Moon Caps - Any Value on 1st gen. Honda Insight? (A-B-A coastdown test)
 

I've acquired a set of moon caps that fit my Insight, but I'm suspicious that they have any value aerodynamically over the stock alloy Honda wheel. Impossible to test on an Insight. Any opinions?

I think you won't notice any gain, because Insight wheels are pretty smooth already. You may try to measure that but you'll add some weight to the wheels so it will affect your FE too but in other direction.

I'd say it'll help overall, but we're talking tiny fractions now, not worth the effort. MetroMPG just did some tuft testing with his insight and the tire area and the flow is pretty darn good around the area.

http://ecomodder.com/forum/showthrea...eel-23529.html

Wonder if they will fit the alloy wheels of an Insight 2?:thumbup:

I wish I'd planned a bit further ahead when I did the tuft test video - I would have stuck on a smooth wheel cover to see what effect, if any, it would have had on the tufts.

As the others have said, I think the gains would be very small.

That said, if I had a set, I'd absolutely put them on.

Yes that would have been nice, BUT I can't imagine the tufts would have been any better with moon caps. The original wheels are so smooth that improvement if any is gonna be really small.

I just got a good buy on a used set so thought I'd grab them:)

I don't think the OEM wheels are that smooth.

They're better than average, sure. But they're full of holes, little spokes between the holes, and those 3 styling "indents".

Well somewhere between your opinion and mine, there is a real answer;)

I just don't know how to get at the answer, since the difference is probably very small and nearly impossible to separate from the noise, given the difficulty of testing on an Insight.

What would you suggest? I suppose that one could do some high speed coast down test from say 70MPH to 50MPH. That would eliminate many of the Insight specific variables that confuse Insight testing. What you think of that approach?

BTW, the Moons protrude more into the airstream than the rather flat stock wheels, so there is that arguement against them. If this protrusion were productive, then Honda could easily have done the same thing with an alloy wheel. I'm just kinda suspicious that flat is better with the other work that Honda did around the fender well.

Your thought welcome:)

I agree the potential for improvement is probably tiny - likely too small to show up above the noise in coastdown testing.

In this case, I really do think a subjective look via tuft testing might be the best way to make a decision.

There may be other reasons Honda may not have gone with a more convex shape - eg. maybe it would have been too likely to suffer curb damage. Just speculating. We know they didn't completely optimize aerodynamics, so it's fair to be skeptical that a particular design choice is as good as it can be.

Tough to say for sure, other than this question gets pretty close to how much gild to put on the lilly. :)

FYI: I'm sure you've seen other Insights with smooth wheel covers, eg. member mattsmalls did DIY covers with fiberglass:

http://ecomodder.com/forum/attachmen...1&d=1273170953

from: http://ecomodder.com/forum/showthrea...tml#post173457

And I've seen moon type caps on other cars, eg Alfred State's competition (efficiency) Insight:

http://img.photobucket.com/albums/09...insightlow.jpg

from: http://ecomodder.com/forum/showthrea...l-20178-3.html

Add one more to the list:
http://3.bp.blogspot.com/_fHAKRh15ey...0/DSC02000.JPG
G/GT record holder at Bonneville.

Interesting, but Moonies are a knee jerk at Bonneville. Where's the wind tunnel data;)

So, if you already have the moon caps, test the buggers! Try something like 80-40mph and see if they show up. Remember, the higher the speed, the more difference aeromods will make and the easier it will be to see them in test data. Maybe try an ABABA?

Smooth and flat is better than smooth and domed. The convex dome sticks out and has more drag than flat covers. As long as you're trying to improve things, go flat and smooth.

I'm pretty sure Phil has posted before that a convex wheel cover/shape is better. I'll search around to see what I can find. Can you point to a source for flat being better?

EDIT: found the post - it's not conclusive.

Not that it would be an ideal wheel from a weight perspective, but for those running stock steel wheels that accept a stainless beauty ring -- would it not be a very simple mod to first fasten (screw/rivet) the Moon (or similar) disk to the beauty ring, and allow the spring tension of the beauty ring to hold the disk/ring to the wheel (it was designed to do so)? This would seem to minimize some of the downside/difficulty in using/mounting/servicing wheels with disks. No?

I think that would work, but why not just buy the clip on Moons. They have little steel fingers and mount like a regular hubcap. I knew one guy who had trouble keeping them on, but others experience is that they work fine.

Only because I have been told the failure rate with the clip-ons actually staying clipped on has not been good. No other reason.

In the case of the Insight, I'm a Honda believer. I deeply suspect that they did the best that they could in the wind tunnel. It is true that the current wheel has some minor styling touches, and some small spoke slits, but it works very well with the current fender design, which incidently is far different than your "normal" car. If I had to venture a guess, and we are all guessing at this point, I'd guess that a FLAT cover would make a small, very small, improvement. If I test I'll do that first :) The Moons stick out way too far.

http://ecomodder.com/forum/showthrea...tml#post332964

:thumbup:

OK, I'm warming to the idea of testing the Moons. I did a measurement with crushed aluminum foil behid the Moons and found that they are only .6" more into the airstream. I'm trying to find a way to mount them temporarily without drilling my wheels - for the time being.

I went out looking for a test course. Richmond is terrible with hills, but I think I found an essentially flat section I can use. I did a few experimental 70-50mph coastdown runs and I think the course will work ok. Perfectly flat would be better, but that just isn't available anywhere around here.

I do have one technical problem. How in blue blazes do I mark the end of the coastdown at 50 MPH? Remember, I'm going 50 so I can't just step out and put a chalk mark on the road;)

Ron the hybrid-battery-repair guy had used velcro to attach moon discs to his Insight, and reported no problems over long distances. I don't know how he stuck the velcro to the wheel or disc though. (Useful, aren't I!)

I'd be more comfortable with something more secure. To avoid drilling the wheel, you could fashion a sort of plug that fits through from the back of the wheel and slots into one of the openings. Then drill/screw into the plug.

You will have to time the runs, not measure their distance. Using a video recorder on the speedometer would be more accurate than eyeball/brain/stopwatch. Changing your speedometer units to km/h should be slightly more accurate still (higher speed resolution).

I still don't think you'll be able to detect any difference, above the normal variability you'll get from multiple runs. Don't let me discourage you though!

MOONs
 

In the mod-data sticky there is some data from GM's wind tunnel development work for their Pontiac Trans Am Firebird at the Lockheed Marietta,Georgia facility.
They did comprehensive evaluations of wheel covers and their 'MOONesque' cover delivered the lowest drag.
Boeing uses them on everything they build for Southwest Airlines,and all the fastest LSR streamliners typically use them at Bonneville,as does Danika Patrick on her Indycars (both in and outside application).
The subtle convexity lowers drag (as with GOODYEAR's streamline tires for aircraft) compared to a flat disc and they are very strong for their thickness and are not subject to flutter.

When I picture this from above I see the back half of a truncated Kamm tail. Perhaps the slight frontal area penalty is more than offset by the air coming together more smoothly behind the tire phantom boattail.

Well, my first attempt to experimentally measure the Moon benefit, if any, failed. Tried to attach them with double sided tape. One flew off at about 60 MPH and I spent 20 minutes looking for it in the grass. Glad it didn't stay onthe road and wind up in someone's windshield:(

I checked all nine pages of that thread and could not find that data.

Are you sure it isn't somewhere else?

LOL, jeez. Find a way to zip tie em to the spokes

somewhere
 

I found what I was thinking about at the mods-data #7 Wheels /Tires.
Larry Shinoda had a secret-weapon MOON-esque wheel cover they tested on the Trans Am 'bird.Compared to an open steel wheel they lowered the Cd by 0.027.
They were lower than the OEM wheel which ended up on the car.I thought I might have posted that comparison.I'll dig it out.

Thanks Phil. Found it.

Quote,"1982,Pontiac Trans Am Firebird wind-tunnel studies show Larry Shinoda's trick aero wheel-covers trim 0.027 off drag coefficient,compared to a open wheel."

The baseline for this test seems to be an "open wheel." I seriously doubt that the improvement would be anywhere near that great if the Insight wheel were used as a reference. It is by no means "open."

Still, I like the moons and will give them a try. Already drilled and threaded my "worst" set of Insight wheels :)

I'm having doubts about the 70-50 coastdown. The problem, as I see it, is that there are multiple forces acting on the car, other than aero drag. I've played around with the math a bit, but I don't see any way to resolve the portion attributable to aero drag. So, it is not possible to determine from the coastdown how much the aero drag is changed. It is true that the "other" factors which I can identify, primarily tire drag and drivetrain drag are constants, but there doesn't seem any way to decide on the fraction, and therefore no way to get very close to an improvement factor for aero drag.

I suppose one could resort to an approximate fraction, and that would result in a derived aero drag change.

Why are you trying to determine Cd? Isn't your primary goal to know whether smooth wheels are better than stock wheels? You don't have to calculate Cd to answer that question.

That said, I will still be very surprised if coastdown testing shows any difference at all (too small a change to the vehicle).

reference
 

Somewhere,I have a case study which summarizes the minutia of Shinoda's tests on the Firebird.I'll find it asap and report back.There were some other wheel configurations tested and you are very correct to think that the MOONs won't 'show' on the Insight like they might for the open wheel.

Moons Don't Help Insight
 

Well the very short answer is in the title.

A slightly longer version, after testing coastdown times from 70 to 50 MPH in bidirectional testing, one cannot discern any improvement or penalty in using Moon wheel covers.

The longer, data version. I selected an essentially level course of about 1.5 miles where I could find convenient turnarounds. I tested over a course of about 3 hours, beginning at 11 am. The data:

No Moons:
East 33.89sec
West 36.01sec
E 32.4 sec
W 38.56 sec
E 35.19 sec
W 37.33 sec
Ave 35.72

Moons
E 34.03
W 35.95
E 36.87
W 36.53
E 37.25
W 37.51
36.35

No Moons
W 38.5
E 36.10
W 37.67
E 35.48
W 38.81
E 35.84
AVE. 37.24

There was a fairly linear minor increase in coast down times as the day became warmer. Anyone who drives an Insight will be well familiar with the improving MPG with temperature. In this case this is entirely dure to the warming ambient temperature and NOT to car warm up, as the car was thoroughly warmed before all segments.

If one averages the first and last segments to take out the temperature gradient, one gets 36.49 sec, no Moons. The Moon cap time was 36.35.

I'm left with several conclusions:
1. Stable temperatures, like still winds would be nice if we could get god to cooperate.
2. The Insight loves warm temperatures.
3. When compared to stock Insight wheels, Moons are a wash - no better, no worse.

Lots of time invested for no improvement in the end, but then there are never any guarantees.

Moons MIGHT improve on the stock wheels at much higher speeds, but that is not the hypermiler's world;)

Thanks for doing that test. Another collection of data for the pile doesn't hurt.

Yes, that is very true and useful in itself. But we modders always hope we have found something;)

I think that if the test were against a different wheel on a different car, then the Moons might show improvements. It's just that on an Insight they seem to be a wash.

I still plan to do some tuft comparrisons when I can get my setup together and get the time and conditions. I'd be interesting if there is any reaction of the tufts.

The Moons look neat, so I'll probably keep them - since I don't seem to be paying a measurable penalty.

There is another issue which might actually be more important to us in the ecomodding community. High speed coast down tests seem to be so full of pitfalls that they may be unreliable in general. Hucho mentions one of the famous aerodynamists who got false results - forget which one. A thorough, honest, and well controlled A-B-A test is going to take 2-3 hours to conduct on any reasonable course with turn arounds, acceleration zones, stabalization zones, etc. During that time, there is a very high porobability that some of the environmental conditions will change, thereby "swamping" the small changes we are trying to gather. I think I learned my lesson.

I agree, but would restate slightly to say the takeaway from this is that testing modifications with likely small effects is difficult.

The larger/more effective mods have a much better chance of showing up above the normal noise of variability in a set of runs.

Yep, but you might still miss the true value even on the big ones. That is one reason that I "copy" the results of trend lines of multiple other testers when it comes to the Insight. (The Insight itself being practically impossible to test.) We know from many tests here that removing both mirrors is worth about 3%. It is sort of a Oracle of Delphi method of seeking the truth;)

The low fruit on the Insight is the boattail. How do I know? Several folks have built and tested boattails and gotten 12-15% improvement on Insights and other cars. The "trend" is to the 13.5% =/- range:)

Trans Am & Subaru XT
 

I dug out the info for the above cars.
*On the Firebird,the vented hubcap showed a 5.7% drag reduction
*The MOON-esque hubcap measured out at 7.6% reduction.
----------------------------------------------------------------------------
With the wind tunnel development of the Subaru XT,the final 'vented' production hubcap showed 0.008 delta-Cd (2.2% drag reduction)
*A completely sealed cover was good for 0.01 delta-Cd (2.75% drag reduction).
----------------------------------------------------------------------------
It would probably require a wind tunnel to discern the difference in performance between 'aspirated' and 'sealed' covers.
On-road testing with the slightest change in any parameter,might slew the scatter-plot enough to 'hide' the very data we were looking for.:o

...true, but it's so much FUN picking the fly·$hjt out of the pepper :D.

Jim,

I read through the entire thread, and did not notice a detailed picture of your particular moons, as this *could* make a difference, depending on how well they are executed.

Jim.

Jim,

I think that simply stating that a tapered tail yields an aero improvement of 12 to 15% is over-simplifying the situation.

Why?

Yes, it's true that the Insight benefits from a tail by about 10% from a gas mileage standpoint, and about 18% from an aero standpoint.

However, we need to keep in mind that on the Insight, the hatchback of the car is already closing down in shape, and that the addition of the tail does not add that much aerodynamically than it would on a car body that does not close at all.

In that case the addition of a tail might be 20% gas and 35% aero, and even higher gains at highway speeds.

Jim.