40kw E-Taljet Dragster. Thoughts?

[MitchJi]

Hi Matt,

Are you planning to produce more units based on the italjet dragster? If so where are you planning to purchase them?

You might want to test the controller with a 3220 and if it works better or more reliably than the HV160, request/demand that Astro make similar controllers that are sized and priced appropriately for their 32xx series motors.

I am against a hub motor for this for the pure reason of unsprung/rotating weight.

Matt

Luke did you mean he could use this hub motor, but not in the hub?:
http://endless-sphere.com/forums/viewtopic.php?f=3&t=41604

Over 65mph on 20s

weighs over 6kg less than I expected coming in at only 14kg

http://endless-sphere.com/forums/viewtopic.php?f=31&t=46898

Hubmonster 94% efficient 7kw rated $539 introductory price

The motor has a Kv of 16rpm/volt. I push 210A peak from a 20s20ah20c RC Lipo pack at get it up to 65mph on the highway with voltage sagging to about 75V. To be sure I was stating fact I just weighed myself and the bike with its current pack configuration and I weigh 265lbs (120kg), and the bike 122lbs (55kg). That doesn't include the 25lbs or so commonly in my backpack for the generally uphill ride home. I'm not into high speed on 2 wheels, but with good aero and higher voltage there's lots of unexplored speed, because it's a 10 pole motor with high quality thin laminating steel of approximately 3 lams per mm...

I've been using about $250 worth of controllers to run 15kw with a totally stress free system.

http://endless-sphere.com/forums/viewtopic.php?f=6&t=48399

HubMonster hits 107mph on the flats on page 6

Luke are you opposed to a simple single stage (one sprocket on the motor and one on the rear hub) chain drive/reduction? With a kv of 16 chain noise should not be a problem.

I seems like this has as much or more power than the big Astro for about 3k less (msrp), not counting the fact that the Astro will require a more complex reduction unit, which might add another $1k-$2k to the price? I realize you don't pay the list price but won't you gain more by prototyping a system that has a price that's affordable? Even if the power is equal the cost savings seem compelling, but what if the performance is better?

Capable of 15KW, the 4535 is the newest and largest motor in Astro Flight's lineup. This 12 pound motor is intended for applications where power-to-weight is of the utmost importance.

Model planes and helocopters? The Astro does weigh about 18 lbs less, but the additional reduction unit will probably add 4-5 lbs? On a 250 lb scooter, with the motor mounted as low as is convenient, will a rider even be able to feel the 14 lbs of additional weight? On your recumbent an additional 14 lbs would be huge, on a scooter I don't believe it makes any difference.

 

[etard]

Here's your competition: :wink:

These guys sink some serious $$$ into these Rucksters, there has got to be a market for your E-Jet

I find scooters to be the ultimate hooligan machine, I'm glad your decided to go this way Matt, it should be a grin maker for sure!

 

[recumpence]

Let me be clear,

This is not going to be what we are looking to produce. This first scooter is an exercise in what is possible with a scooter platform. A scooter is cheap to buy, cheap to insure, and small as well as light. This makes a relatively reasonable project to build. We will judge the market for high performance E scooters by building and showing it off.

The I C guys cannot comprehend the power available in a scooter size electric package. The most powerful Ruckuss I have ever seen used a 600cc maxi-scooter engine and was roughly 40hp. It weighted a friggin TON and was very long (it looked awesome, though). This thing should be way over the top, while being real life useable.

I am determined to use a small, light weight mid motor, not a hub motor. There are, indeed, applications for both. However, for this build, mid motor will reign. Oh, also, I am planning on single stage reduction (motor shaft to rear wheel with one toothed belt). That should minimize complication, weight, and cost. I know this motor is hugely expensive. But, again, I am getting my first test motor and controller at a deep discount for testing purposes. We would need to find a much less costly alternative for production. This scooter is merely a concept that I plan on riding daily. I want to be able to have ground pounding burst acceleration with excellent cruising range, very good handling, and wild looks.

Matt

 

[www.recumbents.com]

Matt, I think this project is an ideal candidate for the next generation of your drivetrain. This one could have a 2-speed Thudster style gearbox that could handle the torque and just be generally cool. That way you could get your off-the-line performance and a high top end speed. Your machine shop buddies need a new project anyway...

-Warren.

 

[motomoto]

Hi Matt, I am doing a project using the Astro 4535 motor also, but don't want to do a thread on it until it is done, kind of how Tench did his hub motor mid-drive.
Everyone has their own idea on your project and it just muddies up the pure intention and choices you have made.
You are a fantastic engineer and we need to find out what this new motor can do !!!
I am a fan of changing gear ratios where appropriate, but I guess if you have big enough power with the right setup, you don't need any gears.
I am sure that whatever you do, it will be a rocket either way.

 

[liveforphysics]

This one could have a 2-speed Thudster style gearbox that could handle the torque and just be generally cool. That way you could get your off-the-line performance and a high top end speed.

That is one band-aid for making the wrong motor choice work for an application. It sadly adds a bunch of non-active-material weight and losses that don't help your performance though.

I guess if you have big enough power with the right setup, you don't need any gears.

+1

This.

It's also inherently the highest efficiency capable option, as it involves the lowest designed-in losses.


If you are doing an efficient single-stage from that Astro to the wheel, you will likely be able to make a serious ripping monster scooter Matt. Even if the motor is a bit short of the radius side of things for my own personal performance tastes.

 

[motomoto]

I visited the Astro Flight facility last week, and the latest version of the motors have a lot more copper fill
plus I held one of the stator plates and was told it was .004 of an inch thick. Paper thin. I think that works
out to 250 plates per inch of stator length. The stator is 3.5" long so that's 875 stator plates in each motor. Wow.

The 4535 is a scaled up version of the 3200 series motor with ultra thin laminations for greater efficiency. It should be
holy shit fast.

 

[John in CR]

If the scooter starts at 120kg you're going to need 2 of those motors for fun performance. Anyone talking multiple gears or CVT's has just never ridden an adequately powered electric 2 or 3 wheelers, because the fun is laying on the throttle at any speed, not just from a stop, and with any kind of multi-speed it's all down hill after the launch. You're not talking about trying to stretch the performance of the smallest motor possible with a lightweight rider like Gwhy does with his.

Also, on a scooter, unsprung weight in the rear wheel is all but irrelevant because due to the small fat tire size they can never handle well on bumpy roads, and the spinning mass of the motor is certainly irrelevant with no clutch and an electric's ability to easily deliver more torque than you can use. The real advantage of using a hubbie on a scooter is the space it frees up for batteries. The Chinese make them by the millions that way, but what do they know?

If you want to make a high end scooter, the place to start isn't with an obviously heavy gasser. Maybe pick one up to copy, but spend the big effort making it lighter by making a carbon tub to replace the steel tubing frame, which also gives you the waterproof enclosure for the big battery pack needed for meaningful range. Then you can reap the weight benefit of your Astro drivetrain once you get the total weight down well below 70kg including a couple kwh of battery. That should put your 15kw will be in the fun range as long as it's not pushing a heavy rider. Saving a few kg with those great Astro motors is pretty pointless if you're putting it on a heavy pig scooter frame that's made of mystery steel tubing, which is why it's so heavy.

 

[liveforphysics]

I am excited Astro is taking the larger motor diameter path, and using super high quality lams and good copper fill etc. That is fantastic to hear, and Im excited to see it run, Im sure the scooter will be great.

I do agree with John though. Once you've experienced big slow motor torque, like 80mph power-wheelies on a bike with a longer wheelbase than a hyabusa, you just realise that this is finally what experiencing the pinical of drivetrain performance. There were tracks I never could use full-throttle on for the whole lap, because even at the end of the straight-aways it was still holding the front wheel up. Thats what I consider drivetrain perfection, the near-effortless ability to instantly control the amount of torque being sent to the rear wheel at whatever level you requested over the complete speed range you wish to operate inside.

Large radius slow motor power offers this, and it's stupidly efficient when the motor is done right. Deathbike's complete drivetrain is > 90% efficiency over the nearly the entire range I operate it in (dips below 90% past 55kW or so IIRC, and also at very light loads).

Big radius slow motors do things like make 400lbs objects accelerate with a crazy 0-60 and still keep you pulling to past a hundred, and not burn-up or mechanically fail. While still cruising around with higher efficiency than nearly any other setups.

I never thought I'd say it, but I am a serious hubmotor believer for hotrod EVs as well as commuters. The current designs still have many additional optimizations in design and a switch to premium magnetic materials before they are going to complete with the Astro or Deathbike's motor efficiency levels, but that doesn't mean it can't be done. It just takes someone willing to put like $300-400 into the large radius exotic cobalt steel lamination stack, and someone to spend all week winding it.

 

[recumpence]

So, let me get this straight;

When running a vehicle that is extremely light weight like a bicycle, a small mid-mounted motor is a wise choice because of the weight savings and weight centralization affecting such a light vehicle, but on a heavier vehicle the added weight of a hub is offset by the added room available in the chassis as well as the ability of a heavier vehicle (generally speaking) to handle even higher torque loads?

Please understand, I am not saying you are wrong Luke (and John). I have a ton of experience building bikes that will blow the tire loose and shoot the front wheel skyward at 60mph. But, I have never built a heavier electric street legal vehicle. I have also built some radical dirtbikes in my day as well as 100+mph RC cars. I have a decent understanding of power per pound, torque versus overall power, etc. But, again, I am green (pun intended :wink: ) when it comes to 150+ pound street bikes.

It sounds almost like the fact that a tiny two stroke is great for a light supercross bike, but a big block is best left to a 69 Chevelle? :wink:

I cannot argue with your experience in electric motorcycles, Guys. So, I bow to your knowledge. I guess the last item for me to point out is the extremely light weight of the scooter I am starting with. It is one of the lightest scooter chassis out there. Also, I am starting with an existing scooter, rather than build one from scratch, to save a ton of time. I do not want to do chassis R&D right now.

Again, I appreciate all the info, Guys. I really do. For this scooter, I am staying with this Astro because of the fact that I have a number of customers who have asked about it. So, I need to test it anyway. Since I have two projects to work on in my spare time (this Astro testing and building a prototype performance scooter), the two seem to dove-tail together nicely.

Matt

 

[Frank]

I'm assuming that the CVT you're contemplating has torque multiplication, right? Do you know how much?

 

[gwhy!1]

Hi Matt, there are lighter scooters but none of them look as good as the italjet frame 8) , but you are only talking 5kg difference so not a real biggy. the scoot I converted was 98kg wet weight before I stripped everything off that wasnt needed and it ended up around 65kg after the conversion but this was with only 10ah worth of 48v lipo battery ( and using the original CVT/swingarm which is maybe around 10kg) so hard to tell how much lighter it could have been if I made a swingarm for it ( maybe would have shaved off another 5kg total )..

 

[John in CR]

Matt,

You already know that power-to-weight ratio is EVERYTHING. The last place you want to try to save weight is with the motor when you're talking about high power. The reason is a larger motor has real advantages, and it's not only in reduced noise and greater torque at the motor. The biggie is the thermal mass and how they can better absorb the excess heat created by the intermittent loads of a vehicle. High efficiency motors are great, but just like any motor they still start at 0% efficiency at 0 rpm. Those motors are designed for high power, but they're really for high continuous power with great cooling air flow over and through them. They're great at that and the very high efficiency makes them able to run at even higher power, because they're making less heat to dissipate from the relatively small surface area of the small motor. Running at the upper end of the scale like that is where any motor is happy as a lark.

Get into intermittent operation like with a vehicle stopping and starting frequently or climbing a steep hill slowing down for traffic or sharp curves and accelerating out of the curves is where the tiny motors quickly run into problems. That's because their limited mass results in rapid spikes in temperature resulting from the excess heat created during acceleration. Assuming the same coefficients of heat transfer an otherwise identical motor of twice the mass will experience far less than half the temperature increase for 3 reasons:
1. It has twice the mass to sink the heat, so that alone would cut the temp increase in half.
2. It has much more surface area, so it can dissipate heat much faster.
3. It can inherently create more torque, so it will be taxed less for the same acceleration and create less heat to dissipate.

That's why with my 15.5kg ventilated motor I can gear it for a top speed of over 100mph on the flats, and without changing anything I can ask it to haul my 150lb+ bike with 250lb+ me up a miles long 20%+ continuous grade without burning it up. I'm not talking about easing up the hill. I'm talking about blasting up an extremely curvy climb blowing right by any vehicle also making the climb, accelerating past 50mph on straights that are long enough, and braking for the curves enough to get several % regen recovery on the way up.

I'm not saying don't do it with that high end Astro, because there's definitely a cool factor of great performance with what looks like a tiny motor. I'm just saying that things change a lot once you start having to push the big scooter loads around, so make a big effort in the weight reduction area. Also, they'll want range, so put a lot of effort into aero too, as well as consider reducing the top speed target. Lower gearing will have benefits all around, and you just don't see scooters riding around at high speed anyway. Quick and zippy is what they will want, and geared low enough even the big weight can leave the motor unstressed.

John

 

[gwhy!1]

Matt,

You already know that power-to-weight ratio is EVERYTHING.
High efficiency motors are great, but just like any motor they still start at 0% efficiency at 0 rpm. Those motors are designed for high power, but they're really for high continuous power with great cooling air flow over and through them. They're great at that and the very high efficiency makes them able to run at even higher power, because they're making less heat to dissipate from the relatively small surface area of the small motor. Running at the upper end of the scale like that is where any motor is happy as a lark.

Get into intermittent operation like with a vehicle stopping and starting frequently or climbing a steep hill slowing down for traffic or sharp curves and accelerating out of the curves is where the tiny motors quickly run into problems. That's because their limited mass results in rapid spikes in temperature resulting from the excess heat created during acceleration.

I'm not doing to start on about using the cvt again but I would like to point out that the rpm/load will be nearlly 100% constant when accelerating when using a cvt if setup correctly.

 

[recumpence]

Matt,

You already know that power-to-weight ratio is EVERYTHING. The last place you want to try to save weight is with the motor when you're talking about high power. The reason is a larger motor has real advantages, and it's not only in reduced noise and greater torque at the motor. The biggie is the thermal mass and how they can better absorb the excess heat created by the intermittent loads of a vehicle. High efficiency motors are great, but just like any motor they still start at 0% efficiency at 0 rpm. Those motors are designed for high power, but they're really for high continuous power with great cooling air flow over and through them. They're great at that and the very high efficiency makes them able to run at even higher power, because they're making less heat to dissipate from the relatively small surface area of the small motor. Running at the upper end of the scale like that is where any motor is happy as a lark.

Get into intermittent operation like with a vehicle stopping and starting frequently or climbing a steep hill slowing down for traffic or sharp curves and accelerating out of the curves is where the tiny motors quickly run into problems. That's because their limited mass results in rapid spikes in temperature resulting from the excess heat created during acceleration. Assuming the same coefficients of heat transfer an otherwise identical motor of twice the mass will experience far less than half the temperature increase for 3 reasons:
1. It has twice the mass to sink the heat, so that alone would cut the temp increase in half.
2. It has much more surface area, so it can dissipate heat much faster.
3. It can inherently create more torque, so it will be taxed less for the same acceleration and create less heat to dissipate.

That's why with my 15.5kg ventilated motor I can gear it for a top speed of over 100mph on the flats, and without changing anything I can ask it to haul my 150lb+ bike with 250lb+ me up a miles long 20%+ continuous grade without burning it up. I'm not talking about easing up the hill. I'm talking about blasting up an extremely curvy climb blowing right by any vehicle also making the climb, accelerating past 50mph on straights that are long enough, and braking for the curves enough to get several % regen recovery on the way up.

I'm not saying don't do it with that high end Astro, because there's definitely a cool factor of great performance with what looks like a tiny motor. I'm just saying that things change a lot once you start having to push the big scooter loads around, so make a big effort in the weight reduction area. Also, they'll want range, so put a lot of effort into aero too, as well as consider reducing the top speed target. Lower gearing will have benefits all around, and you just don't see scooters riding around at high speed anyway. Quick and zippy is what they will want, and geared low enough even the big weight can leave the motor unstressed.

John

Hey John,

Gotcha. Understood.

I totally see your point, John. At this point I am not changing the layout. I will say that you are probably correct about the gearing. Maybe I should hear it for 50mph and have tons of fun. I can tell you that two 3220s geared for 50mph is stupid fast and reliable. This thing is twice that power, like running four 3220s. So, geared for 50mph should make this thing an absolutely absurd mind numbing beast.

I do appreciate the education on all of this. You guys are all respected by me (Gwhy, Luke, John and others). I am a techable person, for sure. I am still stuck on using this motor for this scooter. But, I do see what you are saying about the big hub motors.

At any rate, I find it fascinating how much performance can be wrung from one of these small motors. That is part of the fun. Of course, everything has its drawbacks.

I agree about aero too. I have learned alot about the importance of aero doing mods on my hybrid car. Electric range is DRASTICALLY affected by drag. Gas engines are so innefficient that added aero drag makes little difference. However, with our efficient electric systems, aero drag has a substancial affect on range.

Matt

 

[John in CR]

It could turn out that gearing for 60 or more is fine. The limits depend on terrain and usage with the absolute worst case scenario for any 200lb+ electric scooter regardless of motor type being stuck in bumper-to-bumper stop-n-go traffic. Even going up to very low speed, the repetitive 0% efficient launches build up heat.

Your stuff is always top shelf, so I can't wait to see what you come up with.

 

[MitchJi]

Hi,

Very interesting conversation, I learned quite a bit, thank you everyone!

IMO at least four topics:
Is the Astro a good choice for this project for Matt? Not surprisingly the more completely Matt explains his reasoning and build details the better it sounds. When he stated he would use a single stage belt reduction I was convinced it was a good choice for his purposes. He has had inquiries about the $4k Astro motor and controller combo and has too much integrity to sell them without personally testing it. He will probably make at least $500-$1,000 on each unit he sells if it works out well which seems like a slam-dunk. If he is dissatisfied he can easily swap in another motor, by modifying his motor mounts and changing one or two pulleys.

The 4535's dimensions are approximately 4.5" diameter and 7" length and the motor weighs approximately 12 pounds.

Is the $4k Astro the best fit for an e-scooter build? Probably not. The main justification for the high price is the light weight. Unfortunately it's a solution to a non-existent LEV problem. Most people won't be willing to spend and additional $2.5-$3k to save roughly 10 lb's on a small e-scooter. The best fit for that would be an ebike where you want to maintain the possibility of pedaling. Unfortunately it's too long for most ebikes. If it fits in KMX trikes (like your Yellow trike) it would probably be a better solution than dual 3220's. It would either not require a reduction unit or at least the complexity would be greatly reduced. Similar overall cost and weight, hopefully a more robust controller and totally over the top performance . If Astro wants you to sell a substantial number of their 45xx form factor motors they need to produce them in a range of lengths like the 32xx series. 4510 (3210 equivalent), 4515 (3215 equivalent), and 4520 (3220 equivalent). The other thing they should do is to (assuming their controller is superior to the HV160) is to produce more affordable versions of their controller. Two ways to do that would be to ramp up production and to make some smaller versions.

Is there a much less expensive larger diameter motor that will deliver even better performance that weigh's within about 10 lbs of the Astro?
Probably yes. If you explore that option with your second italjet you might discover that there isn't any need for your more affordable production models to have reduced performance.

Does a hub motor make any sense in this application if you want a light weight, good handling e-scooter? I don't think so. My previous post about using John's motor (suggested by Luke) was based on the erroneous assumption that Luke was recommending it in a mid-hub configuration. I think John explains the advantages very well in his post, which I quoted below. But he overlooks the facts that with sufficient reduction the start-up load (current requirement) will be much less and the 0 rpm load will be much briefer. So for the small penalty of a few pounds of pulleys and a belt, and a very small efficiency loss the system will be much more efficient at 0-low rpm and you can run a much lighter motor, without having to resort to a boat anchor in the rear hub.

You already know that power-to-weight ratio is EVERYTHING. The last place you want to try to save weight is with the motor when you're talking about high power. The reason is a larger motor has real advantages, and it's not only in reduced noise and greater torque at the motor. The biggie is the thermal mass and how they can better absorb the excess heat created by the intermittent loads of a vehicle. High efficiency motors are great, but just like any motor they still start at 0% efficiency at 0 rpm. Those motors are designed for high power, but they're really for high continuous power with great cooling air flow over and through them. They're great at that and the very high efficiency makes them able to run at even higher power, because they're making less heat to dissipate from the relatively small surface area of the small motor. Running at the upper end of the scale like that is where any motor is happy as a lark.

 

[MitchJi]

Hi,

Another opinion on reduction versus no reduction:

http://endless-sphere.com/forums/viewtopic.php?f=30&t=14260#p359886

If you have any ideas on what you guys think an ideal e-bike motor spec would be, I'd like to hear it. We've done some research into the market, so we have some ideas; but I wouldn't mind hearing your opinion. Things like continuous torque, peak torque, and stall torque requirements as well as speed requirements would be interesting to know. Also, there are serious trade-offs between direct drive versus geared. In general gearing make the motor much smaller and lighter, and the reduction in motor weight is much more than the weight of the gears. Also if you are going for lowest weight solution a gearbox is almost essential. From the analysis I've done, direct drive only makes sense if you are going for very high efficiency during cruise mode i.e. >95% electric->mechanical energy conversion while running at full speed. (At 0 speed efficiency is always less, and the lightest way to get high starting torque will always be with a gearbox.

Other applications for direct drive might be extremely long life applications where the gearbox can not be kept lubricated; but I doubt that this applies to e-bikes.

Best regards,
Michael Ricci, P.E.
VP Engineering
Launchpoint Technologies

IMO single stage belt reduction like Matt is using is an excellent (arguably the best) solution. Efficient, quiet, simple, inexpensive.

 

[dcb]

I just wanted to weigh in on the cvt stuff while I was thinking about it.

Using weights and springs and helixes to program your transmission ratio is like using a carburetor in a sea of fuel injectors, only many times worse.

You need something like a throwout bearing to control the ratio (and maybe a controller if manually adjusting ratio is too difficult for you).

These motors make gobs of low end torque and hit peak power fairly early on, you basically want to hold the ratio at its very lowest until you hit peak power, then and only then do you adjust the ratio for best acceleration from that point forward (increase it as fast as you can without dropping peak hp, which relates directly to RPM at full throttle). Fairly simple control logic.

Like carburetors, I have no idea how to get it to do exactly what I want, maybe some detents and a lot more fiddling with spring tensions and weights and whatnot, throw an air bleed in there for fun. But I could sure figure it out with a microcontroller and a ratio lever, hell it can figure it out on its own (just needs the rpm @ peak hp).

And if you leave the centrifugal clutch in there instead of a solid coupling (or use the space/cover to put your throwout bearing thingie), I don't know what to tell you.