diy solar

diy solar

Avoiding Galvanic Corrosion

This aluminum lugs are rated for 2:0 copper wire, meaning they should be able to handle 283A: https://www.gordonelectricsupply.co...eai-iIXQ9HJunhhZE-nywJ1UL2uqESUwaAi5VEALw_wcB

So what do you think about the idea of using a long aluminum bolt with nut,etc to lock the bolt with lug attached into the terminal with Loctite or JB Weld and then torque the lug down into the terminal surface with the nut?

The nut can be loosened if the angle of the lug needs to be adjusted but otherwise, short sections of 2/0 thick-stranded copper wire (not thin stranded welders wire) is used to form connections between cells by positioning in a pair of lugs and tightening down the set screws.

Is this a good solution to avoid galvanic corrosion and preserve the soft aluminum threads in the terminals by only using them once (and never unthreading them)?

The only dissimilar metal is the copper stranded wire in the aluminum lug, but I’ve got that combo used throughout my electrical panels...

The problem with those thingsis that you'll apply some torque to the lug screw which might make it move on the terminal. I'd recommend to use flat aluminium stock like we do with copper.

If you're worried about the rigidity of such busbars then apply the same techniques than with copper: stack multiple thin busbars with a bridge form (I think it's CALB who does the copper version of that if you need examples) ;)
 
The faster you weld, the less the thermal energy.
If they were meant to be bolted, I think they would have been drilled and tapped when the terminals were manufactured, not after the cell was finished.



Technology - Laser
Battery Pack Applications - Cylindrical, prismatic, pouch, ultra-capacitor

View attachment 34970
This is probably the best approach for a one and done application like if you were manufacturing a complete battery to sell, but for a large battery bank a DIYer is putting together I dont see it as a real practical approach. Again, it is probably the best method of attaching a conductor, just not from any other point of view IMO.
 
Actually there is motion due to the fact the cells expand and contract. Even if the cells are mounted tight in a fixture with no springs there will be some motion. Compressing the cells too tight could shorten the cycle life and that's straight from the horses mouth, EVE. I don't see any way of avoiding some motion between the cells unless they are spaced far enough apart with a gap between them to avoid it.
The datasheet indicates that the compression of 300kgf (I've not seen a satisfactory definition of exactly how much that is) extends the cycles the cells will endure.
300kgf sounds like a lot to me. Maybe I'm wrong, but my thoughts are that it is enough to resist movement and keep the terminals immobile in relation to one another.
Screenshot_20210129-080539_Adobe Acrobat.jpg
 
I just snug the fixture when the cells are fully charged.
When the cells deflate they don't move much.

At least that's the new plan after messing up a bunch of terminal holes
The busbars should be loose then retorqued after adjusting the fixture.
 
Last edited:
They will move the same amount from 100 % to 0 % SoC than from 0 % to 100 % SoC, else the pack would grow over time... ?
 
I was somewhat joking but if you lined up the cells up without a fixture you'd save money and time and the terminal holes would never be stretched.
 
Oh ok ^^

Note that in a mobile application you need to fasten the cells anyway so that wouldn't work.
 
Mine didn't have a hex key at the top or a hex shape on the flange, but these were custom made, so you could specify that in your drawing. I was quoted approx £0.88each +vat if ordering 500. I only ordered 17, which cost a total including postage of £128, which made it very expensive. It is the machine setting that is the costly part.

A hex shape would have eliminated my problem of the stripped terminal completely, since i was having to double nut lock at the top to set my 4Nm torque. When i was subsequently 'breaking' the lock, i inadvertently released the 4Nm torque setting below un beknown to me. Of course, having set it with thread locker, it was then impossible for me to break it free and the stud itsel sheared. I'm not actually sure which grade aluminium was used, but you could specify this also. These can be machined out of many metals...see here below
Product overview - SB Cov Ltd (specialbolt.com)
in use....
View attachment 34983
Wow, so these were aluminum M6 to M10 flanged adapters that you had custom milled - I’m impressed.

Since you’ve got that flange, no worries about piercing the bottom and sounds like you have had no issues unthreading the post when loosening the nuts (if anything, the opposite).

This is what Basen, Xuba and the other LiFePO4 cell resellers should have been bundling with their cells...

Where did you get the idea?
 
I think the disadvantages of that flange adapter is that it adds an extra physical contact between the cell terminal and the bus bar.... which is also less conductive since it is stainless steel .... and the surface area on the top of the flange is quite a bit less that the cell terminal.
I like getting up to the M10 size, but seems to have potential to add more resistance overall.
 
The datasheet indicates that the compression of 300kgf (I've not seen a satisfactory definition of exactly how much that is) extends the cycles the cells will endure.
300kgf sounds like a lot to me. Maybe I'm wrong, but my thoughts are that it is enough to resist movement and keep the terminals immobile in relation to one another.
View attachment 35012
You are (wrong).

With the recommended 300Kgf of force (15psi over the end surface of the cell) applied constantly through the full 0% to 100% charge cycle, the cells expand and contract by about 1mm.

The only way to avoid movement is to use a rigid fixture applied at 100% SOC and allow the cells to shrink freely as they discharge (meaning ~1mm gaps will form between the cells at 0% SOC).

Now that I’ve understood that even aluminum lugs are plated with copper or nickel and then tin, I’m realizing that avoiding dissimilar metals is essentially impossible with these cells.

So I’ve concluded the best way to avoid corrosion (galvanic and atmospheric) will be to keep the cells dry on a sealed container (possibly temperature-controlled to 25C).

In a warm dry environment, would you see major concerns as far as galvanic corrosion with tin-plated lugs secured onto the aluminum surface of the terminals using stainless grub screws (threaded posts) and brass nuts?
 
I think the disadvantages of that flange adapter is that it adds an extra physical contact between the cell terminal and the bus bar.... which is also less conductive since it is stainless steel .... and the surface area on the top of the flange is quite a bit less that the cell terminal.
I like getting up to the M10 size, but seems to have potential to add more resistance overall.
I thought he had those milled out of aluminum?

Also, the flange provides another advantage - it prevents the M6 thread from penetrating too deep (self limiting).

My busbars are narrower than my terminals, so the full terminal surface is rarely contacted...
 
So I’ve concluded the best way to avoid corrosion (galvanic and atmospheric) will be to keep the cells dry on a sealed container (possibly temperature-controlled to 25C).

Adding NO-OX-ID is one way many of us are avoiding corrosion from dissimilar materials. Even when I have aluminum to aluminum connections, I'm still going to use this.

 
Adding NO-OX-ID is one way many of us are avoiding corrosion from dissimilar materials. Even when I have aluminum to aluminum connections, I'm still going to use this.

Yes, that too.

But Haugen makes it sound like that alone won’t be enough (at least in a moist environment and given enough time)...
 
I thought he had those milled out of aluminum?

Also, the flange provides another advantage - it prevents the M6 thread from penetrating too deep (self limiting).

My busbars are narrower than my terminals, so the full terminal surface is rarely contacted...
I think you are correct that those custom ones of his are aluminum .... but they do still add an extra contact point in series with the bus bar ....My bus bars are wider than the terminal top .... and I think his are also.
So, and extra physical contact point is added and there is less surface are for the bus bar.
If the flange size were larger that would help, also it might be possible to make up for the lost surface area by using a higher torque that the M10 will allow.
 
Last edited:
Yes, that too.

But Haugen makes it sound like that alone won’t be enough (at least in a moist environment and given enough time)...

Don't let perfect be the enemy of good.

How long are these batteries going to last? How long are you going to keep them in one place, never removing the battery cables?

You're taking things to the 99.99999th percentile. I'm designing/building for no more than a ten year life cycle. I suspect my connections will look pristine at ten years.
 
Don't let perfect be the enemy of good.

How long are these batteries going to last? How long are you going to keep them in one place, never removing the battery cables?

You're taking things to the 99.99999th percentile. I'm designing/building for no more than a ten year life cycle. I suspect my connections will look pristine at ten years.
Well having my cells in a sealed enclosure is easy - was planning on that anyway.

Biggest downside is that if they get too hot through use, I’ll need to add a fan (not a big deal and probably won’t be needed at average discharge of ~20A).

And after that, temperature control is easy to add but something I’ll probably only bother with if I do confirm that my cells lose substantial capacity when charged under 55F.

But yes, perfection is the enemy of good, I agree.
 
Back
Top