2013 Subaru Sti side mirror delete coastdown. Tuft testing too

[jedi_sol]

2013 Subaru Sti side mirror delete coast down tests. Smaller side mirror tuft testing.

I finally got around to running more tests on my car. Side mirror delete coast down tests. The weather conditions changed during the course of the experiment, however I am still going to post my results because I don't want 4 hours of work to go to waste.

Time 12am -4am Pacific time
Temperature 60*F
Humidity: 9%
Wind: calm to 9mph from the northeast
Location: business district, late at night, flat road (slight grade), no traffic.
Method: from a stand still, drive up to 40mph, shift to neutral, coast down to 20mph, them time the results. Bi directional runs to average out the grade and wind.
Tools: Torque pro app, used a graph to measure speed (y axis) vs time (x) axis. Then after each run, I used the graph to see how many seconds it takes to go from 40-20.Very useful!

I have other mods installed (upper AMD lower grill block) but this test is for the mirrors because they are easiest to take off and on.

Mirrors on
Runs A:
39.9 sec
41 sec

Runs B
40.9 sec
41sec

Mirrors off
Runs A
38.5 sec
41.8 sec
47.5 sec (winds started to pick up)
46.6 sec

Runs B
42 sec
41.8 sec
41.4 sec
40.4 sec

Mirrors on again (still windy)
Runs A
45.7 sec
45.7 sec

Runs B
38.9 sec
39 sec

Mirrors off again (still windy)
Runs A
44.3 sec
45.7 sec
46.6 sec

Runs B
39.7 sec
40.1 sec
40.7 sec

Total average mirrors on: 41.41 sec
Total average mirrors off: 42.61 sec

Approx 2.82% increase in rolling time with the mirrors off. That's within the "noise" so in conclusion, this test was inconclusive. Next time I will.definitely need to test on a more calm day and also test at a faster speed, perhaps 60mph down to 30, because aero mods have more impact at highway speeds.

I'm also CAD designing a smaller aerodynamic mirror and did some tuft testing





Stock mirror


Here is the video. Passenger side is the small mirror. Driver side is the stock mirror in the open position. As you can see, the driver side tufts are flailing more violently than the passenger side.

http://www.youtube.com/watch?v=tSaGC...e_gdata_player

[redpoint5]

The video is private and cannot be viewed. Thanks for taking the time to post your results though!

[jedi_sol]

Oops, try now!

[UltArc]

Do you have any highways with elevated sides? I think the purpose is to reduce noise pollution to nearby homes, but it's great for deflecting the wind from the road, moving over it instead.

[MetroMPG]

It bites when testing is messed up by mother nature. Been there, done that. Best laid plans, etc. Thanks for posting.

[jedi_sol]

Originally Posted by UltArc

Do you have any highways with elevated sides? I think the purpose is to reduce noise pollution to nearby homes, but it's great for deflecting the wind from the road, moving over it instead.

Yes, but in so cal, it's hard to drive on the freeway with no other cars around, even late at night!

[Xist]

Originally Posted by jedi_sol

Approx 2.82% increase in rolling time with the mirrors off. That's within the "noise" so in conclusion, this test was inconclusive. Next time I will.definitely need to test on a more calm day and also test at a faster speed, perhaps 60mph down to 30, because aero mods have more impact at highway speeds.

Four runs with the mirrors on, seven with them off, another four with them on, and an additional six with them off. Aside from the fact that you have uneven numbers, I do not know how many more trials you need to establish a pattern. I believe that most modifications would yield a 3% improvement or less.

Keep up the good work!

[jedi_sol]

Yeah, the first mirror on runs were pretty consistent, hence I chose to only do four runs. However half way through the mirrors off runs, the winds started to pick up (coasting times greatly increased). So I decided to do extra runs with mirrors off /on (with windy conditions).

Instead of abandoning the experiment, I decided to put the mirrors back on, this time with the windy conditions. Again, the times remained consistent, hence only doing four runs.

Then a few more runs with mirrors off (windy) to add more ABA data points so that all in all I had runs with mirrors on (no wind), mirrors off (no wind), mirrors on (windy), mirrors off (windy).

[oldtamiyaphile]

Originally Posted by jedi_sol

Tools: Torque pro app, used a graph to measure speed (y axis) vs time (x) axis. Then after each run, I used the graph to see how many seconds it takes to go from 40-20.Very useful!

From my experience, phone/car type GPS isn't accurate enough to pick up small gains. If you watch your GPS speed it will often jump 5km/h at a time. Even in gentle coast down tests 3km/h jumps are typical.

I have a bicycle Garmin unit that will log and stamp at selected intervals down to 0.1km/h. I've used it before in AutoX because I can actually see my lines through corners with it along with terminal and apex speeds. Only now do I realize it would be a great way to do both real world and coast down tests.

I did my first coast down testing last week just to play with the method, I used the Scangauge numbers because it was easiest (these also jump at 3km/h intervals in coast down testing). To my surprise, when I ran the numbers my calculated cd matched the factory figure (although my bigger tyres should have made it worse).

I also tried mirrors folded and could find no difference (these are big mirrors too). It's worth noting that some cars like the Ford Falcon AU were reported to have a lower CD with mirrors than without. The mirrors were designed to help deflect air away from the B pillars. I imagine this is the case on 90% of new cars.