I'm not sure what to make of this exactly... what do you guys think?
 

Rear Wheel Skirts A-B-A Test= mixed results.

So I'll put all the data here and let you guys take a look. I think I'm going to need to test a different day, but here it is:

It's a windy day in Richmond. 14-15mph winds traveling in the SSE direction. The ambient temp was holding at 50 degrees. I was on my usual test area which travels in a relative east-west direction. This leading to half my run being relatively "with" the wind and half "against." I also decided to test at both 60mph AND 70mph (was that a mistake?). The first readings are going from west to east, after the '/' from east to west.

Without rear skirts:

A1 (70mph)- 39.4mpg/38.5mpg [38.95AVG]
A2 (70mph)- 41.1mpg/39.1mpg [40.1AVG]
A3 (60mph)- 47.4mpg/46.0mpg [46.7AVG]
A4 (60mph)- 47.2mpg/46.1mpg [46.65AVG]

WITH rear skirts:

B1 (70mph)- 42.0/36.6 [39.75AVG]
B2 (70mph)- 42.9/37.5 [40.2AVG]
B3 (60mph)- 49.2/43.2 [46.2AVG]
B4 (60mph)- 49.6/43.4 [46.5AVG]

Without rear skirts:

A1 (70mph)- 40.5/39.2 [39.85AVG]
A2 (70mph)- 40.8/38.9 [39.85AVG]
A3 (60mph)- 47.5/46.3 [46.9AVG]


Ok, after crunching numbers:

Total averages indicate that the rear wheel skirts have a -0.7% affect on FE.

However, I decided to compare runs "with" and "against" the wind individually and I came up with this:

With the wind: 4.9% INCREASE in FE
Against the wind: 5.3% DECREASE in FE


It would appear that as long as I don't have a cross wind hitting against my general travel direction then the skirts actually help, but where I'm confused is that if I DO have wind hitting me then it's hurting... I'm not happy.

Redo your testing on a calm day. Wind introduces too much variability to notice small increases. Its hard enough to see a change without wind!

Yeah, I really wanted to get the test done before I leave on a trip to Florida next week. That way I'd know if I should keep them on or not.

what are you using for the readings?
that isn't as much improvement as i would have thought. hope it was the wind.
can you post a pic of the skirts?

+1

Aerodynamics is only part of the parasitic losses, and you're only addressing one aspect of aero drag, so the actual fuel consumption improvement is likely very small from one modification. It takes a suite of aero mods to really reduce fuel consumption.

Keep testing, though preferably on a calm day, or over many days using a consistent route.

Have fun,
Frank

I have a hunch that it has to do with the rounded edges on your rear bumper.
Usually the air besides it is turbulent and the cruved edge has little effect. But on a strong sidewind that turbulence might flatten out, and the air may bend along the corner (the Coanda effect) causing massive drag.

The rear wheel wells add a lot of turbulence, but shielding them over reduces that and increases the drag caused by the corner.
You can easily tape on a cardboard strip or such and see what the effect is.
I expect that the combination of covering the wheel arches and sharpening the corners will have a bigger effect than the combined effect of both measures tested separately.(*)

I was planning to do this mod on my Insight as it was quite sensitive to sidewind, but the effect disappeared by raising the tire pressure so now it would just be for the gains.
But I have not has spare time for the last 5 months or so.
I have bought a sheet of Lexan (which on second thought I will use for something else) and some aluminum stripping to make support brackets and such... but too many plans and too little time.

Tried some ABA testing the other day, but it is hard in a hybrid; no way to ensure the battery involvement is neutral except for pulling the fuse on the whole system.

(*) Like to add, judging from this photo:https://picolio.auto123.com/art-imag...?scale=484x323
that the edge is quite sharp at taillight height but not at the bumper edge, which corresponds to the wheel arch height...

That's a neat theory - I like it. And I agree it's probably worth adding a trip strip at the rounded rear bumper corner.

The new Mirage has this (as do most new/redesigned high-efficiency cars/hybrids):

http://mirageforum.com/imgs/mirage-rear-non-us-can.jpg

Another way to deal with the testing question is to look at the margin of error in your set of results. IE - compare how much variability there is within the A & B sets against the difference between the sets.