to achieve good thermal conductivity ~ 5 W/mK). the cell holder acts as a heat sink which is essential when using some of the newer 30A discharge cells.

Nice work. Will you have to machine the bores of the compression molded part? Or can you mold that bore with zero draft? Might be tricky to get good contact between the cell and the plastic. Perhaps the holder could be 2 pieces and clamp around the cell?

 

[docnjoj]

Wow! If these can be done as kits and are cost effective, I see 18650 cells in my future. Simply great work, both of your.
otherDoc

 

[j3tch1u]

i may be able to sell some of these cell blocks and busbars if it passes proof of concept (ie. heat dissipation and connection resistance/vibration tests). i'll ask my buddy who is in charge of this whether he wants to sell some on the side.

Wow! If these can be done as kits and are cost effective, I see 18650 cells in my future. Simply great work, both of your.
otherDoc

 

[j3tch1u]

paul, no the bores will be molded to slip-fit tolerance which should be good enough for heat transfer. could also use thermal grease but i hate working with that messy stuff. machining this material loses it's "in-plane" conductivity (ie. the surfaces are not as tight).

our production blocks will be compression molded (can't 3d print the thermal plastic as it only comes in pellets and requires high pressure to achieve good thermal conductivity ~ 5 W/mK). the cell holder acts as a heat sink which is essential when using some of the newer 30A discharge cells.

Nice work. Will you have to machine the bores of the compression molded part? Or can you mold that bore with zero draft? Might be tricky to get good contact between the cell and the plastic. Perhaps the holder could be 2 pieces and clamp around the cell?

 

[j3tch1u]

thank you ron. you are correct--we decided to have "open ends" to allow the cells to "float" between the 2 busbars. this equalizes the contact pressure and allows a degree of tolerance.

how hazardous is that beryllium?

another cool thing i found is you can solder to it as well.

Many years ago I was a parts inspector in an aerospace company. One of the jobs we would do on occasion was to pressure test and cycle (roughly 1,000 cycles) sevral types of transducers that involved the flexing of a beryllium-copper disc with a strain-gauge attached to it.

What I'm getting at is that: beryllium-copper in the right alloy is a phenomenal "leaf spring" for these contacts. We've all handled copper wire that bends easily, and once bent...stays in the new shape. As long as the BeCu is not bent past its allowable spec, it will literally cycle millions of times and return to its original shape.

The way that Ben is using these, the contact strips will have a very small flexing to them as the cells get warm (expand) and then cool down.

This is a brilliant solution.

________________________________

edit: this style of contruction will likely not need any goop between the cell-ends and the bus bar contact points, but...a while back when I asked Luke if there was any conductive paste, where a "tiny" dab would ensure solid contact and help avoid oxidation over time, he recommended this: http://endless-sphere.com/forums/viewtopic.php?f=14&t=57810#p862253

 

[docnjoj]

You guys, should you ever get to selling these and production mode could do for 18650 cell what Agniusm did for AMP 20 A123 pouches. I would buy his kits in a heartbeat again, and would try yours to make my own Sony VTC packs. Love them cells. (at least the older versions from Fatpacks)
otherDoc

 

[spinningmagnets]

how hazardous is that beryllium?

Beryllium is very hazardous to breathe in the dust. Wet-cut the pieces (or use a shear), and use a lubricant when drilling the holes. It might be worth having the BeCu pieces water-jetted or laser-cut (by someone else who is familiar with safe machining of BeCu)

beryllium copper presents no known health hazard. However, breathing its dust or vapors, as formed when machining or welding, will eventually cause serious lung damage.[2] Beryllium compounds are known human carcinogens when inhaled

-wikipedia

Also...

Alloy 25 beryllium copper (C17200 & C17300) is an age-hardening alloy which attains the highest strength of any copper base alloy. It may be age hardened after forming into springs, intricate forms or complex shapes. It has superb spring properties, corrosion resistance and stability as well as good conductivity and low creep.