JumpBro! An off-road semi-lightweight 20S(?)1P torque monster

Dinnye · October 1, 2024, 10:00am

This build focuses to use as many parts as possible from the remains of the first iteration of my raceboard - RaceBro.

Unfortunately as that originally race ready all purpose board turned more and more into a raceboard, I no longer felt like it belonged in the woods. This build makes sure that I can still have some fun in the woods.

This will also be a rolling build thread. At the moment I am still waiting for parts.

So, here’s the basics: haero bro stiff deck, pelican im2050, MakerX D100S which has a small external heatsink pointing downwards. The pelican is quite heavy and it’s going to stand quite empty but I already have the parts built out so I’ll just use them. At least the batteries between the two boards will be compatible.

For now I am going to use the BN310 precision channel trucks that also came off race bro

They are pretty nice for traditional channel trucks but are still traditional channel trucks. I am heavily considering changing these out for a set of Tito’s sometime during spring.
I recently stood on my old Matrix IIs and after riding Tito’s for what feels like an eternity (4 months), I could barely turn on them. So we will see how these go. But no money for a second set of Tito’s until spring.

Now, something exciting.

Drives you’ve never seen before. The all mighty @Boardnamics 7:1 ratio straight cut steel drives, of which only 3 sets exist. The other two sets are on @Esk8Hills 4wd monster. They are fairly large and heavy but these will definitely make sure that I won’t be lacking torque.

The motors are going to be Reachers of course, I went with 6375 for some weight savings, 170KV in the v6 version. They should arrive in approximately 3 weeks. Hopefully it will be a nice balance of power and weight. If they overheat too quickly I can grab the 6385s from the front of the raceboard and exchange them with those. I do plan to push a ton of amps and this is 2WD so will see how it goes.

Now let’s talk battery stuff!
This is going to be a high voltage build, to spin up those 7:1 drives properly. However, for weight saving reasons, only 1P. But hold up. It doesn’t mean that it has to be weak does it?

For now I plan to test out the shiny new BAK 45D cells, which offer a tabless construction and (according to it’s datasheet) 3.4 mOhms of resistance. They are supposed to be able to peak 126 amps. My plan if all goes well is to peak 100A. And that’s not the whole story. They are cheeeeaaaaaap. Less than 4€ a cell. Some might say too good to be true. But I am happy to spend 80€ to find out. If the experiment doesn’t go that well I can just a build a second pack using JP40s or EVE 40PLs.

That is however asking a lot from a single P group. I do not recommend others to follow, this is going to be a potentially quite dangerous experiment! Especially during the first few cycles of testing, I will need to constantly monitor the cell temperatures, determine voltage sag, adjust battery current accordingly.

Also, constructing a 1P battery pack with beefy enough connections that it can carry that current without the connection themselves heating up too much is no small feat!

Here’s my plan so far assuming I want to do 20S.
I plan to 3D print cell holders, fold 30x30x0.2 nickel (or if I manage to source some then nickel plated copper) onto the sides, and run an extra wire down the side for some increased current carrying capability, to reduce the current which is going straight between the cells on the nickel and spread it out a bit more. Will it work well enough for this current level? Well that’s part of the testing I suppose. I do have a 100A battery pack tester at work which I might be able to use…

Also. This is arguably not the smartest idea, but I plan to completely skip having a BMS inside the board, but have the balance harness soldered up and use it only for charging.

I am however considering installing one of my spare ESP32s and NTCs onto the pack in order to be able to get real time temperature measurements over Bluetooth. Not sure on this one just yet. But it would definitely be a good idea for gathering data during testing.

At the moment I need to decide if I want to do 20S or 21S. The reason for 20S would be to avoid crazy reverse voltage sag if I pull a lot of braking current which could get me near the limit of the 100V hardware. As these are unknown cells which I plan to push really hard, this would be the smarter call.

But I am lazy. So the reason for 21S would be that I don’t need to adjust the charger’s voltage output, as I plan to share the same adjustable charger with RaceBoi which is has a 21S4P pack, and I need a screwdriver to adjust charger voltage. I am also considering rebuilding that board’s pack to a 20S4P with these cells, but if I found out I wouldn’t have to worry about reverse voltage sag at 21S I might go 21S again on the raceboard, where after a rebuilt pack I would like to be able to push 18kW+. Also 21S is 5% more range. Decisions, decisions.

Regarding the rest, mostly there’s just tires hubs remote and bindings. Gonna be F5 bindings + heelstraps as usual, GT2E remote with a 3D printed case probably, some lightweight plastic hubs, probably plain old mbs rockstar ii, and for tires I am considering 8" trampa mudpluggers and the cheapo clever 8" offroad tires from ali.

Timeframe for completion is roughly a month.

Dinnye · October 1, 2024, 10:01am

reserved

Dinnye · October 1, 2024, 10:16am

Venom121212 · October 1, 2024, 10:26am

Following this one for sure. You’re mad, and I’m here for that.

Dinnye · October 1, 2024, 4:23pm

Got the rolling chassis put together

Ride height is low with the BN channels. I need to decide if I want to keep it this way and cut into the deck to not get hanger bite so early, or use risers.

Weight: it’s missing 2x6375s (950g each) and the probably 20s1p battery, which is going to be about 1.6kg. And bindings / heelstraps, no clue on the weight of those… Maybe 700g total?

So weight is going to be roughly 11.30kg + 4.2kg = 15.5kg. While it’s not the lightest thing in existence it’s definitely not bad. I am pretty happy with that.

It’s going to be very rear heavy though, I guess I am gonna find out how much that matters.

Egtscs · October 1, 2024, 8:51pm

My Boardy Mcboardface is very similar to your 2wd iteration of the racebro in terms of weight distribution (84 cell 21700 battery, pelican 1200, M1AT, 6485s, RTKP) so I think I can speak a bit to how the rear heavy weight distribution effects behavior in the air.

If you’re worried about how it’ll feel off-road doing jumps with substantially more weight in the rear, I don’t think it matters that much, in the air the rear end will generally just follow wherever you direct the front, IME it’s much more important to have the front end lightweight than anything else related to distribution. If you can use the front end to direct the board where you want it to go your momentum will generally work the rear end out to go where you want it to go.
There’s a lot of momentum behind your body when you’re doing jumps and it feels to me (my bodyweight fluctuates pretty heavily between about 75-85kg) like your momentum has enough influence over the board that as long as you don’t go off the jump funny everything else kinda just works itself out

Super keen to see how this build turns out!!!

Dinnye · October 1, 2024, 9:37pm

Thanks! At the moment I don’t know which way I will take the build exactly, there’s a few things I want to try.

I definitely want to give trick jumping a shot. Well at least a short lived shot. For that I would need symmetrical angles, so that if I do a 180 it doesn’t feel as bad riding the other way. And here maybe being super rear heavy wouldn’t feel as natural. For the other use cases I am sure it’s fine though!

I also have a micro motocross track nearby, which I would absolutely love to try out with this board if they let me on

That however requires a bit higher speeds, where I might not be able to get away with symmetrical angles. Especially that I am no longer really comfortable riding channel trucks.

And lastly I want to ride some technical mountainbike trails. Here both symmetrical and split angles work but I absolutely need to have a nimble setup that can turn super tight.

I might be able to achieve the stability I need with symmetrical or nearly symmetrical angles on Tito Dualities with WFBs once I’ll have the budget for a second set. I am actually considering 2x 0 degree baseplates for that instead of a 0 + a 15.

Dinnye · October 21, 2024, 9:07am

Just got the motors, 6375 170KV

What’s left: fixing my gt2e, assembling the gear drive, hook up phase and sensor wires, borrow some bindings from my friends broken board. Short term I’ve got a 12S3P 40T pack that I need to install a BMS on just to try out the board, but probably top speed is gonna be awful. But also I am so unfamiliar to channels now that maybe it doesn’t matter

I’ll be ordering the cells for the 20S1P middle of next week, but I also decided to try the copper - nickel sandwich so I have to order material for that as well. I want to minimize the internal resistance of this battery pack build as much as possible. Copper is 4x as conductive as nickel, so the 0.2 nickel equals about 0.05 copper. I’m considering 0.15 or 0.2 copper. Maybe even 0.3 depending on what I manage to get my hand on and what I manage to reliably weld. More on this when I have everything in hand.

will_manners · October 21, 2024, 9:37am

Excited to see this completed! Especially the high amp small battery.

How are you planning on welding copper tabs?

Jaws · October 21, 2024, 10:08am

Surprisingly, rear heavy is easier to jump than balanced weight/4wd is. Light front end is important for getting some pop!

rich · October 21, 2024, 4:38pm

Yes! We need more boards leaving the ground

Having the most weight on the rear for jumping is no problem at all. At least not as long as you don’t lean back while jumping

From my experience a carvy setup is awful and very sketchy on mountainbike trails. A stiff setup is good for stability but bad for riding curves/berms so the best is something in between but better a bit on the stiffer side.

SternWake · October 21, 2024, 5:03pm

My first Battery, I used a shitty welder and a low quality(Zee) 5.2 ah 3s weld battery using copper nickel sandwich method

The nickel cap works better with nickel plated steel rather than pure nickel.

I was able to get solid welds using a 70 to 85 ms pulse on 0.1mm copper, and 0.1mm nickel plated steel, but had issues with 0.14mm copper, and issues with 0.15 pure nickel even with the ’ infinite slot’ sandwich method.

For reference a 35-40ms pulse was good for 0.15mm pure nickel by itself.

Perhaps a second weld battery in parallel would have allowed solid repeatability with thicker copper and capped nickel plated steel, or pure nickel.

I’m sure an experienced builder with a better welder can pull it off with thicker copper.

Be sure to clean copper well, and use gloves.

I didn’t and dont really need the lower resistance with my low power build, but am mentally allergic to electrical resistance.

I’ve got about 2k miles/3k KM’s on the battery, so, so far so good.

Dinnye · October 21, 2024, 7:32pm

Copper in itself is impossible to weld with the kweld - to weld that direct you’d need an industrial 30k$ welder. The problem is the low resistance and high thermal conductivity.

However with the “copper - nickel sandwich” technique, even the kweld can weld copper. It works by placing a small piece of nickel, or preferably nickel plated steel over the weld area. That creates a lot of extra resistance compared to welding directly on copper, which means once the current flows it creates a lot of heat in a short amount of time to make a spot weld. The best is to the have two pieces of nickel plated steel as separate pieces, in order to force more of the welding current through the material under it, kinda like in this picture.

Since I shared this early far from finished rendering I might as well talk about it a bit now.

I am actually considering building this pack in 3d printed <99 Wh modules in order to experiment with what’s possible for a flight safe battery, as I might fly to esk8con next year if my situation allows. I don’t need to fly with this pack, but to make a bigger flight safe pack a reality it would be easier to start small.

For this pack I have a good amount of extra space to play with, so seeing how compact these modules can be done and if they still allow the performance I want would be a good thing to find out now, without having to go all out on trying to shave off every single mm otherwise I risk that it wouldn’t fit the other board. For this I am considering a 4S1P configuration of 5S1P modules, as a first iteration of a flight safe pack.

But for the raceboards flight pack maybe 6 cell modules are also worth considering, as 6 cells is the maximum per module to stay under the 99 Wh limit, even though they might mean that I need to get more creative with shapes.

Dinnye · October 21, 2024, 7:54pm

Honestly my setup philosphy has always been “loose = good”. And after riding so much on duality trucks, I no longer believe that tight can ever be good.

These trucks just give me all the stability I want, while I can run them wayyy looser than anything else.

This is how loose I run my racer’s front trucks at the moment. Hitting insane speeds on, putting down 0.8-0.9G of acceleration and braking. It’s just stable. I don’t even need to think about weight distribution over the trucks from a stability perspective, only from a traction perspective. I liked a bit tighter setup than this while I ran that, but to have a healthy amount of preload going tighter requires 100A WFB and I haven’t talked to Riptide yet about those.

Back in my channel truck days I also prefered loose setups, even for mountainbike trails, but loose channels and loose dualities on WFB are very far apart.

One thing I can say though, back in the summer I did a bit of offroading on my 2wd iteration of the racer on dualities, and I didn’t mind the loose trucks even for hitting some ramps, although by your standards those ramps were probably fairly small maybe like 45-50cm height with no downramp, but challenging enough for me especially with a 25kg board.

Honestly I am very scared of riding these channel trucks again after exclusively riding dualities for nearly half an year now. But the Duality swap won’t happen on this board until esk8con.

Dinnye · October 21, 2024, 7:56pm

Does your welder report welding current? Would be interesting to see just for reference.

Is this important to avoid rusting over time or for more consistent weld results?

tuckjohn · October 21, 2024, 8:26pm

+1

I regularly ride bindingless on DKP, channels, dualities, and 3-links

Loose trucks saves lives.

Arzamenable · October 21, 2024, 8:47pm

Hey Yoo-hoo, do you have an iPhone? I have and seek thermal camera that has seen better days.

Use a big lunch box, use a diy macro lense and a the camera, extend the lighting cable out of box have phone external and record under actual load, in focused, rigid setup.

Make you cell holders expose areas of interest on the cell cans by design, if you are gonna. Run it hard, print it in PPS. If you can melt that, you melted the Polypropylene pelican. I got CF polypropylene from the ultrafuse line of BASF filaments. It’s gotta be good for the same as the pelican case.

Happy to print you designs and loan the cam. If are you jazzed about the idea, maybe you can use the better one if you promise not to break it, I snapped the first one. Yes, I am on my third thermal cam.

CHADFARTHOUSE · October 21, 2024, 8:55pm

I have a question when making the copper and nickel sandwich. Do you premake the copper and nickel sandwich before welding to a group of cells?
Or do you sort of place the copper and nickel loosely on top at and weld it all in one go?

SternWake · October 21, 2024, 11:21pm

The copper I bought had some sort of oily finish on it to keep it from oxidizing and appearing dull.

When I cleaned it off, and then cut the copper, without gloves, then got sidetracked, when I came back many days later, i could see my fingerprints and some oxidation, which would not simply wipe off.

I wound up having to sand it with 2500 grit to remove the fingerprints and make it shiny.

I assume any oxidation between copper and the battery during welding would mess up the weld or increase the resistance, or reduce the consistency, and whatever they used to keep it from oxidizing at the factory, would compromise the weld too.

My welder is a pretty pathetic inexpensive device, but it works.

I’ve no idea how much energy it delivered to welding pens from battery.

I could elaborate on it, but dont wanna further derail your thread.

I have used 14-17ms pulses to weld the nickel plated steel to the copper, to prevent shifting of the layers for the initial sandwich weld to the battery, but later on decided the strong magnet holding both, and more precisely lowering the welding pens, was one less step required.

Dinnye · October 22, 2024, 1:24am

Mostly final. Needs a small case around each module, and each module would be connected by connectors on the top. All cables come up to the top, nothing runs on the bottom.

There’s zero spare mms length wise. Maybe the case even won’t fit. Would need to be tested.

118mm long / 25.5mm wide per modul. Width might need to be increased so that the soldered on terminal cables would also fit

If it fits length wise (especially after the case…) it should hopefully make 20S3P a possibility by wiring up a 4S3P config of the modules. 20S1P for this board is also possible of course. Along with 2P

No way to fit a modular 4P though.