2008 CRF250R - QS138v3 70H - ANT - Sanyo UR18650-RX 20s20p - Votol EM260

I never saw someone making his own levers. But why not?

Well done!

Do you know how sharp the carbon fiber parts get when they brake? I hope you will never brake one.
I have seen people 3D printing levers, most likely plain PLA as the guy wasn't a mad 3D print geek. The result was pretty nice, good feel and sturdy enough to ride confidently. Personally I am yet to manage to break AliExpress flipable levers so that's my go too.
 
1742572900541.png

I was too impatient to put everything together properly, so here it is after the first test run. I'm really, really happy with the results.

Some quick thoughts, in random order, while the experience is still fresh:

* chain tension will require experimentation, and it's a bit noisy running through the bottom guard
* the sprocket height is like barely too low - i think 1cm would be more than enough to make it completely clear the swingarm under the rider's weight. I might try 15T or 16T front and see how it lays out then
* I had it set up with 100A in normal mode and 200A in sport. I didn't feel much difference between the modes down low, but they pick up in the higher portion of rpm and each has a higher limit. Still useful, I would say. In sport, first thing i did was an unsoliticited wheelie. Clearly even with just ~15kW it has plenty of pull with this gearing. It's going to be very interesting once I start ramping the power up.
* i definitely have to route all the cables better, and it wil definitely gain with some proper side covers
* both wheels are out of true quite significantly
* front suspension started working really well with the extra weight added.

Overall - I feel great about it.

Couple more shots. Look how skinny it is!
1742573302144.png

1742573315830.png

I'm taking bets on the weight. My personal guess is 92kg.
 
Got my first ride in the forest in today. Did about 35km over an hour, with the battery starting at ~90% and ending at ~10%. It's on the charger now and I'll likely leave it a bit longer for the balance to kick in.

The bike is a blast to ride especially on very steep uphills. Even with just 100bA it crawls on anything and offers a lot of control. It does top out noticeably especially in the slower modes, but in H it's barely noticeable for the way i ride, and in S I don't think i noticed it at all (with higher FW). The S mode is decidedly snappier on the throttle, but even on H it's quite possible to lift the front wheel. I'll likely try playing with the settings a bit and see what I can change. I think increasing the top speed on the M mode but giving it a bit less push would be a good combination for all-around riding. I have no idea how much power it was using as my display is still in its breadboard form.

I also noticed that the edges of the battery are quite annoying, so my hunch that smooth side panels are necessary was correct. I'll start working on the forms right away, as well as for the bottom battery mounts and the airbox. The bike really deserves to be finished. At least nothing went loose/flying/caugth in the wheel. Even the chain slider situation doesn't seem too bad - ironically it's the bottom chainguard that's making the most noise. I have no idea why it's being so annoyingly loud, whether the chain is too loose (doesn't feel like it), or maybe it's just crap made from brittle plastic instead of something more rubbery. I actually modeled the thing in CAD before I mounted it, so I might try printing it from TPU instead.

The front suspension that was "recently serviced by the seller" is already leaking... not surprising considering how all the other mechanicals were done. I guess I'll just have to take it to my suspension job and get properly serviced and set up a bit softer - I used maybe half of the front travel today.

And finally, the weigh-in results are in. 53kg rear, 49kg front - a bit more than expected, but most importantly with good balance.
 
I keep looking at the horizontal chainline pics, as while I'd like to make new mounts, I'm not sure what the proper values would be.. I think 10mm of raise looks reasonable - about one chain link.

I tried to get the rear brake operational but I still can't bleed the system. I should've bled it with everything off the bike, but I didn't and now it's being annoying. I predict eventually all the bubbles will migrate towards either end, if not I'll have to remove one end and/or raise one part higher.
 

Attachments

  • chainline.png
    chainline.png
    416.3 KB · Views: 7
What's the chain line look when rear suspension compressed? Hard to photo that, but it looks like it might be better under load.

So cool you progressed from non function to function on this steed so swift. I'm still pouring over my parts and build 2 yr later!
 
What's the chain line look when rear suspension compressed? Hard to photo that, but it looks like it might be better under load.

So cool you progressed from non function to function on this steed so swift. I'm still pouring over my parts and build 2 yr later!
Oh it definitely is better under load. Don't get me wrong, with three spare chain sliders on the shelf, it could probably be ridden like that for quite a while. That's why I'm taking my time with the new design.

I was actually surprised how quickly the time has passed. All in all it took 6 months, but I had a lot of waiting time during winter. I think seeing the motor spin gave me a push to get it running and then start improving from there. A lot of parts I had designed in theory and then they failed in practice. I don't think it's possible to solve all issues of such a build upfront, especially without a precise 3D scanner, so you kinda have to start putting it together at some point and then see what problems arise and fix them.
 
Åh, and one more thing. The Ant BMS app is really quite nice. Indeed no need for precharge resistors needed, it's easy to turn the battery off just from the app, in charge/discharge directions separately, and initiate the auto balance. I like it especially because that means I won't need any extra cables for a switch from the battery, I can turn it off remotely.
Nice build. could you show pic of Ant app setting for precharge page
thanks
 
I keep looking at the horizontal chainline pics, as while I'd like to make new mounts, I'm not sure what the proper values would be.. I think 10mm of raise looks reasonable - about one chain link.

I tried to get the rear brake operational but I still can't bleed the system. I should've bled it with everything off the bike, but I didn't and now it's being annoying. I predict eventually all the bubbles will migrate towards either end, if not I'll have to remove one end and/or raise one part higher.
Your chainline issue is most likely due to the fact that you are using a 428 transmission, rather than a 520 transmission as was intended for this bike (hence swingarm radius and thickness). For now 520 sprockets aren't cheap for a QS138v3 but it's not like you go through sprockets every months either. Are you sure is it worth your time to change your motor location and possibly introduce some unnecessary compromises ?
 
Your chainline issue is most likely due to the fact that you are using a 428 transmission, rather than a 520 transmission as was intended for this bike (hence swingarm radius and thickness). For now 520 sprockets aren't cheap for a QS138v3 but it's not like you go through sprockets every months either. Are you sure is it worth your time to change your motor location and possibly introduce some unnecessary compromises ?
I mean, drawing the new mounts in CAD isn't that big of a deal (it's 2D after all), I just need to know how much to offset. The entire 428 drivetrain is very affordable indeed, offset a bit by my custom rear sprocket. Basically it's a toss up between an expensive custom front sprocket or expensive custom rear sprocket.

P.S. and I'd need to do that work again for the 520 chain anyway, because it's not just about the sprocket top, I wasn't accurate enough with the axle center.

That being said, so far I love the 14:60 gearing. I have 10.07 final drive ratio right now, which is pretty crazy. I don't really ride all that fast (on this bike - I have the EXC for rallies), and for hard enduro and super enduro it's going to help a lot.

The issue is mostly the result of my poor eyeballing. If I 3D-printed a set of mounts I could've probably verified it earlier, but then again having real metal ones was very useful for setting everything up. In that sense you could say they were scratch parts from the get go, and really weren't expensive all things considered. I'm pretty sure if I get it right the second time, I should be able to get virtually the same antisquat/chain pull location as OEM and keep that drive ratio. There's also always the possibility to (very cheaply) go with 15T or 16T in the front, which would raise the chain even more and bring the sprocket diameter closer to the stock 14T @ 520.

Let's settle on "it's for science!" ;)
 
Last edited:
And I just can't stress enough how good the machine is, even in its unfinished state, with the suboptimally programmed controller and loose cables. I finished work at 6pm, have maybe 30 minutes of daylight, and i can spend 28 minutes of it having a blast riding singletrack that's been right under my nose but was inaccessible by any other bike or motorbike. Worth every hour spent.
 
So, I've checked it and it just says 20A in the "system params" page. I don't have a separate page for precharge, I think it's just this one value. I've taken screenshots of all of them if you need something specific, I'll post them later as I'm sorting out photo uploads at the moment.

I was too annoyed with having to unplug the cables all the time, so the bike has a charging port now. I still haven't started on the new controller box, but this will make my life so much easier, as I ride it every day.
charging.png

A 30 minute ride uses about 25% of the battery when I ride alone. That means hard acceleration, steep climbs, but frequent stops. Used like that, I should get 2 hours of fun at once out of it, which is more than acceptable.
 
Last edited:
Oh, and I also got the LHRB to work. Unfortunately @jbjork was right, the pump diameter is too large and it feels like it has zero power, although you can lock up the rear wheel if you really squeeze on it. I'll see what I can do about it.
 
Doing some cosmeticsIMG_20250328_114356.jpg
As unfortunately the polished frame is already getting boot scratches. Looks like another great location for composites :D

I started slowly chipping away at the suspension. Right now the front (other than the fact that he left keeps leaking) is still too stiff in the early travel, and doesn't really use all of its potential. So I decided for 4.0N/mm springs for the front. I'm still pondering whether I should also replace the cartridge pressure springs, but it seems that I'd need to order them from the US or customer make as they seem to be quite an elusive item.

Nevertheless I've ordered new high quality seals from SKF and a complete bushing and seal kit, just for good measure. I'll likely drop all of that to my suspension shop, because that's one part of the bike which while I can do, I know that the shop can do better.

On the back it initially felt nice, soft and relaxed, if a bit low, but that quickly proven to be unstable and kicking on jumps and obstacles. I check the sag and it was obvious that it's missing preload.

I've added two turns and arrived at 24mm static sag (recommended 10-25) and while it's hard to do by myself, about 89mm race sag in casual clothes and sneakers. It seems that the spring is thusly quite stiff (maximum static sag allowed at the minimum expected race sag), which makes it a bit weird that it was diving and kicking so much. I'll check the clicker settings before the next ride as well and make notes on that.
 
Last edited:
share your Votol settings
I definitely will, once I'm happy with them. I'm currently running a very basic configuration I've described a couple posts above. The fine-tuning of the parameters will definitely happen, i just have more important things to fix first.

I'm currently working on an oven for composite making. The nights are still too cold for the parts to be left unattended overnight. I am planning to start my CF production with a couple of original parts and taking the molds almost directly from them, perhaps with small modifications. Since I have the old plastic bodykit, I can cut those parts up and retain the main mounting interfaces, which should give me good results.

The 14T sprocket is steadily eating into the chain slider - the chain needs lubing and it can be clearly heard during riding now. I'll need to tackle the chain height issue soon as well.
 
The past week has been a bit busy, and I didn't really work on the bike, just rode it a bit. I can definitely hear the chain eating the slider, so it will have to be addressed soon.

Instead, I did some work on the workshop itself. I moved my air compressor to the back to make space for a new workbench:

workbench.png

Which, drum roll, will become my new composite oven!

oven.png

I liked my cardboard ones a lot, but jokes aside working on the floor has been a bit tiresome. I hope that with this setup I can take my parts to the next level. I still have to finish everything up, get the vacuum line, temp sensor etc. all hooked up, make the door for it, but it's looking very promising. I plan to use it at 25C at first for stable curing environment at "room temp" in the cold garage, and try elevated temps of 60C for epoxy curing when I get to epoxy moulds and finally CF laminates.
 
That should help out. :)

I still need to rebuild the insulation box for the 3D printer now that it has it's own workbench (built out of a decent crate some fixtures were shipped to my workplace in). What shall we do with a 3D printer...early in the morning! For my insulation I just used lids from styrofoam shipping containers we get fish in every week. Not as good as what you've got there, but it doesn't cost anything but the moments spent not throwing them away. ;) What shall we do with a 3D printer...early in the morning!
 
That should help out. :)

I still need to rebuild the insulation box for the 3D printer now that it has it's own workbench (built out of a decent crate some fixtures were shipped to my workplace in). What shall we do with a 3D printer...early in the morning! For my insulation I just used lids from styrofoam shipping containers we get fish in every week. Not as good as what you've got there, but it doesn't cost anything but the moments spent not throwing them away. ;) What shall we do with a 3D printer...early in the morning!
Hmm, it's interesting that you needed so much insulation. Generally I keep my printer open for PLA as then you generally want to cool it as fast as possible and actually lower the ambient temp (despite still running bed heating for adhesion and shrink control - basically you want to get it from 200 to 50 as fast as possible and then keep at 50 till the end, more or less, give or take). It does of course help for more annoying materials, but then I highly recommend moving the control board out for longevity.

Another thing worth noting is that styrofoam is unfortunately flammable. You should definitely have a functioning temp cutoff when printing in a closed box. PIR fares much better on that front, but o agree it's more costly; the two large panels we used for the oven cost about $100 in total (with about 30% leftovers). That's significantly more than 0 ;)
 
Hmm, it's interesting that you needed so much insulation.
Didn't "need" it but enclosing it makes it work better, and that's what I had available that was easy to quickly build something from. (less than an hour)

Ideally I'd use a few pieces of flimsy clear plastic and a little frame to secure them to to make the door and top easy to open, but I didn't have that handy.

Most of the time I have to use what's at hand and get stuff done quickly so I can get on with other things; I have about a thousand or two hours of stuff I'd like to do in every day, but rarely have even an hour of time I can actually dedicate to what I want to do (outside of whatever I can manage on the computer while laying here waking and dozing most of the time I'm not at my dayjob, so that can *do* that dayjob).

Generally I keep my printer open for PLA as then you generally want to cool it as fast as possible and actually lower the ambient temp (despite still running bed heating for adhesion and shrink control - basically you want to get it from 200 to 50 as fast as possible and then keep at 50 till the end, more or less, give or take).
Didn't know that...I wonder why it helps make the PLA prints stick together better? (I don't know if they're deforming or not; haven't been trying to make anything that needs exact dimensions yet).


It does of course help for more annoying materials, but then I highly recommend moving the control board out for longevity.
Hadn't thought about that part yet, but now that it's in my mind I have an idea about that for the enclosure design if made from the clear plastic.



Another thing worth noting is that styrofoam is unfortunately flammable. You should definitely have a functioning temp cutoff when printing in a closed box. PIR fares much better on that front, but o agree it's more costly; the two large panels we used for the oven cost about $100 in total (with about 30% leftovers). That's significantly more than 0 ;)
Yeah, I think I posted in the printer thread about possibly using one of the pymeter thermal-contorl power strips to monitor temperatures and turn off power to the whole printer if it goes too high. I also have some other cheap thermal control board that could control relays to do that indirectly.
 
Didn't know that...I wonder why it helps make the PLA prints stick together better? (I don't know if they're deforming or not; haven't been trying to make anything that needs exact dimensions yet).

Better cooling allows you to print faster. That kind of makes sense, you need to heat the plastic up to make it flow and then cool down to make it keep the shape. The quicker you can do that the quicker you can move with the print. I'm not exactly sure how cooling rates impact adhesion - I know that a lot of printers don't cool the first layers to minimise shrink and help them stay on the bed, so I guess that might have some impact higher up on the stack, where the contracting layers might pull a bit.

I'd take a look at PETG for functional prints. It solves virtually all of the printing issues of PLA (of course introducing new ones...), has excellent layer and plate adhesion, thermal stability, shock resistance and chemical resistance, and as long as you can print it (which an Ender can, just with a different heatbreak) it's just a much more robust material overall.
 
Tiny update: I took the bike apart again and dropped the suspension off at my shop. I got new HD seals from SKF (they come with double springs on the oil seals, on the outside and inside), a full set of bearings from Prox, and this:

1744221903376.png

A set of 4.0N/mm spring should make the front end much more pliable. Jury is out on whether the current damping settings will be okay for them, but the first order of business is assessing whether the fork is worth rebuilding to begin with, as it was serviced by an amateur before.
 
Hmm, when I made the oven, i sort of forgot that the entire bike won't fit inside :D
form.jpeg
My sticker frame protectors were already starting to peel off, so while I'm working on the fairings, I've decided to try taking the mould directly off of the frame. It would have been much easier when the bike was still in the disassembled state, but still perfectly possible now, just a bit tricky. I got a bit of gelcoat drip, but i think I've wiped most of it out. Here's how it looked at the start of the process:

QuMagie IMG 2025-04-12 (1).jpg

Speaking of fairings, I don't love how the right-hand side stock one looks. It curves very excessively to accomodate the exhaust and I'm afraid I might have to cut out a bit more from it than I wanted. The left-hand side is looking pretty decent, though. I've used the part of the old airbox to create the combined shape:

QuMagie IMG 2025-04-12.jpg
 
Back
Top