Thanks for the update jehu, those cells are very specific, but sounds awesome. Remember guys, those are not vanilla batteries. They are made for very high burst (like a spot welder), or for a high load over a few minutes. They are not made for range or longevity. They are rated for a low amount of cycles (in the hundreds, compared to more than a thousand for a classic 18650). The main use I see for them, in a vehicle, is for a dragster. Otherwise you better go for the classic cylindrical 1-3C discharge rate cells (18650 or 21k cells). It will be cheaper and you'll have better energy density.
@@dennisyoung4631 You're welcome. I love my 14s batteries for my bike. Too bad you missed this wonderfull 12S battery, jehu had: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-ZW6FxARACfg.html Those cells are top notch.
@@piterex7 not at all. Most of jehu sell are pretty good for a power wall. Definitely not those. For a power wall, what you need is the highest number of cycles, everything else doesn't matter. The best at this, are LFP batteries.
Great to see you're rediscovering what we'd confirmed over at Endless-Sphere about a decade ago :) I build most of my Hi-Powered LiPo batteries this way, although I moved on to screws instead of solder. Might see that innovation in a couple of years eh? :D
Tabs on pouch cells are the main heat path and as mentioned below, solder has TONS of resistance. The cells weren't heating up but being heated by the resistors placed on their heatsinks.
Would it be better to spot weld the tabs directly to an exposed copper plane on the PCB? (I've never heard of spot welding directly to a PCB --I have no idea if that'd work 😄 )
There are resistors big enough to handle upwards of 100amp. I've seen other DIYers use multiple when they needed more. Now where you get them, I'm unsure, but you would need either a 300 amp, or 3 x 100 amp resistors. Once you get them, I've seen them add fans and holders. If I come across them again, I'll try and see if they remember where they got them.
Nice board. For applications where the battery needs to be exposed to vibrations, I would prefer to use a more robust connector for the balance port. Like a Dupont connector with locking for example.
Need to build your own load bank. The resistors would be a bit expensive, but you already have the wire to join them all up. Just need a couple large switches to add/remove banks (and _never_ swtich under load.. lol.).
Using some brush on flux would make it flow better. At least that's what I have seen when soldering on larger lugs or pcb traces requiring more heat. Sometimes the flux core just doesn't have enough
The only lithium Ion Cells that i have seen that you can buy that do what they are actually rated to do are "Overlander" batteries. There made here in the UK, so not cheap cells. The high end cells are usually rated around 30C max, but usually 20C, but they will actually do what they say on the packaging. I have used them for years on my model aircraft and you can really punish the crap out of them, I had a 3.2ah 20C rated pack that was old already, that I would pull 60 amps from repeatedly, probably 40 amps continuously, the battery would puff up a bit and get hot, but never ever did it fail, it was old when i bought it and i did a few hundred cycles with it in my E-flite Hurricane and it just kept on chugging along! I would love to see an E-bike build with Overlander batteries in it to see how much punch it would have as they are insane cells.
Hey Jehu, have you thought of making the tab hole wider so the tabs go through it already touching? I bet this would mean, that you need less solder and therefore less heat goes into the cells?
That soldering iron is great, but inadequate for this job. You may want to invest in a soldering station and some chisel tips that have more mass. That will allow you to put enough heat, fast enough to melt the solder without sending too much heat into the battery.
Hey, we used water heater heating elements. Take your DMM and go to Lowe's or whatever so you can figure out ac as listed or pure dc load. Hardly ever needed to submerge them in water. obw... don't use salt water :)
Sounds like a better idea than soldering. Although I can't remember what the tabs were plated with, which could be problematic for spot welding, but certainly worth a try.
Cool to see 20S packs. My electric unicycle runs on 84 volts and is considered bottom of the food chain with the newer wheels being 100 to 118 volts now.
hey man! superior work! is this pack capable to run mid drive motor like QS 5KW 165 ? i mean if you make an assembly of enough cells. sorry if the question is silly, i am absolute 0 in batteries and voltages etc... thanks in advance! and thanks for classy content!
Hook up metering equipment and get some ark welding electrodes (electrode positive) ground negative, and weld something and check the draw current 😉 👌🏻
I'm getting to a point I am going to build a battery bank for rv. Great videos on testing inverter and batteries. 13:00 Your so close to saying it. Make a 118 volt so when loaded it's nominal voltage, and directly wire heaters. Resistance type, same as what you have. If they have basic circuits, most heaters that are ac can also run DC in a pinch. Most electronics rectify the ac anyway to DC for control voltage and heaters are great for DC test too. Or get a load bank tester
Jump starting a diesel engine might be a good test. DC to DC step down converter. Say 72v to 24v. Just replace the existing battery with your battery and converter. Should take at least 200 amps to crank over. Disable fuel on the diesel motor so it cranks longer but not too long. High compression big block might work too. LOL
We should make a contraption box for charging this battery pack. Make a box that is ventilated with in/out fans on opposite side. Then please lets apply this pack practically in an e-bike, and show us the performance. Like - capacity test, temperature monitoring while using it, etc
I wonder if it was possible to overlap the tabs of the cells over one-another on a bare copper patch on the pcb and spot-weld them in place onto the pcb itself... I once had a battery pack in my hands were the tabs were laser-welded to thick copper bus-bars. Clean af but costly to make...
That is exactly the kind of process that these cells need to be made into a proper, well performing battery pack. These were made to be built into an actual battery pack in an enclosure using bus bars of the correct size. It also needs to be contained, protected and slightly compressed. These pouch cells are the most dangerous kind of lithium ion cells, and they should be in something that gives at least a little protection. You can't just slap these things together like this and expect these cells or this kind of slapped together pack to put out 3,000 watts continuous with only tiny solder connections. Jehu should know this. He's been doing this long enough and he wouldn't try to build any other 3,000 watt battery pack like this.
To Speed soldering use a higher wattage iron with larger tip. This allows you to get on and off the joint much faster. In this video his iron tip was way too small.
Also, you are trying to solder dissimilar metals together, the aluminum tab to the copper tab to the copper trace. This is building up a lot of resistance in those solder joints. As shown in the IR the solder points is where the heat was coming from and perhaps, due to the high resistance is the reason you had a flat battery in the pack, its solder point was higher resistance.
@@jehugarcia Thanks for the reply, my apologies for assuming one of the tabs was aluminum. I thought you had mentioned that you were "welding" copper tab extensions to the aluminum tabs in a previous video, then, in this video were trimming the tabs beyond the "weld" point. I am using some bag batteries that have the two different tabs, copper and aluminum... a puzzle...
I like it but would like a 24V and a 48V Version using series parallel connections. Older APC Smart UPS"s can be had for cheap and I believe 2 24V packs would fit into it and provide a very good solution to the Home Lab crowd! I currently use 2 18650 packs to do this and can keep my whole Lab running for about 4 hours. (Warning! if you do this increase the cooling of the UPS as it cannot run for an extended length of time without over heating!
I'm still believing that copper or nickel strips would help the current transfer and reduce the sagging, though it would need to be tested for this theory to be proven
I am going to guess that the 300 amps is a momentary discharge and continuous discharge is much lower. If you try it in your car, mount the battery outside on some sort of standoff arm and have a fire extinguisher handy. I think it may catch fire most likely from any wiring and perhaps from cell overheating if the wires last.
@@Blox117 700A at 1110V is 77,000W. What size DC motor do you need to move a VW bus? Sounds about right to me. You could pull 700A from a bunch of 100Ah cell banks
please try and move the car with the pack! i think with proper thermal management of each cell, this could be quite a potent battery pack for high performance EV's! the only big concern I can think of is thermal management, otherwise youll get some swelling going on
If anyone ever needed any visual proof that that soldering is INFERIOR to joining tabs, skip to @17:21 and look at the heat produced at every single solder point! haha
That's not what they are made for. Those high discharge battery are made for huge burst (50C), or for 10mn high load. Not for a 1h trip with a scooter. They are mostly used in RC models
as per the site specs those batteries are rated 12 amps continuous discharging ,and just temporarily bursts of up to 50C not to tolerate continuous 50C that's why they are over heating ? just saying .
None of these PCBs or the way they are being connected can handle the 200 and 300 amp loads at all. You aren't considering that a 1/4 inch x 1 inch piece of soldier will raise the resistance by a lot, and resistance equals heat. It isn't meant to handle that kind of current transfer. That's why the pack is getting hot, and that's why the voltage losses are so high across the pack. If those connections were made with a piece of solid copper able to handle those kinds of amp loads, and the whole battery tab was touching that copper and each other, it would make a massive difference in how the pack took the load and how much heat was generated and how much voltage drop there was across the pack. This is why education is so important and why actual electrical engineers should be designing high discharge battery packs, or at least someone who considers the most basic electrical principles, like ohms law, should be making high discharge battery packs. It's all good when people don't know what they are doing and they only need 10 or 20 amps from the battery pack. That's a lot less dangerous than trying to build a pack with cells that output 50c and hundreds of amps at almost a hundred volts DC. If it's not designed properly with very low resistance materials, resistance will be high and cause way too much heat, and it's gonna cause a lithium fire eventually. Please, at least consider ohms law and resistance, and how much copper is needed to safely conduct 300 amps of current and consider how much of the battery tabs are in contact with each other to carry that kind of current. Those cells were manufactured with those big tabs for a reason. It's so they can make a lot of contact with a big piece of copper and have minimum resistance.
No one, including J.G., thought those things could handle 300 amps. Your comment about needing a college degree to understand basic electrical physics is offensive, disgusting, and more than a bit elitist. Your own understanding is a bit questionable: The total size of the solder joint we see does not add to resistance as you imply - in fact, it lowers it. That's why everyone (including your vaunted 'engineers') adds solder to long traces that carry significant current - increasing the cross-sectional area decreases resistance. Yes, the resistivity of solder is almost 10x that of copper, but we're talking 18 mm (guessing..) wide tabs less than 3 mm from each other. The wide width and tiny distance between them _decreases_ the resistance of the joint. To calculate resistance, you take the electrical length in the direction of current flow (about 3 mm) and divide by the cross sectional area (18 wide x ...3 mm high?). So 3 mm / (18 * 3). Multiply that number by the resistivity of the material. 60/40 solder is about 15 x 10^-8, so each joint is about 8.3 nano-ohms. The extra solder around the tabs decreases that resistance a bit more (admittedly not much). If we guess at 0.3 mm thick, the tabs' cross sectional area is ten times less than the solder joint, making their resistance _per millimeter_ the same as the solder joint (without extra solder!), and they're a lot longer than 1/8" (3mm). (Yes, I chose 0.3 as a convenient number, but it's probably very close.)
@@chrisw1462 I didn't say you needed a degree to build a battery pack, I said you needed to at least consider resistance and ohms law. If you were to connecting together batteries and each one was expected to move 300 amps of current, then you would need 0 guage wire to carry that much current from battery to battery, right? Why is this different all of a sudden? A battery like this needed substantial bus bars made of copper or Aluminum of a certain cross sectional area to properly conduct this amount of power, plain and simple. I was never trying to be elitist. Im not an electrical engineer, but I am an electrician, and I'm smart enough to know that this was a poorly thought out, bad design. Jehu has been doing this for a long time, and I know he's smart enough to know ohms law. The fact is, he's on the internet selling these batteries to any idiot that wants to buy them, and they are still advertised as 300 amp capable cells. The way that he put these together was subpar to say the least, the proof is in the outcome of the experiment, and these pouch cells are the most likely kind of cells to start a fire. What do you think is more likely? That L.G. did a bad job manufacturing these cells, or that Jehu did a bad job of building a pack out of them with only little bits of solder? Come on man, the tabs weren't even touching. Using lead free solder does not lower the resistance when it's literally the only thing connecting these cells. Be honest with yourself and just admit this was a bad design and cells like these need to be connected together using big chunks of aluminum at the very least, or more appropriately, copper bus bars of a substantial size and thickness. Somebody is gonna buy these Cells and one of these crappy PCBs and attempt to replicate this, and there is a 50/50 chance they might burn down an entire building full of people or something. Like I said, he's selling these things on the internet and any idiot can buy these cells. If he was going to try to help people make something with these packs, He should had done a much more proper job of finding a good way to connect and compress and actually build these pouch cells into a proper battery pack that wasn't dangerous. This, in my opinion, and obviously other people's opinion, is really dangerous. Anyone who bought these cells and PCBs and copied this and doesn't really know what they are doing is creating a potential bomb. You ever see what happens when someone smacks one of these pouch cells with a bat or a hammer? It's a lot of fucking energy, and having 16 of them together is enough to do a lot of damage to everything from a huge building, to a car and the people in it, to a bike and the person riding it. You forget the ways that people will try to use these things. They will be used in ways you probably can't imagine, and I hope nobody gets hurt when they inevitably do something stupid with these cells. This is the exact reason that LG and Samsung won't sell lithium batteries to just anyone. Because they have the potential to be really dangerous in the hands of people that don't know what they are doing.
I want to make a 72volt battery what would be your best battery for the specifications i need it for? i need a pack that is not to bulky and maximum space of 14" W X 15" L X 4" D to sit flat on a scooter I am wanting to build and perhaps connect two of the same packs to make the capacity greater....so what batteries do you have that I can possibly use that can discharge fast for such a pretty quick strong motor ill be putting?.
I have an ebike with a 3kw cyclone. I have to turn the torque down to prevent tearing the bike to pieces. Starting my third attempt now with 60% torque and single speed. Seven speed sprockets cannot hold the power, and the teeth just break off. A Sturmey Archer 3 speed is OK in 2nd gear as that is direct drive. 3rd gear bit the dust. And I had to really work on the Sturmey Archer sprocket to make it drive the hub. Ordering parts now for a single speed hub version. I cannot even imagine what would happen with a 22kw motor.
Those packs are made for huge burst (50C). Or high rate 10mn discharge (we are talking dozen kW). If your unicycle doesn't need that type of load, go for a classic 18650 battery, better energy density, way less expensive
