my wheel skirt fail. basic design issues? testing? other ideas?

[2000mc]

i was planning to make a set of wheel skirts, they seem like they work for most everyone else. attachment i had in mind would be hard to do an a-b-a on, so i figured i would just slap on a cardboard set to test. i made 9 east and west runs, 3 sets w/o skirts, 3 w/skirts, and 3 more w/o. i know my mileage seems to magically keep creeping up from previous testing.

measured w/ scangauge2
runs 1-3 avg 42.78 w/o skirts
runs 4-6 avg 42.92 with skirts
runs 7-9 avg 43.77 w/o skirts
so it looks like the skirts lost me 0.8% (the west run on #9 was an outlier on the high side but after trying to control for that the skirts still lost me 0.2%)

the skirts are about as long / low as i thought i could run with a flat panel. they get within about 1/2 inch of the tire at the bottom edge. while i didnt think that these would do as well as a longer skirt, i was virtually certain they should have had a positive affect. ...my best idea so far is just to retest, other thoughts?

the test skirts


old pic w/o

[Frank Lee]

Maybe that is all they are worth.

I had skirts on the Coupe and couldn't detect any fe change, better or worse (no instrumentation- tank-to-tank).

[niky]

If the mileage is creeping up steadily, then you have to figure out why and get the car stable enough to have consistent results. Could be the car running in on the oil... or temperatures becoming more ideal as the day gets warmer/cooler or as air pressure changes.

How do runs 3 and 4 compare? How do runs 6 and 7 compare?

[2000mc]

Originally Posted by niky

If the mileage is creeping up steadily, then you have to figure out why and get the car stable enough to have consistent results. Could be the car running in on the oil... or temperatures becoming more ideal as the day gets warmer/cooler or as air pressure changes.

How do runs 3 and 4 compare? How do runs 6 and 7 compare?

1 43.25
2 43.4
3 41.7
4 42.7
5 42.75
6 43.3
7 43.75
8 43.15
9 44.4

The individual east west runs have even greater variations, it is surprising how when all the runs without the skirts are averaged out, how close it is to the average of the runs with skirts. Which i guess is why I used 9 runs instead of 3.

I think the mileage keeps going up with trans, engine oil, tire temps, and maybe increased rake as the fuel tank gets lighter. I had thought before that it stabilized after about half an hr, but had since found it only slowed, and haven't found the point where it stops going up. I chose a test area on the far side of town to help warm things up, and I'm guessing by the end of the last run I put over 80 miles on the car since I left home, and the mileage still climbing.

The car burns oil at about a qt/1500mi which I wouldn't think would have any significant affect. Temps/ weather were basically stable throughout the test, which I don't really think was a factor as I would have been more surprised to not see this happening again.

Anyone have other examples of longer tests showing mileage creep, or totally stable for that matter?

[cowmeat]

You can make them lower, bro! They probably don't cover enough real estate to do much as-is. Black Widow's are fully horizontal, and they come within a half inch of the rear tires. The trick is to flare them out ever so slightly at the bottom, which is accomplished when you attach the front and rear lowest screws/clips.

I would test them for a full tank, that'll give you a better idea if they are going to be worth permanently installing. If you're questioning your instant read-outs, there's nothing like a full tank's mpgs to average all the conditions out.

I don't have any instrumentation, but in all honesty mine really don't make much of a difference, maybe because my wheel wells are smaller than the average car's. They definitely get people talking about the car, though! My installation is on here somewhere if you want to see how to NOT do it, lol!

[oldtamiyaphile]

If SG testing isn't conclusive, it's time to move to coast down tests.

[MetroMPG]

I'm with Frank: it's possible the benefit is so small as to disappear in the normal static/variability of testing.

[dirtydave]

Well I don't think it can hurt!

[Big Dave]

Don't give up. If you never fail, you aren't trying hard enough.

The car does look to have rather tight wheel wells. Maybe partial shirting to fill in the gaps (plus Moon eyes) would be the optimum.

[RedDevil]

The skirts need to be really stiff to work right.
The cardboard does not look very rigid, esp. those 2 vertical folds. If it deforms or vibrates that will reduce its effect big time.

Cardboard is usually much stiffer in one direction than the other.
The central layer forms a wave shape between the outer layers, creating channels in between. It bends easily parallel to those channels, not so across them.
I suspect you used it in vertical orientation (the folds seem smooth, like they'd be when folded parallel with the channels)

If you do a rerun with horizontally oriented cardboard skirts that may show better results. Make it a double layer just to be sure.

Even if it works I'd expect a reduction in fuel consumption of about 1 - 1.5 %.
It will be hard to prove it does make a difference.