we disagree too much man. i get your vibe and appreciate the little hostility as its a good flavor when its in moderation and light, but i feel like its too much internet and not enough riding and skating sometimes. bet if we had a little skate sesh and a beverage of your preference it would take the whole thing down a notch and feel good.
if your ever skating Seattle hmu and we’ll see if our particular breed of crazy is compatible. trying to pick on people in the fun friendly “just giving you shit but with a smile” kinda way doesn’t seem to come across to you dude, and some push on the pull door to get your goat which can get old (see ryan ). i bet if we were just talking shop at some nice charge spot it would feel different and get the vibe back.
Last two times I was here it was because others were talking about me with vague insults. I don’t get with that vibe
Reality is data sheets often state u can revive cells. The amount of freakout I got for stating they could be revived was huge. I give what I get… actually not even and had a huge amount of bullshit thrown at me then. Know-it-alls
Ok check my work as im getting to the point of pita to roll back and correct my work, shooting for 100a continuous and burst to 200a pack and my bottle neck is in the p group i had to make in a shitty layout.
18s6p p45b with major space constraints because i want to use what i have for the enclosure.
Solder isn’t complete just in place so I don’t have to pump too much heat. Just doing a little then letting it cool and going back.
The 1, 7, and 13 p groups are the elongated non standard ones. The outside most nickel is 35mm wide .2 and is soldered to the long braid under the pack with #10awg equivalent tinned copper. The nickel here should be good for about 60a easy but 100a in bursts so should be fine to move the current from those two cells.
The braid attaches to the lower wings of the center pair of cells (parallel) which is 30mm .2 and is now carrying the current of all 6 cells and stacked with the 35mm .2 that is on the top of the pack that butterflies open.
Because of how i needed to run the braid underneath i didn’t make the nickel at the center connection for the outside two pairs go both ways like i did for the center two cells and i think i might need to derate it as a result. I just wanted more separation between potentials and this is my compromise.
That bottleneck where 4cells are moving current on the 30mm .2 i think should be considered 60a max continuous and only burst for 100-120a and so I should consider this a 90a continuous and 180a burst build.
Just for next time, you should consider welding copper. The kweld does 0.1 easy, I think it also does 0.15 on a semi decent power source. You need a small strip of 0.1 steel over the copper to create the heat during the welding.
Even just 0.1 copper would be equivalent to 0.4 nickel, so for packs where you have a small cross section for large current, copper makes a lot of sense. Definitely a bit more work than just doing nickel but some situations justify it, and it’s a cool technique to have in your arsenal. It’s my go to method for tabless cells.
That said, during real world use I honestly doubt you’ll be overheating the nickel, unless you are racing. It’s probably more so that there would be a measurable voltage drop coming from that nickel reducing efficiency and output power a bit. Since it’s not like every connection in the pack is made this way, it’s probably alright.
I like that the chart gives us a good idea of what’s good, but where did the data come from?
It’s be pretty cool to see some real data/testing done on the pack configurations that have been pretty standard for some time now.
Taking current ratings for 12mm strip and extrapolating from that just doesnt seem super accurate and I feel like this nickel can work harder than we give it credit for if the rest of the conductors are up to scratch
ya, totally. I was just saying to Gus - I’m building a race board but i don’t race?!? whatever its fun.
just looking to get some input as i was waiting for everything to cool back down after hitting the little braid jumpers and figured Id bounce it around here for an opinion or two. if I was using this type of connection throughout the pack id defiantly get a little copper to play with.
ya, and overbuilt is something I’m guilty of for sure. but its fun so fuck it. when its not fun (like the weather is above freezing and i can go ride) my attitude will quickly change. fuck its cold out today, it was only 20freedom out and all the flag poles have little bits of tongue stuck to them… disgusting
It’s .2mm nickel plated copper folded over .2mm nickel.
I posted it as it seems the easiest way to mix the two materials and not have the solder n copper get in the way for later welding.
How many joules are people using to weld copper with the kweld? My attempts with nickel on top failed and I was already doing what I think are a lot of jules (90)
THIS REPORT IS BASED ON MY TEST RESULTS, RESEARCH, AND DISCUSSIONS WITH PEOPLE IN THE BATTERY INDUSTRY. ALL STATEMENTS HERE ARE JUST MY PERSONAL OPINIONS AND ARE SHARED ONLY FOR INFORMATIONAL PURPOSES.
The full AM04 test report (two documents) is done and posted to my Patreon wall so I can easily keep them updated in one place if anything changes. You can access it here: https://www.patreon.com/posts/bench-test-am04-122489376
The report is huge so here’s a quick overview…
The AM04 is a rewrap of the Ampace JP40.
The AM04 performs a touch worse than the other JP40’s but you probably won’t notice the difference.
There are serious AM04 performance inconsistency issues at high power levels.
There are multiple AM04 datasheet errors and ratings that can cause rapid cell degradation and even cell failure. This affects user safety.
An uncalibrated prototype testing system and inconsistent, slow, and inaccurate temperature sensors were used to set some of the AM04’s ratings.
A video on Austin’s web site falsely claims that it shows AM04 cells being made.
The AM04 has none of the required safety testing and regulatory documents. Pack assemblers, make sure your insurance still covers you if using this cell!
Austin was given a chance to preview the test report and respond with a statement that I would include in the report. I have never offered this to anyone before. Unfortunately, his statement was just an appalling mix of bad science and a sales brochure. Such a wasted opportunity to address all the issues I found.
For more information, please read both documents as the second one has a lot of additional test results.
Damn, that was a great read!!!
Thanks for taking the time to test these properly!
I’ve got a few packs built with the am04 and so far they have been great.
I’m not pulling big amps, so I’m not noticing any issues with the cells yet.
Once again mooch, thank you
Long story short would a CT scan of nickel welds onto a cell be valuable? Maybe to tell if there is contact on the whole area vs just the welds or to see if there’s any effect on the negative side.
Or any arguments that can be settled with a multimillion machine?
On the topic of spot welding and how much of the strip outside the welds, contacts the battery, and optimization there of, I wonder what could be done.
Glitter sells a copper welding flux, that one guy on the ES forums says makes a huge difference in copper to copper welds.
I get the impression it is like a flux and solder, perhaps fills those voids, reducing resistance, allowing less power to achieve solid welds, frying less internal electrolyte.
I’m all for lower resistance where ever possible in path from battery to motor.
If anything going back is more likely to eventually get me a job. I’ll miss a day of school too! When I was there months ago I wanted to scan a tab and a tabless 21700 bc they are very keen for new battery tech and something interesting which they can have a a demo unlike confidential client’s stuff.
They pretty much only scan metal and were doing a big EV battery while I was there today. It’s a fancy version of a CT scanner the new one they were unveiling is the size of an apartment and costs millions.
Posting in this thread cuz it’s active. But I’m open to creating a separate thread if that’s preferred.
Embarking on building my first proper🔸️ esk8 battery!!
For a Lacroix Barrel whose stock 12s4p pcb-based pack is 100% ded following a winter of neglect. RIP
My wish is a bigger and betterer battery. Hoping you brilliant kooks can offer guidance along the way.
clarification: I’ve built several simple packs before for ebike, scooter, and esk8 use. Simple brick packs. I’m viewing this one with a higher level of scrutiny and respect cuz it must be flex tolerant. “A proper esk8 pack.”
Do you know if you have room for that 12s7p in the enclosure? Or are you planning to extend the enclosure with a thick gasket like @Pecos did recently?
For a single stack battery layout, 0.2mm x 30mm nickel can pretty much do everything you’ll need. Cut strip to the length of the p-group, round the corners, prep it with some solder pools, spot weld, fold it over, solder your series connection wires. Simple as that.
Buy several extra P42A’s to test on. You’ll need to dial in you settings/techniques for the specific cells your using, and pull tests are still the best way to know if you’re doing it right.
Fish paper and brain power are typically all you need. But some edge cases can be helped by FR4 sheets or 3d printed separators or whatever.
I strongly recommend a car battery on a float charger instead of lipos. But if you already have l lipos that work well for you, then I’m sure they’re fine.
Have you read my detailed rundown of my welding setup/process? It was in this thread.
). i bet if we were just talking shop at some nice charge spot it would feel different and get the vibe back.
following a winter of neglect. RIP 
clarification: I’ve built several simple packs before for ebike, scooter, and esk8 use. Simple brick packs. I’m viewing this one with a higher level of scrutiny and respect cuz it must be flex tolerant. “A proper esk8 pack.” 