My new way of measuring on-road drag

[Piotrsko]

Depron isn't available in the US of A anymore but just about anything over 2 lb density squared works. I'm using Dollar tree foam board 20 x 30" 3/16 thick ($1 @sheet) but needs waterproofing as the paper and glues are sensitive.

[Yippeekyaa]

After following this thread with great interest I set out to test and confirm my 1 and only areo mod to my 2011 Prius. It is a lower grill air block that is used 8 months out of the year and has been used for 6 years and 160k miles. It has always added/reduced my mpg by 3 with milage logs and notes of when it has been added or removed.
Built a throttle block from scraps of thin plywood and had to keep adding layers until i was able to keep the speeds down to a reasonable level for testing. Found a nice road for testing and the speeds with or without the block were exactly the same at 94 kmph, set the display to kilometers for better accuracy. This was curious as the benefits of my air block have been proven given the time it has been in use.
After some more thought i plugged in my obdii bluetooth dongle and opened my obdfusion app. Previously purchased the vehicle specific add on for my vehicles and set up a screen with all the various throttle position options. Absolute throttle position%, throttle position sensor output%, and throttle position command value which is in voltages.
Here is what I have learned. In the prius drive by wire system it varies the output based on vehicle speed. At idle, foot off the accelerator pedal it registers .8 volts. With the throttle block in place the output varied according to speed. As the speed increases the voltage increases. From a standstill, accelerator pressed down firmly against the stop it registers 1.1 volts until 25kmh, then goes to 1.2 until 45kmh, 1.3 volts at 60 and 1.4 at 75. Oddly, when going up a long hill speed and voltage dropped all the way down .9 volts and a speed of 22 kmh.
Also confirmed the same readings on my 2017 Prius and my mothers 2014 ford escape.

Sorry Julian, but it does not appear to work on a drive by wire system that varies throttle position in relationship to speed. I will update with photos and a videoi made of the screens previously described. Have to go to work.

[JulianEdgar]

Originally Posted by Yippeekyaa

After following this thread with great interest I set out to test and confirm my 1 and only areo mod to my 2011 Prius. It is a lower grill air block that is used 8 months out of the year and has been used for 6 years and 160k miles. It has always added/reduced my mpg by 3 with milage logs and notes of when it has been added or removed.
Built a throttle block from scraps of thin plywood and had to keep adding layers until i was able to keep the speeds down to a reasonable level for testing. Found a nice road for testing and the speeds with or without the block were exactly the same at 94 kmph, set the display to kilometers for better accuracy. This was curious as the benefits of my air block have been proven given the time it has been in use.
After some more thought i plugged in my obdii bluetooth dongle and opened my obdfusion app. Previously purchased the vehicle specific add on for my vehicles and set up a screen with all the various throttle position options. Absolute throttle position%, throttle position sensor output%, and throttle position command value which is in voltages.
Here is what I have learned. In the prius drive by wire system it varies the output based on vehicle speed. At idle, foot off the accelerator pedal it registers .8 volts. With the throttle block in place the output varied according to speed. As the speed increases the voltage increases. From a standstill, accelerator pressed down firmly against the stop it registers 1.1 volts until 25kmh, then goes to 1.2 until 45kmh, 1.3 volts at 60 and 1.4 at 75. Oddly, when going up a long hill speed and voltage dropped all the way down .9 volts and a speed of 22 kmh.
Also confirmed the same readings on my 2017 Prius and my mothers 2014 ford escape.

Sorry Julian, but it does not appear to work on a drive by wire system that varies throttle position in relationship to speed. I will update with photos and a videoi made of the screens previously described. Have to go to work.

Thanks for that. I am not surpised by the Prius with its 'CVT like' power split device - actual throttle angle won't always agree with accelerator pedal position. I am a little more surprised by the Ford Escape (unless it too runs a CVT?), but if that's what you found, well that's it!

Thanks for the feedback.

[Yippeekyaa]

Your welcome Julian, was very disappointing that the throttle stop method will not work on a Prius. My moms Ford has an 8 speed transmission.

[JulianEdgar]

Originally Posted by Yippeekyaa

Your welcome Julian, was very disappointing that the throttle stop method will not work on a Prius. My moms Ford has an 8 speed transmission.

I've added a note to the first post in this thread.

[JulianEdgar]

Originally Posted by Yippeekyaa

Your welcome Julian, was very disappointing that the throttle stop method will not work on a Prius. My moms Ford has an 8 speed transmission.

There is possibly a way of doing this with electronic throttle.

• You will need two people, flat and straight road, decent speed update OBD data.
  
• Drive car at constant speed eg 100 km/h. You will probably be able to do that more accurately than the cruise control, but see which is better.
  
• Passenger reads actual throttle blade position (or total engine load if such a figure is available).
  
• Figure needs to stabilise eg less than 1 per cent variation as you hold speed constant.
  
• Repeat test and make sure figures are very close.
  
• Open all windows. Go back over same stretch of road with all windows open.
  
• Maintain same speed.
  
• Read what actual throttle blade position (or total load number) now is.

If the blade opening or load number has distinctly increased, then you have a base reading for what causes a major (eg typically 12 per cent) change in drag.

If that is the case, you have a go / no go indicator for aero mods - so long as the change with the aero modification is above the variation in the readings at a constant speed.

However, do not apply any maths to attempt to work out percentage changes - too hard when throttle blade position / load number is probably not linear with power (or torque).

(And I wouldn't do it with the Prius yet, in case electric power is also changing.)

Let us know how you go!

[JulianEdgar]

Originally Posted by JulianEdgar

Edgarwit air curtains tested in final form. Throttle-stop drag testing and wool tuft testing. Also includes data on what proved to be the optimal inlet/outlet duct area ratio. My final figure on drag reduction (on still days, anyway) is 5 per cent.

And now I have tested on a day with gusting 25-30 km/h crosswinds. And, in those conditions, the Edgarwits do absolutely nothing to reduce drag. The throttle-stop speeds with and without the aerofoils is identical.

So that behoves the question: do any air curtains work in gusty crosswinds? I have never seen any published research on this, so I have asked an expert I know. It will be interesting to see what he says.

[JulianEdgar]

I now have three sets of data on the 'windows down' drag increase on the Gen 1 Honda Insight. The figures are (all winds largely crosswinds):

• Very windy - 13 per cent drag increase
• Moderately windy - 10 per cent increase
• Dead calm - 5 per cent drag increase

[Yippeekyaa]

Originally Posted by JulianEdgar

There is possibly a way of doing this with electronic throttle.

• You will need two people, flat and straight road, decent speed update OBD data.
  
• Drive car at constant speed eg 100 km/h. You will probably be able to do that more accurately than the cruise control, but see which is better.
  
• Passenger reads actual throttle blade position (or total engine load if such a figure is available).
  
• Figure needs to stabilise eg less than 1 per cent variation as you hold speed constant.
  
• Repeat test and make sure figures are very close.
  
• Open all windows. Go back over same stretch of road with all windows open.
  
• Maintain same speed.
  
• Read what actual throttle blade position (or total load number) now is.

If the blade opening or load number has distinctly increased, then you have a base reading for what causes a major (eg typically 12 per cent) change in drag.

If that is the case, you have a go / no go indicator for aero mods - so long as the change with the aero modification is above the variation in the readings at a constant speed.

However, do not apply any maths to attempt to work out percentage changes - too hard when throttle blade position / load number is probably not linear with power (or torque).

(And I wouldn't do it with the Prius yet, in case electric power is also changing.)

Let us know how you go!

The obdfusion app does have engine load and engine torque output options, it also has a performance option that measures that is similar to drag racing timing system but uses actual data from the obd port. I will experiment with engine load parameters and see what happens. Thanks for the idea.

[cr45]

Originally Posted by JulianEdgar

There is possibly a way of doing this with electronic throttle.

• You will need two people, flat and straight road, decent speed update OBD data.
  
• Drive car at constant speed eg 100 km/h. You will probably be able to do that more accurately than the cruise control, but see which is better.
  
• Passenger reads actual throttle blade position (or total engine load if such a figure is available).
  
• Figure needs to stabilise eg less than 1 per cent variation as you hold speed constant.
  
• Repeat test and make sure figures are very close.
  
• Open all windows. Go back over same stretch of road with all windows open.
  
• Maintain same speed.
  
• Read what actual throttle blade position (or total load number) now is.

If the blade opening or load number has distinctly increased, then you have a base reading for what causes a major (eg typically 12 per cent) change in drag.

If that is the case, you have a go / no go indicator for aero mods - so long as the change with the aero modification is above the variation in the readings at a constant speed.

However, do not apply any maths to attempt to work out percentage changes - too hard when throttle blade position / load number is probably not linear with power (or torque).

(And I wouldn't do it with the Prius yet, in case electric power is also changing.)

Let us know how you go!

Another way of doing this would be to use MAP readings.

If the engine speed at 100 km/hr is say 3000 rpm then first obtain the MAP reading under no load conditions at 3000 rpm.

This is your zero power MAP or MAP_0.

To make the maths easy lets say this reading was 4.5 psi and at WOT the reading is 14.5 PSI.

If the baseline run gives a reading of 8.5 PSI and the run with aero mods gives 8.0 PSI then clearly the aero mods have reduced the drag.

However if we make the assumption that the change in engine torque output is proportional to MAP for small delta P then we can calculate the approx change in road load.

reduction in road load = 100 * (8.5 - 8.0) / (8.5 - 4.5) = 12.5 %

One caveat here is that road load includes drag due to rolling resistance thus the above result is not a direct measure of the aero drag reduction.

In actual fact the true drag reduction would be greater than the road load reduction.

Having said all that your method should be first choice if at all possible due to its simplicity and the fact that the rolling resistance terms cancel out in the equations.