My excellent adventure through some battery builds

[ElectricGod]

Interesting didn't realize the venting mech and how they fail differed.

I like the idea of sub packs for my scooter frame I can slot them down a long bit of box section and only have a small opening like stuffing lipos in rather than all at once allows me to create a real clean design.

Wish I was local I'd be dropping by have a look at the lair your cave make Aladdin blush.

Yeah...it would be cool to have a few more EV friends nearby. I've been trying to create that locally for quite a while.

 

[ElectricGod]

The 3 sub packs are a smidge tighter than the box allows for. My mockup was clearly a little smaller than the final product. I need to still enclose all 3 sub packs into 1 large pack and it already won't fit the box. I'll need to take down those threaded blocks a good bit so I have width and length to clear the packs.

 

[ElectricGod]

I bought some .3mm x 8mm nickel since the KWeld can handle it. The guy I got it from told me about a new welding strip he will be carrying soon.

It's nickel, stainless steel and copper. I asked if he could send me a sample of it so he sent me 12" of it. He has no supply yet so this with a lot of his very limited stock.

I wish I had widths and thicknesses all the same to make things a bit more obvious.

Top to bottom:
SS/Ni/Cu .1mm x 9mm 11.1 mOhm
Ni .15mm x 7mm 20.9 mOhm
Ni .3mm x 8mm 9.85 mOhm

As expected the .3mm nickel has the lowest resistance. But check out that .1mm thick alloy! Not much more resistance despite being 1/3rd as thick.
There is a very slight color difference between pure Ni and the alloy. The alloy is a bit stiffer than the .15mm nickel.
To keep glare from the camera flash off the metal, perspective makes these 3 strips look like they are different lengths, but they are all 12" long.

Close-up on the values:

 

[ElectricGod]

I have some PET tape coming to finish this up, but otherwise it's done.

The dual 10S balance connections are on each sub-pack.

The kapton tape will get covered with PET tape which is a lot tougher. It's on order, just not here yet.

It's partly enclosed in fiberglass sheet. I started bringing out the power wires.

All enclosed...and waiting for the final PET tape layer. The power wires will get tied together later.

 

[john61ct]

Seems silly to me to limit the design to a specific pre-existing box.

It is so much easier to build a custom box than it is to build a great battery pack, focus on the latter, get it as optimised and compact as possible overall, but still well ventilated, within an overall dimensional "boundary box".

Once the pack is finalized, then build the enclosure so it works well for that pack design.

 

[thundercamel]

I view one of the main advantages of 18650 cells as the ability to configure packs into shapes that meet space constraints. Laying out the series/parallel groups is my favorite part of pack design

 

[ElectricGod]

I view one of the main advantages of 18650 cells as the ability to configure packs into shapes that meet space constraints. Laying out the series/parallel groups is my favorite part of pack design

You got it exactly! Space constraints defined the shape and size of the pack. It HAD to fit in a very specific space. I agree there is some enjoyment in figuring out how to max out a given space. I look at 18650 cells like they are lego bricks. How many can I pack together in a given space.

Seems silly to me to limit the design to a specific pre-existing box.

It is so much easier to build a custom box than it is to build a great battery pack, focus on the latter, get it as optimised and compact as possible overall, but still well ventilated, within an overall dimensional "boundary box".

Once the pack is finalized, then build the enclosure so it works well for that pack design.

The battery box defined the amount of room I had to work with and it's already about 50% larger than the previous battery box. I used 100% of the available space.

 

[ElectricGod]

Last night I took apart the original 20S10P pack to get down to the cells. I removed the original dumb BMS and all the fiberglass sheet cladding. This got that pack down to it's minimum size possible. I'll probably reuse the existing balance wires and add a couple 10s balance connectors to them like I did on the 3 20S3P packs I built and have completed. The end goal is a single smart BMS that balances everything.

This battery box is the maximum amount of space available to use. It's pretty full! based on the cell capacities, my add-on pack despite filling the wedge shaped area has the same capacity as the larger cube of the older pack. I have yet to do actual capacity testing for both packs.

 

[ElectricGod]

This is the original pack next to the new pack I just built.

I've rerouted the balance wires and added 6 awg wires to the old pack.

Still waiting on the PET tape but it's all enclosed. I need to yet re-terminate the balance wires.

 

[ElectricGod]

The old pack is enclosed again. I'm still waiting on PET tape to finish enclosing the packs. The old pack got new balance connectors on it after these images were taken.

This balance cable connects the BMS to all 4 packs. It took a couple of hours to make. The 6 connectors that are side by side are for connecting to the new packs.

 

[ElectricGod]

I couldn't quite fit the packs in the box until I filed down these 2 screw blocks. There's about 1/4" gap front to back in the box.

This would have never happened if I didn't slim down the old pack as much as possible. Both packs have 2 layers of foam on all sides and then fiberglass sheet outside that. I'll add padding around the packs so they can't move around, but this is pretty much how it's going to look. Other than wiring things together, there is virtually no room for more anything in the box.

 

[ElectricGod]

I wrapped it all in Kapton tape first. I ordered some PET tape which arrived yesterday so that got added tonight. The packs are effectively sealed up and enclosed. I won't be doing heat shrink around this...no point.

 

[ElectricGod]

Time passes...lol...
I get back to projects and work on other stuff.

These packs have been testing nicely for a while now.
I'm getting expected capacity and it all has been working great!
I was about to pack them up to ship to my friend in San Francisco.
I thought one last check of everything was a good idea.

I don't know what yet, but cell 12 in one of the sub packs has died. It is dead flat and will not take a charge.
They have been been running fine for weeks! Now all of sudden, I have a dead cell.
Grrr! I need to take apart the new pack and see what happened.

 

[john61ct]

Why I want solderless methods to mature.

Is the PET just to add tensile strength and abrasion resistance to the Kapton?

In other words, not necessary if to be used inside a robust padded enclosure?

 

[ElectricGod]

Why I want solderless methods to mature.

Is the PET just to add tensile strength and abrasion resistance to the Kapton?

In other words, not necessary if to be used inside a robust padded enclosure?

Right...PET is tough and scrub resistant, Kapton is not. Thick padding is needed if you use just Kapton. These packs will be installed in an off road bike. They will be enclosed, but they will still see vibration and shaking. Kapton alone was not going to be enough.

These packs were welded with a KWeld. The only soldering done was onto the nickel for balance and power wires. Even that was done well away from the cells. I have been experimenting with solder paste and welding. I put a little bit of solder paste on the cell before welding. The cell then gets a spot weld and solder flow. It works pretty well and the resistance at the spot weld is significantly lower. Weld strength is improved too.

I have 2 theories about what happened:

1. I missed isolating a balance wire from the tops of the cells and one is shorting to the next cell group.

2. I did have a single incident where I had the weld current set too high and I burned a hole through the bottom of a cell. I caught it immediately since it started leaking out the tiny hole right away. I did check for this before I enclosed everything and saw no evidence of any other cells having this issue. Maybe I missed a cell with a leak from too strong a weld?

 

[john61ct]

By solderless I also meant no-weld.

So that it is relatively easy to atomize / reassemble / reconfigure the pack, and also test and replace cells individually as needed.

As well as avoiding the potential damage and failure modes you mention.

 

[ElectricGod]

By solderless I also meant no-weld.

So that it is relatively easy to atomize / reassemble / reconfigure the pack, and also test and replace cells individually as needed.

As well as avoiding the potential damage and failure modes you mention.

The 2 options I have found are:

18650 battery holders: (I've used this option for several years. It works pretty good as long as the cell can't migrate out of the holders.)
Low current per cell
Fairly tight cell packing
Easy cell replacement or installation
Cheap - less than 10 cents per cell


Vruzend kits:
Good current draw per cell
Poor cell packing - the end caps add 2/3 more length to each cell
Somewhat easy cell replacement - pull the retaining screws, pull of an end cap, pull the cell.
Expensive - $1.33 per cell...OUCH!

1. Build a 20S10P pack or 200 cells and the Vruzend option will add $133 to the cost of the pack and can't be reused for other packs.
2. A KWeld spot welder in making dual 20S10P packs at the price of equivalent Vruzend parts will pay for itself and continue to serve well for many thousands of more welds.
3. Nickel strip is cheap and very compact.
4. Spot welding is very reliable.

I see a viable price point for battery holders as long as you secure the cells and don't need more than 5 amps per cell.
Otherwise...spot welding wins. It is very compact, reliable, provides the best Ah density of all options and the welder is 100% reusable for many hundreds of packs.

I didn't mention this above when I talked about using solder paste. If I turn down the power to the point that a spot weld won't form, there is still enough heat to melt the solder (solder weld). In effect, I soldered the cells together without all the heat soaking that an iron creates and damages cells. Solder has less resistance than a spot weld. The spot weld plus the solder makes a more secure connection that has all the benefits of the spot weld plus the benefits of solders lower resistance. A solder weld is pretty secure by itself and you can't accidentally pop a hole through a cell. The failure mode is the solder. You still make several spots for the solder weld.

 

[Ianhill]

Seen the new gen drill batterys ? they are using thin copper foil spot welded.

Surely the copper would have to be enameled for longevity specially on a drill battery in harsh environments, I can not see how the spot welder could blow straight through the enamel at the point of contact without having high enough voltage to break down it's insulation I'm interested in what they have done can it pass down to diy packs or is it inferior in longevity just bare copper with a high level surface finish to help keep it from corroding to fast but yield great results within the 2 years guarantee.

 

[ElectricGod]

Seen the new gen drill batterys ? they are using thin copper foil spot welded.

Surely the copper would have to be enameled for longevity specially on a drill battery in harsh environments, I can not see how the spot welder could blow straight through the enamel at the point of contact without having high enough voltage to break down it's insulation I'm interested in what they have done can it pass down to diy packs or is it inferior in longevity just bare copper with a high level surface finish to help keep it from corroding to fast but yield great results within the 2 years guarantee.

I don't know how enameled copper would work.
I can tell you that gunk on your weld tips will effect how good of a weld you get. Resistance at the probe tips effects weld quality.
I can't see how an actual full on insulator would work at all.
With bare copper, the resistance is so low that you get insufficient heat at the probe tips to melt the copper to make spot welds.
It's a case of too little resistance and the copper just spreads the heat around instead.
What you do is put a section of nickel on top of the copper and then the nickel provides the heat path to weld down the copper to the cell.
What I have read is that stainless steel or nickel weld tips will do the same job as a section of nickel on top of the copper.
The steel weld tips focus the heat at the spots.
I haven't messed with nickel on top of copper or steel weld tips at all so I am basically parroting what I have read.
I have tried welding copper sheet and wire with the copper weld tips and that fails completely to weld no matter how high I set the current.
My guess is they are using bare copper with no enamel coating.
I recently came across a welding material that is copper, nickel and stainless steel. It's far lower resistance than nickel and doesn't corrode like copper. It also welds directly. Maybe these packs are using a copper alloy?
I have a 12" long section of this stuff. It welds nicely.
I used my milli Ohm meter to compare it to nickel and the resistance is not as good as copper, but far better than nickel.
If you imagine a scale 1 to 10 and copper is a 1 and nickel is a 10, the alloy is a 3.
I put a section of the alloy, nickel and bare steel in salt water for 24 hours. The nickel and alloy were untouched. The steel was rusty.
Maybe if they use pure copper, they are coating it after it is welded?

BTW...voltage isn't what creates the weld. It's highly localized current that creates heat at the weld tips and melts the metals together. I have 2 super cap modules for my KWeld. In series, that gets me 16.8v. I could not tell I was welding at nearly 17v as compared to 8.4v from a single super cap module. I now run my 2 modules in parallel if I need loads of current.

 

[Ianhill]

Nice read, yeah i see the voltage is way to low to brake Down the insulation of a enameled conductor they get tested to 1000v i beleive so any cell on the end if that be brown bread so i guess it's a copper alloy specially for the job becuase i can not see any evidence of additional laquer been applied after the weld so it's a straight forward as possible for factory assembly purposes.
Nice to see you already have a sample to test out too never one to hang about when you get something in your head.

 

[ElectricGod]

Nice read, yeah i see the voltage is way to low to brake Down the insulation of a enameled conductor they get tested to 1000v i beleive so any cell on the end if that be brown bread so i guess it's a copper alloy specially for the job becuase i can not see any evidence of additional laquer been applied after the weld so it's a straight forward as possible for factory assembly purposes.
Nice to see you already have a sample to test out too never one to hang about when you get something in your head.

Right...why I dished out for a KWeld! It was the right tool for the job and not some chinese POS.