Insight undertray
 

The Insight is a nice example of function over form when it comes to aerodynamics in production cars. But can it be improved further? Autospeed gave it a shot, and it didn't work.

What went wrong? Well, I think it's simply that a new undertray did not have the potential to make a huge difference on this car in the first place.

Darin, with his Metro's "aero dirty" underside, when compared to an Insight, only saw a 2.2% increase from a full belly pan.

The way they monitor fuel economy on the Insight, in L/100km with one decimal, is quite imprecise. That is especially true when you get in the sub 3 L/100km, where each tenth change is at least a 3% change in fuel economy.

The wheel fairings were a nice touch, although they could have been better designed.

But anyway, thanks to autospeed for their newest aero article, it's always nice to read peoples' hands on experimentation.

that article really did not seem very scientific. Assumed values, lack of equipment, just seems like there are a lot of variables he left open. Plus he was asking for a 5% improvement on a car that is pretty close to maxed out. Id like more data before any real conclusions can be made.

OK Jazzie604, how exactly would you have done it?

Julian, 5% was asking for quite a lot from just a rear undertray if you ask me.

If the wheel fairings would have had rounder edges, if you would have done the same for front wheels (like you did for the 1st gen Prius), and if you had put small boat tails behind both rear and front wheels, along with full rear wheel skirts. I'm pretty sure you would easilly get your 5% right there.

I was looking for a detectable change. The Prius front undertray gave that; the Honda rear undertray did not.

Therefore, as the article states, the rear undertray was, for me, ineffective.

As was done initially on the Prius, the depicted undertray was a trial - if it had made a detcetable improvement, it would then have been further refined.

Re wheel skirts - this was a trial of an undertray!

Hi Julian,

I'm wondering if the edges of the Coroplast (?) like just ahead of the rear axle would have "caught" the air flow?
http://us1.webpublications.com.au/st...110872_9mg.jpg
I was concerned about the sides of my rear belly pan would do this... If they were catching the air, then this might have negated the (possible) gains of smoothing out other areas.

The boundary layer back there would be way thicker than the edge of the coreflute - I would have thought so, anyway

Hi Julian,

I'm thinking that the air can get "caught" above the Coroplast? It might be acting as a scoop...

What about building up the side (curve or just angle) to catch the air and work it down under the panel?

That's what I've been doing on my Jeep to get around suspension and various other things hanging down. Seems to be working on the Jeep. But the under body of mine is a hell of a lot messier.

tfirst of all, great that autospeed continues to explore aeromods, and other tweaks with FE in mind, it's on of the reasons it's in my top 3 car sites next to ecomodder and gassavers!

A few points i was wondering about is if perhaps extending the sides of the new wheel fairings beyond the inner sides of the wheels wouldn't have been a good move. they could have ran the entire remaining length of the car preventing air from getting in the rear wheelwells and keeping the rear wheel wake separated from the underbody flow.

yet perhaps the lack of improvement comes from the wrong assumption.

at this point along the length of the car diffusion might be more important that attached flow. that's to say that the design is so that theres a smooth gradual transition from the fast moving air under the car to the slow air behind it even if this is made invisible by the somewhat untidy underside (although most cars look far worse than the insight)... afterall it's a pressure difference.

imagine a car with a perfectly flat underside parallel to the road.
sure the underside will promote smooth airflow but when the air reaches the rear bumper suddenly the pressure will drop in an instance.

where a car with a slightly rougher underside that slopes up near the end would allow for a much smoother transition in pressure.

if the underside of the second car isn't to messy, perhaps it will have a lower Cd than the first while appearing less aerodynamic to the untrained eye

a flat underbody might server to prevent turbulence, but can it restore it if it already exists? as the boundry layer is likely quite think form the uneven surface before this might make the actual surface on the aft portion of the car less critical. and so all your left with is the original diffusing effect of the average shape of the rear. (on the insight, combined with the rear bumper design this is likely to function very well)

what i'd try to do is add some vortex generators to the front of the rear tray, as these might tidy up the uneven flow so it can reattach to the tray, or add finlets as on a racing diffuser to guide the airflow more.

on the other hand given the results, it's quite fair to draw a line and conclude the insights aerodynamics are very good and the undertray is not going to make the difference that would justify the trouble.

still i think a similar setup on less smooth cars (wich are more common than the insight) might illustrate more clearly if a rear undertray has pontential or not...

anyway, regardless of the outcome i'd like to see more like this on autospeed!!!