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

Nice to see the project going!

I wouldn't advise putting a MAC in a 20" wheel unless you absolutely need low weight. The efficiency will be poor because gear friction will be increased significantly while cruising vs a 26" wheel.

There's aren't many geared motors which are properly geared for 20" wheels in general.

What a beautiful Bike!

Take care riding it as it will for sure be a speed demon.

neptronix said:

Nice to see the project going!

I wouldn't advise putting a MAC in a 20" wheel unless you absolutely need low weight. The efficiency will be poor because gear friction will be increased significantly while cruising vs a 26" wheel.

There's aren't many geared motors which are properly geared for 20" wheels in general.

Click to expand...

I wouldn't put a MAC in the KMX.

I am considering a MAC for the Milan velomobile, which has a 26" rear wheel, and which I need a solution where I can decouple the cogging losses. If I were to put a MAC in the Milan, I'd limit it to 750W/28mph, which would be street legal in the majority of the U.S. but allow me to maintain 30-35 mph completely pedal powered because the Milan is so slippery, and for which I need to be able to eliminate the cogging losses. The motor would mostly be used for going uphill and accelerating from stops. I'd only need about 3 lbs of Panasonic 2170s to power it. The all-up weight penalty imposed by the electric conversion could be under 15 lbs in this configuration if running a Phaserunner controller. I could also have an "off road" mode, bump the power to 1 kW or more, and do 65 mph with 72V wherever I can get away with it.

nicobie said:

What a beautiful Bike!

Take care riding it as it will for sure be a speed demon.

Click to expand...

I ride both of them like I would drive a car. The Milan is entirely unmotorized and I have almost hit 50 mph on flat ground. I've been over 70 mph downhill in the KMX after converting it to electric.

The Toecutter said:

adding a Schlumpf HS drive with a triple crank up front, and replacing the torque sensor with a model that accurately reads the torque at the crank, and not merely estimates it based on tension.

Click to expand...

I look forward to your progress! What kind of torque sensor will work with the Schlumph drive?

thundercamel said:

I look forward to your progress! What kind of torque sensor will work with the Schlumph drive?

Click to expand...

One that isn't directly built into the bottom bracket.

Alternatively, a torque sensing bottom bracket can be chamfered to allow the Schlumpf to fit on it.

Nice work!

I recently installed a geared motor GMAC 10T on a 20" solar trike and I was surprised to see how well the "electronic freewheeling" feature work.

Regen was also pretty good. I reached -1800W in a big hill ( with a Phaserunner controller )

https://ebikes.ca/product-info/grin-kits/gmac.html

adam333 said:

Nice work!

I recently installed a geared motor GMAC 10T on a 20" solar trike and I was surprised to see how well the "electronic freewheeling" feature work.

Regen was also pretty good. I reached -1800W in a big hill ( with a Phaserunner controller )

https://ebikes.ca/product-info/grin-kits/gmac.html

Click to expand...

I can do electronic freewheeling with a direct drive hub motor. I really want the ability to continue pedaling the bike with no losses if the battery runs dead. I want the motor to supplement the rider, and not the other way around. I can almost hit 50 mph on flat ground in the unmotorized Milan SL. Regen is not much of a concern for this one, so I'm really looking for an original MAC 6T. It would compliment the Milan velomobile perfectly, and serve as a 100% street legal setup where the cops can't do shit if I fly by them at 45 mph because I'd be doing it on 100% human power. But I do need a motor for going up hills, given that the automobile traffic is almost never sympathetic when I'm using all of my strength, overheating myself, and causing myself muscle cramps to go 15 mph up a 15% grade while the cars behind me want to do 40 mph. A 750W assist would work wonders toward narrowing the speed differential in these conditions. By running a 72V setup and an FOC controller, below 28 mph it will never see more than about 30V, but without the penalty to thermal efficiency due to the controller being able to match the voltage to the operating point, BUT I could always have an off-road mode for flying down a state highway at 65 mph with 750W of assist plus pedaling whenever/wherever I can get away with it. The Milan is slippery and doesn't need a lot of power to go fast, and is guiding the re-design of my custom build.

Now the KMX, that will continue to use a direct drive motor, mainly because there is no geared hub on the market that can handle the power I want to put to the ground. I'm eventually going to dump 7 kW peak or more to the rear wheel and make this thing accelerate like a car! If I get the drag of the next shell to be twice that of the Milan, that will be a major victory, and I'd be looking at 60 mph on roughly 1.5 kW total power at the wheel, so if the rider pedals with 150W and the battery dumps another 1.5 kW as the motor gets just under 90% efficiency, I'd be cruising along at 25 Wh/mi on the freeway. Granted, the KMX wasn't designed to handle this(I'll need to re-design the steering spindles, axles, and wheels/hubs before it can, at minimum), so it will spend most of its time around 35 mph with the goal of getting a 300 mile range at that speed on a 20S7P pack of Panasonic 2170s, while still remaining very pedalable with a dead battery and the motor shut off by keeping the total mass under 45 kg and having a wide gearing range to allow at the extreme low end climbing a 20% grade at 2.5 mph with a 60 rpm cadence. The upgrades to handle higher speeds can come later. Plus it will have solar panels, and storage space comparable to a small car. Would be a GREAT vehicle for a post collapse resource-starved society.

The concept I am seeking to make reality is low maintenance, almost-free, reliable rapid transport that is comparable in utility to a car while cheaper per mile over the vehicle's lifespan than taking the bus, is easily repaired with widely-available components that could even be stripped off of cheap bicycles, allows one to save time by getting exercise during their commute, is versatile enough to be pedaled faster than a normal bicycle with the motor disabled if need be, and has the ability to haul ass and mess with cars at stoplight drag races when using the motor, the purpose for which this performance is sought being in effort to broaden its appeal.

The next step towards this vision is getting a functioning proof of concept built to the above specs AND operable a daily basis. And to do that, it is going to use your rear suspension kit.

The rear suspension kit for the KMX is on the way.

To install it, I will have to remove the body shell, cut the frame apart in the rear, drill some holes, and bolt everything back together. I'm getting a 3T wind 1000W Leafbike built into a 20" rear wheel to replace the 4T wind 1500W Leafbike motor built into a 26" wheel. While I have everything apart, I'm going to have to address any scraped paint and rust on the frame as well, as it has some of both readily visible. This frame has almost 61,000 miles on it!

I will then recline the seat back further, take measurements, and get to work designing the next iteration of the body shell, based heavily on my Milan SL velomobile.

It will probably be at least a month or two before I can even have the time to start this work. Don't know when I'll have a new body shell ready, but a friend of mine now has a large 3D printer up and running that can print me a body in thirds(nacelle, nose, and tail pieces, like a WAW velomobile). The plastic may not be strong enough to use the printed body for anything other than a mockup, and I will likely have to fiberglass over it on both sides as a sandwich construction to make it strong enough, but with the printer printing the basic shape for me, thousands of hours of work will be saved, and I'll be able to test fit everything before devoting too many hours into it in the event it doesn't work.

If it does work, I'll be able to use it, as well as make a mold of it in case others will want it.

Looking forward to seeing the end result.

Mac in a 26" is good stuff.
If you really like the quality of MAC motors, they also make a lawnmower motor rated for 2kw. It has identical dimensions to the hub but half the poles and spins quicker. Chain drive only and no built in freewheel, of course.

If you want to play with one i could probably sweet talk Damon du of cutler MAC into selling you one as a "sample".

nicobie said:

Looking forward to seeing the end result.

Click to expand...

Me too! This Milan is quite an example to study. I now have a much better understanding of how to seal up all the individual items out of the airflow. Every tiny detail adds up to something of great significance when adequately addressed. I just did a 20 mile ride with a 23 mph rolling average, with lots of 30-35 mph cruising and lots of red lights at the bottoms of hills interrupting my pace. If I end up with twice the Milan's drag and run heavy DOT-rated solar car tires, I'm looking at the possibility of maintaining a flat ground speed of 30 mph on 220-250W at the rear wheel. Meaning, it will be very pedalable with the motor shut off even factoring in cogging losses, perhaps able to do 100 miles in 4 hours on a closed course while lugging all the electronics around but keeping them disabled. And when you turn the motor on, that might mean 1.5 kW from the battery required to do 70 mph with 150-200W of pedaling on flat ground. That means a 2 kWh pack would get nearly a 100 mile range at 70 mph, and if you drop that speed to 55 mph and maintain the same rider input wattage, maybe a 150 mile range.

Plus any 120V outlet would be the equivalent of Level 2 charging. A 240V outlet or a ChaDeMo system would be the equivalent of a Tesla Supercharger. That's the magic of a small battery. The trick is to make a charger that is light and can handle the power needed for this, which my Saitiator is not(although it is a good charger, just not for that purpose).

All theory at this point predicated upon my getting the drag low enough. It's not going to be easy to do. But the possibilities are there. We'll see. The one I built did okay thus far. I'm about to increase its 35 mph range to around 150 miles by adding the second Greenway battery in parallel, likely before disassembling it altogether to install the rear shock.

neptronix said:

Mac in a 26" is good stuff.
If you really like the quality of MAC motors, they also make a lawnmower motor rated for 2kw. It has identical dimensions to the hub but half the poles and spins quicker. Chain drive only and no built in freewheel, of course.

If you want to play with one i could probably sweet talk Damon du of cutler MAC into selling you one as a "sample".

Click to expand...

I need my gears. This lawnmower motor would only work if I invested in a Pinion drive up front with a jackshaft configuration, and I don't want long crankarms which are all that come with the Pinion drives(won't fit in a Milan SL), plus they come with baggage like high complexity making them too difficult to service, or the incompatibility with a torque sensor for PAS. For the Milan, the MAC 6T wind with freewheel compatibility is exactly what I need. It doesn't need 2 kW.

The KMX could make decent use of the lawnmower variant, but I'm intending to dump a lot more than 2 kW to it making the lawnmower motor inadequate. The 2 kW MAC 1-speed would still go quite well in the KMX, mind you, but I'd have to spend a lot of money to accommodate it keeping in mind I want it to be pedalable with the motor shut off and need a wide gearing range. A 3T wind 1,000W Leafbike can handle a lot more power and would be the preferred route to take, unless someone comes up with a more efficient motor/controller system affordable by hobbyists. What both vehicles really need is a synchronous reluctance or synchronous reactance system, which is not available in an e-bike appropriate package. the 3T wind 1,000W Leafbike motor would be the next best thing, at lest for the KMX.

I got ahold of a contact in Switzerland that can sell me a MAC 6T. Awaiting further details. It will have to have a wheel laced up around it by the local bike shop.

I'll have to eventually make a separate topic for the Milan should I decide to go through with this. A 20S2P pack of Panasonic INR21700s and a Phaserunner controller would be perfect for this. This pack would be 704 Wh, make up to 1 kW peak reliably without damage, and only weigh 6 lbs. The vehicle will spend most of its life at over 28 mph with the clutch disengaged and the rider accounting for all motive force, with the motor only coming on while taking off or going up steep hills. The perfect city "car", only needing 750W. I'd likely have a 150-200 mile range in this usage case. Best of all, the total added weight from the EV conversion would only be around 15 lbs, which wouldn't much affect performance at the speeds where the motor shuts off. Operating in the "off road" mode where the wattage restriction remains but the speed restriction is eliminated, I'd likely be able to top out at over 65 mph on flat ground and get about an hour's worth of run time at that speed!

If I get the chance, this weekend I want to build a housing to fit the second Greenway 46.8V 15.5AH pack into my corovelo, running the two in parallel. If successful, I can bump the peak power to 3,000W. 0-30 mph would be quite car-like even with the puny Phaserunner driving it.

I'm also intent on swapping the rear tire of the Milan out from a 559x28mm Conti Grand Prix to a 559x40mm Schwalbe Marathon Greenguard.

Schwalbe Marathon Plus on the front (20" x 1.35") and rear (26" x 1.75") of my trike.

The fronts are rated for 100psi, run 80psi. They are fast and you feel every bump. Have 5k miles on them, will rotate them soon to see if I can get another 2k-3k miles.

The rear is ran at 60psi, don't know how many miles I will get out of it maybe 15k-20k miles The rear had been on for 2yrs no flats.

The trike came with Michelin's Protek on the front and rear. Have a brand new 26" Protek in the shop no mile still mounted original wheel. The front were 20" x 1.5" they road good but the ware was not good and had a flat the first few months of use then replaced them. Wanting to try the 20"x 1.75" SMP on the front but need to mesure my rim width. May go with the 20" x 1.5: SM Greenguards.

Don't know how many miles you get out of a set of front tires/tyres.

ZeroEm said:

Don't know how many miles you get out of a set of front tires/tyres.

Click to expand...

My front wheels on the KMX are cambered 3 degrees, which puts the surface in contact with the road slightly off center. I use Schwalbe Marathon Greenguard 406x40 on the front wheels and a Schwalbe Marathon Plus Tour 559x40 on the rear.

I get about 4,000-5,000 miles out of a set of front tires, factoring in that I rotate them every 1,000 miles or so. The front tires are at the end of their life when a small number of cracks and gashes develop in the tires(usually less than 5) that are large enough to expose the greenguard layer and/or if the cracks spread to the sidewall, which then makes them prone to puncture and failure, and often they are in this end-life state while the rest of the tread still looks new. I have to change them in spite of still looking good when the few cracks/gashes are ignored, because they start getting flat too frequently at this stage, every 100-200 miles or so. Before they get to this point, flats simply don't happen with them(one exception was when I hit a 1" nail at 40 mph with only 1,000 miles on the tire, but a new tube rendered the tire still usable to its normal end of life).

The rear tire has over 45,000 miles on it and the tread is still good with no cracks or gashes and I'm liable to get another 10,000-20,000 miles out of it. I've been using the same rear tire for almost 3 years. I'm very satisfied with it. I have a Schwalbe Marathon Supreme 559x52 with over 10,000 miles on it that still looks brand new and never had a flat rear while I used it, but I swapped it out because it had terrible traction in the rain and I'd have to get out of the velomobile and drag it onto a sidewalk and uphill from a stop because I couldn't get sufficient traction to go forward when trying to pedal it on slick ground, and the trike was unmotorized then.

Out of all the front tires I've tried on the KMX, the Marathon Greenguard have given me the longest life AND the lowest rolling resistance. I've also tried Maxxis Hookworms and Schwalbe Trykers, neither of which were as long lasting or as efficient. I can't tell if the Marathon Plus Tour or the Marathon Supreme is the longer lasting tire for the rear because neither have ever been worn out. The Marathon Supreme definitely rolls more efficiently than the Marathon Plus Tour, but has no traction in inclement weather(do consider I have very little weight on the rear wheel), but the Marathon Plus Tour has adequate traction even riding in the snow provided the snow is not higher than about 2".

The Milan has a new set of Continental Contact Urban 406x32 on the front, and a used/worn Continental Grand Prix 559x28 in the rear. The rear will soon be replaced with a Marathon Greenguard 559x40 I have on hand if it will fit, and if it doesn't, I'm getting a Continental Contact Urban 559x32 for it.

Thx, no one really talks about their front tires on trikes. Wished I could adjust my camber, have not measured it but wares on the inside more. Over the years have adjusted my toe in where i'm happy with it. When it is set straight it shimmies and wares the tires out.

ZeroEm said:

Thx, no one really talks about their front tires on trikes. Wished I could adjust my camber, have not measured it but wares on the inside more. Over the years have adjusted my toe in where i'm happy with it. When it is set straight it shimmies and wares the tires out.

Click to expand...

My front tires also wear on the inside more. My toe-in is about 1/16th of an inch, which is about as close to centered as I can hope for while still retaining straight-line stability at high speeds. At over 50 mph downhill, if the road is smooth and even, I can take my hand off the steering and it will stay straight and centered.

First six months in learning mode could not take one hand off. Can drive now with one handed at speed. Direct steer takes some practice and a steady hand. I under stand people wanting USS (Under seat steering), easer to learn. Let my brother drive my trike once and scared him and never road it again.

I meant to say "hands", not hand. That being said, I can ride one handed with direct steering, but I don't often do it. Not worth the risk.

I really liked the under seat steering of the Thunderbolt trike I restored. It was extremely stable and easy to control. The direct steering of my KMX is darty, and before I added the front suspension which widened the track, it was extremely difficult to keep straight over rough roads. I once hit a metal road covering that was put in place of some pavement that was removed, which caused my trike to go airborne and one of my hands was jerked off of the steering. When the trike landed, it was a struggle to get my other hand back onto the steering without losing control and crashing because I had also shifted out of my seat and could have easily tipped the trike. And I was doing about 25 mph. Not a scenario I ever want to encounter again. It wouldn't have been so bad if I had the body on it at the time, but it was before my first shell was built. With the front suspension and widened track, the KMX is a lot less perilous to operate, almost on par with the indirect underseat steering of the Thunderbolt I had before it. Almost.

I just melted an XT60 on the Phaserunner controller. I was running 30A battery current and 1500W. The controller still functions, but the XT60 plug from the battery to the controller is melted to the controller. For the near future, I'm going to dial the maximum power down to 350W or so. I have more than 10,000 miles on this controller. I didn't expect 1500W would damage it in this way, and was planning to later run it at much higher amounts of power. Now I know that this won't be prudent. I'm going to repair the controller, cut down the power allotted, and then upgrade to a more powerful controller. If I electrify the Milan, the Phaserunner is going into it, with a 750W limit, and no regen.

The rear suspension arrived more than a week ago, but I haven't had the time to install it. It's a big job that is going to require disassembling the vehicle, but I use it for work so that won't be happening. The Milan is waiting on a new rear tire to arrive, so my backup is also out of commission until then. Eventually the circumstances will be such that I can get this rear suspension installed. I might have to wait a month or two.

I finally got the melted connectors between the battery cable and the controller terminal separated. It wasn't the plastic that melted. Surprisingly, it was the copper terminals themselves that welded together in limited spots, and they had corrosion. I've been riding this vehicle in all weather conditions, including hundreds of hours of operation in the rain, so I'm not at all surprised by this in retrospect, but I had thought the plastic itself was melted, and I'm surprised that the plastic around the copper wasn't the culprit. The controller was almost hot enough to boil water yesterday.

Looks like my controller is completely salvageable, with just a bit of cleaning. It still works just fine.

Would dielectric grease help for the future?

99t4 said:

Would dielectric grease help for the future?

Click to expand...

I'm sure it would. Is there any in particular that you would recommend?

Have not used enough of it over time to make a good recommendation. Bought the large economy toothpaste-sized tube of a commonly available brand at the FLAPS. Seems to work, that is, nothing bad is happening.

A wheel building stand is on the way. Two new KMX hubs arrived yesterday, along with a new rear derailleur hangar(it has over 60k miles on it and is slightly bent, and needs replacement). A hydraulic brake system and thicker rotors were ordered last week. A Leafbike 1000W 3T wind motor built into a 20" bicycle wheel is also on the way, and once I install that new motor and the rear suspension, I will be able to temporarily take the body shell off and take some measurements in preparation for the next body shell design and to finalize the roll cage design and begin construction of both.

I'm intent upon having some DOT rims and Mitas MC2 tires on all three wheels sometime this winter. A Schlumpf HS drive, roll cage, new body shell with improved aero heavily based upon my Milan SL velomobile, more accurate torque sensor, will also be added during or shortly after.

The idea being, to build a motorcycle so aerodynamically efficient one can pedal it to 40+ mph in a sprint with the EV system turned off in spite of all the weight(expected vehicle weight of 90-110 lbs), but then turn everything on, and it now accelerates like and can reach freeway-illegal speeds like a car, with gearing sufficient for human power being able to add thrust at all operating points the vehicle is capable of as well as climb a mountain at 2.5 mph with the motor disabled at a reasonable cadence that will be easy on the knees.

Theoretically, 20-25 Wh/mile @ 60-70 mph with light pedaling should be possible, if I end up with a CdA 2x that of my Milan SL. That would be the equivalent of close to 1,500-1,700 MPG at that speed!

We'll see. Getting enough time off from work is definitely an issue, as I still haven't had the chance to install the rear suspension.