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

[The Toecutter]

Seems the faster you ride the heaver the trike needs to be! Just a thought.

That has a lot to do with weight distribution and frame geometry. A longer wheelbase can mitigate the tendency for a trike to lift the rear wheel when braking hard, and so too can shifting the rider and battery more toward the rear. While shifting the weight back adversely effects handling making the trike more tippy in corners, a wider front track and longer wheelbase both can mitigate this in order to expand the size of the triangle that the center of gravity must be contained within, and so too can a more reclined seat angle mitigate this.

After converting mine to electric with a heavy hub motor in the rear wheel in 2021, it no longer dived during hard braking.

This changed last year once I took the body shell off because it skewed the weight distribution in such a way that having my battery underneath the front boom was no longer viable and the trike then dove forward and lifted the rear wheel while braking. So I completely took the EV drive system apart after a few rides, because without the weight distribution set correctly, it was too unwieldly to ride, even though cornering capability improved and it seemed to become tip-proof, braking hard was too dangerous. The cornering improved so greatly that I could do donuts with this thing and slide around without tipping. But during a hard stop, it risked rubbing the battery mounted underneath the front boom against the pavement every time it dove. Now that I've added a rear suspension and extended the wheelbase, I expect this issue will be mitigated. I know with certainty it will be mitigated once I have a new body on it, but I can now put the battery safely behind the seat due to the extended wheelbase, and eliminate the issue. In fact, on the next body, like the Shell Eco Marathon cars, I intend to have the battery, controller, and motor all separated behind the rear firewall behind where I sit.

 

[ZeroEm]

My battery is in the bag sitting on a platform, has straps from the platform to anchor the battery down.
Battery is about 15lbs weight is almost centered between the front wheels. It does sit toward the rear a few inches. Not in the picture is saddle bags on the rack that carry a bladder of water, tubes and two or three tools. Pump is attached to the seat in the picture, as a water bottle that was replaced by the water bladder.
Trike is not tippy, can slide the front wheels in a hard enough corner but going to run wider tires next tire change. It needs to be steep hill with hard breaking to try to lift the rear wheel can not do it on flat ground.

 

[The Toecutter]

It also appears that your seat is more reclined than mine when I still had the shell and prior to installing the rear shock and reclining the seat more. That helps things tremendously. The rider accounts for most of the laden weight after all, in your case as well as mine. Changing the rider placement in any way dramatically alters the dynamics of the vehicle.

I'm certain that reclining my seat back further, enabled by going to a 20" rear wheel and drilling new holes in the seat posts, is what accounts for most of its new-found stability in corners. So much so that mounting all the electronics other than the computer and PAS sensors out of the cockpit area and behind a firewall behind the rider, is viable without my weight distribution getting altered in too negative a manner. Fire is going to be something I worry about less because it won't be occurring between my legs or on a battery held to the boom by c-clips, zip-ties, straps, and the body shell itself, nor will it be one of the first things that comes in contact with an object in the case of a frontal collision.

I'm looking into possibly getting a Rohloff hub mounted as a mid drive to the frame, with a single speed sprocket on the Leafbike motor, so I no longer have to worry about rear derailleur clearance. I could have an 18/32/44T up front, combined with the 550% range of the Rohloff, and get all the gearing range I seek, while retaining a fully sealed chainline for what would be both chains. The Schlumpf won't work with my torque sensor, plus using it was predicated upon retaining my rear derailleur to get the gearing I was after. Without a rear derailleur to worry about, I can have a true 5" or more ground clearance all around without having to consider the rear derailleur as the exception, allowing a perfectly flat and smooth undertray(not even my Milan has that!), and more confidently go over bad roads at speed. I'll still need to redesign the spindles for higher-speed riding.

Depending upon how this project turns out, the one after it might use a full-suspension trike designed to accept Grin's all-axle motors high-speed wound in the front wheels plus a Leafbike powering the rear, so that I might build an AWD 20 kW beast of a velomobile/sports car that does 0-60 mph in under 4 seconds with which to troll Hellcats with. Unlike my current project, being able to be operated unassisted on a dead battery as fast as is typical of velomobiles will not be a focus, so the cogging torque losses of 3 hub motors won't matter, nor will the added mass necessary to beef everything up for sustained triple-digit-mph cruising speeds matter. And it would be an absurdly fun vehicle offering dynamics you couldn't get out of any car in existence due to its low mass. As long as pedals remain functional and there is a "street legal" setting, it could still pass as a street legal e-bike when I want it to. Plus AWD would make winter/snow riding a lot less perilous.

 

[ZeroEm]

Everyone pushes just to run one big motor on the rear. Everyone seen so far are great for smoking the tire. Most of them end up moving weight backward to get more traction then they tip the trike over. Not enough weight on the front. Thinking a pair of Grin all axel's would give more traction and add weight to the front end where it needs some. This much power is not really needed but it would be a stop light King.

My front end unbolts and have started thinking of making a front suspension that would bolt on to my trike or any performer with a bolt pattern like the X26.

 

[The Toecutter]

Everyone pushes just to run one big motor on the rear. Everyone seen so far are great for smoking the tire. Most of them end up moving weight backward to get more traction then they tip the trike over. Not enough weight on the front. Thinking a pair of Grin all axel's would give more traction and add weight to the front end where it needs some. This much power is not really needed but it would be a stop light King.

I ran a simulation.

Assuming a 6T wind Grin All-axle on each front wheel running 180V and 150A max phase current with a 4.5 kW input limit, with a Leafbike 5T in the rear with a 10 kW input limit and 150A max phase current, ferrofluid in all three motors, a 20.5" wheel diameter in each wheel running a Mitas MC2 with a Crr value 0.009, CdA value 0.010 m^2, and 320 lbs laden weight, I get the following:

MPH/Time
10/0.4
20/0.8
30/1.5
40/2.2
50/3.2
60/4.2
70/5.6
80/7.2
90/9.1
100/11.8

82.9/7.7 (1/8 mile)
101.5/12.5 (¼ mile)
106.2/17.6 (Top speed)

Drop 30 lbs, and it will theoretically do 0-60 mph in 3.8 seconds, 1/4 mile time of 12.0 seconds at 103 mph. THAT if it can be achieved will be able to play with a Hellcat. THAT is what I eventually want. Hellcats are all over my hood and I want something I can troll them with, with all of a puny 19 kW. It would be quite a fun "bicycle", and I'd operate it on the street with a 750W/28 mph limit when pedaling, at least most of the time. If the cops catch me operating it illegally and try to pull me over, I don't think they will be catching it(provided I don't wreck or have a mechanical failure).

A modified Catrike Dumont would be a decent base for this. It uses thick, beefy 20mm spindles. I don't need it to fold, so the frame can be reinforced in the areas where it is weakened to accommodate folding, and a proper roll cage could be permanently affixed to it with a monocoque shell built around it.

All of the electronic parts needed to build this exist off the shelf today.

A LoneStar battery would provide ample power with a pack weight of under 15 lbs. I might be able to up the power on the front motors depending on how well they handle 180V, and perhaps increase the phase current on the rear to 250A. The Leafbike 5T and Grin all-axle are available from overseas retailers. The PowerVelocity controllers are now available and at 1 lb each and I'd have one going to each motor.

Depending upon construction techniques, the finished vehicle, unladen but ready to ride, robust enough to handle its capabilities, would likely be around 150 lbs.

Yeah, this thing would be insanity if it ever got built.

And to top things off, it would be cheaper than a moped to run, even cheaper than taking the bus on a per mile basis, only needing about 20 Wh/mile to hold 70 mph on the highway, and with everything shut off and accounting for cogging torque losses would still be marginally pedalable.

If cars are to have a long-term future on this planet without committing ecocide, something like this is it. And if it can screw around with modern musclecars and compete with them, it will be able to attract the lowest common denominator over with more money than brains as well as someone who needs simple/cheap transportation as well as environmentalists. There'd be a broad appeal in a vehicle like this, even if it would initially be tailored to a niche role. I bet Luke Workman would approve.

All of this is highly theoretical at this time. I still have to get the next phase of my KMX set up, and that's a much more modest goal(albeit still ambitious when compared to a typical ebike build). I'll be lucky if it runs 15s in the 1/4 mile and possibly still top out at over 100 mph, but it will be very pedalable with the motor shut off and it would be wonderful to be able to fly around at 40+ mph in a sprint using just my legs when desired.

 

[ZeroEm]

Grin's V2 Fst is Kv 12.118, 4T leafmotor Kv 12.33 or 12.53. Could try to find 22" front wheels and and run a 24" rear wheel with a 4T leaf motor they would be close enough to run together. Just don't know if there is enough tire selection's for the 22".

Have been thinking about the A123 26650 batteries in a 24S setup. That way could run a pair of Phaserunners for the front motors and some think like a bac2000 for the leafmotor. Build a battery case that would mount under and to the sides of the frame. Like some of the other trikes. The ah would depend on how big it is made.

I'm not that interested in speeds above 40mph. In town would not go above 28mph. If got back into riding out of town again, higher speeds would be nice but then would need a bigger battery for at least 80 miles. Have made a 60 mile round trip on my current set up (24ah) with out running the battery out at 30mph. could have strapped my second battery to the rack to go farther but can not ride that machine over 4hrs so a little faster would be nice. More miles in that 4hrs.

I would like a fast 0-30mph. Don't think drag racing at my age is a good idea. Will leave that up to you.

 

[E-HP]

Everyone pushes just to run one big motor on the rear. Everyone seen so far are great for smoking the tire. Most of them end up moving weight backward to get more traction then they tip the trike over. Not enough weight on the front. Thinking a pair of Grin all axel's would give more traction and add weight to the front end where it needs some. This much power is not really needed but it would be a stop light King.

Formula 1 trike! Since the rear tire on trike doesn't need to lean, have you looked into a flatter or wider profile tire to increase the contact patch and traction? I don't know if they exist, or if there would be too great a hit on rolling resistance.

 

[ZeroEm]

by E-HP » Jan 23 2023 11:48am

ZeroEm wrote: ↑Jan 22 2023 9:57pm
Everyone pushes just to run one big motor on the rear. Everyone seen so far are great for smoking the tire. Most of them end up moving weight backward to get more traction then they tip the trike over. Not enough weight on the front. Thinking a pair of Grin all axel's would give more traction and add weight to the front end where it needs some. This much power is not really needed but it would be a stop light King.

Formula 1 trike! Since the rear tire on trike doesn't need to lean, have you looked into a flatter or wider profile tire to increase the contact patch and traction? I don't know if they exist, or if there would be too great a hit on rolling resistance.

It's mostly fun, main reason for keeping something like the leafmotor is being able to pedal. the bigger motors go down to single gear so might as well put on pegs.

Spreading the weight out makes a stable trike. Everyone talks about making it wider, then you can not go thru gates or between poles on the trails or even the trails . Could widen mine about 2"-4" then would start having issues. This is total width not per side. Think Toecutter is two wide now. It your going to go bigger then go big. My trike is only 35lbs, a big trike would be closer to 100lbs.
At this time my 15lb battery is between my front wheels. For a bigger battery it will need to be moved back under the seat, then will need to move my seat forward to keep enough weight on the front wheels. So adding motors to the front will solve somethings.

 

[The Toecutter]

Everyone pushes just to run one big motor on the rear. Everyone seen so far are great for smoking the tire. Most of them end up moving weight backward to get more traction then they tip the trike over. Not enough weight on the front. Thinking a pair of Grin all axel's would give more traction and add weight to the front end where it needs some. This much power is not really needed but it would be a stop light King.

Most of the weight is the rider. I set mine up to where I'm seated as close to the front as possible.

Being a stop light king is the idea behind 3 motors. It would be very impractical, especially if pedaling on a dead battery and having to power through the cogging torque losses of 3 motors plus all of the weight this setup would entail. I wish they used lighter/more efficient motors with < 0.2mm laminations so that it wouldn't be so bad, but alas. I'd be looking at 35 lbs of the trike's weight being dedicated to motors with what I can buy off the shelf today, when much lighter for more peak power is possible. No one makes high quality hub motors with modern tech, and the offerings we get, even the premium ones from the likes of Leafbike or Grin, are stuck with technology and construction techniques from the early 2000s.

Formula 1 trike! Since the rear tire on trike doesn't need to lean, have you looked into a flatter or wider profile tire to increase the contact patch and traction? I don't know if they exist, or if there would be too great a hit on rolling resistance.

I'm going with Mitas MC2 16x2.5" tires on my upgrade. This is less for traction, and more for not having a blowout when I'm careening down the highway at close to double the speed limit.

Spreading the weight out makes a stable trike. Everyone talks about making it wider, then you can not go thru gates or between poles on the trails or even the trails . Could widen mine about 2"-4" then would start having issues. This is total width not per side. Think Toecutter is two wide now. It your going to go bigger then go big. My trike is only 35lbs, a big trike would be closer to 100lbs.
At this time my 15lb battery is between my front wheels. For a bigger battery it will need to be moved back under the seat, then will need to move my seat forward to keep enough weight on the front wheels. So adding motors to the front will solve somethings.

My KMX is 39" wide from wheel to wheel. It fits through bollards onto bike trails and even on sidewalks, but just barely so. I have about 2-3" on each side going through bollards. It went bigger because the aftermarket suspension necessitated it. I'd have preferred to keep it more narrow to keep frontal area and thus overall CdA value down, but I needed suspension.

 

[ZeroEm]

If I was that worried about weight would lose 75-100lbs. the weight that bothers me is that last straw that breaks my trikes back.
wheel weight does not bother me much. Do want full suspension.

 

[The Toecutter]

Here are pics of a functioning hydraulic braking system in this vehicle:

 

[uPET]

New brakes look good, how easy is it to bleed the system down?

 

[The Toecutter]

New brakes look good, how easy is it to bleed the system down?

I haven't had to. It came pre-filled with DOT3 fluid and worked flawlessly without me having to bleed anything. I'm sure eventually the day will come where I must.

 

[uPET]

I see I am thinking about taking a couple of front callipers and combining them by making a "T" connector, so they connect to a single brake lever. When there are two paths I wonder how easy it is to bleed down. I suppose you have to bleed one arm and then the other and hope nothing floats back while doing so.

 

[The Toecutter]

I've tried this beast at 200A phase current with a 48V pack of A123s gorilla-taped to the frame a two months ago. 10 kW.

Holy crap! Doing donuts in intersections was effortless. With the reduced center of gravity from the modifications I made to the seat position enabled by the rear suspension, I can't make this thing go up on 2 wheels or tip anymore. I was skidding around in intersections and parking lots testing the tires for grip, and wore them both down to the Greenguard layer in short order. The new DOT wheels are getting Mitas MC2s on them.

I'm currently waiting on parts to arrive to upgrade to 72V, so that I can put this BAC4000 controller to its full potential. Molicel P42A batteries will be used for this upgrade, starting with a 20S6P setup, of 72V 25.2AH. I'll dial the phase current back to 150A, and it will remain 10 kW peak. Then it will be time to upgrade to a PowerVelocity controller, and I might be able to re-arrange it into a 30S4P setup for more top speed.

Should this work out to my liking, I will be looking into perhaps getting some Grin all-axle side motors for the front wheels, and a controller for each of those. I might decide to make a completely different trike for that configuration though, because it would not be very pedalable with the power shut off any longer. My only gripe is that these motors' fastest wind is only 5T, so I'd need to get a new Leafbike motor for the rear appropriately wound to closely match them, or alternatively, find a more powerful motor.

There is, of course, a setting to limit it to 750W/28 mph while pedaling, to keep it "street legal" where class 3 ebikes are defined.

 

[nicobie]

My favorite motor was a 5T. However I was running 24s at the time. Went faster than I needed to go.

 

[The Toecutter]

My favorite motor was a 5T. However I was running 24s at the time. Went faster than I needed to go.

Originally, it had no suspension and rim brakes. You gave it front suspension and cable-pull disc brakes.

Yeah, from a vehicle dynamics standpoint, I wouldn't trust it over 35 mph. Looking at the setup, it probably does 60 mph. Ballsy.

 

[The Toecutter]

I got ahold of Grin. They're considering offering me two 3T wind versions of the All-Axle side motors, at 20kV. I'm probably going to pull the trigger on that purchase.

I'd have a Phaserunner running each front all-axle side motor, and my Leafbike 3T wind at 17.5kV would be in the rear. So if I go from 16" DOT to 19" DOT rim for the rear rim, I could have it matched in speed for the front motors which will be built into 16" DOT rims. Because the front motors use L1019 connectors, I will be restricted to controllers that can interface with this connector, so I'll be stuck with a 72V Phaserunner powering each front motor. But the Leafbike in the rear has an ASI BAC4000 controller running it, a controller also limited to 72V. The larger rear wheel and winding of the front motors should allow 85 mph in this configuration, BUT it will have AWD and do 0-60 mph in about 5 seconds, assuming each All-Axle/Phaserunner is restricted to 4kW and I get 10kW from the Leafbike/BAC4000.

That may end up being the next iteration of this build... Damn wouldn't that be fun?

 

[99t4]

Because the front motors use L1019 connectors, I will be restricted to controllers that can interface with this connector, so I'll be stuck with a 72V Phaserunner powering each front motor.

Why couldn't you cut'n'splice in whatever connector you wanted?

About the dual hubmotor front wheels: Just curious (part of a thought exercise) but would not exactly equally balanced L-R hubmotors power forces cause unwanted steering effects?

 

[amberwolf]

Because the front motors use L1019 connectors, I will be restricted to controllers that can interface with this connector,

Why not just add that connector to whatever controller you want to use? (for instance, I bought an L1019 extension cable from Grin to cut in half, so I can use the GMAC that has one with any of my controllers, and I can use the Phaserunner6 that has one with any of my motors).

 

[The Toecutter]

Why couldn't you cut'n'splice in whatever connector you wanted?

About the dual hubmotor front wheels: Just curious (part of a thought exercise) but would not exactly equally balanced L-R hubmotors power forces cause unwanted steering effects?

Why not just add that connector to whatever controller you want to use? (for instance, I bought an L1019 extension cable from Grin to cut in half, so I can use the GMAC that has one with any of my controllers, and I can use the Phaserunner6 that has one with any of my motors).

I could add a different connector later on, but if I make this purchase, I want something as plug-and-play as possible in the short term. My spare time to get anything done is very limited, I need to keep my vehicles usable with as little downtime as possible, and I've never met a connector I've liked...

 

[amberwolf]

I've never met a connector I've liked..

Can't argue with that. I've dealt with electrical connectivity problems most of my life with all sorts of devices and systems, most of which were connector problems of one type or another.

If I ever finish the Lebowski-HondaIMA stuff, the (very short***) phase wires will bolt directly to the controllers with crimped ring terminals.

***(probably short enough they'll have to be unbolted to remove the wheel if I'm still using hubmotors at that point)


I definitely understand the limited time--my usual problem is that I have even less money than I have time, so most of my project ideas never happen (the cheaper version takes more time than I have, and the quicker version takes more money than I have).


For later PnPability, you can make adapter cables from whatever connector a controller comes with to whatever cable a motor comes with; it's not the ideal thing because it adds two connectors plus a short length of cable between them, but it simplifies wiring them up as you aren't messing with the actual motor or controller cables.

 

[The Toecutter]

I definitely understand the limited time--my usual problem is that I have even less money than I have time, so most of my project ideas never happen (the cheaper version takes more time than I have, and the quicker version takes more money than I have).

I've always had either one or the other. Whenever I came close to having both, there was always something happening in my life that removed one or both from the equation... To say it is frustrating would be a massive understatement.

I'm planning to order the all-axle side motors built into 20" bicycle rims. Thus will allow me to get it set up for AWD ASAP, although I won't be pushing its performance at that point. Then I can pay to get some custom nipples made for the spokes and transfer the motor and spokes that come built into the Grin hub/wheel to a set of 16x1.5" DOT rims and fit the Mitas MC2s on them. I have a good truing stand to do the job.

I'm going to have everything built nice and stout for sustained highway cruising. I have a set of hydraulic disc brakes up front that the builder says are good for 40 mph by themselves, plus a mechanical disc brake in the rear, and I'll have regen on all three wheels. The latter might make high speed braking viable by taking some of the load off the front brakes, rendering them mostly for emergency stops.

The pieces for the roll cage need to be finished and welded up, and a safety harness installed.

Chalo loves it when people try to build cars out of bicycles, so I think he's really going to like this mockery of a car once it sees fruition. His advice helped me put the first one together. I was a complete newbie at ebike building, so I think my first build turned out great in that context. A 45-50 mph capable first build that I didn't kill myself in. And I promise to mock plenty of expensive cars at stoplights with this thing. To the casual observer and to the operator, it's a race car. To the cops, it will be a functional "bicycle" or a "class 3 ebike" depending on where I'm at. I'm going to make it look all post apocalyptic/cyberpunk/Satanic and shit too.

 

[The Toecutter]

Grin got back to me. It will be a few weeks before they have the 3T wind Grin all-axle side motors available. Running a Phaserunner to each at 72V/96A/4kW and the Leafbike 3T in the back at 72V/250A/10kW+field weakening using the ASI BAC4000 should allow 0-60 mph in 5 seconds. Faster once I change the controllers/wiring for the front motors to allow 150A phase current.

The 72V/25.2AH/1.8kWh pack of Molicel P42A using a Lithuanian weldless kit and just the single Leafbike 3T is amazing. Last year, I raced a V6 Dodge Charger at a stoplight and blew its doors off, at least until it overtook me somewhere around 30 mph, using just 46.8V/200A phase current. A bit too scared to top it out on 72V at the moment. Not everything is set up for that yet, and having regen on the front wheels is kind of necessary given my hydraulic brakes aren't rated for more than 40 mph.

I want more acceleration dammit!

I'm definitely going to order the parts to convert the Milan. Going to make a topic on that.

 

[The Toecutter]

The new super-fast wind all-axle motors have been ordered.

I look forward to making this thing into something truly retarded. LFP once told me he'd like to see me have 20 kW in it. So maybe sometime in the near future, it will! The 20S6P battery I put together with Agnuism's weldless kit and some Molicel P42A can do that, continuously. I haven't tried to get more than 10 kW from it yet, but should it hold up to 20 kW, I may very well permanently spot weld the pack together with copper sheet and nickel strips. The BAC4000 can run field weakening, so I can make my rear 17.5kV Leafbike 1500W 3T match what will be 20kV Grin All-axle side 3T motors up front in speed. 72V may well be enough for 90 mph or thereabouts, which is plenty.