I did a little experimenting with a slot punch from aliexpress on 0.3 copper
It was achievable, but tedious and I doubt the punch would last very long. I’ll personally just be upping the power on the welder, the amount of work to punch all the holes would not be worth the effort IMO
ordered 0,3 copper, 0,4 copper, 0.1 stainless steel and I have all sorts of pure nickel/ nickel steel. etc.
My goal is to use 0.4 copper preferably without a sandwich with flux. if that doesn’t work I will use either nickel plated or stainless steel sandwich.
We shall see what the P60 can handle. Only thing left to order is some flux. found some 40grams for 28 dollars + free shipping.
But hopefully I can order for similarly priced @500amp.
I was planning to try to cut slots in the nickel for my next battery build using a dremel cutting wheel. I’ll report back on how that goes. I do also have a bit for it that calls itself a “milling” bit, but I think it’s more intended for plastic and wood than even soft metal.
It should be ok to charge.
you’ll want to watch how it discharges. and you’ll probably want to raise the vesc voltage cutoff
start and end to avoid dipping that cell too low. (it will likely drop faster than the others )
if it stays out of whack you’ll want to figure out why. broken weld to one of the pgroup? breaking balance wire?
That group discharges faster regularly. It’s not a new thing.
Has been at least a couple months.
So probably broken weld.
Or more likely one bad cell out of a group of 7?
Broken nickel and a bad cell have the same tell if the cell isn’t self discharging. Id crack it open and check the connection very thoroughly hoping for the easy fix. Much more likely on a flat pack vs a brick though. Otherwise definitely up the cutoff start if it’s not going on the bench soon
What should I set the cutoff to?
I know it depends on a few variables, such as settings and the load.
But I have a 3s1p holder which i can easily remove and replace the cells, and have cycled the snot out of some old reclaimed cells in it.
It has a hard cutoff at 2.8v, and every time the LVD kicks in, and I later measure the cells, it is almost always 3.476, 3.041, 3.468v, eerily similar to the screenshot above.
That weak cell which i have marked is always 3.04v when I check sometime after the lvd kicks in, but the other two cells are then 3.52, then 3.61, the 3.7. It happens surprisingly quickly, like the capacity loss is exponential rather than linear, at least on the old reclaimed early 2010’s LG cells I was hammering.
I guess it is ok as long as the p group rebounds to above 2.5v with load removed, but I might go a little more conservative with a higher hard cutoff voltage, as if it is a weak cell rather than a broken weld or compromised balance lead, that p group will likely start sagging harder faster as it seems to degrade at an accelerated rate, in my limited experience.
agreed.
that suprisingly quickly is even on new cells. mostly all the cells we use have a curve where at some point near the end it just starts dropping voltage fast.
I hypothesize that the cell’s capacity degradation rate at near the end of its usable lifespan follows a similar curve, in that once the capacity decline starts noticeably falling faster, it then goes into a nosedive, like the vapcell f63 and samsung 58e in Pajda’s cycle life testing.
This is a very well built pack, The pictures do somewhat speak for themselves.
(what i expected. )
.Hey guys I’m planning on building my first pack for my e-bike and I wanted to see if anyone had any tips
1.) I was planning to build a 20s4p configuration bc my motor is 72v. (Using rs50’s). I’ve seen a few people 21s builds and so I was thinking of going 21 for a little extra capacity and bc I ride a lot in super cold temps so want to avoid voltage sag as much as possible. Is this a good idea or should I go 20s?
also are there any online tools that help you design pack layouts?
2.) was planning to use a 150a Jikong bms from AliExpress for $80. Does anyone have experience with this one or any other recommendations?
3.) Curious what the best material to use for the strips is? Nickel copper sandwich? Pure copper? Looking to build the lowest resistance/ highest performing pack possible.
4.) is the p20 welder plenty or worth it to pay a little more for p30?
Thanks!
Bypass the BMS the for discharge. I’d go with a lower current JBD wired for charging only
Need a couple more details, are you using a vesc or a more standard ebike controller?
What’s the wattage output you want? Assuming 150a because that’s what bms you mentioned but could be other factors not mentioned
Fusion 360 is the 3d modeling software im using now but i don’t know if there is something more specific as i haven’t looked much.
If you aren’t planning on doing a lot of battery building the p20 would likely work but the p30 is just going to do more welds in less time before overheating as well as just a bit more material thickness. Both seem like good welders but i haven’t used either model specifically but have a p60. @SternWake has the p20 maybe can chime in about his experience
There is specific battery building software that I have seen @DIY500AMP.COM use in his videos.
I forget what it is called, but you can likely find it on his youtube channel, where he does one or two videos on how to use it.
search: testing the new https://electronicsworkshawp.com/ software
I don’t know if BMS are often or rarely bypassed for discharge on ebike like they usually are, on DIY esk8. The bms doing its job on discharge can injure a skater.
My P20b has welded copper as thick as 0.35mm, under 0.1mm stainless steel, using brazing paste/flux., at near max power. gear 945, double pulse
0.2mm copper under 0.1mm stainless steel with flux is about gear 385, without flux gear 550.
The AwithZ p20b will get hot doing a bunch of welds at these power levels, quickly.
So while it can do 0.3mm copper, one is going to have to give it a cool down break after a certain number of welds.
The cool down factor has not been a factor for me as I have only built smallish batteries, slowly, so far.
Oh yeah I guess I just assumed the bms needed to be rated for a certain discharge but my controller drawing/monitoring that so it really shouldn’t matter?
And thanks so much for the input on the welder. I certainly won’t be building anything very quickly as I’m just getting started and I’m not very bright honestly.
The e-bike is a cyc motor kit and honestly not sure if it’s vesc based or not. What even is the other option besides vesc is it sine wave? lol.
I do know it’s definitely rated to handle 72v and about 7000w so I’m guessing 150a is well overkill and like I said in another comment not really sure I need to manage the battery output that carefully.
So most e bike controllers are gonna be proprietary fw/hw and not open source like vesc. If it doesn’t advertise as vesc based it’s not likely. Vesc isn’t necessarily the best for bikes anyway but im not expert in the slightest.
Running bms discharge bypassed is a double edged sword : it has the advantage of not putting you in a bad position if the bms trips and killing eskate regenerative breaks or faulting and giving you street face, but it relies on the vesc or other means to protect the battery from over discharge. It doesn’t make sense to do that with an ebike in my personal opinion because you have mechanical brakes and a fault is only annoying and not dangerous. Also the protection of the battery by the controller is inferior imho and more likely to let you accidentally damage the battery.
you are spot on about 150a bms for a 100a controller though, you would be well into the safe zone and also not nuisance tripping with 50% headroom like you were planning
