Honda Insight Forum banner

1 - 20 of 64 Posts

·
Registered
Joined
·
222 Posts
Discussion Starter #1
My prius-owning, fellow-HPV-riding, efficiency-conscious, chair-of-the-local-university-physics-department friend has explained to me the increased wind resistance as temperature drops and it is very significant and far more so than rolling resistance even at moderate bicycle speeds.

I wonder how much I'd have to slow down in the winter to completely offset the additional drag due to the colder air? I will ask my physics friend and see what he says.
 

·
Registered
Joined
·
182 Posts
Everyone seems very certain that the air resistance is a much more important factor than rolling resistance. But nobody cites any evidence. Brian, how does your experience biking in the winter allow you to differentiate between extra wind resistance and extra rolling resistance? Paul, your physicist friend can easily calculate the increase in wind resistance, but there's no fundamental formula that tells you how the characterisitics of rubber change with temperature. So ask him how he knows how that side of the equation changes with temperature.

After digging for a long time, I did find some data on rubber characterisitics and, and I also figured out how to do a quick and dirty experiment that would be able to show whether there was a significant change in those characteristics with my bike tires. That was for a discussion on a winter-bicycling list. I'll try to find that data to share with you, but in the meantime I'd be interested to hear what basis you have for your confidence that air resistance is more important.

Charlie
 

·
Registered
Joined
·
222 Posts
Discussion Starter #3
I will certainly be quizzing my physics friend for all the dirt, but in the meantime (he is on vacacation this week),

is it your allegation that for some reason rubber tires have more rolling resistance when it gets colder out? Now there is some data I would like to see, since my practical experience with rubber is that it is less flexible and "grippy" when it gets colder out.

Anyway, at anything over 20 mph or so (and it increases dramatically with your speed), rolling resistance is a joke compared to wind resistance. Certainly your web searches must have shown that.
 

·
Registered
Joined
·
224 Posts
Hi Charlie,

The effects of aerodynamic drag as a function of drag coefficient, air density, and velocity, are probably reasonably well documented for automobiles and even conventional bicycles; and the information is probably accessible via the internet.
In contrast, quantifying the factors that affect rolling resitance of tires (such as viscoelastic properties, road surface roughness, and tire deformation) would involve considerably more reseach IMHO.
My assumptions, posted just about a year ago on this thread, were based on qualitative observations. In other words, I don't have quantitative data. :oops:
While the rolling resistance analysis might be an interesting subject for a graduate thesis, my simplistic answer would be that while bicycling at temperatures below 20 deg. F, I'm more concerned about how to maintain circulation in my extremities than about the rolling resitance of my tires. ;)
-Brian
 

·
Premium Member
Joined
·
4,942 Posts
ethicalpaul said:
Anyway, at anything over 20 mph or so (and it increases dramatically with your speed), rolling resistance is a joke compared to wind resistance. Certainly your web searches must have shown that.
IIRC from what I've read in various technical papers on this subject 40 MPH is the "magic" number when the other losses of friction and combustion efficiency are included.

I'm sure bicyclists would disagree :!: :p ;)

HTH! :)
 

·
Registered
Joined
·
33 Posts
is it your allegation that for some reason rubber tires have more rolling resistance when it gets colder out? Now there is some data I would like to see, since my practical experience with rubber is that it is less flexible and "grippy" when it gets colder out.
Tire traction/grippiness is not the same as rolling-resistance. Tires designed to be gripper tend to have more RR, but the traction itself is not what causes the RR. Rolling resistance is primarily from the tire rubber deforming. The fact that the rubber is more flexible at higher temperature means that less energy is needed to "bend" the tire sidewall and tread components as it rotates.

This is also why higher tire pressure leads to a drop in rolling resistance (tire holds a "rounder" shape). A studded or treaded tire will have much more RR than a slick tire as the bits of tread each deform.

The reason there are no specific data for tire rolling resistance in part has to do with the fact that rolling resistance coefficient values need to be determined experimentally and cannot easily be predicted. Also, rolling resistance coefficient also varies based on the driving surface, as deformation of the road surface (ie dirt roads) or evacuation of water on a wet road, all require energy, and will not affect all tires equally. Without some sort of systematic test for all tires similar to the EPA driving cycle, rolling resistance between tires cannot be easily compared, and even then, those values will vary dramatically based on the bearings used on the car, alignment characteristics, and many other factors.
 

·
Registered
Joined
·
182 Posts
Data!

I located the data I had found on rolling resistance vs. temperature.

On the Bayer Rubber web site, I found a curve of "tan delta" vs temperature for two different kinds of tire rubber. This was back in 2004; I couldn't find it again on their web site (it appears they sold some of their rubber business to a German company), but I did save a pdf of that page. PM me with your email address if you'd like me to email a copy.

Tan delta is a measure of the energy lost to hysteresis as the tire is flexed. It's a relative measure, that also depends on stiffness--it's the fraction of the energy needed to flex the tire that is lost to hysteresis as the tire. Thus, the rolling resistance should be proportional to the product of the stiffness and tan delta.

(If that didn't any sense to you, and you care to dig into the details, I recommend this explanation:
http://darwin.nap.edu/books/0309094216/html/42.html
from "Tires and Passenger Vehicle Fuel Economy: Informing Consumers, Improving Performance -- Special Report 286", (2006) Board on Energy and Environmental Systems (BEES), National Academies Press, p. 42)

The plot of tan delta for "s-sbr" rubber with silica instead of carbon black is at about 0.2 at 20 C. It drops to 0.15 at 30 C and goes up to 0.25 at 10 C. Then below 10 C it rockets up reaching 0.5 at 0 C and 0.7 at -10 C.

That means that even without factoring in stiffness, the rolling reistance can increase dramatically at lower temperatures.

More later on my own measurements, and on the relative importance of rolling resistance and aero drag in the insight in general.

Charlie
 

·
Premium Member
Joined
·
4,942 Posts
WOW :shock:

This IS some high tech discussion of tire rolling resistance. Thanks for consending all the technicals to a "plain" english sentence Charlie.

And this has also been one of the undefined unknowns of our commonly expereinced cold MPG losses. Some of the other more obvious are combustion effifiency losses during warm-up and with a "colder" engine.

As such I'd hate to see this discussion become lost in the clutter of a hot air intake thread so I'm splitting it out into its own thread. :)

Please let the group know what you discover as your work progresses. :)

Sincerely,
 

·
Registered
Joined
·
182 Posts
my own experiments

Thanks for the thread split--I was thinking that might be a good idea.

Before I found that Bayer Rubber data, I did some crude experiments. I was mostly interested in bike tires at the time, so these are using bike tires and inner tubes:

1) Bounce test. I dropped a wheel with a semi-knobby MTB/hybrid tire from three feet and measured the bounced height, and assumed that most of the loss in height was energy absorbed by the rubber in deforming and springing back. 50 psi.

2) Spring test. I hung a weight (cement block) using an old inner tube. I expected to bounce it up and down and count bounces, but it damped out too quickly, so instead I set it wobbling in a way that alternately stretched different pairs of the four segments of tube it was hanging from (I'm not sure that's clear but the details aren't too important).

Results:

50 F garage:
1) Bounce: loss of 8 inches out of 36 inches drop height (i.e., it came back up to 28 inches).
2) Spring test: about 10 cycles of joggling before it damped out "pretty much completely".

After perhaps 15 minutes at 5 F outdoors:
1) Bounce: Loss of 14 inches (ie. came back to 22 inches).
2) Spring test: 2.5 cycles.

The next morning, after sitting outside overnight at between 0 and 5 F:
1) Bounce: Loss of 16 inches (ie came back to 20 inches). With air added to bring the pressure to 52 psi, loss of 14 inches (ie came back to 22 inches).
2) Spring test 2.5 cycles.

Conclusion: Very roughly, rubber hysteresis losses appear to increase by 2 to 5 x going from 50 F to 5 F.

Note that what can often happen is that if the tires are very lossy at low temperatures, the loss generates heat in them, and thus to some extent solves the problem. In a big, heavy vehicle, the losses are big enough to heat the tires pretty effectively. On a bicycle, that doesn't tend to happen much, so the high rolling resistance in the cold is a bigger problem. I imagine the Insight is somewhat intermediate in that regard--significant self-heating on the tires, but not as much as in a big car. And even with the self heating, the tires are still colder than they would be in a warmer environment, since the self heating happens then too (although to a lesser extent).

Charlie
 

·
Premium Member
Joined
·
4,389 Posts
Interesting test results, and I concur that it is a good experimental way to quantify the difference.

Intuitively I would expect the real world losses due to temp to be much less on a tire that is run at 45-55 PSI vs 35-40 psi, the deflection is less.

The tire heating in winter is somewhat offset as the tire hits the cold road ahead, I have one of those non contact IR temp probes, and will do some tire temp rise experiments when it gets cold.

Air density changes between winter and summer is a big factor as well. Any pilot will tell you that the take off speed in cold air is much shorter than when it is hot.
Getting real numbers is the problem.
 

·
Premium Member
Joined
·
2,897 Posts
Since this topic already has a technical disclaimer.... I thought I would give my 2 bits.... anyone who knows / understands this stuff better than me fell free to correct we any where I made any mistakes.

Conclusion:
Colder Air requires more force to push the same Cd object threw.


Reasoning:

Increase in Air Density effect #1

Colder Air is more dense than warmer air at the same pressure
Density = Pressure / ( R constant of Air * Absolute Temperature K )
this is the general formula for any gas if not Air just change the R constant to that of the gas being used.

-------------------------

Increase in Air Density effect #2

The addition of water vapor to air ( making the air humid) reduces the density of the air, which may at first appear contrary to logic.

This occurs because the molecular mass of water ( 18 ) is less than the molecular mass of air ( around 29 ). For any gas, at a given temperature and pressure, the number of molecules present is constant for a particular volume. So when water molecules ( vapor ) are introduced to the air, the number of air molcules must reduce by the same number in a given volume, without the pressure or temperature increasing. Hence the mass per unit volume of the gas decreases, hence the density reduces.

The magnitude of the effect is determined according to the absolute humidity rather than the relative humidity.

Winter colder air has less water in it ( drier ) than warmer summer Air , thus more dense.

---------------------------

Increase in Air Pressure #1

warmer air rises and lifts the atmosphere causing lower air pressure, colder air falls and pushes down causing an increase in air pressure.

The extra surface heat in summer generates a continental region of low pressure, whilst in winter, the colder descending air gives rise to dominant high pressure anticyclones

----------------------------

So winter Air is more dense than summer air due to 2 factors ( Heat and Moisture ) , and the colder winter air has a higher air pressure than the warmer summer air.... high density means a heavy more mass of air to move with the car and more pressure also means a heavier mass of air to move with the car.

So as the temperature drops 3 factors increase the mass of Air to move with your car.

----------------------------

Insight central data for the Insight http://www.insightcentral.net/encyclopedia/enaero.html

In physics, the drag equation gives the force of drag experienced by an object due to a fluid that it is moving through. The force on a moving object due to a fluid is:

F = -1/2 * p * v * v * A * Cd

where
F is the force of drag,
p is the density of the Air
v is the velocity of the car
A is the frontal Area ( Insight = 1.9 Square Meters )
Cd is the drag coefficient ( Insight = 0.25 )

In winter at a certain velocity v the A anc Cd of the Insight are fixed... but the density of the air increases as the temperature decreases by the 3 factors listed above , temp , humidity , pressure

since these effects continue with the square of the velocity ...

while greater rolling resistance from a lower temperature will decrease at greater veleocities due to the outward force of the spining tire making it harder to compress the tire and thus at faster speeds your tire is more ridged... plus at higher speeds the tire heats up more which reduces the flexing losses of the tire.....


----------------------------

At higher speeds aerodynamic drag will always dominate..... these don't have to be very high speeds ... anything above 40MPH the Aerodynamic drag will always dominate ..... as for speeds bellow 40 MPH .... that gets trickier.....
 

·
Registered
Joined
·
66 Posts
Although these may not be "hard numbers", this should help push us in the right direction. See this graph I came across showing air density at various temperatures at STP:

http://www.ce.utexas.edu/prof/kinnas/31 ... ensair.gif

At 32 degrees, air density at STP is approx 1.3 kg/m^3.
At 68 degrees, air density at STP is approx 1.2 kg/m^3.
At 114 degrees, air density at STP is approx 1.1 kg/m^3.

If I'm doing my math right, the force of drag at 68 degrees is about 9% higher than at 114 degrees ambient. At freezing, that doubles to about 18%, due to ambient temperature alone.

I don't have anything to back this up, but it certainly stands to reason that fuel efficiency is linearly proportional to the force incurred from drag. (either aerodynamic, or frictional).

I have kept a paper log of my fuel efficiency for every tank since I've purchased my insight. Unfortunately, I never recorded the dates of when I filled up. Only miles on ODO, miles/tank, gallons, and FCD indicated MPG. However, the data is as follows:

The first 4 tanks averaged around 73.6mpg. I'm willing to throw this data out due to me "learning the car" and the motor breaking in.

The next 6 tanks averaged almost 85mpg exactly. IIRC, most of this driving was during the early spring months in Phoenix, with morning temps cool enough to use the heat, and afternoons in the upper 70's.

The last 6 tanks are my summer tanks averaging 95.2mpg. All of these ttanks were during what I call "hard summer" here in AZ. consecutive 100+ degree days, with 90+ degree mornings. The weather is starting to break now, so it's a good demarkation of the data.

What does all this mean? Well, the math seems to imply that the increase of drag due to wind resistence between early spring and summer here in phoenix is somewhere around 9%. I'd think that I'd expect at least that much of a FE loss due to the ambient temperature.
My spring fuel efficiency figures are 10.5% lower than my summer FE figures.

SO, the data I have certainly seems to say that Aerodynamic drag is THE dominant factor to FE swings in various temperature climes. Not to say that rolling resistence isn't a factor as well, but air density is clearly more than a 50% contributor.

I can post all of my tank FE records if that is of interest to anyone.
 

·
Registered
Joined
·
33 Posts
Infidel said:
SO, the data I have certainly seems to say that Aerodynamic drag is THE dominant factor to FE swings in various temperature climes. Not to say that rolling resistence isn't a factor as well, but air density is clearly more than a 50% contributor.

I can post all of my tank FE records if that is of interest to anyone.
This is what I have suspected as well. I believe the problem with the rolling resistance should more or less "fix" itself after driving for a while, as the tire will tend to heat itself from all the resistance, so it's somewhat of a self-correcting problem. The fact that the Honda cars vent their underhood air over the tires should help evem more. That however may be a large cause of the "warm up" time penalty, aside from the engine/drivetrain components being warm.

The groups investigating the rolling resistance along with the colder temperaturs are usually bicyclists -- in bicycling, rolling resistance is a much larger component of the total drag due to the lower speeds, and as they are at lower speeds, the tires don't have as much of an opportunity to warm up.
 

·
Registered
Joined
·
222 Posts
Discussion Starter #14
I disagree with your statement about bicycles. Rolling resistance is so very very low with bikes that it is overcome by wind resistance at about 15 miles per hour, and drops far far out of sight as speed increases.

I race recumbent bikes and we are very vigilant about wind resistance. Rolling resistance gets none of our attention or concern. With any modern equipment you are good to go. Our tires are at a minimum of 120 psi.

Here is a recent race:
http://www.wisil.recumbents.com/wisil/r ... ctures.htm

I also ride with some upright racer types, and even they rightly know not to give a squat about rolling resistance too, despite their dark ages mentality about bike design :)
 

·
Registered
Joined
·
182 Posts
ethicalpaul said:
I disagree with your statement about bicycles. Rolling resistance is so very very low with bikes that it is overcome by wind resistance at about 15 miles per hour, and drops far far out of sight as speed increases.
I guess that refers to one of my statements but I'm not sure which statement you have in mind. I don't think I made a comparison between wind resistance and rolling resistance on bikes in this thread.

But in order to give you something to disagree with, I'll go ahead and make a few statements :).

At usual riding temperatures, I agree that somewhere around 15 mph rolling resistance and wind resistance are about equal. Exactly where depends on the riding position and the tires, but I imagine it's always between say 10 and 18 mph. For racing, that means that wind matters a lot more. For slow riding, wind resistance and rolling resistance might be more like equal in importance.

My interest in bike rolling resistance was due to my experience riding in temperatures below 10 F, where I noticed it got difficult faster than I expected based on wind resistance. Since I was running knobby tires, and riding pretty slowly, that already meant that rolling resistance was much more of a factor than it would be in road racing. Then add the 2x to 5x increase that I found, and it can become very significant.

Charlie
 

·
Registered
Joined
·
222 Posts
Discussion Starter #16
Double-Trinity said:
The groups investigating the rolling resistance along with the colder temperaturs are usually bicyclists -- in bicycling, rolling resistance is a much larger component of the total drag due to the lower speeds, and as they are at lower speeds, the tires don't have as much of an opportunity to warm up.
Yeah, see, that's what I get for listening to Insightful Trekker about cutting down on the quoting. Me, I am a huge fan of quoting because then you really know what the heck the person is talking about. [mod note: READ the rules, its not personal] {Ethical Paul note: I'm not saying it's personal}

So up there is the quote, which appears in the post just before mine. It was a response to Double-Trinity, not you chrs.

_________________
2006 MT Blue #643
Complete fuel usage log: http://ethicalpaul.com/insight/
 

·
Premium Member
Joined
·
278 Posts
ethicalpaul said:
increased wind resistance as temperature drops
I've been told the exact opposite by a motorsport aerodynamicist! :eek:

At first glance it appears a no-brainer, colder air = more dense = gloopier = harder work to push through it.

But apparently that effect is outweighed by something more significant.

A major contributor to good aerodynamics is smooth LAMINAR airflow over the surfaces and particularly in the wake at the rear. When this becomes TURBULENT, areas of low pressure at the rear increase drag massively.

When you're driving along, you have a thin "layer" of air nearest the car surface that is low speed and laminar (the boundary layer). In the next layer out, it is a bit faster. The next layer is faster still (relative to the car), etc etc.

The ideal situation is that these laminar streams do not mix. When they do, that causes much increased turbulence and drag.

When the air is cold, it diffuses less between the streams and improves the laminar flow. When it is warm, it diffuses more, leading to less stable laminar flows and more turbulence and more drag. (Unlike liquids, when gases get warmer the viscosity, or stickiness, INCREASES).

SO: cold air = more laminar flow = less drag.
 

·
Registered
Joined
·
222 Posts
Discussion Starter #18
Well, Clett, I think your friend is dead wrong, but here is what my physicist friend said:

I'd hate to use the term "dead" wrong, since there could be some circumstances or certain range of conditions where this is correct. First question would be, what speed range and what temperature range is this claimed to apply in? Second question would be, is this claim supported by any empirical data, and if so, what type of experiments? Anecdotes are not evidence in a case like this.
He says that his diffusion effect "outweighs" the density effect. I know how to quantify the latter, but not the former. Air density comes into the drag coefficient in first order. That means the drag force is in a simple proportion with the density. All else equal, the density is in an inverse proportion to the temperature, but on the absolute scale, ie., the real zero is at -273 C. I would be very surprised if the influence of temperature on stream mixing would be large enough to overcome the density increase, but as I say, it is possible, and possibly shape dependent also. Turbulence is complex. It could be a large effect if most of your surface was really in laminar flow, but I don't think that is the case even for cars in the real world.
So until I see any data, I think your aerodynamicist is nuts.

I can tell you that fully-faired bicycle speed record seekers all make their attempts during the hottest part of the summer, and I can assure you it's not because of comfort that they choose this time of year, it's because that's when they are fastest! (They run in Battle Mountain, Nevada).

_________________
2006 MT Blue #643
Complete fuel usage log: http://ethicalpaul.com/insight/
 

·
Registered
Joined
·
1,032 Posts
Laminar flow is almost impossible to maintain on a production car. If you read up on solar powered cars, they make a huge effort to reduce drag, and as a result they have to resort to eliminating ventilation holes and stickers on the front of their cars, polishing the surface, eliminating waviness in the contour, etc. They're able to maintain laminar flow for the first couple of feet, which significantly reduces their total drag, but there is no way that a regular production car with huge radiator openings, rough paint, bugs, etc. is going to maintain laminar flow.
 

·
Premium Member
Joined
·
2,897 Posts
I think the first and major mistake is thinking that laminar flow is not already included in the Cd number.... laminar flow or lack there of is one of many factors that go into calculating an objects Cd which ultimately then determines the objects net drag at a given speed at a given fluid density of the air and at a given frontal area displacement:

F = -1/2 * p * v * v * A * Cd

where
F is the force of drag,
p is the density of the Air
v is the velocity of the car
A is the frontal Area ( Insight = 1.9 Square Meters )
Cd is the drag coefficient ( Insight = 0.25 )

additionally as others have said production cars never get Laminar Flow as it is far too easy to break it up with even dirt and stickers being on the car.... you would need a smooth skin all the way ... no seems... no exterior mirrors... no windshield wipers .... no bumps in the paint ... no stickers .... no front grill .... exterior radio antenna.... etc...etc...

The other mistake is that the idea that greater air density would help to increase a Laminar Effect... It is the opposite.... If an object is aerodynamic enough to get an laminar effect ... and no production car is..... most air planes are not aerodynamic enough to get a Laminar Effect .... Increased Air Density and Increased Air pressure and Increased Air Speed will increase the turbulent laminar disruptive effects of any surface skin defects.... thus making it harder to get a Laminar Effect not easier....

Cd numbers bellow the Insights 0.25 are hard to get .... bellow 0.2 is extremely hard to get.... and Laminar effects which come up for Solar Race Cars become a factor when you are trying to get Cd numbers under 0.1
 
1 - 20 of 64 Posts
Top