Extreme efficiency?

[recumpence]

The typical riding I do is pedalling to mantain 20mph on the flats, then use the motor for accelleration and for hills (I live in a hilly area). If I run the motor ONLY, I can get 13wh per mile if I really nurse it. If I just crank the throttle and average maybe 25mph or so and not worry at all about efficiency, I normally see 18 to 21wh per mile. Now, if I pedal on the flats, but throttle hard whenever I feel like it (bashing around like I am riding a motorcycle), I almost always see 15 to 17wh per mile.

That is really the way I like to ride. I like to not worry about my efficiency, but still pedal alot for exercise. If I do that, I can be sloppy with the throttle and still get good efficiency (of course, that is not what this thread is about).

Bill, I would love to see you try one of my drives. I expect to have my materials this week (they'd better be here this week! ).

The drive can be setup for huge power handling, up to 5kw, or for efficiency. The most efficient setup would be a single stage using a 9mm wide belt. The lease efficient setup would be a 15mm wide primary belt run tight with a chain second stage. That is what I am running on my recumbent. It is setup for high power and 40mph top speed.

Matt

 

[paultrafalgar]

aehemm... what'ya wanna tell me, Paul?

-If you wanna say I lie, discussion ends here for me. I'm not here to prove anything, but to share experiences
-If you think we didn't measure acurately, hmm, as I said, we have 2 bikes, both equally equipped, both with a CA
-If you think, the CA is off, I don't think so, because it would mean we get more than 576Wh out of our 48V/12Ah YESA-batteries
-If you wanna say that we did pedal a good amount, true, but the motor was on all the time where it made sense (read: NOT down the hill), but also our eyes were every 10 seconds on the CA
.

Raphael, absolutely no offence intended
It's the 4th option: your amp hours per mile seem lower than anyone else here, I just assume that that was due to athletic pedaling that's all. Sorry!

 

[mi7d1]

Fascinating, Bill. How much does your battery weigh? I'd love to see you repeat that with some Lithium on board. I'd lay money on there being a big improvement in Wh/mile.

77 lbs (35 kg) velomobile
48 lbs (22 kg) batteries
15 lbs (7 kg) Crystalyte 406 motor
220 lbs (100 kg) my weight

360 lbs (163 kg) minimum total weight during the race.

I do have some LiFe batteries on the way. I'm part of the Headway group purchase with Joshua G.
The thread from last years ePower Challenge can be found here:
http://endless-sphere.com/forums/viewtopic.php?f=3&t=4426&st=0&sk=t&sd=a

 

[mi7d1]

I did pedal during the race. How much I can't say but I don't have the strength nor stamina to compete otherwise, I'm by no means an athlete. The results I posted were not normal. Using the velo for commuting I get around 18Wh/mi in city ridding. It was fun to see how much impact stop and go ridding has on my bike. I'm looking for 40 mph for the race in May 2009.

Matt, I have a few questions about your system. First a disclaimer. Take it easy with me, I only have experience with hubmotors. I'm not sure how the rules for next years ePower Challenge will change. The club (OHPV) is trying to get watt meters for all racers to use so that racers wouldn't be held to a certain division based on their battery. The LiFe batteries I'm getting are 36V 20Ah Headway packs. I plan on running them in a series/parallel configuration on my hubmotor. According to the rules from last year:
http://www.ohpv.org/HPC/EventsNews.html

we will also be limiting the motors used to conform to a 750 watt continuous power rating that should be visibly marked on the motor casing, no switchero of stickers folks. Too many of us know what to look out for. The 750 watt limit is to “honor” the Fed output limit and will showcase to the public what amazing things can be done with that amount of assist. Of course you can certainly use a less powerful motor, it could be an advantage!

The idea was that they didn't want any Perm or Etec type motors. How would your system running 36/72V on a motor rated at 750W or perhaps 1000W fare? Could it be competitive? With my C'lyte 406 at 36V and a 20A controller I'll get 20 mph and at 72V I'll get a little over 45 mph according to http://www.ebikes.ca/simulator/ My velo is geared from 47-147 gear inches using the 60t chain ring and 11-34 seven speed megarange freewheel. The bike should do 40 mph at 100 rpm with my gear setup (165mm cranks, 38-559 rear tire) according to http://www.sheldonbrown.com/gears/

I'd prefer to have my e-assist drive the velo through the gear train like your setup. I've had my share of broken spokes and am tired of the weight of the hubmotor. It's no fun to pedal the weight around without the assist and I've had to do that a few times. I have no doubt that your system is ideal for me commuting but how about racing with it? If you want to take this discussion off board because the subject at hand is changing please PM me. I have a little over six months to get ready for the next race.

 

[Raphael]

Ok, a few numbers.

Calculation (c), measurement (m):

It's always Wh/km.

0% slope (flat):
@20kmh: 5.964(c), 5.951(m)
@30kmh: 7.469(c), 7.55(m)
@40kmh: 9.575(c), 9.576(m)

I have no measurements on slopes, but for reference, always @30kmh:

@2% slope: 18.047
@4% slope: 28.613
@6% slope: 39.153
@8% slope: 49.655
@10% slope: 60.108

This is all calculated / measured based on a recumbent bike with following specifications:
-Rolling resistance: 0.009
-All-up-weight: 112kg
-Drag coefficient cw: 0.3
-Biggest cross sectional front area of the vehicle in m2: 0.45

Maybe you find it interesting.

Raphael
When you know the theory, you can optimize your practics

 

[recumpence]

I am not an engineer. So, I do not have any math to throw at you. However, I will give you the advice I can;

To achieve the greatest efficiency, you need to maximize your motor with your final drive looking to keep the motor in its peak efficiency right at your targetted top speed. For instance--- I will use my bike as an example. My recumbent requires a touch under 500 watts to mantain 20mph on the flat. However, my motor's peak efficiency is up near 3,000 watts. I am also running a very high drag reduction system on my bike because it needs to be able to handle the occasional 4,000 to 5,000 watt burst. So, my 500 watts to run 20mph is really horrible. It should be about half that. I know that because my no load power useage at 20mph (running the motor with the rear wheel off the ground) requires 200 watts. So, at 20mph, I am only 60% efficient because I am wasting 200 watts just driving the system. Now, if I went to a motor with a max efficiency in the 500 watt range, I could then build a much more efficient drive system that would handle maybe 800 watts or so, and gear the bike up a touch and probably see 30mph with that same 500 watts.

So, you really need to maximize everything for the exact range of speed you are looking for to get the ultimate efficiency.

Anyway, it would be really easy to achieve 750 watts out of even a small RC setup.

Matt

 

[skydog]

"My question is how much can some very basic aerodynamic structures improve efficiency at say 30mph with zero wind? ... Anyone have a ballpark estimate of how much efficiency can be gained for a 30mph speed with minimal fairing structure? Could it be as much as 20-30% benefit."

I used to play around a lot with bicycle and recumbent aerodynamics and produced conventional bike fairings about 2.5 X the size of the Zippers. On my 15 mile commute I would increase speed about 10% roughly from 18-20 mph. The faster I went the greater the % increase. The really interesting thing was when automobile traffic passed nearby, the fairing would act like a spinnaker sail and when done right, would increase as much 4-5 mph for short bursts when traveling above 20 mph. Although I am a middle of the pack rider, I could maintain 28-29 mph on PCH. It was kind of like drafting someone. Very beneficial especially in a headwind.
The spandex side panels seen on some lightweight fairing designs are really effective and the porosity of the fabric doesn't seem to diminish efficiency so much.

Overall, aerodynamic efficiency on ebikes essentially means HP, or less required. The record for single rider HPV is about 82MPH/ approx. 1HP. If that same recumbent was unfaired, the top speed would be about half.

Incidentally, Astropower used to sponsor bike conversion competitions in the early 80's and made the components available very reasonably to competitors at IHPVA events. There would usually be about 10 or so conventional bikes racing various courses at about 30mph. Usually powering the front wheel.

I recently purchased a used Intense M1 and will be converting it to an ebike and will be trying out one of my old fairings along with a kevlar tail section I'll mount on the rear to house some gear. All my bicycle life I've always built everything as light as possible. Now I'm converting the heaviest bike I've ever owned and looking to increase its weight!

 

[John in CR]

Skydog,

Aerodynamics is far more important than 20-30%. Play around with the calculator linked previously to better understand. With same tires and weight on the flat terrain the difference is all aero and can be more than 50%, even at 20mph and far greater at 30.


Matt,

Your 500 watts doesn't sound right at 20mph. That would be 25wh/mile. 15wh/mi is 300 watts. After playing around with that calculator, the big gains are in aerodynamics, so for my coast-to-coast on a single charge bike build one of the biggest design aspects is aerodynamics. I don't want to go fully faired, but it will definitely have some kind of front fairing and a tailbox. On the flats I'd like to get to 300W needed at 25-30mph to reduce the battery requirements and shorten the ride. It will give us extremely low power usage if we need to slow down to 15-20 to make sure we complete the almost 200 mile ride.

John

 

[mi7d1]

"My question is how much can some very basic aerodynamic structures improve efficiency at say 30mph with zero wind? ... Anyone have a ballpark estimate of how much efficiency can be gained for a 30mph speed with minimal fairing structure? Could it be as much as 20-30% benefit."
<snip>
Overall, aerodynamic efficiency on ebikes essentially means HP, or less required. The record for single rider HPV is about 82MPH/ approx. 1HP. If that same recumbent was unfaired, the top speed would be about half.

Incidentally, Astropower used to sponsor bike conversion competitions in the early 80's and made the components available very reasonably to competitors at IHPVA events. There would usually be about 10 or so conventional bikes racing various courses at about 30mph. Usually powering the front wheel.

Have a look here for some power requirement ideas:
http://www.kreuzotter.de/english/espeed.htm
To give you an idea according to the site with off road tires it takes 284W for a mountain bike to do 20mph but only 94W for a Quest velomobile. My velomobile, a WAW is not quite as efficient as a Quest, but very close. At 284W the Quest would be doing 33 mph.
I was in Battle Mountain just over a month ago and watched Sam Whittingham ride the Varna Diable II to a new record for the flying 200m he did 82.33 mph. There are a lot of photos from the WHPSC on my flickr site and some videos on my Vimeo site. See my sig for the url.

It surprises me that there was electric races at an IHPVA event. It took the club I belong to just about three years to find insurance coverage to allow e-assist races at the club sponsored event.

 

[recumpence]

Skydog,

Aerodynamics is far more important than 20-30%. Play around with the calculator linked previously to better understand. With same tires and weight on the flat terrain the difference is all aero and can be more than 50%, even at 20mph and far greater at 30.


Matt,

Your 500 watts doesn't sound right at 20mph. That would be 25wh/mile. 15wh/mi is 300 watts. After playing around with that calculator, the big gains are in aerodynamics, so for my coast-to-coast on a single charge bike build one of the biggest design aspects is aerodynamics. I don't want to go fully faired, but it will definitely have some kind of front fairing and a tailbox. On the flats I'd like to get to 300W needed at 25-30mph to reduce the battery requirements and shorten the ride. It will give us extremely low power usage if we need to slow down to 15-20 to make sure we complete the almost 200 mile ride.

John

That is continuous power useage at 20mph (it is almost always 480 watts at 20mph cruise). If I accellerate to 25mph up a hill, then coast unpowered down to 15mph on the flat, then accellerate up the next hill, my wh per mile required to average 20mph goes way down to 15wh per mile. See what I mean? I know it seems counter intuitive. But, again, my motor sees its best efficiency at high output levels.

I am using a handlebar mounted watt meter (Eagle Tree Micro logger with display). I monitor that display alot.

What I am saying is, my bike is not built for efficiency, it is built for high power. Yet, for a street E-bike, if ridden right, it is pretty good overall (good efficiency). However, I am sure I could double or tripple my efficiency if I targetted a specific speed and built the drive system to max out at that speed.

Matt

 

[skydog]

"Skydog,
Aerodynamics is far more important than 20-30%. Play around with the calculator linked previously to better understand. With same tires and weight on the flat terrain the difference is all aero and can be more than 50%, even at 20mph and far greater at 30."

I agree aerodynamics is far more important. The numbers I provided weren't from a calculator but from hundreds of miles of onroad testing and not fuly enclosed but with a "partial fairing". My own enclosed trike from back then (mid 80's) hit 61mph on the flats.

 

[skydog]

Bill wrote:
"It surprises me that there was electric races at an IHPVA event. It took the club I belong to just about three years to find insurance coverage to allow e-assist races at the club sponsored event."

The IHPVA also allowed me to run the first solar car race in the US in 88' which helped a spawn lot of activity @ GM and Honda. It was a pretty cool time for the IHPVA, kind of an organized technical anarchy.

 

[mi7d1]

Matt, If you system were set up to do say 30mph, as I understand if it were run slower than the efficiency would be lower. Could it be run faster than the 30mph or would the motor heat up from too much. What I'm trying to understand is in setting goals for your system setup does one aim for the middle range or the maximum speed desired?

 

[Raphael]

Aerodynamics is far more important than 20-30%.

As you say, it depends totally on speed, but even more on slope.

My bike as described above has a rolling resistance of 9.89N.
Aerodynamic drag is:

@10kmh: 0.63N
@20kmh: 2.50N
@30kmh: 5.63N
@40kmh: 10.0N
@50kmh: 15.6N

So, total resistance, as there is no grade resistance on flats, is:
@10kmh: 10.51N
@20kmh: 12.39N
@30kmh: 15.51N
@40kmh: 19.89N
@50kmh: 25.51N

So if you look at it like this, if you lower either your frontal aera or your drag coefficient to it's half (and that's a LOT!), therefore lowering your aerodynamic drag to it's half, it would give you a total advantage of:

@10kmh: 2.97%
@20kmh: 10.09%
@30kmh: 18.13%
@40kmh: 25.14%
@50kmh: 30.62%

So, at 50kmh, it's almost 1/3. That seems to be a lot and may be for racing, especially on the flat. But now let's take a look what happens if we have some grade resistance from a 4% slope:

Aerodynamic drag: the same as above.
Rolling drag: the same as above.
Grade drag: 43.91N

So your advantage with lowering the drag to it's half on a 4% slope is:
@10kmh: 0.57%
@20kmh: 2.22%
@30kmh: 4.73%
@40kmh: 7.84%
@50kmh: 11.25%

I can calculate this for any slope, bike and speed you want. Just tell me the numbers.

Don't wanna say it's worthless to work on aerodynamics - my main hobby is aeromodeling in which I'm pretty successful, and there's a lot about aero there obviously - but on a bike, there are other factors that are (for commuting) by far more important. For racing, of course every 1/10th percent counts

Raphael

 

[cerewa]

The way to make aero efficiency go much higher is to get yourself a full coverage fairing similar to the varna diablo. (There are a number of copies of that fairing, and one or two Varna molds where a person can make his/her own fairing.)

http://images.google.com/images?q=varna%20diablo.

Energy (per mile) used to overcome air resistance goes up with the square of speed, so...
assuming you use 5 watt hours per mile to overcome air resistance at 20mph
you will use 5 * (2-squared) = 20 watt hours per mile to overcome air resistance at 2 times the speed ( 40mph )

What this probably means is that the electric car that gets 18 wh/mile at 50mph could probably do far better, maybe half as many watt hours per mile, if it were going 20 miles per hour.

Matt didn't make his bike for maximum efficient operation at 20mph. If he had, he might have chosen a smaller motor (less energy required to just turn the copper coils and magnets against each other) and gone for a full fairing, which is what the most efficient electric cars and human powered bikes use.

 

[John in CR]

Raphael,

I'm not sure I understand your point. Of course the work required to get up grades is generally more significant than getting through the wind, but I see them as unrelated and quite different. Wind resistance and rolling resistance are losses that can't be recouped. Going up a grade is different though, since what goes up must come down, and you recoup the uphill price when you go down except what you in braking on the downhill or increase wind losses by coasting down faster.

If you're talking about power requirement then I understand, but not so much if you're talking about energy capacity requirements. This is an important topic for me, because I'm in the planning phase for building a bike for a 180 mile ride on a single charge. I see aero as extremely important, since I don't want to ride at less than 15mph, so it will be vital to the build. Power is a critical component for me too, because there are some mountains to cross. I see it as 2 separate engineering issues. Aero must be addressed to obtain overall efficiency at reasonable speeds in order to reduce battery capacity requirements, which also impacts power requirements because batteries have significant weight. Power must be addressed because there's no way I'm going to rely on just the addition of pedal power to get up the 10+ mile climb with a 4-5% grade, must less the shorter climbs that have much steeper grades.

If I'm missing something, then please help me understand.

John

 

[dnmun]

why build a bike with such a long range? is there no electric power in between where you could recharge on the road at a justin junction? maybe you can change your route to reach one, and just carry the charger with you rather than more batteries. i think justin was averaging about 60km between charges, on the flat.

 

[John in CR]

why build a bike with such a long range? is there no electric power in between where you could recharge on the road at a justin junction? maybe you can change your route to reach one, and just carry the charger with you rather than more batteries. i think justin was averaging about 60km between charges, on the flat.

Caribbean Sea to the Pacific Ocean on a single charge is the goal I've set, which I don't believe has been done with any EV. If I lived in Panama instead of Costa Rica, it would be easier at only 100 or so miles by road, but here I'm quite familiar with the roads and the highest pass on the route is about 30min from my house so I can fully test the rig.

John

 

[Raphael]

Hi John,

I just wanted to let everyone here know, that improvements in aerodynamic are a) not easy and b) will not have such a big effect as it first seems to be. If you have no recuperation, you don't have any advantages driving downhill with improved aerodynamics. You don't recup nearly everything when going downhill what you spent uphill.

Look at it like this:

-Drive 2km on the flat.
You will have a drag of 12.39N for 2km, total = 24.78N during 2km.

-Drive 1km uphill 4% and 1km downhill 4%
You will have a drag of 12.39N aero-/rollingdrag plus 43.91N grade drag making for a total of 56.3N drag for 1km and 12.39N aero-/rollingdrag going downhill with a grade force of 43.91N you can recuperate. Don't expect to be efficiency of recuperating on a bike to be more than 50% in reality. So you'd use 56.3N for 1km uphill and -9.57 for 1km downhill. That's a total of 46.74N during 2km.

=You're still going with a lot less effort driving on the flat than up-/downhill with recuperation. Try to optimise your route with that in mind. Only if the distance is about half going over the mountain than around it and only if you have a very good recuperation and almost never use your friction brakes, you should choose the slope.

If you are willing and able to give me some numbers of your bike and your planned route, I can calculate to a VERY exakt level how much energy you're going to need. You don't need 50% reserve as you wrote above. 10% should be plenty.

Raphael

 

[paultrafalgar]

Justin's overall regeneration for the whole trip was 2.4%. If you want the individual regeneration for any particular leg of that journey (which is more like your proposed trip) look at his spreadsheet. Personally, I think regeneration isn't worth the effort of organising it.

 

[John in CR]

Rafael,

Thank you and I will take you up on your offer. I already picked up a GPS and very soon I'll get all of the road profiles and have all the data necessary to plan the route. One advantage I have in going east to west is that the downhills generally have a shallower grade, reducing brake use. BTW, I was only thinking of a 20-25% cushion in battery capacity not 50%. The bike itself is currently totally up in the air. Over the next month we'll be testing some different drivetrains before finalizing and building the actual form of the bike, but a full fairing is out of the question in this tropical sun.

Early recommendations are greatly appreciated.


Paul,

I'd love to have regen braking to recoup as much as possible, but it's not something I'm banking on. I will have plug braking at a minimum, because I'll definitely need it especially for one dangerously curvy downhill stretch of a few miles.

John

 

[skydog]

Just an FYI:
This is from my good friend Michael Lewis who recently set the 1 hr. record for Electrathon.

Rob:

Are you logged into the blog? You can tell them that its now 58 miles, which
used 960 w/hrs, or 16.55 wh/hr/mile. I wonder if there is a measure of watt
hours per mile per hour?

 

[mi7d1]

Just an FYI:
This is from my good friend Michael Lewis who recently set the 1 hr. record for Electrathon.

Rob:

Are you logged into the blog? You can tell them that its now 58 miles, which
used 960 w/hrs, or 16.55 wh/hr/mile. I wonder if there is a measure of watt
hours per mile per hour?

Now we have some stats that don't involve pedaling influences.

Any friend of Mr. C. Michael Lewis is a friend of mine. I had the pleasure of meeting him in Sept at Battle Mountain, Nevada. He was on his way to the salt flats the following day and slowed down enough to autograph my shirt (first photo)

Here's a better photo of his electrathon. Jason Erickson on the left is who I crewed for at Battle Mountain. Scott Wilson on the right was my crew mate and he helped me all week long. It was my first time crewing for a streamliner.

 

[skydog]

Thanks for posting those pics of the Eracer, it really is a masterpiece.
He clicked off a conservative 100+ MPH at Bonneville last month. He could've gone faster but was concerned about frying his motor.
Plus he had an 80MPH "episode" on an earlier run where he lost control/traction.
When you met Michael, did you also happen to meet Fred? He's Michael's electronics guru.

 

[mi7d1]

Thanks for posting those pics of the Eracer, it really is a masterpiece.
He clicked off a conservative 100+ MPH at Bonneville last month. He could've gone faster but was concerned about frying his motor.
Plus he had an 80MPH "episode" on an earlier run where he lost control/traction.
When you met Michael, did you also happen to meet Fred? He's Michael's electronics guru.

Michael and another electrathon racer were heading to the salt flats for a race and then to some other track for another race. As I recall being told one racer was expected to do better at the salt flats and the other at the race track. I don't know the other racers name but was told that these two exchange record titles back and forth. It's possible that I talked to Fred. The name doesn't ring a bell but there were so many people that I didn't know at the World Human Power Speed Challenge.

Another thing Michael is know for is his art work. He made the design for the WHPSC this year and in years past.