Nice. I’m prob going to order a bulk of the 21700s in a few weeks for two boards. Maybe I’ll grab a set of the 30q for testing when I do. I’m all about budget building.
My plan is to build two packs, and be able to swap them on my short commuter board and then run them both in parallel on my long board for long rides.
Thats sounds good… I was thinking a similar idea but having like a modular top mount box that had everything and I’ll id have to do is swap the box and.plug in motor wires and go…
It’s be easier if you had your ESC in a separate box.
That way, all you’d be doing is dis/connecting the battery plug. And the esc/switch/phase wires, etc wont be touched at all. (May be more of a hassle depending how crammed it is inside)
This reply for me? Assuming so… I like the thought of just a swappable battery box only… But if everything goes its one unity for 2 boards… Sure I might have to run detection each.time I swap it… But not inconvenient… Nor would I be swapping daily between boards… I feel like one boaed would be the “special occasion” board…
I will have separate vesc. Only batteries will be swapped. And it will be quick. I’ll post a thread when it’s done, I haven’t seen it used in esk8 yet so it’s my secret for now
Any update with these cells?
I was just browsing ali and saw that they recently made new listing selling these as 4800mAh-5000mAh at price $4.65 when ordering 20pcs.
Yes sorry was.out of town last week…then promptly got on the “Mexican flu diet” the last few days lol… 2nd video almost done AND I got a metr module coming for my unity so hopefully in a week I’ll have some good telemetry data to add to ride testing… But aside from that I’m still getting 13-16mi with 25ish per cent left with max batt amps at 70 of 75…
And then?
Sorry I’m not being a lazy ass… One day after I thought the mexican flu was gone I got strep so no talky for 5 days then had to.leave again for a trip BUT I should have my metr module by wed I think so we’ll have telemetry by end of week… I did 1 voice over for the discharge data today and the 2nd discharge test vid files are on a diff drive so I couldnt add that to vid 2 which is waiting only on that…however… Real world test the last couple days… Commuting 9mi from Brooklyn to manhattan… Some stop n go traffic and normal city bs left me with 55% left… So I feel for commuting I could make a round trip with out charging…
iirc cells typically improve after a few cycles as well correct? When you put your board specs into the calculator what kind of mileage did it give?
good news! the metr pro came today and I installed in the unity… will try and squeeze a ride in tomorrow around central park or something if i can… bad news…no big local rides around my area until after the weekend =/
This will burn the pack sooner than later. Although it is a short piece of nickel. The heat generated cannot dissipated enough to the copper and cell. I highly recommend to improve your design. That piece of nickel can only handle max 8-10A. Please take the advice seriously.
Are there actually some real world test to confirm this?
Doesn’t the current run through the entire length of the nickel strip?
I don’t think it will fuse as the fusing current is still higher than what the OP will ever draw from that pack.
But higher resistance in the nickel only can lead to the fact that the pack gets out of balance more „fast“.
In fact it will work, but the design idea with the copper doesn’t make much sense if you don’t do the copper from cell to cell 
Yes, but 
let me pull out my Picasso skills 
Like in a water pipe which is slimmer on one part, the speed of the flow gets higher
Or let’s make it more electrical.
You charge with 10A.
Once you use 12Awg for the charge wires, the other time you use 22Awg.
Which wire gets warmer?
I checked this table against my own bench testing with a primary injection tool (up to 70A).
The table is spot on.
Of course we are talking about continuous current. While the loads during riding vary a lot.
However your battery pack should be designed for certain base load.
In the battery above you are playing with fire…
But do you think there is a difference in heat if you were to series the connection directly to the next p group as opposed to what is done here?
And I mean one connection then not two.
If it’s one it would be probably the same.
Hard to say from the picture, but the length from cell terminal to the copper x2 is probably the same length if you would go from one terminal directly to the other.
Of course he wouldn’t have the same flexibility but good to know it wouldn’t matter.
Basically it’d be fine if there is at least another nickel strip on each side connecting it all together.
it was kind of a hybrid idea off of the pcb’s people use and copper braid… in the past ive use silicone wire to make series connections but since these were 20700s and my particular case there is zero room for anything to be inbetween the pack or on the sides…if you watch the video you can see the tight fit… but with that said I could add an extra nickle strip up top on each pack to connect to the braid once Im done testing… this isnt a forever pack although its been running great so far… more to come on that later… but I already have a a bunch of 30qs coming for another build so eventually Ill probably pass this pack off once done testing and maybe add a lil wire upgrade for it.
like wise… i guess i could have made one long strip up top and used silicone or copper just like a pcb
What is the length of material used for each category? Current capabilities are relative to length. Longer strips= higher resistance= higher heat= lower current.
