Supermileage Projects

[taylorp035]

I figured that I needed to put the "Treadmill Dyno" thread to rest, but since there are a handful of people that seem to enjoy my project, I decided to make this thread that will house all the new stuff, which there is plenty planned 

The old thread: http://fieldlines.com/board/index.php/topic,144470.0.html


After achieving an impressive 1011 mpg in sub-par driving conditions and a stock engine, this year we plan on taking 1st place.  We have the resources (well at least 10% of our competitors anyways) and the smarts, so there isn't much stopping us.  I figure this year's car had 1/2 the drag of the 2010 car, so I think we could of had 1200-1300 mpg if our last run was good.  I also believe that our porting and polishing actually lowered our 1,500-3,000 rpm efficiency, so we would of been better off with a stock engine (and the tests proved it...).

We are not yet 100% sure yet if we are going to build a new body, but there are many improvements to the various components that we will be doing anyways, starting off with this one:


Larger Version:  http://i55.tinypic.com/34o3e5e.png


This is going to be our new rear sprocket.  Last year's car used a 12:100 chain set up using #35 non-roller chain.  Max efficiency couldn't of been higher than ~94%, and probably closer to 90-92%.  This year, we are going with a 12mm wide carbon fiber timing belt, which has a max efficiency in the 98-99% range.  If we can add 5%, it will be well worth our effort.  We are going to build this sprocket ourselves at work (CNC shop) out of a 1" x 19" x 19" piece of 7075-T651 aluminum (good stuff )  It has 180 teeth, which allows for a larger driving sprocket on the engine, which is where we were loosing most of efficiency with the chain.  It will weigh 2.2 lbs and be 3/4" thick by 18" diameter.

The belt was just ordered, so it should be here by next week.  According to Gates, it can handle 300 hp at 4000 rpm on the driver sprocket..... yeah 12mm wide    This should be completed before school starts.


The next project is a really big one:  The Engine.  Obviously we can do better than a stock engine, and we can double our efficiency if we do it right.  Last year's car car had too much power, so much so that the car did a few moving burnouts during the fuel economy runs at 1/2 throttle (SMV's aren't supposed to get their back ends out  ).

So the solution:



Compared to a stock:




This a a briggs 3.5 hp 148 cc engine that now has a honda gx35 35cc piston in it and had been de-bored and chromed.  Due to the extra long stroke, we now have a 53.61cc engine, which makes it easier to have a super high compression ratio.  We have plans on making a 13.5 :1 compression ratio, which should raise the thermodynamic efficiency another ~10%. 

There have been dozens of plans to make a rocker arm set up of over head valves, but I have a different plan that is very different:  Rotary Valve Engine.

Now, you may have thoughts of 2-strokes and sleeve valve engines running in your head, but this is totally different.
http://home.people.net.au/~mrbdesign/PDF/AutoTechBRV.pdf  - F1 team that helped develop a modern version of this.

I came up with this idea all by myself, but it took a friend, my brother, and I 6 weeks to find another engine like this.

Basically, it has a rotating pipe that acts like a valve.  It has several advantages, some of which include less air flow issues, less moving parts, shorter engine height (big one for us), high compression ratios, sideways exhaust, less friction, and easy to build.  There have been people who have made their own versions and once guy actually got better mpg's with the same block and had slightly more hp.  He ended up putting 30,000 miles on the engine.






Sealing obviously is the hard part, but I am very confident that we can overcome this, especially with 3,000 psi Teflon and carbon rotary shaft seals (something others haven't tried).  The pipe will be the diameter of the socket in the photo.  The spark plug is shown too.



Lastly, there are some major plans to do a better job with the treadmill dyno this year, starting with chucking it up in a CNC machine with shaft power and rpm readouts.  I also want to get a timing belt on it so there is less issues with that.  Last year's tests were majorly skewed by the limited power of our Bridgeport, so the RPM's were off.  Maybe direct drive if we are ambitious straight from the treadmill motor to the engine is in store.  Our new room has a high temp exhaust fan, good for 1500F, so engine testing will be a piece of cake ( especially while classes are in session ).

[Tritium]

I like that compression ratio. I would like to try one of those engines on my homebrew alcohol fuel. 

Thurmond

[Madscientist267]

I'm thinking that the socket is going to get in the way of the piston, and that valve isn't going to make contact with the seat...  

EDIT - And I have NO idea what the hell you are thinking by putting the spark plug there....

Steve

[taylorp035]

^^^ That's just for scale / designing purposes.  The valve and sparkplug set up will be much more like the drawings.

If you look closely at the bottom photo, you can see the combustion chamber.  The top of the piston is "dished" and the head is milled out a little.  We have about 4.2 cc's to work with, which makes designing really tough for an OHV engine.

The crank, piston, and rod is already done, all we really need is the head.

[Madscientist267]

You're looking at a smaller compression ratio... Gonna squeeze the excess out with a flatter head?

L-head is a BITCH to get the compression out of... Most of your ratio is in your valve clearance.

Steve

EDIT -

And DAMMIT *I* came up with the rotary valve concept... 20 years ago... I was SICK when I saw it on the web some time later....

WISH I had sent that one to the patent farm... 

Steve

[wdyasq]

EDIT -

And DAMMIT *I* came up with the rotary valve concept... 20 years ago... I was SICK when I saw it on the web some time later....

WISH I had sent that one to the patent farm... 

Steve

And you would have had the experience of an expensive trip to the patent attorney's orifice to show for it as as far as I can tell, nobody has used it commercially.

Even its' use in a hyper-mileage vehicle would probably be a use not governed by patents act as it is not commercial. 

I've looked at many 'patents' that should have been thrown out due to 'prior art'. Some of the art was mine. If I do good searches, I find others had similar ideas .... before I had mine, most of the time.

In engines, superchargers date to pre-1910. So does the overhead cam. I've worked on pre-1910 American engines that have variable valve timing.

I know I have seen rotary valving on 4 cycle "Otto" engines.  Or, I should say the drawings.  I can't remember when it might have been. I know I haven't researched internal combustion engines for near 30 years.

The really large advancements have come with computers being able to tune the engine many times a second.  CNC control of machining and assembly eliminates human error and inattention. Material science allows different seals and better .... cooling, sealing .... damn near name anything dealing with mechanical interface.

Working with the many variables like Taylor is doing takes dedication and desire. I wish him the best. When will the choice of going from spark ignition to compression ignition occur? When does the power to drive an injection pump for compression ignition (Diesel Cycle) become more efficient than a carburetor or fuel injection in spark ignition (Otto Cycle)?

I'm getting a headache just thinking about intake runner size and length and how they relate to valve timing and efficiency. What style exhaust system will work best? ....  And all of this stuff is interactive! Taylor should be grateful he can setup spreadsheets or computer programs that can help analyze a whole fist-full of these interactions quickly.  Just an exhaust calculation on valve timing, exhaust tube length and diameter, collector size, displacement, CFM and such took quite a while, even with a slide rule!

Well, good Luck Taylor ... hope you win a lot and learn more,

Ron

[Madscientist267]

Yeah, there's actually a use for them; they've managed physical manifestation at least once...

I can't recall exactly where now, but I think it was for a racing engine.

I don't know why the concept didn't take off with average everyday engines. Probably for the same reason that a Wankel engine was plagued with issues in it's early days. As much as the same, I figured (at the time) that sealing would be the downfall. Shrug - no idea.

It's a beautiful concept though. 1 moving part that replaces dozens (on a full scale engine). Not so much on a single cylinder. Came to me when I was fixing a lawnmower engine one day.

My biggest issue was that I had no capital to throw at it to design a prototype, but it worked on paper! And apparently in real life too! 

Yeah - here it is:

http://www.thetruthaboutcars.com/2007/12/coates-spherical-rotary-valve-system-let-the-rivet-counting-begin/

This wasn't quite how I envisioned it, but close enough. My prototype on paper was more or less a pipe that ran all the way across the head, rather than the individual 'valves' being obvious. But the concept is still exactly the same. 1 moving part. No springs. No rockers. No lifters.

Mine also used the same channel for both intake and exhaust, but I can see where using dedicated ports would be beneficial (in terms of timing, 'recirculation', and the like).

I'm not greedy. Would have sold it off for a cool mil or so. Whatever it took to pay for the patent, prototype(s), and a little profit.

Guess it just wasn't meant to be.

On a lighter note, Taylor, you've proven you have some serious talent there dude. I personally can say that I enjoy reading your posts. Lots of inspiration for all kinds of things!

Steve

[Bruce S]

Madscientist267;
 Could not have said any better other than when can I send my 5Hp B&S up to rebuilt into one of these 
There once was an article very well done in popular mechanics (I think) that had a guy building his own compression ignition for ICE units tried following it but missed some of the mags and now can't find the mag in my pile of old ones.  . Maybe this Internet thingy that is only a fad will help  .

Taylor, keep them coming they are very informative. and Again
Thanks for taking the extra time to post these.
Bruce S
   

[taylorp035]

When will the choice of going from spark ignition to compression ignition occur? When does the power to drive an injection pump for compression ignition (Diesel Cycle) become more efficient than a carburetor or fuel injection in spark ignition (Otto Cycle)?

Somewhere around 14 or 14.5:1 at least.  Some bigger schools have proven that anything past 14:1 can't be built.  For some reason, I want to say you would need 18:1 to do a good compression engine, but I have not found a good source for it.

As for the efficiency of the carb / injector,  our EFI system uses only about 30 to 40 watts, depending on pump pressure.  The fuel pump only uses about ~15 watts, since it is a tiny brushless motor.  We have the world's smallest fuel pump (about 2/3 the size of a roll of pennies).  Else, we would have to get one that is 6x larger, meant for a 4-wheeler, which uses a ton more power.

The EFI system is very good at getting the fuel to air mixture ratio correct, so that helps with efficiency.

20 years ago... I was SICK when I saw it on the web some time later

My design was used a good 60 years ago, but sealing technique have been improved steadily over the years.

CNC control of machining and assembly eliminates human error and inattention.

It is nice for making fancy curved combustion chambers.

I can't recall exactly where now, but I think it was for a racing engine.

I don't know why the concept didn't take off with average everyday engines.

There was an F1 team (Bishop rotary valve engine), that used this, but F1 decided to ban the design after 6 years of work on it.  For high reving engines (18k rpm), the rotary valve has some serious advantages.

I'm getting a headache just thinking about intake runner size and length and how they relate to valve timing and efficiency. What style exhaust system will work best? ....  And all of this stuff is interactive! Taylor should be grateful he can setup spreadsheets or computer programs that can help analyze a whole fist-full of these interactions quickly.  Just an exhaust calculation on valve timing, exhaust tube length and diameter, collector size, displacement, CFM and such took quite a while, even with a slide rule!

I wish I could use air flow programs and other fancy things to model an engine, but it would require someone who already knows it.

Our approach to this will be to build several intake pipes (and maybe heads) and test to see which ones are the best.  Since the pipe is relatively easy to build, modifications should be easy.  Last night, I found two 4" XL sized timing pulleys, perfect for a dyno modification.

You're looking at a smaller compression ratio... Gonna squeeze the excess out with a flatter head?

Maybe, but we already have two of the Honda pistons.  It seems like we have a decent amount of volume to work with (~4.28cc's).  IF it were 3 cc's, we would be in trouble.  If the piston were flat, we would save another 0.88cc's.


The ideal thermodynamic efficiency will go from ~50% to 60% because of the change in our compression ratio.  The extra 10% should go mostly into driving the crank, so maybe we can hit 15-20% efficiency, which would mean 2000-2500 mpg.

[taylorp035]

The belt has arrived     




It fit the wood model perfectly, which is quite impressive since there is almost no room for error with a belt like this.   The belt has some really weird qualities to it, starting off with the fact that it is has carbon fiber for the strands, but also that the teeth are made of this hard blue plastic, which is super slippery.  The teeth are specially designed not to rub on the sprockets, unlike normal square toothed timing belts.  It definitely looks like it was worth the $47.  On the large box it that it came in (fully uncoiled), it says not to kink the belt or hang it from a nail, for fear of breaking the carbon strands.



The rotary valve designing is going well.  We have built two prototypes so far out of wood on the 4 axis mill, but it is obvious that I made an error when calculating the angle on the new one.  I guess it is back to the drawing board.  Maybe I will get a photo of it on here once I make a good one.

[taylorp035]

Here are my current thoughts on how the valve timing should be.  The straight lines are the rotary valve and the curved lines are my measurements from the stock engine.  The relative height means almost nothing on the graph. 




And the full sized version here :  http://i56.tinypic.com/k99fn4.png   (1,213px × 717px,  91kb)


The valve overlap should be minimal, since the engine rpms are most likely going to be between 1500 and 3000 rpm.  If we do manage to get a higher gear ratio, it would still be under 4000 rpm.  The stock engine's efficiency curve was still peaking around 3000-3600 rpm, which was too high.  After we ported and polished it (the engine that was raced) the peak efficiency was around 4000 rpm, which means it was actually worse than the stock one.

The flat areas are a nice benefit of the design.  We don't yet have a good feel for how large the valve holes should be, but something a little smaller than a stock 148cc.  We could have intake and exhaust holes the same size as the GX35, since it ran up to 8000 rpm, but we are also up to 53 cc's.

[electrondady1]

this rotary valve tech, is a great area for your team to explore Taylor.
one source of ideas may be in supercharged or turbo charged cam profiles .
i'm thinking along the lines of the need to reduce valve overlap under those conditions.
in a book on turbo charging, (i basically assimilated the whole thing), they mentioned that "economy cams"are preferable to normal high performance cams  because of reduced valve overlap.
as well ,roller lifters are use in order to allow steeper cam profiles on high performance engines.
you can do away with this problem as there is no camshaft.
the rotary valve tech should allow you to open and close the valves much faster and have them stay open much longer.
 

[taylorp035]

i'm thinking along the lines of the need to reduce valve overlap under those conditions.

Right.  At such low rpm's, any excessive overlap is going to allow unburned fuel to go out the exhaust.

I am also concerned that my intake timing may be cutting down on my compression ratio.

I wish I could find some clear cut information on valve timing theory, but most of the stuff I have found is for high performance engines

The rotary pipe should allow for better mixing as it enters the chamber.  The air flow should be good too, but it really doesn't matter much at the low rpm's.  Friction, compression ratio, and the quality of the burn are much more important.


Yesterday, I saw the specs on University of Ottawa.  38 cc's, OHV, OHC, dual spark plugs, 16.5:1 compression ratio, water cooled cylinder sleeve!  I'm not even sure how you can fit all of that into something that small.  Too bad their car didn't get a good run in.

[Madscientist267]

I'm anxious to see how this thing turns out.

It was always a fascination to think about when I was pondering it. At the time, obviously, I had no idea that anybody had ever done anything like this (and it was before this here webernets)... I even called it something different actually - "rotary port"

Either way, I'm excited to see someone with the talent, tools, and resources available to them tackling this. Even if it IS an "older" technology, it isn't prevalent, and I can't really understand why. It seems so promising, and I somehow feel privileged to be able to watch it unfold (provided you keep posting)...

One thing that I had a problem with (and seems as though may be amplified for your design) is such a small displacement becomes a problem since the port must open and close in such a small space. The problem for me was figuring out how to make the 'valve' open and close quickly enough without increasing the diameter of the tube (ie vector issues). To solve this, I had an alternate design that only used one 'port', shared between the intake and exhaust, with only one port in the cylinder for the gasses to enter/escape from.

This allowed the size of the port to be larger (allowing more flow), and then the rate at which the port opens and closes can be effectively increased by adjusting it's shape.

This doesn't come without it's own set of problems, the first to mind for me was that then there would be exhaust recirculation (from the intake pulling the contents of the port's volume in the tube left over from the last exhaust stroke), and another is that it will complicate the head design to accommodate the valve timing. Another was what I called 'vacuum shock', basically the idea that the exhaust leaving the port will (can) cause a vacuum to form within the port as it leaves into the manifold or exhaust pipe (as the case applies). Basically the same thing that causes a two stroke to suck the exhaust out of the cylinder and pull fresh mix in. Only different.

Just some things to think about, I don't know how much use any of that will be, but they are some things that I thought about while I was looking at it on paper.

Like I said, I can't wait to see this unfold. It will be interesting to say the least.

Steve

[electrondady1]

don't know what restrictions/limitations they place on you.
my nephew has a high performance ignition , just one plug but it fires multiple times during the power stroke.

 you must use  a very lean fuel air mixture .
is detonation or preignition a problem ?
perhaps a bit of water injection to counter that and reduce the exhaust gas temperature .

i recall a news item i saw just once during the first energy crisis.
the guy had an injected Porsche.
he mixed gas and water 50/50 using a surfactant.(looked like dish soap to me).
the fuel became an opaque  white liquid .
he swore it doubled the mileage with out any loss of power.
here's a link to some thoughts on fuel emulsion
http://www.rexresearch.com/gunnrman/gunnrman.htm


since there is no valve clearance you could use a very small combustion chamber,
 wicked compression

if your exhaust is closed and the intake is open on the intake stroke why wouldn't you get a full cylinder.?
 



 
 

[bj]

  Just for info, another "rotary" valve is the Aspin valve.  Basically, it is a cone, which is the combustion chamber, with one port,
and it rotates.  The port uncovers the intake, exhaust, and spark plug at the appropriate time.  First used around 1937.  Advantage
was non-reciprocating valve train, and very small combustion chamber.  I expect life might have been the disadvantage at the time.

[Madscientist267]

It's just like the first generation Wankel issues; the seals just about have to be the weakest link.

Materials have improved since then though, and from what I can tell, the new (ie RX8) engines don't have the short seal life that the RX7's originally did.

The version I came up with was similar to the Frayer & Howard at the top of this page:

http://www.aqpl43.dsl.pipex.com/MUSEUM/POWER/unusualICeng/RotaryValveIC/RotaryValveIC.htm

Biggest difference is that the intake and exhaust on mine were 180 degrees apart, perpendicular to piston travel, and the rotor had a 90 in it instead of being straight.

Not that it really relates all that much to this in particular, but is still a rather interesting design; the Mellors version just past the half way mark down the page. I'm surprised this actually worked. I would have thought that at higher speed, the maltese crosses would have had a tendency to overshoot and cause binding/time jumping problems. But apparently, it worked despite various contempts to the contrary. Go figure.  

I guess it's true at this point, there isn't really anything else to invent. Just improvements to old inventions... LOL

Steve

[taylorp035]

Thanks everybody for the feed back.

I guess it's true at this point, there isn't really anything else to invent. Just improvements to old inventions... LOL

You have that right.  I really don't see any issues with our design right now, and I think any sealing issues can be solved with a combination of some brass rings, rotary shaft seals, bearing seals, and close tolerances.

since there is no valve clearance you could use a very small combustion chamber,
 wicked compression

We are going to design it with a thick head gasket (say 0.040") and then later try out a 0.020" gasket, which will raise our compression by quite a bit (0.606 cc's out of 4.2cc's!!!!).  We will build it to a 13.5:1 at first and then change the gasket, which should move it to about 15.8:1 ratio and then dyno test it to see the results.

i recall a news item i saw just once during the first energy crisis.
the guy had an injected Porsche.
he mixed gas and water 50/50 using a surfactant.(looked like dish soap to me).
the fuel became an opaque  white liquid .
he swore it doubled the mileage with out any loss of power.
here's a link to some thoughts on fuel emulsion
http://www.rexresearch.com/gunnrman/gunnrman.htm

Maybe this guy needs to take a basic thermodynamics course and prove his point.  As it says, you can run it hotter, but that's sketchy.  There is no way we could do that on our car, since it is all aluminum and the fact that the engine is going to be quite cold anyways since it only runs for 3 seconds.

if your exhaust is closed and the intake is open on the intake stroke why wouldn't you get a full cylinder.?

If you look at my graph, the intake valve is still closing while the cylinder starts to go back up right after the 540 degree mark.  The little bit of overlap should allow for the momentum of the incoming air to further raise the compression ratio (and increase the volumetric efficiency).  But I we get it wrong or wait too long, the cylinder will push the vapor right back out the intake valve.  Where the purple line hits the bottom, the cylinder probably is still ~ 93% of the way down.

One thing that I had a problem with (and seems as though may be amplified for your design) is such a small displacement becomes a problem since the port must open and close in such a small space. The problem for me was figuring out how to make the 'valve' open and close quickly enough without increasing the diameter of the tube (ie vector issues). To solve this, I had an alternate design that only used one 'port', shared between the intake and exhaust, with only one port in the cylinder for the gasses to enter/escape from.

From my measurements, it seems like my design opens and closes fast enough (based on my graph).  The idea of a single hole had crossed my mind too, but I then figured the exhaust would push air back into the carburetor and cause a mess.  I wish we had direct injection, which then you could probably get away with that.

[REdiculous]

Can you guys use an electric motor?...

There's a guy claiming 10wh/mi consumption @ 22mph, no pedaling, mostly-flat, on a full suspension mountain bike using a fairly common direct drive hub motor. Using CARB's numbers for regular gas that equates to 3260mpg. Using the EPA's numbers, 3370mpg.

iirc, the fuel you use is better so his numbers would be even better...near 4000mpg.

[taylorp035]

I have thought of this very hard before.... here are some of my conclusions:

1) Yes, you can do this.  It is a popular was to solve the main transmission problem (useable left hand drive bicycle hubs are extremely hard to get).  It also smooths things out.

2) An electric motor generator combo allows you to run the engine at it's peak engine efficiency.  But this is hard to justify when you run the engine for 3 seconds.  If you had a smaller engine, then there would be very large gains here. 

3) You could in theory get rid of the flywheel and use the generator as the rotating mass.

4) Peak efficiency for a stock engine like ours would be in the 3-5 hp range, if not more.  If I could put the power down, I would shoot for 5-7 hp and just beef up the connecting rod.

5)

here's a guy claiming 10wh/mi consumption @ 22mph, no pedaling, mostly-flat, on a full suspension mountain bike using a fairly common direct drive hub motor. Using CARB's numbers for regular gas that equates to 3260mpg. Using the EPA's numbers, 3370mpg.

iirc, the fuel you use is better so his numbers would be even better...near 4000mpg.

So he is using ~ 36,000 J per mile.  We were using ~ 10,000-15,000 J per mile.  BUT, remember you need to factor in the engine efficiency (9-25%) and the transmission (<98%).  Now punch the numbers into the calculator.  So his claims are accurate.... and this is a good demonstration for why I like the idea of home charging electric vehicles (the power plant + grid is much more efficient than you ICE).

6) The downside to all of this is the generating losses and the drive motor losses.  I figure the best we could reasonably assume would be an 85% generator and 85% drive motor, which equals .85*.85 = 72%.  That means you need to make up for that 28% in the engine and transmission.  And remember that it doesn't mean you need a 128% engine efficiency, it would be greater.  Some brushless motors could be around 92-94%...

7)  If we were allowed to use batteries to store our generated energy (using a joule meter to make sure you aren't cheating), then you could greatly increase your mileage.  First, you could have a smaller drive motor and the car could do a constant 15 mph (saving 10-20% for sure).  Also, you have less losses in spinning up the flywheel.  UNFORTUNATELY, I asked to do this, and the officials said no.  Eco-marthon has many other classes of vehicles for all kinds of cars.


Maybe more will come to me later.

[taylorp035]

Traction control is easier with electricity too.  You can also do 3 wheel drive.  Monitoring the power usage levels and performance is super easier.  With 3WD, you could put down closer to 7-10 hp   and make a real rocket ship.

Electronic braking would also be an option.

[zap]

...useable left hand drive bicycle hubs are extremely hard to get...

Like this?

[REdiculous]

1) Yes, you can do this.  It is a popular was to solve the main transmission problem (useable left hand drive bicycle hubs are extremely hard to get).  It also smooths things out.

I'd use a hub motor so there's no real need to drive the left side. I don't see a need to drive either side if you use a hub motor and do the generator thing or ditch the engine. You can get better efficiency and more power from an RC motor but it's not nearly as cheap or easy.

Some brushless motors could be around 92-94%

There are brush motors that get that as well. It'd be way overkill but look up Agni....you'd be done with the race by the time everyone else got 100 yards, lol.

7)  If we were allowed to use batteries to store our generated energy (using a joule meter to make sure you aren't cheating), then you could greatly increase your mileage.  First, you could have a smaller drive motor and the car could do a constant 15 mph (saving 10-20% for sure).  Also, you have less losses in spinning up the flywheel.  UNFORTUNATELY, I asked to do this, and the officials said no.  Eco-marthon has many other classes of vehicles for all kinds of cars.

OK, the rules/officials say no..lame. I don't know how it works but what I was imagining was just totally ditching the gas engine and replacing it with a 1kwh battery pack and a hub motor. If you have to cover a certain distance just use a pack big enough to make it. All it would take to measure consumption is a Cycle Analyst. You'd have to know the energy content in a gallon of the gas they use so you could compare directly but even if you don't have that you could take a handicap and compare to regular gas.

1011mpg is good and all, but even my homemade $500 junker ebike gets about 1100mpg (EPA) when tortured, or about 1500mpg when ridden with efficiency in mind (and that's not pedaling).

With 3WD, you could put down closer to 7-10 hp   and make a real rocket ship.

You can do 10HP peaks with a single RC motor on the rear wheel. There's a guy on another forum that does that regularly. There's another guy that strapped an Agni on a custom bmx frame and pushed near 70HP.

[taylorp035]

...useable left hand drive bicycle hubs are extremely hard to get...
Like this?

Sure... but will the teeth be able to take 3.5 hp?  For a few hundred starts.

1011mpg is good and all, but even my homemade $500 junker ebike gets about 1100mpg (EPA) when tortured, or about 1500mpg when ridden with efficiency in mind (and that's not pedaling).

Yep, and we could get 15,000 mpg if you used the same rating system.  The drag coefficient and frontal area of a bicyclist is really bad, and couple it with a large ratio transmission from an rc motor.... leads to high power usage.

You can do 10HP peaks with a single RC motor on the rear wheel. There's a guy on another forum that does that regularly. There's another guy that strapped an Agni on a custom bmx frame and pushed near 70HP. Wink

 Maybe you should look at Castle Creation's new Mamba XL system.... 10 hp continuous... meant for 1:5 scale buggies.  70 hp would smoke your tire through 60 mph easy.... I have seen 25 hp and that was way past outrageous.  There was once a Canadian electric go-kart that pushed 200 hp... his forklift motor gave out after that.  He could spin the tires pretty much the whole way down a 1/4 mile strip up to 100 mph.

[zap]

...useable left hand drive bicycle hubs are extremely hard to get...
Like this?

Sure... but will the teeth be able to take 3.5 hp?  For a few hundred starts.

That I don't know... not even sure what "teeth" you're talking about

[REdiculous]

Yep, and we could get 15,000 mpg if you used the same rating system.

I don't think so. Your numbers would go down, not up, because I'm comparing to the energy content of regular gas. Check out the MPGe wikipedia page. I'm pretty sure I'm doing it right but corrections are welcome...

With regular gas, it looks to me like an ebike that's using 33.7wh/mi is achieving 1000mpg and a car that's getting 33.7mpg is using 1kwh/mi. I'm comparing tank-to-wheel, not well-to-wheel, because everyone compares the EPA stickers and few even know CARB exists.

Not all fuels are the same so if you were comparing with E85, 24wh/mi = 1000mpg and with diesel, 37.9wh/mi = 1000mpg. Obviously, the better the fuel, the more energy you can use per mile and still get high marks.

The drag coefficient and frontal area of a bicyclist is really bad, and couple it with a large ratio transmission from an rc motor.... leads to high power usage.

The terrible drag of a normal bicycle is why this is interesting to me. Even without any real serious effort at efficiency, 30wh/mi is pretty normal. RC-based systems can be very efficient...one of the main proponents (who produces RC reduction units) has reported as low as 12wh/mi @20mph on an Actionbent Midracer, which is like 2808mpg. On a Catrike 700 he said he's getting 14.4wh/mi average @20mph; 2340mpg. Now imagine what you could get from that if you tried one on your car where the drag is less, the wheels better, etc.


Anyway, since going electric isn't a real option, I'm interested in seeing what you guys can do to wrangle more efficiency out of your engine. It looks like you got some good ideas so I hope it works out.

[taylorp035]

Now imagine what you could get from that if you tried one on your car where the drag is less, the wheels better, etc. Wink

Tires are a very, very important part of the equation.  With our supermileage car, about 75% of the drag is caused by the tires.  A person riding a bike would be much different and the percentage changes a lot with the speed (since air drag is related to velocity cubed).

A typical bicycle tire has a rolling drag coefficient of 0.006.  Racing bicycle tires with no sidewall protection and 200 psi maybe be a good bit better.  Our tires range from 0.0025 to 0.0008, almost as good a locomotive wheel.  Of course, all of this is relative, since sideways forces caused by taking a turn or driving on a banked curve greatly increase rolling resistance.

Anyway, since going electric isn't a real option, I'm interested in seeing what you guys can do to wrangle more efficiency out of your engine. It looks like you got some good ideas so I hope it works out.

Thanks    We are going to need all the help we can get.  Today, we managed to get some better lab access with some really good machines, so things are looking up.  Our sprocket has been delayed a bit (due to an error at the metal distributor), but I foresee it being made before Wednesday.

I don't think so. Your numbers would go down, not up, because I'm comparing to the energy content of regular gas.

There are a lot of ways to play with the numbers.  In terms of rolling resistance and air drag, our car dissipates about 2.7 - 3.0 Wh / mile (Calculated and verified).  If you consider the conversion of the energy in the fuel, then you need to factor in the ~ 10% efficiency of the engine + transmission.  Now you are looking at 30 Wh / mile, but that's not fair since it is gasoline, not electricity (I know you understand this already, but for those haven't put as much thought into it    ).


12 Wh per mile on a bike without supermileage tires seems pretty good to me.  Our total car + driver weighs only 245 lbs, so that needs to be added to these biker's Wh reports. 

Jeez, I could make 12 Wh easily with my little solar panel.... 1/2 kwh battery pack wouldn't be too bad either.  You could buy two 37v 7.7 ah 65/130 C Lipos that charge at 12C... charge slow with R.E. and then charge with 230v 30A when you are on the go in 5 minutes



 

[REdiculous]

If you consider the conversion of the energy in the fuel, then you need to factor in the ~ 10% efficiency of the engine + transmission.  Now you are looking at 30 Wh / mile, but that's not fair since it is gasoline, not electricity

That's the thing though, you don't consider the conversion inefficiencies. We don't care what you waste or what the fuel is, just what was available (in BTU/kwh/kcal, whatever) and how far you went.

If you get 1000 MPGe and your efficiency is 10% you're using 33.7wh/mi but you're wasting most of it so what you can put to the ground is only 3.37wh/mi. If I get the same MPGe at 80% efficiency I'm still using 33.7wh/mi but I'm putting down 26.96wh/mi because I'm not turning so much to heat.

[taylorp035]

About 20 working hours into it....  for reference, the aluminum is 18"x18"x0.75".  7075-T651



Only one major flaw so far... we shall see if I can compensate for it tomorrow.   .... Let's say the spokes are no longer 90 degrees apart and we haven't taken the finish pass yet. 

I ended up redrawing the whole thing tonight for the new angles.

[bj]

   Hopefully you can correct on the cleanup pass.  It is a beautiful piece though.

[taylorp035]

It looks like my careful measurements and new drawing are correct.  It should be done tomorrow.  We should have 12 more programs to run, each averaging an hour to run and about 25,000 lines.

I don't have a picture of it being round, but it is now   Starting from 24 lbs, we are now down to about 5 lbs... on the way to 2.2 

[taylorp035]

We were so close to cutting some teeth today.  Apparently 3 hours isn't enough to compensate for 4 "blue screens of death" by our main cad computer   I had to manually click over 1000 lines twice, in order and not any extras... which all ended in vain after the blue screens.  This is day 6 of machining the sprocket, average a good 8 hours each day.  I think we are up to 18 programs that we have run already.

[bj]

  It just keeps looking better and better.  A load of work though.  You'll sleep better when the teeth are on.