mounting batteries

Hi guys,
i guess most of us do have a lot of batteries left that are not total useless, but they are also to bad to put into a power wall.
I am talking about batteries in the range of 1700-1300 mAh.
The reason we don't use them is because building the pack takes a long time.

Why don't we build the pack so that we can exchange every single cell, without destroying the whole pack?
All we need todo is touse on one sinde the common 18650 holder and on the other side slightly bigger holes so that we can pull the batteries thru. On this side we use for each cell a spring. The springs are on the coverand we do the fuses and the busbar on the backside of the cover.

This way we just need to open our cover and which some or all cells, and fix some broken fuses.
With this design it's worth to burn thru also the not sooo good cells.

So why is nobodey doing it like this? Where is the weakpoint of designing it this way?

Because it cost to much time to fiddle with low capacity cells most People want to build once and then leave it

I think in the long run it takes less time.
So there is no other drawback, than that it takes more time?
If so, then i will do it this way-

Looking forward to seeing your progress and prof of concept.
We all love a few progress pics on here.

There are plenty of users who are using a non-soldered/welded design to hold their cells so they can be replaced easily. The biggest issue and hurdle is, finding out "which" cell is the bad one in a pack in the first place. To do this, you need to add a lot of smarts into the design to monitor either each cell, or just a few cells to narrow down finding the problem.

And a cell can go bad by just not holding as much capacity (bringing that parallel section out of balance with the others faster, thereby making the bms work harder) or go short and pop the fuse. In most cases, it's the former, where it just stops performing as well, so it's hard to find the problem cell.

One of the weak points in using springs for contact is pour contact. A spring doesn't make no where near as good contact with the cell as a solder or welded joint. There are other options like using a tab, but it's still not as good. Another problem is that most springs are silver plated steel. These is a pour conductor of current, especially at higher amps (this would be amps as low as 1A, maybe lower) and will begin to heat up. Instead, a beryllium based spring should be used. There are a few on here who use those for testing rigs to get super accurate test readings. Search for "Beryllium" on the forum and you'll find several threads.

My project is kinda like what you're talking about. I use soldered fuse wires, but a defect cell can be pushed out and replaced easily. Whole process takes like 5 mins (from taking pack offline, to back online), provided the cell is visibly bad, and you have a voltage-matched replacement cell handy.

You 3D print your holders by yourself right?
Isn't it too expansive to 3D print them?

Replacing cells in the common plastic holders arent hard either.,

Tombery said:

You 3D print your holders by yourself right?
Isn't it too expansive to 3D print them?

Click to expand...

"Expansive"? The maximumI can print on my 30cmx30cm CR-10 3d printer is104p.

I'm guessing you meant "expensive": Not at all, but it takes lots of time.
My 104ppack (2x brackets, 4x clamps) costs about $3.50 using cheap $16/kgPLA filament.
A 100p pack using theubiquitous 4x5 brackets (10pcs) costs $6.50 on AliExpress.


Replacement with my design is a littleeasier because:
* the tabs holding the cells in place is thin and flexible - no need to break them off
* the busbar isoff to the side - does not get in the way when pushing out acell
* slightly more loose fit -even slightly thicker cells (double sleeved, leaked&swelling, etc) can be pushed out easily

It certainly makes life a littleeasier, but the biggest hassle is still taking the pack offline/online, finding the bad cell*, and soldering.
*so far, all my bad cellswere leaking electrolyte or had blown fuses, thus making identification quite easy.

Hey I am working on a system that may do this. My goal is to be able to quickly and easily change, add, or remove "Fused" cells. I will be working with recycled 18650s from laptop packs. So, I know I am going to want to fuse every single cell. I will also want to be able to add more as I go, and remove any problem cells as I go. (Although I am testing the cells before using them)

The design I am working on is being 3D printed using PETG filament. The actual battery contacts are simple leaf-spring battery contacts that snap firmly into the 3D printed battery sleds. Fuses and other connecting wire is soldered on the back of the battery contacts. But, the 18650's are just inserted and removed just like any other leaf-spring battery tray.

This is a modular system design. Each module can hold 12 cells. The modules basically hang on the wall using a special 3D printed brackets that also provide buss wire management.

So the thought is that you load the module with 12 cells at your bench or wherever you are comfortable. Then you take the module over to your powerwall and hang/snap it on a bracket. Plug in an XT connector, and you are done. If you need to remove the module, you just unplug the XT connector, and remove the module from the wall. If you provide enough space between the modules you can probably do things to them without having to remove them. But, I am going to hang them VERY close together in my project.

I literally purchased a 3D printer 3 weeks ago just to work on this project. I am currently printing and testing prototypes. As soon as I feel like I have the design working well, I will start a build thread with pictures and all of the details. It is taking me a little time because printing with PETG has a learning curve. Heck....all of this has a learning curve.

I've thought about various solderless approaches but they all involve buying more than bus wire, fuse wire, and solder. The price really starts to add up as you get to 6,000 cells.
I find that its pretty quick to take a pack offline, un-soder/lift the fuse wire up from the cell, and lift off the 4x5 18650 plastic battery holder. This exposes all the cells to review, volt meter for errant voltage, or even pop them out and retest.
For individual cell replacement/additoion - its easy to break the tabs of 4x5 18650 cell holder and desolder/push one out or push more in and solder up. Soldering does not seem to have hurt any of my cells.

The effort is not really that difficult vs how often you do it and even a solderless system will require unscrewing this or that or popping cells out of their holders etc.

Agreed.....totally. Solderless is more expensive and more complicated. But I am treating this as a hobby so don't mind spending a little more to enjoy it. For other people who are trying to be cost effective that may not be a good solution.

Its like building a $1000 solar generator for camping when you could just spend $400 on a nice generator.

clupton said:

Agreed.....totally. Solderless is more expensive and more complicated. But I am treating this as a hobby so don't mind spending a little more to enjoy it. For other people who are trying to be cost effective that may not be a good solution.

Its like building a $1000 solar generator for camping when you could just spend $400 on a nice generator.

Click to expand...

Agree as well! Part of all this is that its FUN and of course you can make it'prettier' etc.