Motorcycle class Hub Motor

[liveforphysics]

The winner, team Agni, posted a time that beat the 50cc record for the track set in 1966.

"Barber led the TTX GP from the early stages and crossed the line at the end of the one lap race with a time of 25m 53.5s, 87.43mph, just inside the 50cc TT lap record of 86.49mph, set by Ralph Bryans in 1966.

Barber, riding for Team Agni in the Pro class, finished a whopping 3m 11.43s ahead of second place Thomas Schoenfelder on the XXL Racing Team entry. Schoenfelder clocked the fast time through the Sulby speed trap of 106.5mph, the only rider to exceed 100mph.

Barber said: “I’m really pleased for Cedric (Lynch – the creator of the winning machine who has been building electric bikes for competition since 1981). He really wanted to beat that 50cc record and we’ve done it."

Please tell me, what's so great about that? Aren't these electric bikes supposed to be much faster than a 50cc equivalent (50cc's only get, what, 15 hp at max?)?

It could be up to 15bhp if it were on Methanol, but this pre-dates power-valves found in modern 2-strokes, so it would be a very very narrow little peaky powerband if it had 15hp. With the right gearing selection, that does work fine though.

How was it so fast on such little power you are asking? Because those bikes were basically bicycles fitted with an engine, which let them all be under 100lbs, even with garbage 1966 metals and crap drum brakes. You get anything under 100lbs, and you give it 15bhp, and it goes pretty damn well. My E-bike is a perfect example of that.

Here is a picture of a 1966 50cc GP racing bike to give you an idea of how bicycle-ish they were. The men that raced on crap rubber with crap safety and crap metallurgy were all heros.

But, now I'm getting this thread way off topic... Sorry!

 

[Jeremy Harris]

Something that is, more or less, relevant to this thread on Marks new hub motor is the comparison between the way power and torque is delivered by electric and IC motors, though.

Race vehicles spend most of their time accelerating as fast as possible or decelerating as fast as possible, with little time overall spent just going flat out. A road vehicle typically uses about 15% or so of it's maximum power as an average, a race vehicle might use around 60 to 70% as an average. If you want to get maximum acceleration from any vehicle, then you want to keep the engine delivering maximum torque, as that's what's going to make it accelerate. A race vehicle may never get to use it's maximum power at maximum speed, I know I used to gear my old race cars to just run out of revs at the end of the fastest straight, so the gearing was then low enough to allow good acceleration out of the corners, where it really counts. The maximum power figure is really a bit misleading for an IC engine, in my view, as what's needed is a combination of enough torque and the right gearing to get best acceleration (I know that power = torque x rpm, but once past peak torque on the rpm curve there's nothing much to be gained in terms of practical performance, even though power is still notionally increasing when measured on the dyno).

Unfortunately, IC engines only deliver maximum torque over quite a narrow rpm range, and worse still, produce little torque at low rpm. We get around this by using gears and a clutch, but gears are always a compromise, as a high peak power engine will have a narrow power band and need more gears, which causes more losses and more lost time from changing gear.

Electric motors have the potential to be better, as they give maximum torque at zero rpm, and maintain high torque levels through a fairly wide band. The net result is that, although the power figures are lower, the effective performance when accelerating can be similar, or even better. Obviously losses are lower with no clutch or gearbox (or perhaps fewer gears), but there are also no losses from the compromise in gearing that a narrow power band IC engine always has. Race performance is ultimately down to how much of the acceleration time can be spent with the motor delivering the highest possible torque levels. Maximise this and you get a race winner, even if the max power as measured on a dyno looks to be lower than that from an IC engine.

There's also another factor that's more to do with the rider/bike interface. To get good performance from an ICE bike needs a lot of skill, primarily in keeping the engine as close as possible to that magical maximum torque level at the rear wheel. This means juggling gears and throttle continuously to make up for the non-linear characteristics of the engine. Substitute an electric motor, with it's wide torque band, and the rider workload drops, leaving more capacity for picking the right line and thinking tactically on the track.

Jeremy

 

[liveforphysics]

All very true Jeremy. I also am always setting my cars and bikes up to have gearing limited top-speeds to maximize power off the corners 4.9 FD's rock

I agree it's definitely an advantage to never need to interrupt power delivery to shift. It's also neat to have torque from zero RPM, but due to having no gears to multiply torque, and efficiency in the single digit range, I'm definitely going to have to give the launch advantage to anything with gearing. I'm able to apply all the torque to the rear tire that the bike can handle without flipping over with about every track oriented bike on earth for at least 0-25mph (or 0-120mph for my GSXR1080).

As far as the bike/rider relationship, I honestly don't think it distracts the rider in anyway at all. Maybe it's all the super short gear ratio shifter karts and dirtbikes I raced as a youngster, but the clutch/shifting/gear selection process never even enters my head when I'm racing. The foot and fingers just do there own thing, totally by themselves, and always manage to keep the revs locked right in the fat of the powerband when I want to accelerate. My mind is left to things like watching for corners where the guy in front of me looses ground, to look for opportunity to jump on his weaker line choices to get me the pass. lol, I think more of my brain is fixed in watching the guy in front of me for slow/weak spots rather than even racing my own bike/kart.


I have boatloads of respect for you Jeremy, and you too Mark, and I wish nothing but the best for you in racing Jay!
You guys all are aware of the strengths and weaknesses, and now it's time to focus on how to milk the available strengths to the fullest, and look for ways to mitigate the weaknesses. I would love nothing more than to see an electric bike out kicking-ass on the track, and it sounds like Jay is a great choice for a rider to make that happen.

Best Wishes,
-Luke

 

[Mathurin]

It could be up to 15bhp if it were on Methanol, but this pre-dates power-valves found in modern 2-strokes, so it would be a very very narrow little peaky powerband if it had 15hp. With the right gearing selection, that does work fine though.

How was it so fast on such little power you are asking? Because those bikes were basically bicycles fitted with an engine, which let them all be under 100lbs, even with garbage 1966 metals and crap drum brakes. You get anything under 100lbs, and you give it 15bhp, and it goes pretty damn well. My E-bike is a perfect example of that.

Here is a picture of a 1966 50cc GP racing bike to give you an idea of how bicycle-ish they were. The men that raced on crap rubber with crap safety and crap metallurgy were all heros.

But, now I'm getting this thread way off topic... Sorry!

18 hp here:
[youtube]EAx1KUIb9Zo[/youtube]
but it's an 50cc modded to 70cc, dosen't count. However the dude recons with better fuel and no air filter they could get 21 bhp and scaled down to 50 cc that would be 15bhp, seems to confirm what you're saying.

Also this video is awesome, building of an euro style 50cc race bike:
[youtube]wvjvWoORCos[/youtube]

 

[markcycle]

Yesterday I did a 20 mile ride on my test course. All I do is repeat this same course time and time again so I know what to expect and can compare even small changes.

I can say The newly designed 602 does not have a heating problem using a 220 amp (1minute) 100 amp continuous Kelly controller hottest rotor temp reached 52C, hottest stator temp about 80C. This was with using Regen on every stop and playing a little cat and mouse with a Ninja 650 (Female rider) traffic can be a wonderful thing sometimes.

The changes I'm using with this round of testing on the 602 wheel, are 110/90-18 tire instead of a 100/90-18 tire. This was a mistake and will be ordering a 110/70 tire for the wheel.

Acceleration:
from 0 - 15 MPH she feel a little sluggish though I'm having no problem staying with traffic, but I find myself at full throttle till about 15 MPH after that you can feel her take off (if not I would be kissing a bumper)

Top end
Haven't found it 65 MPH with head room. I'm going to take her on the parkway today and explore her top speed, since I no longer fear overheating its time to start this round of testing.

Efficiency
Not as good as the 603 because I am spending more time in the motor's lower RPM inefficiency zone. Being a high RPM motor (compared to the 603) with a unexplored top end at over 70 MPH stop and go driving pays a small efficiency penalty of about 10 watt-hour/mile more compared to the 603. I was still staying around 100 watt-hour/mile. For the 603 that number would be 90 watt-hour/mile for an equivalent ride.
I can go along at 40MPH @ 28 to 33 amps flat road little to no head wind

Can we please keep the thread on topic, thanks

Mark

 

[liveforphysics]

I've very pumped to see the topspeed! Let us know voltage and current when you reach the topspeed. I'm all ready hunting around for a feather weight BMX to chop up and mount this hub in. It sounds like fantastic and silent fun!

Great work!

-Luke

 

[Jay64]

Can we please keep the thread on topic, thanks
Mark

That other info is cool, but please start a new thread on it.

Mark, do you have thermal sensors in all the motors, or is that a mod. If it is a mod, will it still be in the motor that we will be testing at the race track? Obviously I think it will just be a blast to get the bike out on a track and have fun on it, but I really would like to use it as a serious testing session too. I want to try to record as much info as possible. Do you think 144v would be decent? That is the biggest controller I can find at Kelly. Would 40ah be good for testing or would you require more? I'm working on getting more batteries ordered up real soon here. Trying to get as many as I can, but not sure if I can get more the 40ah worth for right now.

 

[grandmasterE]

Great project!

I'm curious to know how the wheel is held from sliding back and forth on the 'inner' axle. Is the axle stepped, or do you use spacing 'sleeves' on the inner axle?

I'm thinking of a single sided application for this, and with the 60 mm inner axle it may be ideally suited.

What is the final weight? What gauge spokes, and what spoke count does your final version use? Are you able to mount larger rims (26")?

Again, great work!!

E.

 

[markcycle]

Can we please keep the thread on topic, thanks
Mark

That other info is cool, but please start a new thread on it.

Mark, do you have thermal sensors in all the motors, or is that a mod. If it is a mod, will it still be in the motor that we will be testing at the race track? Obviously I think it will just be a blast to get the bike out on a track and have fun on it, but I really would like to use it as a serious testing session too. I want to try to record as much info as possible. Do you think 144v would be decent? That is the biggest controller I can find at Kelly. Would 40ah be good for testing or would you require more? I'm working on getting more batteries ordered up real soon here. Trying to get as many as I can, but not sure if I can get more the 40ah worth for right now.

Every motor will get a Temp sensor, the one Kelly recommends and I have a chart to translate resistance of the thermistor to temperature, I will help you with this, no worry.

144 volts (is that a 4P_44S pack) will work out very well, we'll just need to be sure our voltage sag under load isn't going to hurt our performance. But that's what testing is for. 40AH; well it depends, if we pull 6 -7C on the batteries what will be the voltage sag. Can headway cells do 10C Peak and 5C continuous. This is what we need to know, I think Doc has the data we need. We need to hold the voltage at 10C to 130 volts or so.

Mark

 

[markcycle]

Great project!

I'm curious to know how the wheel is held from sliding back and forth on the 'inner' axle. Is the axle stepped, or do you use spacing 'sleeves' on the inner axle?

I'm thinking of a single sided application for this, and with the 60 mm inner axle it may be ideally suited.

What is the final weight? What gauge spokes, and what spoke count does your final version use? Are you able to mount larger rims (26")?

Again, great work!!

E.

Spacers
Single sided will put excessive load on one bearing. I can make a custom cover for single sided app. but I do not recommend using the wheel as is. Spokes are 9 gauge 36 spokes Larger rims can be mounted but remember the wheel size is equivalent to changing the gear ratio of a chain drive bike. Bigger wheel more speed less torque. I would not go to a bigger rim. I worked hard at getting the right balance of speed and torque for a 350 to 400 pound bike using a 17 or 18 inch rim.
Finale weight is about 35 to 40 pounds without the tire.

Mark

 

[markcycle]

First Parkway Run
Images are the data
113 watt hour/Mile average for the run
Motor barely got warm not even worth measuring a total surprise. I think the tire got hotter.
My son was behind me (support car) he was impressed with the acceleration its hard to give exact numbers, from the GPS data which may be slow because of sampling rate under 10 sec from 8 MPH to 57 MPH.

Limiting factor was the controller not in terms of amps but above 65 the controller started a vibration it was a weird feeling it was as if the halls signal was to fast. I have to talk to Steve at Kelly controllers. I might need to change the halls supply from 5 volts to 12 volts.

Mark

 

[todayican2]

Ok, dumb question, I know but ill ask anyway...

If I were to use this as part of a hybrid, with an i.c.e. driving a sprocket mounted to the hubmotor, brake disk on the other side.

when the i.c.e. is driving the hub, its independant of the electric motor right?

so for example, if the ice is driving and no amps being applied the "electric parts" are seeing 0rpm?

 

[markcycle]

Ok, dumb question, I know but ill ask anyway...

If I were to use this as part of a hybrid, with an i.c.e. driving a sprocket mounted to the hubmotor, brake disk on the other side.

when the i.c.e. is driving the hub, its independant of the electric motor right?

so for example, if the ice is driving and no amps being applied the "electric parts" are seeing 0rpm?

Well yes and no
The electric parts so to speak see the voltage the motor creates because the motor has become a generator. To do it correctly you need to put a small current through the controller into the motor to eliminate cogging drag or the gas engine will see a small increase in load due to this drag. Maybe not a big deal but it needs to be understood within the context of your design.

Mark

 

[Jay64]

I'm thinking of a single sided application for this, and with the 60 mm inner axle it may be ideally suited.
E.

YES!!! Mark, I believe I PMd you about this very thing at the very beginning. I want to work with the current design first, but eventually I would love to do that single sided project too, with another motor if we can modify it.

 

[Jay64]

Apparently there is some delay in getting the Headway cells. I am still working on getting them, but I guess it is out of our hands as far as when the group buy comes in. I am looking at getting 180 cells for the 144v40ah pack. Should be about 119 lbs of batteries. This motor does have regen right? I like to use a LOT of engine braking when I am racing and I would like to work on a technique I'm thinking about for use with electric regen.

 

[markcycle]

Apparently there is some delay in getting the Headway cells. I am still working on getting them, but I guess it is out of our hands as far as when the group buy comes in. I am looking at getting 180 cells for the 144v40ah pack. Should be about 119 lbs of batteries. This motor does have regen right? I like to use a LOT of engine braking when I am racing and I would like to work on a technique I'm thinking about for use with electric regen.

Sure does use Regen I use it for almost every stop. Regen is a controller function not a motor function. All the Kelly Controllers have the Regen option.

Mark

 

[johnhead@frontiernet.net]

Hello Mark,
For those of us who would prefer not to reinvent the wheel so to speak, would you be able to list the specific components that you are using in your conversion besides the motor? I would like to know what specific other items that I will need to order. Throttle, controller, switches, guages, battery chargers, BMS, etc. A schematic would also be appreciated. I probably won't need this for a few month's however I'm sure many of your "clients" would also like this information?
Regards
John

 

[mad_as_hec]

I have been closely following this one. Bought myself a HONDA VT250F non-runner this week for $100. Keen to find out final details on this project including $$$. And postage to Australia.

 

[markcycle]

Hello Mark,
For those of us who would prefer not to reinvent the wheel so to speak, would you be able to list the specific components that you are using in your conversion besides the motor? I would like to know what specific other items that I will need to order. Throttle, controller, switches, guages, battery chargers, BMS, etc. A schematic would also be appreciated. I probably won't need this for a few month's however I'm sure many of your "clients" would also like this information?
Regards
John

Most of this has been talked about in the thread but for review
Throttle: I prefer Magua but any Hall throttle will work with the Kelly controller
Controller: Kelly Controller any brushless model depending on voltage and current your system is set to run at I'm using a KBL12251H,24-120V,250A,BLDC Controller/With Regen

Switches: ? used what came on the Lifan motorcycle, really didn't add any.
Gauges: The CA from ebike.com
Battery: This decision is all yours. I've used HI-Power and Thundersky. I have had good luck so far with both brands but my performance may not predict your results.
BMS: Not my area of expertise
Battery charger: I'm using thundersky chargers but again I'm not in the charger business and haven't tested enough of them to form a educated opinion.
schematic: Connect the motor as per Kelly Controller connection diagram. I will translate the phase wires and Hall wires to the Kelly controller diagram. For other controllers it may be trial and error method to getting the motor connected correctly.

Mark

 

[Jay64]

I have been closely following this one. Bought myself a HONDA VT250F non-runner this week for $100. Keen to find out final details on this project including $$$. And postage to Australia.

That will be an excellent bike for a conversion. Kinda small chassis, so would probably be perfect for this motor to leave the chassis room for batts.

 

[markcycle]

I now have over 50 parkway miles on the Lifan with the MHM-602 motor.

Here is some data
Top speed 75 MPH @ 94 volts and 100 amps (head down behind windsheild) 65MPH sitting upright
Power usage: 100 to 112 watt-hour/mile
Motor temperature after 12 miles of continuous parkway miles air temp about 70F
Outer rotors case temperature 57C
Stator temperature 82C
No thermal runaway observed
No extra cooling sealed motor

Looking forward to testing the motor on the track with Jay64 riding at 200 to 300 amps, I'm hoping for somewhere between 100 and 140 MPH depending on how slippery the bike is.

Mark

 

[Jay64]

Mark, that just brought something back to my mind from a while back. I know this really eccentric older guy that build racing hulls for boats and does a lot of electric builds. I had spoken to him a while ago and he wants to help me create a very slippery set of bodywork. He claims it will greatly increase the range of the bike. I'm sure it would help for top speed too. However, I just don't know how well it would do in the fast left/right transitions. When you have a bike that likes to cut straight through the air, you end up not being able to change directions too well. I don't want to get into all of that in the first test, but it is something that could be done in later tests.

I'm stoked to hear about the 75mph numbers. I'll try and work a double draft at some point just to give some crazy numbers. 8)

 

[todayican2]

"Top speed 75 MPH @ 94 volts and 100 amps (head down behind windsheild) 65MPH sitting upright"
WOW, call me crazy, but that seems like a huge deal (Said the guy who wants to build a fully faired trike)

 

[RoughRider]

that are very good numbers...75mph is something you can work with...

 

[Jeremy Harris]

My other major interest is designing aircraft, so I can offer some insight into reducing drag that might help.

Aerodynamic drag is proportional to the square of speed, but power needed to overcome it is proportional to the cube of speed. This means that doubling the speed means four times the drag and eight times the power. Even small reductions in drag make a big difference to the power needed for any given speed.

Aerodynamic drag is linearly proportional to the frontal area and linearly proportional to a thing called the drag coefficient.

Frontal area can be reduced by adopting the lowest profile you can, there's not much scope to do more than this on a normal bike.

The drag coefficient is dominated by the shape of the back of the bike. Anything that makes the back end more of a smooth teardrop shape will reduce it. Smoothing the front will have an effect, but less than making a nice tidy rear end.

Handling in crosswinds is the problem. The moment you add side area behind the bike centre of gravity you get a situation where the bike will want to turn into the wind. The reverse is true for side area that is in front of the bike C of G. Teardrop fairings tend to have more side area at the back and consequently make the bike feel odd when it hits a crosswind.

One way around this is to use a cut-off rear fairing that deliberately creates a converging turbulent area behind the rider. Whilst not as effective in reducing drag as a long teardrop fairing, it's better than the sharp transition that occurs at the riders rear end. There are good examples of this in current fast bike designs, plus there are loads of cars that have cut-off rear ends that show how this works. The Prius shape is a really good example, as are most of the current crop of fuel efficient hatch backs.

I'd suggest that a nicely faired back-protector, plus a broad backside rest and tail fairing, coupled with a low riding position that allows the elbows, knees and head to be well tucked in, might give a worthwhile benefit. A light half fairing at the front to smooth the transition would probably be worthwhile, too, and would add some side area up front to make the bike feel more balanced.

Hope this helps, I look forward to seeing the results. I reckon this motor will just transform motorcycle conversions once it's fully sorted.

Jeremy