how to help my matrix

[MetroMPG]

I believe the "reset" option is in the "trip" menu.

EDIT:

FYI, I've tested wheel skirts on 3 vehicles, and the change ranged from very small (couple of percent) to undetectable at 55 mph. The faster your test speed, the more likely you'll be able to measure a difference.

[Weather Spotter]

thanks for the tips!
all the roads around here are 55mph but I could hop on the freeway for a 5 mile stretch and go back and forth.

how many runs should I plan for? I will add a few to make up for bad runs.

[Weather Spotter]

Update:
Both skirts are made and test fitted.
Both front side brackets got a second dose of epoxy and it is curing.
One is painted and clear coated the other one has it's first coat of clear coat on.
I am planning on doing a test drive to make sure they stay on later today.
I may also get to doing some ABA testing today.

How long should my steady state tests be (miles each way) in order to make a valid test?

[Weather Spotter]

update:
Both skirts are done!


On the car:








I also took it for a drive and they work (stay on and do not rub).

Testing data to come soon!

[Weather Spotter]

I ran two tests today, the first one was a 2 mile each way steady speed MPG test, the second was a coast down test from 55MPH.

My data is not what I had wanted but it is something. I think I still need to work on my testing methods.

I will post up my summaries here and do a second post with the raw data later.

Here is the graph for the steady speed test: (Each test data is vertical showing the range of results)


data chart:


Testing notes:
winds changed a lot during the 2.5 hours of testing. They started out a 8-12 mph out of the S,SE. Part way through (A2) they were calm. Then they went back to 5-10mph out of the S,SE. My road was East West.
Temp, 84 RH over 80%, it started sunny then got cloudy.

Speed was 55MPH using cruse (same set point throughout).
The road I used was not that flat, had several high and low points, also the road was well rounded or crowned.

I warmed up car by driving 15 miles before testing.
A= normal no skirt
B= with both wheel skirts on
1W= run 1 West
1E= run 1 East
Each way was 2 miles of testing.
Measurements done with scan gauge on current trip mode, hit rest at start and wrote down number at end of two miles.
Getting good data points going east was harder due to traffic coming that way more often in the afternoon.




And the coast down test:


Chart with raw data:


Testing Notes:
Start at 55mph by cruse, hit reset on scan gauge and clutch at same spot.
Star was still warm from other test (cooled off for an hour during dinner).
I warmed up the car for 4 miles first.
all are .9 + x car lengths, measured by my eye (and by where I stopped).
Since I started at the same spot the end spot is also a valid measure.

[Weather Spotter]

After looking at the steady state data for a while, I noticed that my with skirts data was about 1% lower than the base line. It is well within the Standard deviation of the data sets though, but I think that it is lower. I think I know why as well, the wind. I noticed that during the 3 hours of testing the wind went from not much at the start, to gustier on my B test. For the A2 set of runs the wind was almost none, only a few times did I see leaves move. For the B2 set it was back to windy (it rained 20 min after I stopped testing). Since I was looking for a change at the margins of the data I think that the wind (and time) invalidated the results.

Other ideas?

what should I change if I try the test over?

[COcyclist]

Thanks for posting your test results. It takes a lot of time to do right and can be frustrating if weather conditions don't cooperate. The coast down graph looks encouraging. What does the 7.25 number signify? Also, were you able to leave the car running and cruise on, but disengaged so you could just "resume" to get to the same speed for the A-B Scanguage test? I hope you can find a calm day to try another set of tests.

[Weather Spotter]

The 7.25 is the number of car lengths past .9 of a mile. It is aprox done by eye, but I used the same start point so can judge the difference in the ends.

Yes I did leave the car on and used the same cruse setting for each run. I hit resume before my start spot and gave it time to steady out, then put the clutch in at the end.

I think that today it is a lot calmer out, I might do some more runs. this time I will do it on a flatter road. I think I also will only do 1 mile each way to save time.

other hints?

[gone-ot]

...wind and terrain elevation changes (even slight changes) can affect one-way numbers. To reduce this, back-to-back, two-way tests are preferred (I did automotive-tank fuel economy tests on the GM and Chrysler XM-1 tanks at the USArmy's Yuma Proving Ground).

[Weather Spotter]

All my testing was both ways. start go 2 miles, wright down data, turn around and run 2 miles back to start. the elevation should have averaged out, but the wind was not steady, it was shifting, changing speeds and gusting.