APL's Pressure Contact Battery box

APL

100 kW
Joined
Aug 6, 2018
Messages
1,113
This thread was split off from the custom ebike build thread, found here:
"APL's V4 Cruiser Build"
https://endless-sphere.com/forums/viewtopic.php?f=6&t=111159

Well, with lots of head scratching and figuring I guess it looks like a 60v pack is going to be the best fit, and not
weigh a ton. Vertical cell options give better space models, but seem a bit unsafe in my opinion,.. cell weight is
best supported from the sides.

A 16S - 12P pack fits nicely and gives me two more P-cells that will fatten up the throttle a bit more. 60v nominal
65v full charge should get the gearing I want, and at 20+ pounds, it's a good in-between weight. (192 cells)

I'm not a big fan of welded cells though, to me it's like buying a new hat and then shooting it full of holes. :shock:
Admittedly, welded cells are the best way for production, and the strongest, safest, and compact way to do it,
but I'm a DIY guy, and I don't like the idea of not knowing what shape parallel cells are in, or replacing any single
cell, or all the cells for that matter. Just rubs me wrong.

Solderless is not the best option for many bikes, but just happens to be a shoe-in for this one, given the square box
and the two side plates.

Looking around to see what's available, and has been done first, and I ran across spinningmagnets most excellent
Electricbike.com three part article on the subject, :thumb: .. as well as many other builds here on the ES.

Electricbike: https://www.electricbike.com/introduction-battery-design-1/

After a great crash course on all the do's and don'ts of solderless building, I put on the thinking cap and fired up the
CAD machine.

Here's the latest version,.. changes every day, but you get the picture. Copper rivets, copper series strips, all soldered
of course. Hopping to use a hard-rubber base sheet that will give each cell just enough movement to maintain contact,
and a soft closed-cell rubber that will keep even pressure across the board. Cell spacers on both sides.

Solderless battery pack.png

Packed in a fiberglass box that has thru-screws. And the bikes side plates will help keep even pressure as well.
I have to test some of these materials for their fireproof abilities yet, but this is the direction it's going so far.
We'll see what reality has to say about all this. :wink:
 
Your bike looks like Art. Nice job.

Will follow the battery build with interest. have a build coming up that would work better with a custom battery to tune motor speed.
 
Thanks ZeroEm, I'm really happy with the way the bike turned out, other than being a bit heavy, she's a beauty.

The battery build keeps changing, but thats a good thing,.. the latest improvement is a solid copper strip for
series and parallels, instead of hundreds of small pieces and wires. Should dissipate heat better and is easy to
make.

Also, instead of foam rubber to provide contact pressure, I'm thinking cast silicon, which is fire proof, (I think),
and will seal the box better from rain.

Series strips.png

Anyway, just hunting down pieces and parts for now, and giving the brain a workout.
 
Exactly,.. I haven't found anything that comes close to hitting the bottom of the bike yet.

DIY has got to be the cheapest, I would think. Weld-less has the advantage of changing out cells. But it still isn't cheap,
this is a big pack and it's always running into big $$$ with every plan I make it seems. The price of copper is brutal. :shock:

I tried burning some of the rubber sheet that I had, and it was a fail. Rubber is the best fire starter around! :lol:
Plus, it gets gummy when it gets warm, which is not good for contacts.

So, after searching for days on McMaster's, I wound up with PTFE Teflon sheet for a choice. Its good for 500 degrees, and
flexible even at 1/8" thick. Not to expensive either. So with the silicone, Teflon, and fiberglass,.. all the materials are
high temp, and should make this a fairly safe pack.

I only need it on one side of the pack so that's even better yet, the other side can be rigid. The contacts only need to move
enough to make up any slight differences in cell length.

i sent off for some basic supplies so I can do a few test's before any real money gets spent,.. so we shall see if this works.

I decided against the copper sheet for series connections for now, because it's an expensive way to go and might be too
rigid for the contact posts. Braided flat wire is much cheaper and flexible, and it will handle big amps!
A lot more work though,.. this thing isn't going to be easy because of that. :cry:

Weldless crossection .png
 
APL said:
The price of copper is brutal. :shock:

Because of that, copper theft is a huge problem. Thieves will rip the wiring off the studs of an unguarded construction site. It's a big problem for electric utilities too. This is an older article, but the problem is worse now:

https://www.oe.netl.doe.gov/docs/copper042707.pdf

I've heard of generating plants that were mothballed temporarily, and when they went to repower the plant, they couldn't since all the copper wiring was stolen from the cable trays.
 
Doing my best to avoid building batteries. I like reading about you doing all the hard work and the end result looks good too.

It would be nice to replace a few of the worst cells on a big pack as it ages. I like big batteries, have a pair of 72V 24ah for my trike, cost $1,800 couple of years ago. Thought that was a good deal then.

To customize the shape and tune your speed with the volts would be another plus. There is not much to select from. So I will keep reading and you keep talking me into it. :|
 
That's OK, I enjoy inventing and building, and I don't expect anybody else to do it,.. but it might give somebody an
idea that will help with another project, and lately I'm having a hard time talking myself into it! :oops:

After reading E-HP's article on copper demand, production, and theft, it's easy to see why the price of these parts
are spanking me so hard. (maybe I should invest in copper)

The problem comes with so many parts, 192 batteries is close to 400 ends, times two or three parts, gets into the
thousands and adds up to hundreds of dollars every time.

Well, are we electric bike guys or not? You have to pay to play, and the cost of a decent controller is about the same
so I guess I should just suck it up so long as it gets me what I want.

Wisconsin winters are long, and being able to store all the cells I have as singles is super safe, and helps me sleep at
night.

So, lately I've been downgrading the spec's from the original plan a little, and going with smaller rivets and wire into
a more reasonable unit. No sense going overboard given the experimental nature...and it should still turn out nice.

I should be getting into some experiments soon,.. as parts arrive.
 
Nice packs $$$! Didn't see any dimensions though. I like cells like that, they could be made like a fuse, just plug
and play. Probably lighter with no steel cans, and no wasted space like 18650's have.


Well, I got the 1/8" PTFE sheet a few days ago, and it's a bit stiffer than I wanted so I sent off for some 1/16" stuff.
I'm sure I can use the other sheet in projects by and by, Teflon is amazing stuff!

It finally dawned on me that I don't really need all the rivets and washers and wires, if I can bend the copper series
strips into staples, push them through the sheet and then bend the ends over.. I'll have to put slots the sheet, but
that shouldn't be too hard.

Screenshot 2021-09-03 205128.png

Hmmm, I'll have to try this out and see if it works,.. the staple forms a shorter path between cells, and it's way
cheaper and easier to make than before. My cells are not rock stars, so 30ga should be more than enough, but a
person could go much thicker with it. :wink:
 
Fooled around with it a little today, with some .025" that I had. Double bent the ends and bent them in instead of out,
makes for a longer strip that way but seems to work better.( 8mm x 75mm)

Staple 1.jpg
Staple 2.jpg

Kind of simplistic but that's a good thing, and I'm liking the way it's going so far, flat and tight. Reminds me of the old 'D' cell battery box's, and I guess that's what it is. The Teflon slots easy with a flat punch too. One step forward.
 
Or maybe not.

Making 400 staples is not something I look forward to doing, and slotting fiberglass sheet can't be easily done. Plus,
I don't really like the square heads the strips make,.. too close to the cell case.

I finally received some 3/16" flat head rivet samples the other day, and they look damn near perfect for the job.
The flat 'heads' have a thick enough height from the surface, and will handle all the amps a cell can put out. They're
a bit costly, but save a bundle of work, and all I have to do is drill holes. They also fit the cell caps perfectly.
(had to get the long ones for samples)

Flat head rivet.jpg

Thinking of going back to the copper parallel sheets again too, for the same reason, just need to drill holes and
solder. The newest idea is to drill larger holes in the middles to make it more flexible.

Screenshot 2021-09-13 125441.png

So while I'm going back and forth on all that, I'm working on the things that I know I'll need. I bought some lower grade
XX phenolic sheet for the battery box, about a third the cost of the good stuff, but still plenty good enough for the box.
Butt ugly color, but paint covers everything.

Phenolic sheet (1).jpg

Also bought a big box of 4x5 cell spacers, and the plan is to glue several layers together to make one big block. These
things are cheap enough, and it makes sure that the cells can't move anywhere except 4mm to the side.
Overkill?.. Sure, it's what I do. :wink:

Spacer stack.jpg

I want to use a dozen or so steel allen head thru-bolts to clamp it together, and the glued cell holders form a plastic
tube between the cells that makes sure that the bolts can't touch anything or go anywhere.
They're the last things in, and the first things out.
 
Yes, I like the idea. I've seen it done before but it never clicked,.. now I see it plain as day. Thanks! :thumb:

I'll have to do some experiments, and figure out the backing sheet details and such, but I think it has a good
chance of working well in this situation, and would be the cheapest & easiest way to go yet.

Still working on the spacer block part at the moment, but almost done.
 
My effort at shaping copper sheetIMG_20210905_170131625_HDR.jpg. Bet APL has one of these.
 
3d printed press dies can work well, ideally with the printing just acting as a support for solid metal formers at the load points but using the printing to support metal plates with solid density printed material works ok. The part below has done about 120 plates for 2 poles of 26650 cells from 0.75mm copper (split copper pipe) using m6 cap screws and m12 washers as the formers, no fatigue whatsoever in the body but the guide needed changing a few times.

Also linked is a configurable openscad file for the same kind of press, allows setting of heights, spacings, diameters etc. Only does 2 pole plates with indents at the same height at the mo but things like that wouldn't be too hard to add if there's interest.
 

Attachments

  • 2021-09-24 11.45.02.jpg
    2021-09-24 11.45.02.jpg
    216.1 KB · Views: 2,301
  • 2021-09-24 13.41.12.jpg
    2021-09-24 13.41.12.jpg
    213.2 KB · Views: 2,292
  • 2021-09-24 13.40.27.jpg
    2021-09-24 13.40.27.jpg
    220.7 KB · Views: 2,292
  • stamp_0.2.scad.tar.gz
    1.4 KB · Views: 29
Awesome tools and ideas!  Thanks stan.distortion, your 3D press looks great, and that's some hefty copper,
very precise! Don't have an arbor press mxlemming, but I have a big hammer...

I tried pounding on some copper with a hammer and a punch yesterday, to see how it would go, and get a feel for
the stuff. So I hacked out a die, ground up a punch, and beat on some .025"... and it formed like butter in no time.

Now I need to see how far it will push. I'd like to put a sheet of .032" fiberglass over all the dimples, and maybe
another sheet over the backs too. Still too not clear on all that, but one step at a time.

Finished gluing the spacer block together,.. could have 3D printed it, I know, but it's a big piece and that's a lot of
wear. Had to belt sand the nubs off of one side of the 4x5 spacers, but 8 layers leaves 3mm of the cell exposed.

Spacer block.jpg

It's a pretty strong block too, and I think bonding the battery box (narrow) sides to it will make a stronger center
unit. The two large side plates & contacts can be replaced or exchanged  (S + P options.)  :?:

Now that the spacer block is done, I'll try beating up some more copper for a few days and see what shakes.
 
If you're still thinking of plates or strips, lots of cells covered by a single busplate then I'd look into rolling rather than pressing to do indentations. The one I linked is fairly repeatable, a pressed plate will locate on the bolt heads fairly accurately but not accurately enough to trust dimensions after a dozen or so indents. Wouldn't be too hard to make a jig to improve that and I've no doubt it could give very good results with care but it would be slow going.

The feed on a pillar drill would probably work ok for pressing those indents btw, not all that much force needed for 2x 1mm deep, 10mm dia indents in 0.75mm copper and single indents would be no problem at all even to the cheapest and nastiest pillar drill. Finishing matters more, making sure the pressed faces are dead flat and that would be easy enough to do with emery paper and a plate of glass even for quite big sheets.

Plating is essential imo, those plates above should be capable of over 100 amps when clean but only a fraction of that when they oxidise and nickel plating is working out very well with those using just white vinegar and a pinch of salt as electrolyte, doesn't leave a very bright finish but it's giving a smooth, clean and thick deposit.
 
I'm glad you mentioned plating, I've been thinking about that. Since I'll be taking the pack apart every year, it
wont be too hard to clean the contacts each spring, and it should last at least through the summer.

But eventually it will wear things down, so if I plate them with something it will last longer. I was thinking
silver,.. but nickel might be better because it doesn't oxidize.

You say you've had good luck? What voltage, amps, and time have you been using?
 
Thanks guys!

Interesting, yes, I'm not sure if this is going to turn out, but I'll have welding to fall back on, so it can't turn out
'too' bad,..(hope) :wink:

Using silicone sheet as a spring is a worry,.. not sure if it will stay a spring, or lose memory and take a set. Still not
sure how many thru bolts I'll need either.


Started making a 24 hole anvil to try to do one strip of copper. I used the spacers to scribe some steel and milled
some pockets out. Now I need to round-out the top edges and put some locating pins in it. Start out with small
copper pieces until/if it works out, hopefully.

Steel milled[4343].jpg
 
if using a disc of some type of foam rubber looks like it would be useful, Poron "closed-cell" foam sheets were tried in the past with success.

Just an option...
 
APL said:
Thanks guys!


Steel milled[4343].jpg

Is that tool steel? Could it be?

Gonna take some power to stamp that out.. but all you are doing is upsetting the metal, right? Not punching through.

Cool.

If you want to try that Poron foam rubber stuff ( Mr.SpinningMagnets suggestions above... ) let me know. I got a ton. Its used against cells in EV modules. ... pay for a flat rate SASE and Ill stuff it up for ya.

I like strapping things with tension. Personally, its easy for me and my strapping tools. Retains tansion well.

I make punching dies alot. I got big plans for mass punching of tinned copper strip plate for bussing.
 
Thanks for the Poron suggestion, I'll have to give it a try if the silicone doesn't work. I looked up the Poron specs on
McMasters, (because they always show the specs), and see that its the most springy stuff they have, with good
memory. (190 degree max temp.)

Poron; https://www.mcmaster.com/86375K112/

DogDipstick, it's just cold roll steel, nothing fancy. the plan is to hand punch each dimple...it doesn't take much force,
but they do tend to rip out on the sides if I don't get things just right. Might be a challenge doing all 24.
Depends on the depth of the dimples I guess, so that's where I'm at right now.

Strapping could work too, with the right system... kind of plan B right now, but I'm thinking about it. :)
 
Well, it worked pretty darn good the first try out,.. used some WD40 to help with grabbing, and then gave it the
peening side of the hammer first, and used the punch last. Wants to curl up of course, so I had to make a female
punch to flatten it out again.

Still have to make the center holes, and notch the sides of the strip to clear the thru bolts, but at least it's looking
do-able.

Test strip.jpg

I could just use some paper washers and call it good, but I'd like to do a little better than that. The plan is to try
.032" glass board and drill hundreds of holes in it for a front insulator, but that's pretty labor intensive, and probably
pretty weak. I would need a backing sheet to 'sandwich' the strips and bring back some strength, yet still be flexible.

Paper insulators.jpg

So more experiments are needed, but at least it's moving forward so far,.. dimples are looking good anyway.
 
APL said:
I'm glad you mentioned plating, I've been thinking about that. Since I'll be taking the pack apart every year, it
wont be too hard to clean the contacts each spring, and it should last at least through the summer.

But eventually it will wear things down, so if I plate them with something it will last longer. I was thinking
silver,.. but nickel might be better because it doesn't oxidize.

You say you've had good luck? What voltage, amps, and time have you been using?

Sorry, haven't had a chance to catch up since that post. This is the instructable I'd used and it worked out very well, not bright but perfectly suitable:
https://www.instructables.com/High-Quality-and-safe-Nickel-Plating/

Haven't gone into much fine tuning yet, still working on the battery box design but 5v was about the highest I could go without the finish suffering (rough/stippled) and about an hour was giving a good, thick deposit. Uneven coverage could be an issue, not too hard to get a good finish just where it matters on the terminals but it would probably need turning/repositioning a few times to get a good looking finish, good overall coverage.

Another tank to strip the surface (reverse the process to clean off oxides etc.) makes a big difference and can give two tries at getting the nickel salt solution right, despite my best efforts it ended up slightly cloudy at one attempt and that left a much duller and darker finish that a perfectly clear solution gave and it can take a couple of days to get enough nickel in solution for a volume of about 2 litres. All the gear is still set up here so if you need any other info I can test it out pretty quick, if I get the chance I'll throw an ammeter on it but it was very low, well under half an amp for about 70 cm2.

EDIT: Tinning with solder is another option and if I was starting from zero I'd probably go with that rather than plating, not sure how good corrosion resistance would be on the terminal faces though, doubt it would be an issue but plating is the tried and tested method and I had a heap of nickel strip so went with that instead.
 
Back
Top