Corovelo
 

1kWh
3000w

Probably need another day or two of assembly. :cool:


https://i.imgur.com/5oiAMGs.jpg?1

There are several similar designs like this. See recumbents.com for another version. These are quite cost effective. I use white Gorilla tape and 5/8" OD PVC hoops to stabilize some of my shells.

Took a break from assembling the body to play with my phaserunner - previously I had been running a no-name Chinese controller with the partially faired trike, and was topping out around 40mph. Unloaded, wheel speed peaked at ~45mph. Tweaking the field weakening setting, I was able to get a motor speed of ~5100rpm, or ~51mph. Going this far had very undesirable effects on power consumption, but a very slightly field weakening effect seemed to actually improve (unloaded) power consumption at max RPM.

Turn signals and running lights to come soon!


https://i.imgur.com/ty9Rld2.jpg

https://i.imgur.com/bzCOnsR.jpg

https://i.imgur.com/HcHAn7n.jpg

https://i.imgur.com/W5TYQaW.jpg

https://i.imgur.com/z9pungO.jpg

Don't forget the lap belt.

That thing is badass. It looks simple to make, but will more likely than not be very effective at reducing load.

Let us know what specs you find it to have when you can. Max speed, CdA from coast down testing, dimensions, weight of trike/shell, ect.

Thanks for posting this.

Tuft test tuft test tuft test! :D

I'll get around to it, thanksgiving has had me away from my projects.

I will say that it's nasty-sensitive to crosswinds now. The bare trike, though it used more energy, was far more forgiving when it was windy. I find with the faring I have to keep it under ~20mph when there's more than the faintest breeze, or it gets a bit scary.

A lower nose and a rounder cross section will help. It needs mainly more down force on the nose.

I can't take too much off the nose because I need room for my feet, but I'm hoping this will help:

https://i.imgur.com/ovONWfN.jpg

https://i.imgur.com/CdvNhO3.jpg


In other news, I've been fussing with getting my controller programmed correctly. It doesn't particularly like that I have a motor which hits 4-5000rpm, and regularly throws phase overcurrent faults when the rear wheel loses traction. I also have what appears to be a faulty Cycle Analyst, and have been working with the manufacturer to see if I can fix it, before I resort to sending it back for RMA.

Every little bit helps.
https://www.bing.com/videos/search?q...092B&FORM=VIRE
Compare with sailplane fuselage of Bowlus BA-100.

The changes Grant suggested seem to help with high speed stability. Will be getting wh/mi numbers soon.

Yesterday evening, I put together my light control box. I'm still missing the 3-pin JST connector I plan to use for the turn signal, and the actual physical switches which will be installed on the handlebars.

This will be used to power and connect a motorcycle headlight, tail lights, brake lights and turn signals.


https://i.imgur.com/28U8zKZ.png

https://i.imgur.com/Ws6ZbPk.jpg

https://i.imgur.com/8XxQGZS.jpg

https://i.imgur.com/I24OggB.jpg

Are you using a MAC hubmotor by chance? Any plans to add ferrofluid?

Yep, it's a MAC. Pretty sure ferrofluid would be a disaster in a geared hub motor. My reading suggests that it adds significant drag (and lost efficiency) even in direct drive motors, and geared motors have a lot more in the way of moving parts inside. And anyway, I'm not having any heat problems with 3000w fed only for brief periods. I destroyed my first bearing hitting a pothole at close to 50mph though, and have had issues with the phaserunner. It's not happy with a 5000rpm motor running at both high current and voltage, and it's going to take a lot of tuning to minimize instantaneous phase overcurrent faults. I think if I were to do it again, I might go with something else, maybe an infineon controller. Or, considering how unrealistic it is to pedal at higher speeds, maybe a direct drive motor with the phaserunner.

Working turn signals and brake lights:

https://youtu.be/aOjCmHE5Q7Y

Do you know what winding it is? 6T? 8T?

A direct drive motor can run at much lower rpm. If you have a 26" rear wheel, 388 rpm is 30 mph.

I like the 1500W Leafbike motor.

https://endless-sphere.com/forums/vi...p?f=30&t=66489

With 48V, roughly 90% efficiency at 1kW and 390 rpm in a 26" wheel, the manufacturer's specs having been confirmed by 3rd parties on endless sphere. Without any mods, they can reliably and repeatedly handle short bursts of 4 kW and 1.5kW continuous, but with modification(fans, ducting, ferrofluid, ect), 10 kW peak is possible. It's also very inexpensive at under $300. My only major gripe is that it is 16 lbs of additional rotating mass with which to accelerate while pedaling up to speed should you choose to turn the motor off or run out of battery or for whatever reason need to move it with the motor disabled.

Supposedly, its drag isn't very noticeable when pedaling.

Mine is 8T and is the latest gen, with composite gears and a temperature sensor. I cracked it open after a few hundred miles and the gears all still look good, but I'm probably going to run it at slightly lower power to protect clutch and gears. Geared hub motors feel a lot more torque-y at low RPM than direct drive, and tend to have better efficiency at lower speeds, so I probably don't need nearly as many watts anyway.

At first glance the Leafbike motor looked like any of the other eBay hub motors, but the efficiency graphs suggest otherwise. I might snag one to try.

That MAC motor you have is probably around 75% efficiency in your application, so the improvement in range and reduction of heat will be very noticeable.

Without suspension, I wouldn't want to do more than 20 mph on some roads where I'm at. Any data on the efficiency of the shell yet?

I broke one of my motor bearings so I'm waiting on a local bike shop to get one in. ^^

Always something.

Preliminary data!

20mph - 15-16wh/mile
30mph - ~21wh/mile

Temperature was just about freezing, and there was a crosswind gusting to ~10mph which was pretty scary at 30. :D

EDIT:

So that's effectively 2,250mpge and 1,600mpge respectively.

So roughly 630W from the battery to do 30 mph. That's not bad for such an inefficient motor. With the leafbike motor, I predict that will drop to 550W, or 18 Wh/mi. And that's without any human power to aid things further, correct? You could set it up as a pedalec and provide about 1/4 of that power needed to do 30 mph without much exertion, and drop that figure further.

I would need to get a different set of sprockets. Might well do that.

Leaf bike motor should be here within a week.

How do you feel about adding a few tablespoons of oil to the motor housing, to help with transferring heat from the stator? It seems preferable to me to cutting vents, given that I'll be riding in salt and grit.

For clarification, does this mean you believe that power to the drive wheel is about .75 x 630w = 472 w? Just wondering because for a normal unfaired upright or even an unfaired tadpole it can take 650-700 watts to hit 30 mph.
JJ

That is pretty darn good! The PEBL is consuming around 30Wh/mile - aka 1123MPGe.

With a top speed of 20mph. ;)

Night ride by the beach:

https://www.youtube.com/watch?v=ICRT...ature=youtu.be

I came across this a few days ago:

Eternal Cycle Works - Info

I might do a second build inspired by v2.0 after I finish my initial installation. Mine would still be open-wheeled and narrow shelled unlike the designs in the link above, but use a lot of the same build-techniques and styling ques for the shell. Ecky, for your build, you could copy it outright or make your own tweaks to it. I remember reading somewhere that v1 allowed its rider a 19-20 mph average speed(including stops, hills, traffic, ect), fully human powered.

You could try it. I haven't powered this motor yet myself and I would first investigate where it was generating the heat. What kind of oil, how many kJ can that volume of the oil handle before it changes state, and how many J worth of cooling will it have per unit of time at given speeds?

If I were going to go through all of that trouble to concern myself with heat tolerance instead of just using it stock, I would install ferrofluid for purposes of heat transfer and absorption capability at the cost of a very tiny efficiency loss(if it's even noticeable). Then I would later drill holes and vents and install fans in all the necessary places(shown in that ES topic pertaining to this motor) so I could load more power into it without damaging or destroying it. This motor has been pushed to 10 kW for seconds at a time by third parties on that site, and that potential for peak power output is probably the biggest advantage to modifying this motor. It makes a lot less heat than a MAC if what I've read is correct. I will eventually know first hand what kind of heat it makes under load when I get my pack installed(it won't be for at least a few more months as the body is still in progress).

Otherwise, I'd leave it alone. From the experiences of others, it appears to be reliable for 1-1.5 kW for the duration of the pack discharge depending on ambient temperature, and 4 kW for a few seconds at a time. Will that little bit of oil on the case make a difference in that? With a high enough phase current limit set by whatever controller you have, 4 kW could possibly move you like a sport bike from 0-10 mph, haul ass like a decent performing car from 10-20, and still perform like a slow, out-of-tune car from 20 to top speed. If you need any more performance than that, you might as well go all out and rebuild the motor for it if you're going to be worrying about excess heat at all.

10 kW is perfect for a few seconds at a time when/if you get your motor capable of tolerating it. Before it's ready to overheat after taking off from a stop, you're at the max vehicle speed limited by gearing, voltage, and/or rpm. Then you're load is whatever is necessary to retain cruise at that speed which may be at or under the motor's continuous power rating if you get the aerodynamics right(unless you're going uphill, in which case you have to pay more attention to heat).

The power to drive the rear wheel is whatever it is. Ecky gave us preliminary power versus speed figures. 75% electrical to mechanical power conversion is probably a close guess to its actual real-world operating point at Ecky's quoted speeds. Coast-down tests can be used to get Crr and CdA numbers, and the computer used can provide more data pertaining to motor/battery energy consumption(as long as it is properly calibrated).

472w is a good guess for rear wheel power requirements based on his numbers. In contrast, the leafbike motor plus controller/battery should be around 85% efficient combined for the same operating point, meaning 550w or so would be the new estimate for power consumption. The estimate is crude and doesn't account for increased inertia losses or different bearings or the fact that the MAC can be mechanically decoupled whereas this Leafbike motor cannot. Nor does it account for any accessory loads that could also be draining rhe traction pack.

However, with a sufficiently aerodynamic faring, much greater efficiency is possible than what Ecky is quoting. There exist electric velomobiles that can cruise at 30 mph on flat ground for under 10 wh/mi.

After playing with GRIN's motor simulator, I'm thinking the difference in efficiency won't be nearly so large as you're predicting. It appears that the MAC will actually have an edge in efficiency while getting up to speed, and in some hill-climb scenarios, while the Leaf motor pulls ahead in cruise, but only slightly. I'll get regen braking with the Leaf motor, which may make up for the difference in efficiency in a stop and go "city" drive cycle.

Climbing a hill WOT:

https://i.imgur.com/wBGqISn.png

https://i.imgur.com/liWedU3.png



48km/h cruise:

https://i.imgur.com/jPLwp5n.png

https://i.imgur.com/xftFabM.png

Subscribed! (Delayed response...)

Putting fluid in a motor casing is not an industrial practice. I would suggest mounting metal wheel disks to the sides of the motor as fin area and reducing drag on the wheel assembly. Heat comes from the current in the windings under load.

A lot on Endless Sphere recommend it for geared motors, because there's virtually no contact between motor and casing. The sides of the motor can be cool to the touch, while the windings are in excess of 100c.

Either that, or drilling holes in the sides and installing fans.

Probably won't matter until after winter though. :turtle:

EDIT: Some interesting reading on potted motors, ferrofluid and oil cooling:

https://endless-sphere.com/forums/vi...06995#p1106995

Interesting. I didn't run a simulation with this tool with your parameters and was taking a guess based on earlier simulations I did using this tool where I used parameters suited for my requirements when I was initially considering a MAC(I was going for fastest overall 0-30 mph acceleration time possible with the controller I planned to use, while allowing a 50 mph top speed, which at 30 mph cruise, greatly hurt efficiency). The increased thermal efficiency for your parameters versus what I assumed they would be makes a big difference. However, I am curious if that simulator factors in losses from the composite gears, or just spits out the motor's efficiency by itself based on what it is calculated to be?

So based on your simulation's parameters for 30 mph, 83% eff for the MAC, 87% for the Leaf. Your preliminary 21 wh/mi at 30 mph figure is drawing 630W from your pack. At an assumed 5% electrical losses in non-motor parts(controller, wiring, ect), assuming 83% efficiency for the MAC, we get 497W at the wheels to move at speed. The Leaf motor at 87% efficiency should lower the power from the battery pack to 601W or reduce your energy consumption at a steady 30 mph by 1 Wh/mile, using the same methodology that I used earlier to estimate a larger difference between the two.

1 Wh/mi is still noticeable. There's a lot more to be gained by working on the aerodynamics though.

The simulator seems to be over-estimating your power requirements when comparing with your preliminary data, but then again your vehicle's actual CdA/Crr parameters are still unknowns and that will modify the simulator's performance, temperature, and range calculations in an undesired direction versus the real results.

I think the regen will be a huge asset in stop and go city driving or for reclaiming range while going down long hills when not wanting to exceed the speed limit.

Please, get some videos showing us your acceleration, and have a speedometer of some sort in the video as well when you do it so we can see it climb up in speed!

I think 0-30 mph in under 7 seconds is easily within reach right away for you with the Leafbike motor(and much faster is possible later). You will want to modify the dropouts with torque arms before you abuse them too much. These motors(as well as your MAC) can twist them, and then you're in the market for another trike, so be careful.

I began building and modify DC permanent magnet motors when first racing slot cars in the 1960's. To stabilize the armature windings at high RPM (+100k ) we would apply epoxy. The field magnets were held in place with spacer clips and the motor shell was used to concentrate the the magnetic field. Air flow was primarily axial. Automotive alternators and most industrial motors I have worked on in robotics follow this practice.

Heat conduction of the stator or field windings passes through the casing to the external heat sinks. Axial air flow may be increased by a fan blade mounted on one end of the armature. This is common on electric drills. Since magnetic field strength follows the inverse square rule great care is given to minimizing the air gap. Dynamically balancing the armature and installing high quality bearings reduces the range of motion of the armature in the air gap.

The inquiry becomes how best to minimize temperature rise in the windings which reduces current flow and shortens motor life through the breakdown of wire insulation. The other question comes in discovering the relative merits of ferrofluids in the air gap. Does the improvement in field strength outweigh the viscous drag as the air gap is reduced?

Scored these on eBay recently:

https://i.imgur.com/uakOAOI.jpg

They look a bit better than the coroplast and black duct tape covers I had before. :rolleyes:

Waiting on some new tubes before I remount them.

Spokes on the inside, or is the outer shell the wheel itself?

They're plastic covers which go over the spokes:

https://ssli.ebayimg.com/images/g/86...fM/s-l1600.jpg

I was looking for smooth wheel covers and originally finding a lot of expensive carbon fiber stuff for racing bikes, usually made in Germany or Japan, when I came across these old plastic ones people used to put on BMX bikes.

Have you made any updates to this project?

I'm about to convert mine to electric. It is on its second body shell. I'm curious to see if this one has evolved.

None, sorry! I've had very little time other than to work on my house.