Coast Down Testing - 2000 Honda Insight

[euromodder]

We finally get to see you backlog of fill-ups - impressive

Equally impressive is the fact that your mid-2010 highs, are lower than your current errr ... winter-low of only 90mpg .


A real shame Honda isn't producing an updated version of this Insight.

[jime57]

Originally Posted by 3-Wheeler

Hi Jim,
There are several bicycle sites that I have investigated, and rolling resistance seems primarily a "road/tire" function. Tire carcass deflection is a large factor.Jim.

I have noticed that really slick, silent asphalt produces higher FE than stoney, rough surfaces. But, this should not be a factor in your testinng since you always used the same road section.

Can you elaborate a bit of your comment about the "road/tire" function?

Can you draw any decent estimate of fuel economy difference in your tail on/tail off. data? I'm not sure how to draw conclusions from the fuel logs since we don't know when the tail was on and when it was off.

[3-Wheeler]

Originally Posted by jimepting

I have noticed that really slick, silent asphalt produces higher FE than stoney, rough surfaces. But, this should not be a factor in your testinng since you always used the same road section.

Can you elaborate a bit of your comment about the "road/tire" function?

Can you draw any decent estimate of fuel economy difference in your tail on/tail off. data? I'm not sure how to draw conclusions from the fuel logs since we don't know when the tail was on and when it was off.

Jim,

About "road/tire" function, please see here for start....

The Recumbent Bicycle and Human Powered Vehicle Information Center

I have seen postings of how tire resistance is constant with speed, and then John Tetz shows that for his recumbent bicycles, that this may not be completely true.

As for the tail on the car, the car did not have a tail in the winter of 2010. It was not made yet.

Also, note that I discussed the underbody smoothing panels were only providing 70% coverage during this time.

Hope this helps, Jim.

[3-Wheeler]

Originally Posted by euromodder

We finally get to see you backlog of fill-ups - impressive

Equally impressive is the fact that your mid-2010 highs, are lower than your current errr ... winter-low of only 90mpg .


A real shame Honda isn't producing an updated version of this Insight.

Hi Euro,

I thought of you, when I was putting the finishing touches on the tail last fall.

Remember when I was opening up the gap between the tail and the car to match the rest of the body, and you mentioned, "Arg, now the drag will go up"?

Well the last several pictures in my posting on the tail show that this gap has been closed up again.

Thanks for the nudge!

Jim.

[jime57]

Thanks for the bike article. I read about half of it and intend to study it in more detail, but I'm a bit under the weather.

It was particularly interesting that he identified the cold weather effect similar to one I identified by rolling my Insight downhill from my house. I still haven't fully digested his discussion of the variability of Crr but I'll back to it.

I still think that in your case, ambient temperatures for your otherwise well controlled coast downs may have been a problem.

I think I kinda misspoke my question about the tail. I was simply asking if your fuel logs could be analyzed to give a fairly decent measure of fuel economy improvement. I know it isn't the way you wanted to go at the analysis, but it would be helpful to all of us if you did the analysis - which wouldn't be very difficult. Problem with someone else trying to do that analysis is that we don't actually know when the tail was on and when it was off. Of course, there may be other changes which will have distorted the data. Just asking

[aerohead]

Jim,attempting to reduce the data from the last trip has forced me to revisit all the dynamics a vehicle can undergo which affect mpg,and it is a constant reminder of what you've had to fight over all this time.
So we all owe you the big thank you for good fight.
When CAR and DRIVER tested my CRX we had the luxury of a dead flat,tree-lined and windless straightaway at Chrysler's Proving Ground,where we could test un-molested by traffic.
C/D carries there own weather data acquisition hardware with them,so those variables were covered.
The Daytron Messtechnik optical fifth-wheel offers zero rolling resistance of its own,and the on-board computer was automatically recording all the speed/time data at 1/10th second intervals for later data reduction.
Speeds were started from above 70 mph,then coastdown data was captured between 70mph and 20mph.
The low velocity contour would help nail the power absorption coefficient for the tires to be compared later against the speed trace to 'see' the aero effects.
The car had been weighed on four, drive-on scales and was accurate to the pound.We knew the test mass with driver and recording equipment to the pound.
10,back-to-back runs were conducted as quick as the car could be
u-turned,with a weather check at the conclusion.
We ended up with Road Horsepower at 30-mph,50-mph,and 70-mph.
Knowing the R-R power absorption coefficient obtained,R-R power could be subtracted from the Road Load to isolate the aero contribution,and then by knowing the frontal area of the CRX,the Cd fell out of the data.
C/D used a custom program on a HP programmable calculator to do the data reduction.
--------------------------------------------------------------------------
By using the Proving Grounds,we didn't have to deal with traffic,elevation,curves,wind spectra,etc.. With our own weather station those criteria was accounted for.As testing goes,it was good as it was ever going to get.
In hindsight,the only hitch,was the attachment of the fifth-wheel to the passenger door.
It was only when I reduced the C/D data that I realized that the fifth-wheel itself was aggravating the Cd,with its own frontal area and composite Cd.
No regrets,and it was a very interesting experience watching my car go around the high speed oval at top speed.
Since GM has moved onto a military reservation to escape the poveratsi (sp?)
perhaps they'll sell ecomodder.com their old Mesa Proving Grounds for cheap.
Then we can do our own proper testing out of harms way.
I want to personally thank you for all the hell you went through for all of us.
And I'll keep thinking of ways we can do these exotic tests and maintain our sanity.
Your a hero! Thanks,Phil.

[Ken Fry]

Originally Posted by 3-Wheeler

It will take some dedicated equipment to record enough data, and with enough consistency, to really add value to our aero endeavors at EcoModder.

Jim

I agree. It will also require a much higher starting speed, particularly with an aerodynamically clean car. If the car is clean (or heavy for its frontal area) then rolling resistance has larger effect, and for the differences in deceleration to show up a greater speed differential is helpful.

Repetitions on the same day going in both directions are helpful to filter out wind effects. Working on a still air day is almost essential, because cross winds increase drag when going either direction, both from tire distortion and disruption of flow due to undesirable apparent wind direction.

A fifth wheel is easy to set up, and bicycle speedometers (the magnetic counters) give consistent, repeatable speed indications.

Rolling resistance is fairly easy to measure directly, by careful towing, but requires removing axle shafts, etc. With a known rolling resistance, then aero drag can be calculated more confidentally.

Unless you have really awful tires, your Crr cannot be .0137 -- so this figure is taking into account final drive and transmission friction, which is not perfectly constant -- part of it is hydrodynamic. The usual coastdown test calculations assume one constant drag, and one exponential.

But at very least, consistent measurement speed is required.

Thanks for the impressive work and tireless efforts!

[3-Wheeler]

Originally Posted by jimepting

Thanks for the bike article. I read about half of it and intend to study it in more detail, but I'm a bit under the weather.

It was particularly interesting that he identified the cold weather effect similar to one I identified by rolling my Insight downhill from my house. I still haven't fully digested his discussion of the variability of Crr but I'll back to it.

I still think that in your case, ambient temperatures for your otherwise well controlled coast downs may have been a problem.

I think I kinda misspoke my question about the tail. I was simply asking if your fuel logs could be analyzed to give a fairly decent measure of fuel economy improvement. I know it isn't the way you wanted to go at the analysis, but it would be helpful to all of us if you did the analysis - which wouldn't be very difficult. Problem with someone else trying to do that analysis is that we don't actually know when the tail was on and when it was off. Of course, there may be other changes which will have distorted the data. Just asking

Hi Jim,

OK, finally got around to looking at pictures taken of the car with the tail on and off and have updated the fuel mileage logs to reflect this....

Tail was taken off car at 06-05-11 and found water logged

Tail was put back on car at 09-25-11 after painting with primer.

Happy number crunching Jim.

Jim.

[3-Wheeler]

Originally Posted by aerohead

Jim,attempting to reduce the data from the last trip has forced me to revisit all the dynamics a vehicle can undergo which affect mpg,and it is a constant reminder of what you've had to fight over all this time.
So we all owe you the big thank you for good fight.
When CAR and DRIVER tested my CRX we had the luxury of a dead flat,tree-lined and windless straightaway at Chrysler's Proving Ground,where we could test un-molested by traffic.
C/D carries there own weather data acquisition hardware with them,so those variables were covered.
The Daytron Messtechnik optical fifth-wheel offers zero rolling resistance of its own,and the on-board computer was automatically recording all the speed/time data at 1/10th second intervals for later data reduction.
Speeds were started from above 70 mph,then coastdown data was captured between 70mph and 20mph.
The low velocity contour would help nail the power absorption coefficient for the tires to be compared later against the speed trace to 'see' the aero effects.
The car had been weighed on four, drive-on scales and was accurate to the pound.We knew the test mass with driver and recording equipment to the pound.
10,back-to-back runs were conducted as quick as the car could be
u-turned,with a weather check at the conclusion.
We ended up with Road Horsepower at 30-mph,50-mph,and 70-mph.
Knowing the R-R power absorption coefficient obtained,R-R power could be subtracted from the Road Load to isolate the aero contribution,and then by knowing the frontal area of the CRX,the Cd fell out of the data.
C/D used a custom program on a HP programmable calculator to do the data reduction.
--------------------------------------------------------------------------
By using the Proving Grounds,we didn't have to deal with traffic,elevation,curves,wind spectra,etc.. With our own weather station those criteria was accounted for.As testing goes,it was good as it was ever going to get.
In hindsight,the only hitch,was the attachment of the fifth-wheel to the passenger door.
It was only when I reduced the C/D data that I realized that the fifth-wheel itself was aggravating the Cd,with its own frontal area and composite Cd.
No regrets,and it was a very interesting experience watching my car go around the high speed oval at top speed.
Since GM has moved onto a military reservation to escape the poveratsi (sp?)
perhaps they'll sell ecomodder.com their old Mesa Proving Grounds for cheap.
Then we can do our own proper testing out of harms way.
I want to personally thank you for all the hell you went through for all of us.
And I'll keep thinking of ways we can do these exotic tests and maintain our sanity.
Your a hero! Thanks,Phil.

Hi Phil,

Well thanks for the very kind words!

I really agree with you about the need for having a very refined location for doing these coast down tests.

I do recall reading in one of the national design publications about one of the NASCAR teams actually using an old railroad tunnel to test the Cd of their car for more repeatable conditions. Obviously there is going to be issues with compressed air in the tunnel, but at least they can measure refinements to the their vehicle shape on a relative basis.

Phil, I'm certainly not giving up on coast down testing just yet. I am attempting to learn microprocessor programming, which is going quite slowly right now. Eventually I would like to build a circuit board that takes the car speed pulses from the transmission (VSS) and process them to allow proper coast down testing, with hopefully the required accuracy.

If you run across any of those old data files mentioned above, please let me know.

Jim.

[3-Wheeler]

Originally Posted by Ken Fry

......Unless you have really awful tires, your Crr cannot be .0137 -- so this figure is taking into account final drive and transmission friction, which is not perfectly constant -- part of it is hydrodynamic. The usual coastdown test calculations assume one constant drag, and one exponential......

One other forum member calculated that my Crr should be closer to 0.008 or so, and this seems reasonable if it's warm outside.

The above Crr of 0.013 that you are referring to is probably fairly accurate because this was measured in the dead of winter, during some very cold temperatures. Obviously the Crr in the winter will be at least 20% greater than in the warmer months in the summer. This was pretty obvious from all the coast down tests, whether I published the results on EcoModder or not.

Surprising, the Cd did not change as much, since the equations listed on the Instructables consider barometric pressure and temperature (air density) and thus correct for any change in these values, giving a reasonably stable Cd number, no matter what season the coast down testing was recorded.

At this point in time I have no reason to either agree or disagree with my findings, but simply publish what I found, so for me a more thorough investigation will need to wait until a more refined process is developed.

Jim.