My first ebike build: custom-built KMX-based electric velomobile

[The Toecutter]

I got the brakes stopping as they should and shifter working as it should.

I still prefer riding the Milan, but at least I have the KMX for when I know I'm going to have to go across bad roads. I always had to take detours around my normal routes with the Milan, and it's gotten banged up a bit on the underside, although it's hard to say how much of that was from the two hit and run drivers, including the Karen bitch in a white SUV who deliberately rear ended me and flipped it at 30 mph after tailgaiting me for many blocks(after inspecting the damage in the rear, I'm certain she hit me).

I don't know when I'll get enough time off of work to actually overhaul everything on the KMX the way I intend. I have $2,000 of parts laying around for it waiting to be installed. Still need to spot weld my new battery together as well.

Then comes the arduous task of translating my body shell design to reality. A friend of mine has a 3D printer that can do cubic meters, and he's almost got it calibrated properly. I intend to print the new body shell in pieces out of plastic, then wrap it in fiberglass with innegra. There will be a removable nose and tail mounted to a midsection that is slipped directly over the frame.

I also need to re-engineer those steering spindles. Thanks to uPET, I now have spares.

I'm building a scaled-up Milan with ~6" ground clearance, 39" front track, exposed front wheels flush with the airflow over the body, roll cage, and with clearances appropriate for 2.25" DOT tires as well as a rack and pinion steering system. The goal is 225-250W to hold 30 mph.

If it works well, I'll eventually try to build a monocoque of the design and ditch the KMX frame altogether(I'll still keep whatever bike I build out of the KMX). Then build an overpowered AWD electric car out of it.

To think, a little bit over a year ago, I was washing dishes at the local COVID den. My resume had been languishing around for 3 years and I could scarcely get an interview, and now my current employer at an engineering firm wants me to do way too much work because I'm good at it. Still not putting my greatest talents to use though. "Capitalism", eh?

 

[ZeroEm]

Problem with work is, it's never enough then they work you into the ground.

You seem to be getting a handle on these trikes. Always wanted a KMX, now they seem hard to get and i'm not into wanting as much power of late.

Will stick with my Performer it seems to handle my body mass ok. Would like a Velomobile for the out of town roads maybe some day. It's hot down here would need some AC or just ride it in the winter.

 

[The Toecutter]

If I had the resources and time, I could do a hell of a lot more than mess with pre-made trikes.

When I was in college, I used to spend 16 hours at a time reading SAE publications in the library. I've learned a lot of things that I never got to so much as even experiment with. I have a lot of theoretical knowledge, but comparatively little practical knowledge. The entire reason I went into engineering in the early 2000s is because I wanted to help make affordable EVs a reality. The technology really was ready before then, albeit not as good as it is today. I could never get hired in the industry because I didn't have the money to put anything together and prove I could make a conversion, and then once I finished college, I had to take a job I was not interested in in order to pay my student loans off. Nowadays, I don't have industry experience in that field so getting onboard to an EV manufacturer has proven next to impossible. I basically wasted 20 years of my life, in spite of every effort not to. Had I dropped out of high school and sold dope, I might have had the money to put a conversion together, but then I'd have lacked the permission slip to work called a degree. What a wonderful society we live in!

The most rational transportation OR performance vehicle, IMO, would be designed starting with a velomobile as a base, and adding to it or scaling it to the size needed. Instead, the automotive industry universally starts with an unaerodynamic, oversized 3-box shape and then tries to "optimize" it. It's the difference between a 30 mpg car and a 100 mpg car, where each can do all of the same things, except one is more expensive to operate and consumes more resources over its life span than the other does.

And if you go to a one-seater vehicle, lots of interesting performance and efficiency possibilities avail themselves because the loading placed on the chassis and its components is greatly decreased. Those Eco Marathon cars that get X,XXX miles per gallon could most certainly be translated into a practical/relatively safe road-going vehicle capable of safely sustaining highway speeds. In fact, we already have vehicles with similar form factors that can be readily bought today: motorcycles and autocycles. And I think there's a market for such a thing turned into an ultra-efficient car if enough can be produced in volume to get the per-unit cost down to that of a motorcycle, especially with the direction gasoline prices are going. It is also a way to keep components cost down if the goal is making an affordable EV.

In the early 2000s, I came across Cedric Lynch's custom recumbent streamliner bike build. At the time, he was doing 60 miles range per charge with a single 12V lead acid battery. He didn't have to worry about his battery pack being too expensive to install or maintain, and had the right idea all along. As natural resources become more scarce, more people are going to realize this, but by then, it will be too late to do anything about it...

 

[The Toecutter]

I already put 50 miles on it since getting it operable. I took it to a car show on Sunday and it drew a lot of interest. $0.15 to do 150-200 miles is rather difficult for any other vehicle to match or beat, and with $4.00/gallon gasoline, that sort of economy has a lot of appeal. Wish I had the time to refine my velomobile into something producible and sell-able, then I could start selling them. Haven't had the time to get the baby step of completing it out of the way. At least I have lower-middle-class money though, and since I live in the hood on the cheap, plenty of it left over. It's how I was able to afford a used Milan SL in 3 months of savings last year, especially with that student loan out of the way. Next step is to get my 20 acres in the boondocks and set me up a Ted Kaczynski style shack with some solar panels and a homemade wind turbine. I could run the entire dwelling off of the CALB pack in my GT6 and upgrade the GT6 to some better batteries and a Tesla drive system.

At the car show, I also saw a really nice Triumph GT6 with an LSX V8 engine in it. The owner claimed 11 second 1/4 miles, limited by traction, 25 mpg city and 30 mpg highway if he kept his foot out of it, and including roll cage and chassis mods a weight of only 2,200 lbs. A really nice combination. If someone built a streamliner of that size and weight, math suggests 45-50 mpg highway might be possible, with a big block V8 powering it. My electric GT6 is not yet ready for the road, even though it runs and drives, and until I install a Tesla drive unit, won't be nearly as fast as that. But damn that LSX GT6 was a delightful little deathtrap to see at the car show.

 

[uPET]

Problem with work is, it's never enough then they work you into the ground.

You seem to be getting a handle on these trikes. Always wanted a KMX, now they seem hard to get and i'm not into wanting as much power of late.

Will stick with my Performer it seems to handle my body mass ok. Would like a Velomobile for the out of town roads maybe some day. It's hot down here would need some AC or just ride it in the winter.

KMX are still around. They got knocked sideways by Covid and then the company got taken over by a different owner. So they are in a state of re-organisation. They are still happy to ship stuff to the US, although from recent experience, it seems to take a long time to arrive at it's final destination after it arrives in the US. Not sure sure if that's a local delivery issue or something to do with customs clearing.

Anyway, I have just bought a KMX Cobra kit! It'll be a slow build, but I hope to follow in Toecutter's footsteps and end up with an electrified velomobile. I'm thinking of long range cruising rather than high speed. We are also limited to 250 "rated" Watts over here, so rapid acceleration with high power motors etc is not an option. I'd like to be able to do 250 miles without a recharge if possible.

 

[The Toecutter]

I've got a hydraulic brake system with a motorcycle brake lever and DOT3 fluid reservoir on the way. I'm done with the Avid BB7s and all of their issues. I lost an adjustment knob at speed and only had one side of my front brakes functional after as a result. That was not fun. I'm going to relocate one of the front BB7 calipers to the rear so that I can still have a parking brake and an emergency brake if the front brakes ever fail for any of a wide list of possible reasons.

I'd like to be able to do 250 miles without a recharge if possible.

Very doable. It's going to be interesting to see what you put together.

When I get my next body shell on, I expect energy consumption to drop by close to half, in spite of all of the upgrades the trike will have(heavier wheels, DOT tires). Eventually, I'm hoping to get 400 miles range @ 30 mph on a ~2 kWh pack with ~150W of pedal effort added in as well as some solar panels contributing.

 

[The Toecutter]

Since I am unlikely to be able to get to my friend's shop anytime soon due to my work schedule, I spent today going to hardware stores and getting all of the tools I need to hack the frame of the KMX apart and get that rear suspension installed. I have to do the job in stages, since once I finish my employment for the day, I will only get about 1-2 hours of daylight to work. Considering I have to remove the body shell from the trike, and then modify it so it will fit back on after the suspension is installed, this is likely going to be a 10-15 hour job in total, not counting the time for the paint inside the frame to dry after I cut it apart, drill holes, clean, and paint.

Hoping to get this done within the next 2 weeks. I've been keeping this rear suspension in a box in my room for the better part of a year now.

THEN I can finally take the measurements I need to finalize the next shell design, so I can get going on that. I need to make a 3D model, generate a .stl file from it, and have my friend's 3D printer print it in sections. Then test fit to the trike. If it works, it will be wrapped in either carbon fiber or fiberglass on both sides and permanently installed.

I also have a buttload of parts from Grin on the way. This thing is getting some solar panels. I might be able to get 40-50 miles a day @ 30-35 mph from sunlight alone, and that would be totally bitchin'!

 

[The Toecutter]

I was able to get the body shell off the trike without having to cut apart or destroy anything other than zip ties. I took the tail off, then the rear wheel well cowling, then the rear faring, then the nose, and after detaching the midsection, it slid right off.

Hopefully I can get this rear shock installed before the weekend is over. I'll put the new 20" 3T wind Leafbike motor on the rear wheel as well so I can take measurements with the seat reclined further back.

With these measurements, I can begin work on the new body shell. I may be able to modify the existing one to fit the new dimensions in the meantime as well and put it back on the trike, although that will depend upon what my measurements show.

Looking forward to testing this part. Using a 48V pack, the new 3T wind motor should allow a similar top speed in the 20" wheel, even on the naked trike.

 

[uPET]

That's great to have the KMX back on the road and now getting its long awaited upgrades

Can't wait to see it!

At the weekend I started assembling the bits of the Kobra kit I have recently bought. It doesn't come with any brakes other than some brake handles. What sort of braking set up would you recommend?

 

[The Toecutter]

What sort of braking set up would you recommend?

I've had good reliability out of the Avid BB7 cable-pull disc brake system. They did require adjustments multiple times a month, and I didn't trust them to perform consistently at any speeds over 35 mph(I had a number of panic stops with them from 50+ mph, but I had to adjust my steering direction to compensate for any bias, after quickly tapping them before the full braking effort was used). I had BB5s before that and they were garbage. Those are the only two I've personally tried on the KMX.

I'm looking forward to having a hydraulic braking system in the coming months so that I won't need to adjust the brakes so frequently.

 

[uPET]

What sort of braking set up would you recommend?

I've had good reliability out of the Avid BB7 cable-pull disc brake system. They did require adjustments multiple times a month, and I didn't trust them to perform consistently at any speeds over 35 mph(I had a number of panic stops with them from 50+ mph, but I had to adjust my steering direction to compensate for any bias, after quickly tapping them before the full braking effort was used). I had BB5s before that and they were garbage. Those are the only two I've personally tried on the KMX.

I'm looking forward to having a hydraulic braking system in the coming months so that I won't need to adjust the brakes so frequently.

Yes, I've heard cable disc brakes need a lot of frequent adjustment. What do you think will be a good hydraulic option? Would something like the Shimano SLX M7000 be any good? What will you do about needing two front brakes on one lever? Can this be done easily with hydraulic systems on the market?

 

[thundercamel]

I've had good experiences with Tektro HD-M290 and HD-M285 brakes with Shimano B01S organic/resin pads. They use a screw to hold pads in place instead of a silly cotter pin like Shimano. Opposed pistons are the way disc brakes should be, so they don't require frequent adjustments. I've only ridden a trike once though, so my experience is with bikes. I'd say they provide enough friction for emergency stops on an ebike, but maybe get something larger that can shed more heat for going down a mountain, or use regen.

The green Tektro E10.11 organic/resin pads were just OK, and the black A10.11 semi-metallic/sintered pads would not produce enough friction. The other annoying thing is the first pre-bled lever, caliper, IS adapter and 160mm rotor HD-M290 kit I bought on Amazon in 2019 was $35. Now they're $50 on ebay for the HD-M285, and $47 for the HD-M290 without adapter or rotor. Amazon is all over the place.

 

[Chalo]

Hydraulic disc bicycle brakes are mostly a disgrace, but the good ones are super expensive and still retain a lot of the drawbacks of sh*tty disc brakes.

Consider that every seal leaks, and that the high pressure seal on hydro brakes is right next to the pad.

Avid BB7, TRP Spyke. Nobody is developing better cable discs, because the dingdongs only want to buy hydros. So these are the options if you want first class braking that doesn't grease itself into oblivion.

 

[The Toecutter]

Yes, I've heard cable disc brakes need a lot of frequent adjustment. What do you think will be a good hydraulic option? Would something like the Shimano SLX M7000 be any good? What will you do about needing two front brakes on one lever? Can this be done easily with hydraulic systems on the market?

I have zero experience with hydraulic disc brakes for a trike, but I placed an order on the following:

http://kmxsuspension.epizy.com/Trike Twin hydraulic disc brake.html

I'll see how well it works once it is installed.

None of the widely available bicycle hydraulic brake systems are what I consider adequate for my application. I have an uninstalled hydraulic brake system for a recumbent trike that I bought, but the fluid reservoir is far too small and the components much too fragile for my application, so it goes unused in favor of the current BB7s. The hybrid motorcycle/ebike brake system I ordered in the link above looks promising, especially if I pair it with ATV disc rotors.

 

[The Toecutter]

I tried my 1500W 3T out in a 20" wheel at 3 kW today with a thumb throttle. It wanted to go sideways and was uncontrollable at full throttle, which is only 96A phase current. This is with a Schwalbe Marathon Plus 20x1.75" in the rear. Without the body shell on the trike, it got up to over 40 mph on a 46.8V pack. I did not top it out, either. I was riding it rather recklessly, without the body shell, so my Wh/mile figures approached 40.

Even at 3 kW, this new 3T motor is retarded. Wheelspin is an issue. It's also rather frightening to operate without a body shell. The ride was short, only about 5 miles, but when I was finished, the motor was cool to the touch.

So glad I got this motor! I just wish it were more efficient and could take MOAR power than the 7-10 kW peak people are running it at. Once I get that Mitas MC2 on the rear wheel, things are going to get very interesting. I'm highly doubtful it will hook up properly at 10 kW and 250A phase current, but who knows.

This motor is still awesome, relative to what else is available. Highly recommend.

Looks like I will be able to have my neck resting approximately 600 mm from ground level. This will place my overall ride height at an almost identical figure to the Milan. Except I will have a wider front track and a larger contact patch triangle to hold my center of gravity in place while cornering. It's quite possible I'll be able to corner all out and break the traction of my tires, rather than flip over. Maybe 0.8-0.9g lateral acceleration is possible? I'll find out!

Finalizing some things before I permanently install the rear suspension(especially removing rust and re-painting the trike, as the 65k miles it has endured since 2016 in all weather conditions have taken their toll), but everything is looking very promising. It will be out of commission again soon once I hack the frame apart in the rear. The holes have been drilled the last 2 weeks, but I haven't had a large enough chunk of time(and good weather) to carefully hacksaw it apart.

 

[99t4]

It's quite possible I'll be able to corner all out and break the traction of my tires, rather than flip over. Maybe 0.8-0.9g lateral acceleration is possible? I'll find out!

Schwalbe Marathon Plus are not known for stickiness. Longevity yes, stickiness no.

 

[The Toecutter]

Schwalbe Marathon Plus are not known for stickiness. Longevity yes, stickiness no.

The base Marathons were sticky enough that I could get the trike up on two wheels with relative ease instead of skidding, before installing the smaller 20" rear wheel allowed me to recline the seat further back. It was good for about 0.7g on Marathons where the tires were not the limiting factor, on a 26" rear wheel and a less reclined seat, but haven't tried that test with Marathon Pluses.

 

[The Toecutter]

I test-fitted the suspension and everything is looking good. Tomorrow after work I'm going to play around with different seat angles and positions. I want the most reclined angle possible that still allows me to see over my knees while pedaling, in the interest of the lowest possible overall vehicle height and getting the best frontal area possible with the configuration. It's looking like I'll have to shift the seat away from its current position a few inches towards the back of the trike in order to do that, which will still likely keep me inside the triangle drawn by the contact patches of all three tires, and have the side benefit of stabilizing the vehicle during hard braking.

I'm waiting for some Loctite 262 to arrive before I make the installation permanent and add air pressure into the shock and tune it.

Once I know what geometry I'm looking at for the seat and boom, I can get everything adjusted and then take all the measurements necessary to establish what clearances I need to design the next body shell around.

If it turns out in order to recline the seat back far enough to my liking that I can't see over my knees, this vehicle will have to be more like the Vector Trike and less like the Milan SL. However, I'm aiming for it to be more like the Milan SL. I want to minimize the amount of light/heat magnification inside the cockpit in the context of getting the best possible frontal area. It's going to be a fine balancing act between a bunch of different variables.

It looks like with 20" tires on all three wheels, I'm going to have a 4.5" ground clearance all around, a bit less around my rear derailleur section. If that proves to be problematic, I may have to purchase a Rohloff Speedhub and set that up as a mid-drive mounted to the frame turning a series of sprockets before another chain transfers the pedal output to a single sprocket at the motor, which would eliminate the rear derailleur altogether and give me the option of an extremely-wide gearing range, while retaining my torque sensing bottom bracket.

All kinds of possibilities of where this next design could go at the moment, even though I'm after a generalized design. I won't know what the parameters are until I can take measurements with the new setup ready to ride.

MAYBE I'll have the trike functional again in some from next weekend. Looking forward to trying out the full suspension.

 

[The Toecutter]

I have air in the shock and the suspension is tuned:

Once the Loctite 262 arrives, I will take the screws out one by one and apply the Loctite, then re-insert.

I still have to put the seat back on, get the brakes/shifting working again, re-make the battery box and mount it, and hook everything back up.

Hoping I'll have it rideable this weekend. I won't be able to use the old body shell by the looks of things. I'll make a new one much more slippery once I have all the measurements I need.

 

[The Toecutter]

Loctite 262 just arrived. Will finish the installation tomorrow night. Maybe have it rideable again this weekend. This is going to be cool. Looking forward to riding with full suspension.

 

[ZeroEm]

When I ride i'm looking at my feet unless the gaze is down the road a bit.

 

[The Toecutter]

For me, the sweet spot is while my head is in its default resting position, to see just over my feet and knees. This allows me the lowest possible frontal area/drag, without sacrificing visibility. I do want to minimize the amount of light I need to allow into the vehicle in order to minimize heat magnification/retention, so seeing above the body and looking at the road is the ideal, rather than having to make large sections of the body transparent(like the Vector Trike). But the more reclined/lower I get the vehicle, the lower my frontal area, and the lower my center of gravity. Getting that sweet spot was a challenge for my last designs, just as it will be for this one. An extra two inches can mean the difference between it being pedalable or not, or between having visibility or not.

When I ride i'm looking at my feet unless the gaze is down the road a bit.

 

[ZeroEm]

Inch or two makes a big difference. 4.5" of ground clearance should be good. Think I have around 4" at the lowest point but mine is a bent. It's lowest point is 6" or more in front of the rear wheel. Was thinking about 24" front wheels but going to go with a 24" rear 20" fronts first and see how it goes. Look a rockets, the frontal area is larger than the body so the air does not drag all the way down, just on the front.

 

[The Toecutter]

I cut down the seat posts and drilled some new holes to lower the seating recline. This thing is going to be as low as my Milan, but with a much wider front track, a similar wheelbase, and relative to the front wheels the seating will be closer to the axle line than the Milan. Meaning, much better cornering than the Milan, and the Milan was already decent at that.

The next body is being designed off of the Milan if possible. A scaled version of it designed to fit my new dimensions. Even the way I get in and out of it will be like the Milan where I have to lift myself in and out. I'm going to build a rear firewall and keep the battery, motor, and controller behind the operator for safety reasons. With the seating extremely reclined and the rider as far forward as possible, it shouldn't adversely effect the center of gravity appreciably, and it will all be sprung weight. The only unspring weight will be the rear of the frame-turned-swingarm, rear derailleur, motor, wheel, tire, and freewheel. Even the tail section with the trunk space will be lifted and lowered by the action of the rear suspension like the rest of the body. Hopefully the unspring weight stays down to where it doesn't adversely affect cornering.

A lot of design inspiration from Shell Eco Marathon race cars and various solar cars went into this concept.

If I'm going to do highway speeds with this, safety does greatly increase in priority. Maybe I'll have the next design iteration of the roll cage this weekend so I can start designing the body shell around it.

 

[The Toecutter]

Took the first ride in the streets with full suspension. The battery needs to be split up and attached under the seat, so this ride was entirely pedal-powered, since the battery is no longer mounted. Because of the rear suspension lifting the seat and lowering the front when I get off the trike, and because of bumps lifting me at speed, underneath the boom is no longer a viable place to have the battery mounted. On the next shell, I'm getting a rear firewall and all the electronics, with the exception of the CA3, torque sensor, throttle, and 12V lighting/signals, is being isolated from the rider area. My seat is now as far forward as possible and is as reclined as possible while still allowing the rear shock to travel without interference.

I must say, going downhill at 30 mph and deliberately hitting deep potholes showed me that I barely feel them! With the more reclined seat, tipping it is much more difficult to do. I deliberately cornered tightly at speed and it didn't go up on 2 wheels. It wanted to skid instead. This is encouraging. I want to be resistant to tipping when cornering hard. Previously, this build could still tip if I was careless, and have had it up on 2 wheels a few times(mostly deliberately) before removing the body and doing the recent changes.

Pics below:

Very happy I did this mod. It was necessary for the speed and performance I intend to eventually have with this machine.