Thanks. No one else seems to have a name or source for it. Looks/feels very similar to silicone putty. I would question it's thermal transmittance ability if it was made of silicone. Not a good material for thermal transfer.

@@JamesBiggar An example product is Laird tflex HR400. Other similar ones on Digikey. Tacky silicone sheet, modified for good thermal conductivity - 1.6 W/mK vs 0.03 for air gap.

Nice. Not arguing an air gap is best, just questioning the logic behind using silicone. Without modification it's notorious for having very low thermal conductivity, lower even than air. Works great for apps that require high heat resistance. But if it's modified to be more conductive and it works, then it works. I might just buy some larger sheets and give it a try 👍 www.engineeringtoolbox.com/thermal-conductivity-d_429.html

Just be careful that not all the pad materials are as electrically insulating as others. I think most with ceramics/aluminium oxide are in the megaOhm ranges, but the even higher thermal conductivity sheets are also very electrically conductive.

@@JamesBiggar Thermal pads are very common in PC hardware. It is used for memory modules in graphicscards, ssd controllers etc. These are fused with thermal conductive materials. I doubt its very useful in your case though. The pad is very thick Almost 10 mm by the looks of it. And it is connected to a slab of aluminum without fins etc. I won't expect it to do much in this case, maybe a 4, 5 degree C temperature difference at best.

shades of tony- he never stops talking

Saved almost $2k though 👍

I agree, this process is called "annealing" and makes it more ductile in case it needs to be preformed. In any case, I was impressed by the competence and explanation of each phase, congratulations indeed!

You can also just disconnect the bms from both packs and run to a new 72v BMS, *cough* bac4000..m

Lol, super monotonous for sure. I can't stand tasks like that. But with the cost of new M50lt's being $8/each at the 18650 battery store vs paying $600 for three of these used modules that still have >98% capacity left, that works out to almost $2k in savings with extra parts to boot. So I put up with it lol

At the end when you explain the bms settings, there is a pile of cells unused on the table. Are these extras or duds? If extra why not add them?

Thanks!

Спасибо!

I've been using noise reduction in Audacity.

Np, thanks for watching!

Thanks for being a long time subscriber!

Yes, already did - important part of the planning process ;) They'll be fine. Thanks!

@@JamesBiggar I thought so too. I saw this video and went to battery hookup so quick. I’ve been sleeping on them since the stoped carrying the 32650 Lifepo cells. These modules are a great find!

I would like to know where you purchased the nickel roll from. Thanks again.

I'm no pro, just an ambitious enthusiast. But thank you!

Thanks!

Yes, as long as you size it accordingly.

@@JamesBiggar what you mean by sizing? Does that mean getting the correct awg number for the battery?

The nickel strips and BMS were only rated for 30-40A. I wouldn't count on the bus bars handling much more than that.

@@JamesBiggar Thanks. I've been eyeballing these packs for weeks hoping 2 of them would make a great cheap 20s moto pack without much work. Figured the BMS would have to go but I was unsure about the nickel. They are still an amazing deal but no shortage of work involved.

Yes. I was wrong about the grey material in the video, it could be a good candidate to bridge the gap between the cells and a heat exchanger, but I think it should be thinner for better heat transfer like another user pointed out. I used liquid cooling for the buggy motor too, but the motor was already built for it. I just had to add the pump and radiator. This is quite a bit more complicated, but I'm considering it.

Mentioned at the beginning of the video. You should watch it. It'll help you as much as it'll help me 👍

Links in the description of this video: kzfaq.info/get/bejne/hJl9q8dqmZeqhKs.html

I wouldn't be worried about it with 5P, especially if you're pulling current from the middle of the pack and the load is equal to or lower than the max continuous rating for the cells. The original modules that I used were suitable as they were because the load from the bots wouldn't have overloaded the top cells. They just would've degraded faster than the others over time. But the top cells would have been overloaded in my bike for sure, which necessitated the switch.

​@@JamesBiggar Thanks for the great information! I'm planning to build another 20S 12P pack for my main scooter (the 18s5p is for a friend), and ill be pulling a lot of current, hopefully it doesn't go too crazy then. I'll definitely consider the whole bus bar option for that. I'm using samsung 50S 21700 cells and pulling a max of 20A from each cell (they are rated for 25A but higher if cooled), for not a long duration.

​@@JamesBiggar I have a big concern. In the 20S 12P pack, i only have 10 cells horizontal space. This means I'll have to connect two 10s12p packs in series. For a 12P pack, when i connect it in series, if I just use a singe long copper plate it'll definitely cause some major issues as current would prioritize the few cells that are closest together. should I solder on some thick gauge wires to the + of the first 10s12p, and - of the second 10s12p, then join them together that way to mitigate this problem?

Already did in the video.

From what I understand, you are. A BMS should already monitor the battery correctly through the balance wires, not the negative power lead - that's just a transmission cable. I would assume that if an extra resistor was required to work properly, then they would have built it into the lead or included it separately...would they not? Kind of strange to be in business for so long selling a defective product that's become really popular otherwise, wouldn't it be?

*Never put resistors on BMS sense leads* - they have no effect on sense voltage if there's no load, and as they are used for balancing, they have to carry some current without dropping voltage, wasting heat and ruining balancing capability. Why would anyone do such a thing?

Thank you! It didn't make sense to me, but I'm no engineer.

$600

600 no no, aliexpress will cost you $130 for the one in this video, but there is even more affordable ones on there, just be careful when choosing

I wouldn't pull more than 40 or 50A through it. I haven't measured its exact thickness, but I know it's no more than 0.2-0.3mm between the cells, and the main bus bars 'might' be 1mm, and I don't think they're pure nickel or copper (or nickel plated copper). They're very stiff, not soft like nickel or copper. They could be nickel plated steel. You'll want to double check and measure their thickness because that will tell you what the pack can handle - the bus bars would've been built for the current the pack was built for. A 1mmx10mm pure copper strip can handle 49A. Bare in mind that the og modules are built so that the top cells in each series string are taking most if not all of the load vs spreading the load with the reversed bus bar configuration that I used, so they're at a lot more risk of being damaged from overcurrent and are going to degrade sooner than the rest. Each M50 LT cell can only handle 10A continuous and 20A peak, and being in a series connection together, the top cells will each bare the full load, not together. Ie: if you want to pull 50A from the pack, then you're probably going to pull close to 50A through each cell between each series connection at the top. With the arrangement that I used, that 50A load would be spread out to almost every cell in every string so that each only has to deliver a fraction of that current because it's forced to flow from top to bottom through all of the parallel connections to enter and exit the pack instead of just flowing across the top through the series connections. I would personally rearrange the bus bars so the pack lasts longer, but I think if you keep the load under 40-50A then they'll probably be fine as they are. The original bms can't handle anymore than that, so the conductors likely won't either, and I definitely wouldn't push the cells beyond that without making some changes. Use the ampacity chart here for reference: evolveforums.com/attachments/ampacity-20-10-20meter-20-20low-20temp-jpg.2903/

@@JamesBiggar awesome! Thanks much for your input!

Triangular? Not sure what the benefit of trying to pack a triangular battery into a rectangular robot would be. Triangular batteries fit better in triangular spaces. If you mean a staggered cell spacing in the battery to pack them tighter, then there's also a good reason that isn't usually done either. Packing cells too tight causes more heat buildup and lower performance efficiency. Some cell chemistry doesn't produce a lot of heat and others do. This is the type that does. The space gained by staggering cells is marginal compared to the gains in performance efficiency. The space available in the bike can accommodate a battery 3X this size, if I had a budget to build one. So space isn't an issue in this case.

@@JamesBiggar thanks for the reply, I was referring to the cell staggering. Hadn't thought about heat dissipation

@@m1llie_ It's called "hexagonal packing", though it can make triangles!

They didn't do a very good job. Can't pull my strips off like that after using the kWeld though ;) kzfaq.info/get/bejne/hJl9q8dqmZeqhKs.html

As they are in the video, each of the modules that I built measure 10" wide, 14.75" high (not including the terminal posts), and 3.25" deep.

@JamesBiggar thanks! for some weird reason. I just got notified of your answer, and it has been over a month weird yt shadow ban

Being able to output 250A doesn't mean they should be. The cells in each pack can't even handle that kind of continuous current without risking serious damage, especially with the factory bus bars oriented as they are. The M50lt's are only rated for 14A continuous, 20A peak. 14 * 11 cells in parallel = 154A. At 72V nom, that's only ~11 kW. My bike motor can do 12 kW continuous, hence part of the reason why I broke down 3 modules and paralleled 5 more cells together in each string so the larger pack can handle at least 220A. Refer to the cell spec sheet and bare in mind that max power is not sustained power. But this is just the cells. One of a few reasons why a battery built for low load demands needs to be rebuilt to handle high loads. If you consult a nickel strip ampacity chart and do a calculation on the cross-sectional area, the factory nickel can't handle that kind of current without serious risk of a fire, either. More heat also means less efficiency. Heat is energy, and the more energy wasted creating it, the less you have to play with. Electric powertrains should be kept

Use a towel or something between the iron and the plastic so the plastic doesn't melt and it should help, just be careful not to short it out, and make sure there's no water in the iron.

The tape and old nickel plates? Tape goes in the trash bin, nickel goes in my scrap metal bin. Everything else is reused or saved for future projects.

Takes a lot of heat to get that bus bar hot enough to tin and solder properly. More than my iron and patience can provide, anyway.

Golf cart for my grandfather. Too heavy, too little energy for high performance machines.

@@JamesBiggar I just got a superbeast pack and turned it into a 12v 96ah for a solar generator I'm building!

It's been parked since last November, no place to drive it in the winter other than my driveway. The battery has only been cycled a few times, but in the 6 months it's been sitting in the cold, there's only been a 0.9% voltage drop. So that's a good sign at least. We just bought a couple of fields this spring so I've got lots of ground to tear around on now when I get a chance to revisit the project. I want to upgrade the suspension to make it a bit stiffer and raise it a few more inches so I can still hit the trails and maybe some jumps without bottoming out, though. I was planning on taking it for rip next weekend if I get caught up on work. We'll see. But we're definitely not done with it. Expect to see more at some point.

Correction, my multimeter wasn't working properly. There's only been a 0.002% voltage drop over the winter. Had it out for a quick drive today. Still working fine :)

Oh? Do explain. I might learn something.

Nope. If I had put any more in parallel then I would have considered cell fusing. But the old packs didn't have them with 10P and they're LG cells, not generic no names from who knows where, so I'm not too concerned about it.

Besides the obvious fact that an Ant doesn't have built in forced air cooling to counter their fallacious current ratings, which would make them larger like this Daly? Because I don't trust companies that don't rate their gear properly because they want to give the impression it's more powerful than it actually is. A 250A Daly is rated for 250A continuous and 400A peak, as advertised. A 250A Ant bms is rated for 250A peak, not 250A continuous like it should be. Understanding how much surge power a device can handle is helpful for design, but in reality no device works under surge power constantly because conductors and components can't sustain surge power for very long. A few seconds at best. That's why electrical components are rated in continuous current because that would be the normal operating condition. Any device pushing peak for too long is doomed to fail, so rating devices on it without explaining the difference to the avg DIY'er is stupid at best, dishonest at worst. Early marketing didn't clarify that with Ant - you were left to guess until push back from consumers and regulators forced them to start posting continuous current rating. When a company is intentionally vague to artificially inflate ratings in the consumer's minds, that's a giant red flag for me. Especially when they're claiming peak values that are twice what their competition's is, from bms' that are smaller than their competition. It doesn't add up unless they're using space age material with different properties than standard conductors that only they have access to. Which I highly doubt. I don't want to deal with a company like that.

This is a decades old debate that no one will ever win. I try not to use solder when it's not required, especially on lugs. The connection as it is is superior to a solder joint in terms of conductivity, since solder itself has a higher resistance than copper and any used to connect a lug to the bar would increase the resistance in that connection just as it did when I soldered the copper wire to the nickel. Not a lot, but if it doesn't need to happen then there's no sense doing it imo. Those screws and epoxy kill two birds with one stone, the same as solder, without increasing resistance.

...also, a 1/8x3/4" copper bar would take a lot more heat to solder to properly than what my 100w iron can provide.

I have to disagree with the conductivity part. A mechanical connection when looked at under a microscope makes very small connections due to surface linearity not being perfect. Solder is not a perfect conductor, but because of the sheer volume and the way it alloys itself to the metals that it solders too creates a higher conductive connection than a mechanical connection. There's nothing stopping you from putting some solder paste underneath your lug and mechanically torquing it down the way you did and then applying a little bit of heat. I usually use 150 watt to 200 watt soldering iron for connections to thick copper like that with no issues.

I'd have to disagree on that one. If solder were superior, then every permanent electrical connection would be soldered. Torque is important for mechanical connections, whether we're talking about mashing a lug into a soft copper bar or mashing strands of wire into a solid mass with a crimp. Don't torque it properly, then it's not done properly. From my experience, I've always measured higher resistance from soldered joints vs mechanical. But again, it's a decades old debate that no one will win unless you want to provide some data from the results of said microscope analysis on a properly torqued connection vs solder. Perhaps my soldering skills aren't as good as I thought? It's possible. I'm easy to sway with hard facts if anyone has them ;) Each has pro's and cons and we could argue about anecdotes all night. I think mechanical is better, but drilling and threading like this takes time. Pick your poison my friend.

I hope it didn't sound like I was trying to debate you because that was not my intention. I'm a circuit board guy with 30 something odd years of experience in consumer and industrial electronics so soldering is my gig. Now that we started down this road I think at least for myself I have to figure out which is better. So I think I'm going to grab that micro ohm meter we have at work and do some testing. Cheers

V*AH=WH

....and a lot of $ in a time when we're all struggling and have bills to pay. These projects aren't cheap or easy to create, hence the timeline. Especially after buying two batteries for this one. If it were a manufacturer that was prototyping this bike from scratch as I am, then it would be a 2-3 year process, not 5 months. And they'd have millions of $ at their disposal to do it. Thanks for sticking around, but budget determines everything and I'm just one guy. I can only spend so much at a time ✌️

@@JamesBiggar iya terimakasih sudah merespon❤saya akan slalu menunggu setiap perkembanganya,benar itu pasti menghabiskan uang yang banyak juga dan proses yang panjang,semangat saya akan mengsupport dengan terus menonton video anda,terimakasih

Lol, future ev project. Something more laid back than a street bike.

Lol, ok...