After this video i made the board pickup quicker, or to try to level faster. Currently speed capped at 20kms amps set to 80%. Kickback activates at 95% speed.
4 Likes
That looks really legit dude. Good work.
1 Like
thats interesting! I wish to have access to those but totally zero clue on how its going to connect xD..
while my goal now is to fully make use of the sensors and circuits from the V1 hoverboard (1 mobo 2 sensor type), so that anyone wants to dabble into the conversion will be easy. (i’ll write up a full tutorial of the headaches I went through after i’ve succeeded)
so far I have successfully written a rudimentary-onewheel-like code based off EFeru’s code. Somehow i hit all the high difficulties microcontroller models, so was quite a headache to debug everything.
anyway, anyone interested can take a look at my github, same username, i’ve forked a version and modified the code to fit onewheel, can look into the dev_onewheel_variant branch for more info.
Current progress is building the physical machine, i’ve done the motor in the wheel, right now my 3d printer is printing all the parts to be assembled (templates & frame components)
Some of them took inspiration from others, but most of them are designed from the ground up. Its taking a bit long due to alot of prototyping and long printing time due to it being so big.
2 Likes
Some updates:
Going well:
- General One Wheel shape achieved, whats done:
- Bumpers (with handle and wheels attachment)
- Circuit Box (Not waterproof)
- Most of the wiring harness holes & channels
- Both motor going to the same direction
- Most of the drilling and hardware outfitted
Not going so well:
- Battery Bump Issue (No solution yet, i am not confident enough to change its shape)
- Footpad not started yet
- Wheel tube issue, when i filled it with some air, its causing a bump on one side (could be also attributed by not able to find the correct sizing tube for the wheel)
- Bench trial using the firmware that I’ve uploaded to Github isn’t that smooth and there are still issues to finetune
- Some of the 3d designs not as easily maintainable and not waterproof (likely wanna redesign into a single enclosure as compared to parts fixed to the aluminium frame)
anyway, anyone have suggestions on how to proceed forward, please let me know 
1 Like
Definitely start with the enclosure. If you can get the enclosure sorted so you have all the constraints the battery can be reformatted to fit what you have and take some of the guesswork out. Post some pics and id love to armchair quarter back some ideas
Honestly, version one is already sorted, however battery reconfiguration is the problem xD..
I am constantly circumventing the part where I need to reconfigure the battery..
Here is the look at the problem
Potentially planning to add height to the frame and bolt it with the footpad..
I don’t know yet, anyway, if i can figure out keeping the configuration, i can hook up dual batteries for longer range with a parallel connector (probably won’t work charging at the same time)
Ahh ok. Ive got one of these pack in my recycling. Ill pull it and see what a re configuration would take. Can you give me exact dimensions of the space you have with or without the the extension and ill see what it would take so you don’t have to open yours ups.
Edit :10s2p once you have dimensions ill crack my junk on open and take lots of pics.
What’s your battery experience?
1 Like
I’ve quickly draw out the negative space of the battery compartment,
Personally I have 0 experience in reconstruction / creation of battery pack (although I’ve gotten the welder and nickel strips, but have yet to dabble further into them.
If we are to proceed to reconfigure it, I would say the top 4 batteries are the ones to be taken apart,
its connected to four tabs on the BMS, is it safe for me to just pry any of the tabs off the 4 batteries on top?
- Do i need to start positive or negative or it doesn’t matter?
- Are metal tools okay, i just make sure don’t touch anything else when in contact with one of the tabs right?
- After its off, can i reattach them with longer strips of nickel to its original spot with plenty of heat shrink on the nickel strip as the “shielding”? just like a copper wire has its plastic case
1 Like
Perfect, ya there may be a way to just move that section. When i finish work ill take a look at the pack and make a little step by step. It shouldn’t be too difficult as a first project but i dont want to say exactly until I’ve confirmed what’s going on inside. Many of these little packs were put together very poorly from the factory and you will definitely want to get some neutral cure silicone and or some b7000 adhesive to add some structure to the new configuration
It’s just waiting for me on the workbench to play with 
1 Like
Thank you! I do have fiberglass tape at the ready, as the separation and structural hold.
Preferably if I can use some fishpaper as separation, but I haven’t committed to buying new cells and building a bigger pack to fit the negative space as designed
All my B-7000 glue somehow dried up in their original tube, I am not sure if I got a bad batch or poor storage conditions. Worst case I do have hotglue 
, not ideal i would assume
However the original goal was to do as little as possible to original hardware of the hoverboard, would you say reconfiguring the hoverboard battery is a doable step for new comers looking into a quick one wheel build?
Ya totally, biggest issue is to make sure everything gets locked down after arranging into the new configuration so it’s more vibration resistant and any shortfalls on the nickel is remedied.
What welder setup did you get? Ill try to keep the rework within its abilities
About metal tools - make sure to protect the connections that are around where you are working and all should be well - i use up a lot of electrical and masking tape as im working just to cover and prevent shorts or accidents as i go. Dont rush and protect as you go and you can do it safely no problem.
1 Like
My setup currently is the handheld version, i tried joining up to 0.2 thick nickel strip, seems strong enough, haven’t yet to try on an actual battery
Below is the exact shape and make
So workspace protection is just as important as the tools used to work on the project, just a question for the worst case scenario, because I stay in a condo, if a fire does happen, would dumping the battery on a big vase filled with dirt be enough to contain the heat and fire until it dies down?
Anyway, my backup plan, really gonna be wasting a lot of plastics
“18650 high quality cell”
I recently tore open a fully functional premade Chinese 7s2P pack to inspect its build quality and rebuild it.
I was pleasantly surprised to find DMEGC 20P cells which are a decent low capacity power cell, and they were all within 25mV of each other, and I could not see ANY sign on the BMS that it was a balancing BMS. It had a couple hundred cycles on it aver 2.5 years, So the cells being that close in voltage with no Balancing ability on the BMS is a good sign the cells were well matched, and consistent.
My other premade pack teardown was a 7s1P which used 2000mah ‘BFN’ brand cells rated at only 10 amps CDR to 60C , and the build quality is what destroyed that battery. They missed when spot welding steel strips to the BMS circuit board. I made 2 3s1P packs with those cells and they work OK, but I have to manually balance them fairly regularly, as one cell hits 2.5 when the other are still ~3.16v, and when i discharge them this low then they are also more than 30MV apart in the 4.15v range, and this delta just grows every cycle to where the bms cuts charging current when some 2 cells are 3.85 and one is 4.24.
The DMEGC pack ‘bad’, was that they used 8mm wide nickel plated steel strips, and some of the cells only had a single pair of welds where most others had at least 2. This appeared to be adequate for the ~ 10 amp max load put on each cell within the pack, but must have been heating up pretty well at that rate.
The balance wires were routed pretty well, until they got near the BMS, where they were a tangled mess, and some were in direct contact with the main Positive tab’s sharp edges There was heatshrink, but it was not covering the whole of the strip coming off the main +.
I had previously razored open the heatshrink in this area, and saw this, and had covered that sharp strip with fishpaper dots and immobilized/ separated everything with thin strips of kapton tape, and am very glad I did so, as I got a lot more use from that battery afterwards..
Being a brick pack that cant flex, only vibrate, this was effective mitigation against balance wire shorts.
Removing the Nickel steel strips was pretty easy, but you will need to lift and roll the strips off, and you don’t want to deform the cell can by prying and pulling straight up with a lot of force, but by lifting a corner grabbing with needle nose plyers and rolling the strip sideways, and breaking the welds from the strip.
On the cell’s positive tabs, rolling the strips off can easily lead to a direct short, sparks and soiled underwear. Have a metal bucket and sand nearby. I put Fishpaper on my needle nose plyers so that when rolled, it could not short the cells.
To lift the strips from the battery cells enough to grab with a needle nose plyer, I used a 4mm wide super sharp low angle chisel, to which I also added fishpaper on the bottom flat side starting a few Mm away from cutting edge. One can push it in, under the strip and rotate it to pry a bit like a screwdriver to lift the strip so one can grab it with needle nose plyers and roll it off.
After removing the strips there will be nickel Nubs left on the cells which must be removed. Just laying another strip on top of these nubs and hoping one can press down hard enough to get them in contact with cell body and then achieve a strong reliable weld is a bit fool hearty.
Sidecut/ flush cut plyers can remove a lot of the nubs, but I then use my dremel with a diamond cut off wheel on its cutting edge and try and hit ONLY the remnants of the dots. Use good eye protection and magnification so you can see better. You might be able to get the chisel back in there and lift more of the strip away, rather than abrade it totally away.
I’ve seen other bits used on the dremel to remove these nubs, like a stone barrel or wheel, but I feel I can achieve the most precision with the edge of a new diamond cut off wheel
On the positive sides of the cells I like to set up a vacuum attachment while the dremel is running so the conductive dust does not really get under and in between, and then try and suck out any that might have anyway, and then compressed air.
I have also used a punch to cut a hole in the fishpaper somewhat smaller than the positive tab of the battery and just added that to cell top before grinding to prevent conductive dust from getting under the positive tab. On 21700’s, 18650 fishpaper rings work well but on 18650’s one needs a smaller Inner diameter ring to block dust from getting into and under the vents.
in terms of welding new strips to the cells you will need to experiment on some other cells (preferably discharged to 0.0 volts) until you can at least achieve the same tear off strength as you experienced when disassembling the battery.
The tear off tests are the gold standard in determining whether your welder settings are acceptable or not, and after a lot of practice and experience one can see and feel when the electrodes sink into the strips during the weld whether the settings and technique are achieving acceptable welds, but as a novice this will be completely foreign to you.
Overall my 7s2p DMEGC 20P rebuild went well. I used 0.2mm copper under 0.1mm stainless, even though that is very overkill for these cells, and what I will ask this battery to do from here.
I had not planned on a total rebuild, but when I saw some cells only had one pair of welds and they looked weak I just tore it all apart.
There was some really nice fishpaper in there I reused, and some of it appeared to have extra adhesive to act like double sided fishtape, and the Adhesive was super strong, but still came off cleanly. I dont know whether there is actual double sided adhesive fishtape available, or the assembler used another product to make it act like a double sided adhesive fishpaper tape.
Rebuilding a battery is good practice, but is More work than building a battery with new cells, Quality cells, with a good reputation and far higher specs one will need in their application, in terms of continuous discharge rate.
After I rebuilt my DMEGC 20P 7s2p, I built another 7s2P with EVE 35V and went from 103.7 watt hours to 178.85 wh.
The DMEGC 20P power cell pack stays cooler delivering 300 watts, but the EVE 35V energy cell pack delivers nearly double the capacity at 200 watts and does not overheat. I can ask it for 300 watts, but not continuously from fully charged to fully discharged.
if I had to do it over I would have used EVE 30P cells, an economical choice that would stay cooler and deliver only a few % less energy than the EVE 35V.
So when you tear into your 10s2P, hope for some Decent cells inside, and hope they are all within 30 millivolts of each other. If they are some NoName brand, and the build quality is questionable and they are not balanced very well. maybe just consider this pack as practice and then build a pack with far higher quality cells.
Take lots of photos when tearing it down. Use a red sharpie marker to mark the positive sides of the cells first.
2 Likes
Ok so started playing here. First the construction seems pretty decent, no fish paper rings but the plastic insulator seems a little more substantial than basic so im gonna let it slide.
Picked and peaked at all the welds and all the accessible ones seemed good and solid enough. Not ultra durable overkill like i like but ok.
The cells themselves are cheap junkie cells. Only thing i could find was this reference Here on second life storage and solar. They are cheap low capacity. It’s fine just not gonna be a rocket and likely won’t last very well. For the price they are fine.
You’re right about just taking of the 10 and 9 groups, gonna put them on the side for simplicity.
Sniped the 4 off the top after mark which went where, cleaned up all the edges and left the nickel in place. I don’t want to damage the cell cans and so im just gonna weld right on top of the flattened steel/nickel strip. You could clean it all up like sternwake but i feel like this battery in this application is too close to the good enough/ not worth it border line.
Seriously consider a new pack from scratch - these cells will let you down in the not so long term. As a “hey let’s test this” and a absolute minimum dollars invested it will get by. These batteries new are 5s of dollars sometimes so it’s great practice though.
I arranged the cells and welded a long strip to the +9/-9 group series connection before i folded it into a straight line and glued with b7000 them taped with tesa 51036 into position. Gave it 20 minutes to start to set. Folded the nickel with some Flat jaw pliers and welded it to the bms pcb.
Making sure to have a good layer of fish paper or tesa tape between the nickel and anything else i did the other two connections. Then i went over all the loose bits and bobs with some more b7000 and taped up all the exposed electrical connection points.
Went over it all again and found a dent on one of my cells - made this for the recycle bin from the recycling bin so i can’t complain. If i was gonna keep itni would shrink it and do a charge cycle to make sure it is doing good. All in all it was pretty simple to dial the welds and move this guy around
1 Like
thank you @Pecos @SternWake, your posts gave me the confidence i needed
I did it, final outcome looks like a bomb (literally)
Wrapped in capton → automobile cloth tape (i believe its tesa?) → fibre tape
Step 1 (Honestly sparked 3 times during this operation)
Step 2
Step 3
Step 4
Final
1 Like
46 volts?
A 10S battery should have a full charge voltage of 42v.
2 Likes
I am surprised too, i remember leaving it at 41v before performing the operation..
I pulled this online the exact model BMS and marked the location was when it sparked during the operation for Step 1, do you think that might be causing the issue?
If its unsafe to use the battery, i’ll probably recycle this and get a new one
I dont see the BMS acting as a voltage booster.
I think it is a stronger possibility you wired the 2P packs as 4s1p cells in Series.
So basically you wired up a 8s2P battery inline with a 4s1P battery
1 Like
After reviewing the images, i think this part the most potential accidental parallel wiring here
From White Number Image 1 → 3
Referring to Red circled 1, the face and attachment is as such, the blue line at image 2, is nickel strip tacked on top before wiring it onto the two tabs on the BMS
I figured its alright because the original nickel strip is already connected onto the 4 battery?
@Pecos is ur battery still alive, do you also see v46? xD
1 Like