Experiment: smooth wheel discs tested A-B-A - 4.6% mpg improvement @ 65 mph - Fuel Economy, Hypermiling, EcoModding News and Forum

Concrete · 08-03-2008, 02:41 AM

I drive a 2nd gen S10 (2000) extd. cab and I've got some data on wheel covers

noticed changes in scangauge readings after installing covers
So I went out in search of A-B-A and (~250 miles later) found +1.3 mpg
On ~28 mpg runs - That is about 4-5% - depending on how you want to figure it
(I've got lots of info on methodology & data if anyone cares to see it)

This is in the ball park of aeroheads posted data for wheel cover savings:
http://http://ecomodder.com/forum/showthread.php/aero-mods-data-change-cd-change-installment-7-a-2684.html

I realize most vehicles (even S10s) may not see that kind of improvement
but have stock "swirlies" and they are very un-aerodynamic
where I live here in Kansas - there are bunches of S10 with these wheels
they aren't bad wheels for stock - aluminum & only ~15 lbs
for those of you who have not seen what I am talking about I have attached photos

also I have an after pic of the covers on the truck
As for my covers - I'm sure you are desperate to know who makes these aerodynamic wonders....
check out pic 3 - Walmart 16" pizza pans - $2.97 ea.

MetroMPG · 08-03-2008, 11:34 AM

Concrete: encouraging results! Yes, please, post your methodology.

I agree with your assessment of the OEM alloy wheels; they are far from the aerodynamic ideal. And the worse the starting point, the more likely smoothies will make a measureable improvement.

Concrete · 08-03-2008, 03:35 PM

The story-

been reading at Ecomodders and slowly modding truck
here is where the truck is at for this testing:
Tune up, Bed cover, E-fan, high end tire pressure, upper grill block.
Scan gauge, driver adjustment

with wheel covers I started to see +30mpg regularly - streight & level - A milestone!
but wheel covers were not suppose to make much difference, right?
I had only put them on so my wheels were not cupping air into the wheel skirts I intended to make
I had not done much aero testing so I was starting from scratch

So I spent an hour and a half driving around trying to get good high speed data
A-B-A-B-A-B.... it became a test of will - nearly 100 miles later I had developed a method
and I got three runs I trusted A-B-A
31.6
30.6
31.7mpg
not very rigorus, but I had convinced myself I had ~1 mpg drag reduction

The Method
Here is where I better explain what my proceedure looks like

basically it is a running start with curise control set at 65mph
scan gauge average calculator is reset at a milage sign post
route is 4 lane interstate - level but with two over passes - gentle curve 90 degrees
outbound is 3.3 miles - return is 3.1 miles - nearly symetric
clover leaf turn around is not in test to reduce the varience I was getting
nearly symetric path back for a two way run - scan gauge reset for return trip
all controls are set the same - vent on - fan#2 setting - radio & lights but nothing else
tested late at night for consistent temp and low traffic
tested in no wind conditions (yes it can happen in Kansas sometimes)

"wasted" a lot of gas - but now have a great baseline for future improvements

I should have stopped - But I didn’t
I knew to convince the Ecomodder community & thus get some data out on wheel covers
Even more testing would have to be done - so I did it

Data to follow

Concrete · 08-03-2008, 04:08 PM

+150 miles later heres the data (raw)

................................out bound........Return..................Average
With wheel covers..........28.5.............29.2
...................................29.8........... ..30.7
...................................29.2........... ..29.7
...................................29.4........... ..29.8
...................................29.7........... ..30.1....................29.6 mpg

Wheel covers removed....28.0.............28.2
...................................28.0........... ..28.5*
...................................28.0........... ..28.2
...................................27.7**......... .27.7**................28.0 mpg

With wheel covers..........29.0.............29.5
..................................28.8............ ..28.0 **
..................................28.7............ ..29.5
..................................28.8 ..............29.4...............29.0 mpg

(sorry about the odd table format - got to be abetter way- but I don't know it

)

A-B-A average 29.3 mpg with vs. 28.0 mpg without

* scan gage was reset 0.2 miles into course due to possible semi interference, so this run is .2 miles shorter
** I may have punched it too hard on acceleration - the ECU seems to stay in a higher performance/ lower FE mode for ~90 seconds or so after +50TPS - My bad - not very Hypermiler like. very noticable loses in real time at beginning of course - not too much difference in the averages over the course.

Variances noted, but all data is used in findings

MetroMPG · 08-05-2008, 11:29 AM

So on average, you saw a 1.3 mpg improvement @ 65 mph, or 4.6%

It's a lot of work to go out and collect data like this. Nice job.

Re. the cruise control - did you set it once and leave it "on" for all A & B runs, and just cancel it with the brake?

Bror Jace · 08-05-2008, 10:51 PM

That's terrific!

So, how did you attach the pizza trays?

Concrete · 08-06-2008, 01:06 AM

Quote:

Re. the cruise control - did you set it once and leave it "on" for all A & B runs, and just cancel it with the brake?

yes, the whole test is "on the fly"
when I enter and leave the course - the only thing I touch is the scan gauge
I found that any other influences created too much error - even a brake tap
everything is setup and in place at least ~1/2 mile before I start taking data
it helps the ECU settle in to cruise
If I had made the pre-run longer - I probably would have fewer asterisks
but it took long enough as it was

I haven't run standard deviation calculations ( I have an aversion to statistics

)
but I have gone this far - I guess I should dust the brain cells off and just do it

Concrete · 08-06-2008, 02:41 AM

Quote:

So, how did you attach the pizza trays?

Disclaimer:
there are better ways to attach wheel covers
especially if you have steel wheels - you can just drill and tap them

but in my case I tried to use the existing hub cover
drilled a hole for a 1/4" screw in the center of hub cover

two-part epoxy two washers & a nylon locking nut on the inside of my hub covers
first washer is for spreading load and giving the nut a bigger foot print
the second washer I fashioned a hex shaped hole to keep the nut from spinning use epoxy on it too (see pic #1)
held it all together with a screw while hardening (screw was oil coated to keep it from being glued on as well)

so in essence I have a tapped hub cover (see pic #2)

benefits: I can remove my wheel covers and it looks stock except a 1/4" hole in middle of my hub cover

down side: the wheel cover can spin since there is only a center fastener
(between a tight fit on the covers in the wheel and notches for wheel weights - that has not been a show stopper - but I have one wheel that will click occasionally on a hard stop)

Arminius · 08-06-2008, 03:29 AM

Thanks for your work on this project. It's good for people to see some empirical measurements.

cfg83 · 08-06-2008, 04:02 AM

Concrete -

I had the same idea because I have the "center caps", but I didn't execute it because I was worried about wobble and them unscrewing themselves. If you could find reverse screws and nuts, and kept track of the placement, you could install them so that they are always being tightened when the tires roll in the forward direction.

I'm also a big believer in extra washers to "spread the load",

.

CarloSW2