Thank you @rosco @Linesflag
Do you guys use Kapton tape along side with Filament tape? Seems like the Filamenet is stronger.
They donât look potted. Frankly, the bare cell look doesnât inspire confidence at all.
But you do you.
2 Likes
Thanks for the feedback! Seriously considering reworking & changing future packs as by the looks of it is sure doesnt. I have gotten that comment elsewhere as well.
However there is a little over a full sized tube of silicone in there (nearly 400g). And appart from having used this battery layout since 2018 without a single issue. I have done plenty of testings on a scrap enclosure & deck with and some other test pieces. And wrapping the p-groups with shrink-wrap or fish paper makes the adhesion to the enclosure worse as there is less surface area for the silicone to stick to the P-group & that is the bottleneck in terms of the adhesion.
I am afraid of limiting the cells ability to vent if I add silicone to the positive side of the P-groups by further submerging it. guess I could add more to the top and along the sides so the P-groups are more cohesive if on cell where to get lose out of the P-group.
What I am considering is adding some adhesive neoprene or similar on top of the P-groups. In my testings the enclosure totally fails before the silicone bond fails as there air gap above the cells is small enough to not let the cells move a long enough distance and peal up from below. (even when the epoxy formulation & the orientation of the carbon is put in place to achieve a higher elongation at break.) And it is pretty much impossible to compress each segment far enough to have the cells come in contact with the bottom side of the deck (well it takes alot of force as there is nearly 2mm thick carbon in the bottom of each segment and it has a shape to it to further make it more rigid)
Btw there is a layer of glass fibre on the inside of the enclosure and there is an added think layer of epoxy at the bottom of the deck.
So far only one board has gone out to a local customer in Gothenburg. And I will take his board back in a little bit anyways to swap out the motors (as he uses motors that are not Boards of Sweden branded & he needed a 8mm axles for the gear drives and mine are 10mm)
If you where im my situation how would you go about it?
I value your opinion alot!
2 Likes
I thought the same thing at first glance but I did notice that the enclosure itself does provide physical separation of the p-groups so I donât think a full wrap is really necessary. My one comment would be what is insulating the cells from the balance leads as below. This looks like a good place for some strategic fish paper wrapped around the four p-groups adjacent to where the leads run imo. If u were very specific about where u placed ur fish, it wouldnât affect adhesion or the cellsâ ability to vent.
2 Likes
Not 100% sure if I followed correctly.
But to clarify as it is a little hard to see. The velcro holding the balance leads are glued down to the enclosure. So the balance leads canât flop about and get pinched.
Not sure if that was what you where getting at or not?
Yes, itâd not entirely visible that theyâre physically far enough away. Personally, Iâd run fish along each side if the p-groups adjacent to that corridor that the leads are running along. Glue fails sometimes and when Iâm thinking of battery safety, one of the things I always try to do is envision what would happen if one of the wires came loose (broken solder for e.g.) or if one of my primary measures fails, what secondary measure would I have in place. My $0.02. It would cost you almost nothing in time and materials and would provide more robust isolation/insulation and accordingly, safety.
2 Likes
Yeah you are right wont hurt adding that.
I might do a layer of fish above the p-groups that curves down towards the middle where balancing leads runs along.
that way I keep the bonding surface between the P-groups and the enclosure. While also adding a little more of a buffer on top and towards the balancing leads.
And it wont be as scary to look at. 
2 Likes
Something i forgot to ask yesterday
According to mooch escale, the P28A is better than the P26A, why do we still use the P26A?
Second question
Any opinions/things i should consider about building a wooden compression box for a battery system?
would be with copper tabs from agniusm, as theyâre perfect for the job and fishpaper between the S groups (and on the cell + terminals)
It would be two boxes of
9P length and 3S height
3P length and 3S height
I imagine that i might need button top cells for the S groups to safely connect
The deck is a choochoo from alessio, which is pretty short and stiff
A sketch to show what i mean (not drawing balancing leads as this would be a mess)
it would have space to house everything that way and would be easily maintainable in case something goes bad (which happened to me twice (of 3 batteries) 
)
the tabs would probably be glued in with some super glue and if possible physically separated, just in case
The boxes would be closed from all sides, with just the openings for the terminal tabs and balancing leads
As itâd be only 18mm in height, i doubt that the wood could do anything against the compression
can you buy P28A? can you buy P26A?
thereâs your answer
according to nkon, currently only P28A
Any master builders willing to advise if all looks ok before shrink wrapping this battery up? Bms works and voltages are good. Just want to make sure I havenât done anything that may cause issues later.
Still have to wire positive from battery to charge connector but waiting on inline mini blade fuse holder to arrive.
Got 2 questions as well, will I be OK adding the fuse to positive on charge or is it better on negative going to the bms? And I am going to be using a 3a charger so is a 4a fuse ok or 5a better?
5 Likes
Looks excellent.
Usually I put the fuse on the positive lead. You could easily go up to 10a, I wouldnât want something that close to my actual charge rate, personally. I charge at 6a through a 10a fuse
2 Likes
Thanks man, learnt a lot from this thread. Must have also checked the balance leads about 50 times with a multimeter just to make sure they were still in the right order lol.
I have other fuses so can go up a bit, just read something g a while back but canât find it now about going a certain percentage over the charge voltage, and with 4a fuse being 33% over the charge current of 3a it seemed fitting. Dont want to blow the fuse regular but also didnât want it too high.
- Might be the only one of the two that is available.
- Might be available at a lower cost.
- The P26A has a good track record and is still a great cell.
- Might have a P26A pack and want to replace cells with same type.
- My customers might not be familiar with P28A yet and trust P26A.
- Mooch could be a Molicel shill, selling himself out for a pizza to pimp the P28A.
11 Likes
in that case: Samsung 30Q it is!
(Iâd buy you a pizza, but i need an income first)
2 Likes
Co-sign w @ShutterShock. Looks nice and clean. Not sure if this is your first pack but it looks a LOT better than my first attempt!
Re the fuse, I also fuse the positive charge lead and the rule of thumb is ~2x the current youâre charging at. But a 10 amp fuse will suit most peopleâs needs. Also, make sure your fuse is rated appropriately for the voltage. So for a 12s pack, 51 volts DC or more.
1 Like
Thanks @Common_good, yeah 1st pack but not first time soldering electronics so that gave me a bit of confidence but li ions are a whole nother ball park.
Got a10a fuse too so will go with that rated at 58v 
1 Like
Same here. Really clean work!
2 Likes
Iâm not entirely sure thatâs a good thing⌠Iâm not an engineer, just riddled with anxieties that usually get the better of me. But I appreciate the kind words.
Well considering the rest of what youâve said, I donât have much else to retort. Seems sound enough, and I imagine you know the INS and outs of your deck and enclosure better than I do.
My main concern remains still, the series connections and their soldering. I understand what youâve said about quick joints, high quality etc etc. However, when nickel is welded to the cell, and theyâre jutted up against one another, the heat transfer is still significant. Solder joints between the cells, when on the negative end, is still directly on the cells because of how the can is. Positive ends have more leeway it seems, but the butt of a can is essentially just the whole cell in terms of where to solder.
I donât have a direct fix for how to do it, just that I would personally look for another location to solder.
I also realize that there are countless packs out there that have similar construction and function fine. And that no one can really assess what kind of damage the specific soldering does. But itâs food for thought anyway.
Hereâs the butt of a P42A, just for illustration. Two of these jutted together, with a solder blob between, means the tinning of the nickel and/or the soldering of the wire puts heat right on those corners.
3 Likes
Well I respect you judgment more than most engineers thatâs for sure.
However Not to be the annoying one but I beg to differ just a little tiny bit, I agree it is not ideal.
But I guess it is the pvc wrapping that goes around the negative side that lifts the nickel out from the can just a little bit that makes the difference. Or at the very least it is not a solid connection between the nickel and the negative side until you get to the actual spot weld. And it is enough of a heat break to make a surprisingly big difference. It is a world of difference compared to soldering directly on top of a bare cell.
Feel free to test for your self and il eat my word and some more.
Anyways that is what I have found, if you are stupid enough to test that sort of thing in the first place.
1 Like