Solderless battery project

[Daurema]

I have been working for some time on an easily repairable battery. This model is a 14s10p, suitable for running a small electric vehicle or an inverter for outdoor activities. It can easily draw 1200w to full discharge without overheating.
I was wondering if it would make sense to adapt it for fixed energy storage with more cells in parallel? What do you think ?

 

[Oleksii]

Show close view of contacts please.

 

[Korishan]

To be honest, you will have a lot of resistance between the cells and the contacts. There's no real way around this problem. The added resistance will cause heat generation under normal/heavy loads, thereby wasting energy.

The bolts don't have even enough placement and also not enough placement to have good contacts on all cells. I can see in the image with the cells these are repurposed cells and not new ones. So this causes another issue. All the cells are not uniformly the same dimensions.

The only kind of connection that would possibly be good enough under these situations would maybe be beryllium springs. But they aren't cheap in mass scale.

Sad to say it, but you probably wasted a lot of money on that pcb I hope you don't have a lot of them, and/or you only plan on pulling light loads. By light, I mean <200mA per cell at max load/charge. Otherwise you'll run into problems. Ohh, just thought of another issue, corrosion/electrolysis. Depending on what kind of contact you are using, it could cause electrolysis issue. But there will most likely be a corrosion issue either way.

 

[Wolf]

@Daurema
I like the idea and concept would make for a very tidy pack design,
I am assuming the traces can be beefed up if someone would choose to build a 80p pack instead of a XXsXXp or in your case could the BMS be eliminated and just build 140p pack? Also could cell level fusing be incorporated?
Anyway interesting design.

@Oleksii
I believe they are some kind of spring contact. Not quite sure of the current capacity and or the Borg problem "Resistance is futile" would be interesting to check with a thermal camara under a heavy load. If they are the solid bridge as in the 2nd drawing which I believe they are considering the dented pos caps on the cells conforms to the contact. Not quite sure I would be happy with that.
I would be more inclined to go with a flat contact such as a gold/silver plated button.

Wolf

 

[Daurema]

I don't know how you can assume that without testing.
Beleive me or not but I can pull out 2A per cell without any heat.

 

[Daurema]

Would you be interested in an open source version?
I would like to edit it, but I'm looking to see what configuration would be most useful to most people and for what purpose?

 

[Korishan]

Wait, are those being pressed so hard they deform the top of the cell???


If so, that's not good

 

[Daurema]

Wait, are those being pressed so hard they deform the top of the cell???


If so, that's not good

It is the original welds that were made with too much pressure.
I have checked the heating of this cell and I have not seen anything unusual. The tolerances are respected.

 

[Korishan]

It is the original welds that were made with too much pressure.
I have checked the heating of this cell and I have not seen anything unusual. The tolerances are respected.

By depressing that cap down you are blocking the built in venting of the cell. If the cell cannot vent properly, instead of having a rocket on your hands, you'll have a bomb. That's the issue with this design.
I looked at your web site, and if you plan on mass marketing and selling these, you really really need to make sure there is absolutely zero chance that cell damage can occur. Because even if the end user puts too much pressure and dents a cell, and it goes dynamite style, there will be no way of knowing that they caused this problem and you could very well have a lawsuit on your hands.

You asked for recommendations/input, so that's what we're here for. We don't want anyone to be injured, or sued. So we'll do what we can to give constructive criticism.

 

[Wolf]

It is the original welds that were made with too much pressure.
I have checked the heating of this cell and I have not seen anything unusual. The tolerances are respected.

Well if all the specs check out then it seems to be a decent concept. I however am more inclined to go with some form of contact button on the PCB. JMHO.
Also cell level fusing is very important to at least me if not all of us esp. since we are using reclaimed cells.
Aside from the practicality of having a portable 25Ah or so (depending on the installed cells mAh rating) I would be looking to harden the construction with some form of <0°C cell heating for less than optimal charging temperatures/conditions.
Cost of course is another factor as we are DIY kind of guys.
Each of my 14s80p battery constructions without the cabling and BMS is around $650.00 (or about $46.00 per 80p pack) and I have (4) 14s80p batteries. $2600.00.
I have however gone beyond the normal construction techniques.
So depending on the cost of the PCBs, plus there still has to be some form of protection on each side, is it cost effective?
Here is the breakdown of one of my 14s80p parts costs not including the cells.

Holders J-Chen Fireproof Material 4x5 Cell Spacer 18650 112 X. 50¢, $56.00 Cell Level Fuse Sheets $7.00 per pack X14, $98.00 Pack buss bars 100 feet 8 Gauge Bare Solid Copper Wire, $90.00 0.0625" x 0.5" 9 feet Copper Rectangle Bar Buss connection, $10.00 Wear-Resistant Nylon Tubes 15feet, $21.00 Nylon Threaded Rods 18 feet, $60.00 Nylon Hex Nuts 150, $18.00 Plastic Washers 150, $21.00 Solder 1lb $45.00 0.250 THICK CLEAR POLYCARBONATE RECLING MASKED SHEET X28 Side protection $240.00 Total​ $638.00

​

This of course does not include the Lugs, Bolts, Nuts, (oh yea I'm one) Cabling, BMS, Breakers, Fire proof wood treatment, and everything else to make this a safe and electrically feasible battery.
Wolf

 

[OffGridInTheCity]

Agree with comments above in general and in particular when @Wolf lists "cell level fusing". This is one of the top critical features (in my mind) for safe operation of larger, indoor, lithium-ion batteries.

Perhaps you could add a gap/solder points for each cell on one side of the PCB so one could apply (solder) fuse wire of their choice.

@Wolf listed the cost of building batteries. If I do a gross calculation of overall cost (~$200/kwh) minus the average cost of cells (~1.30/cell) I get a *rough* figure of $800 per 260ah @ 48v battery of non-cell 'stuff' to make the battery functional within the powerwall. Definately in the ball-park of @Wolf's more detailed numbers.

 

[Korishan]

Perhaps you could add a gap/solder points for each cell on one side of the PCB so one could apply (solder) fuse wire of their choice.

I was thinking this as well.
Perhaps could even have an initial thin trace that's only meant to handle 2-3A max with a pad on either end of it. This way it comes "pre-assembled" with built-in fuses. I don't know the exact width and weight of the trace needed to make this this kind of fuse, but I'm sure there are calculators that can figure it out.
That way if the DIY'r is wanting to use that rating, then they just keep on trucking with assembly. However if they need beefier solution, they can add a fuse on top and solder to the pads of either end of the trace-fuse.

 

[Oleksii]

Show close view of contacts please.

@Daurema is there a chance that you share it?

 

[J_Mack58]

I have been working for some time on an easily repairable battery. This model is a 14s10p, suitable for running a small electric vehicle or an inverter for outdoor activities. It can easily draw 1200w to full discharge without overheating.
I was wondering if it would make sense to adapt it for fixed energy storage with more cells in parallel? What do you think ?

Not too easy to repair but better then removing nickel strip, I see the Daly BMS so that will indicate at least the parallel string with the bad cell(s). I like but the ultimate for me is battery repair with the battery still operational.

 

[Wolf]

the ultimate for me is battery repair with the battery still operational.

Well then you would have at least 2 batteries in parallel each individually connected to a buss bar via a circuit breaker.
If Battery 1 needs servicing then trip the breaker and service away .
Battery 2 would still supply the load or be charged whilst you happily service Battery 1.
Wolf

 

[italianuser]

@Daurema I like the concept, a lot.

All the comments do make sense, for me the most difficult point is the cell contact.

Consider my scenario in which I use recycled cells.

How can you tell if a cell isn't making contact? I just checked the battery I'm mounting and I can measure, in some points, a height difference of around 0.5mm (0.2"); that happens because some cells have a bulge on the negative side and are so higher than the adjacent cell. Plus, the adjacent cell could have a nickel strip fragment I didn't remove on the positive side so that could add an extra 0.2mm of height difference (total 0.7mm/0.28" height difference).

I think it would be a great thing to have modular PCBs, for e.g. 4x5 cell PCB which I can mount together to make any size I need. A 4x5 cell holder gives me a 20P module.

I think your best customer could be the one which uses new cells, maybe ebike battery makers?

 

[grundholm]

think it would be a great thing to have modular PCBs, for e.g. 4x5 cell PCB which I can mount together to make any size I need. A 4x5 cell holder gives me a 20P module.

I have made some modular PCB's - still didn't test them... I have no experience with springs etc. so my idea was to spot weld the nickel strip and then solder it to the PCB - then either use a SMD fuse or glass fuse and add a decent busbar...

 

[italianuser]

I have made some modular PCB's - still didn't test them

Uhm, I didn't understand the part of the nickel strip and the soldering...

You have great ideas and a very good ability to make professional products, I had a look at your website. Who is your customer?

If you want to get the end user market: for me as a user it must be easy to create batteries for my needs, today it's a 14s10p, tomorrow it's a 7s20p... other users might want a 14s80p.

So, a part from the points other users spoke about (safety, cell contact, etc), for me the modular system seems the best product: let me mount the battery like LEGO. A modular system is easier to sell because you can satisfy the needs of more customers and costs you less money for producing it and making an efficient warehouse (500 modular 4x5 versus 500 fixed size, 100x 14S10P / 100x 14S20P / 100x 7s10p / 100x 7s20p / etc).

What is your exact market placement, you can't have a product for everybody, I think. For a DIY guy like me your product might be too expensive.
I love this product, but I won't buy it, too expensive:


(https://www.amazon.com/Sara-u-Balancer-Module-Battery-Protection/dp/B08L6CQGD4)

Possible customers could be maybe e-bike battery producers, professional battery repair shops; also to high end users.