Hi Everyone, Beginner Builder Here

[Burkeomatic]

Hi everyone. I am a little ways into battery building and I wanted to learn more about it so I signed up for an account here. I am mainly building RC battery packs (and wanting higher current than nickel strips can provide) but I also want to build an E-Bike as well.

I am interested in learning/doing the copper/nickel sandwich method. I quickly realized that my fleabay spot welder which struggles with .15mm nickel wouldn't be up to the task of welding copper.

I was guided to buying a kweld, and I have the basic kit, but it is a little intimidating and I was just looking for some help and moral support getting started, and you all seem like a good bunch, so I decided instead of lurking I would sign up.

I bought the kweld, but nothing else, and I was told that my Lipo packs should be enough in parallel.

This is roughly the advice I received.

Calibrate with one 3S/6000 Lipo and weld with 2 in-parallel. 2 will over-current the Kweld preventing it from calibrating. I don't use the K-cap add-on. Just use the basic kit. Use quality Lipos.

These are the lipo packs I have which are know to be of quality in testing on that side of the world.

Liperior 5000mAh 3S 35C 11.1V Lipo Battery With XT90 Plug for RC Plane / Heli / Car / Boat

The Liperior battery has been developed using the latest in Lipo technology. Our research and development team has been working tirelessly to come up with a superior pack for you that delivers on quality, power, durability, and price.

rcbattery.com

I did not purchase the K-cap or anything else, and I am trying to avoid having a big honking car battery laying around.

I just wanted to double check with you all, and see if I am headed the right direction. I don't want to blow anything up and I hate being electrocuted. My batteries thus far, while ugly, have been functional, and I built a 4s2p molicel p42 pack with triple nickel stacks that hold up in a pretty powerful twin motor I have that can pull 35-40amps, but the triple stack nickel is the worst! And after reading, that is really less than optimal, and I would like to learn how to do it the best and cleanest way I can. I'm not opposed to buying more gear if need be, but of course would prefer not to.

It looks like I'll need 50-70 Joules to pull this off from what I have read so far.

 

[amberwolf]

There's a number of ways to sandwich nickel and copper together, including one in JonesCG's most recent thread using copper tape that is then welded (or soldered? can't recall) to the nickel, greatly improving thermal and electrical conductivity.

Also remember that nearly all of the current flow in your system passes thru the actual welds*** so if those are not low-resistance and of sufficient cross-section, they represent a barrier to current flow and a source of heat and power loss.

(***since it is unlikely with most people's weld process that the interconnect strips will be tightly held against the cell end surface area evenly enough to make a low-resistance current path--it *could* be done but I don't know that anyone has created a welding fixture to do this; it would have to combine both probes into a fixture that presses over the entire surface area of the interconnect/cell interface evenly and firmly while performing all of the welds, which would require as many pairs of tips as you want welds at that spot, built into this fixture--you don't want to move the fixture on the interface area, to keep the results consistent...and you would need some method of measuring the pressure on the interface to ensure it is enough to force contact over the entire surface area of the interface zone...but I don't know how much difference it would make to the overall cell-interconnect resistance. Some, certainly...but how much? Is it worth doing?).


With any of the battery powered welders, it should help if you keep the leads from the supply battery to the welder as short (and fat) as possible; this reduces inductance and resistance. The more parallel cells in that battery the lower it's internal resistance, so the lower the voltage sag and the easier for it to absorb voltage spikes and noise from switching (which if not taken care of can overwhelm the FETs in the welder and blow them up).

There should be plenty of tips in the Kweld thread(s) for improving / upgrading it for specific uses, and troubleshooting for various problems in various uses, too. (I don't have one, so no direct experience with the specific setup, but have kept reading the new posts about it to keep up on it in case I ever do decide to build things requiring one).



Also, when building any pack for a specific use, remember that the "C rate" of any given cell is not the best performance you can get out of it--it is the *max* performance, and it will perform significantly less well at or near that rate than at lesser rates, with more voltage sag and more wasted power as heat, less capacity, etc. So for best performance, use cells as if their max c-rate is perhaps less than half of stated value.

There are sites like lygte-info.dk with testing data on quite a lot of brand-name cells. (unbranded cells are pretty much pointless to do this with, because you can't tell what cell you actually have and thus can't compare to the testing data--with all the counterfeits and recycled garbage cells out there of brand name cells even that is tough until you test them yourself to compare directly to the testing sites' data).

 

[Burkeomatic]

Ok, I have it put together, what would be the best way of doing 2 xt-90s in paralell? What gauge wire should I use and should I find crimp/butt connectors or try to solder it?