The Daq Civic aero project continues . . . slowly
 

Awhile back I started my 1st thread with pics and descriptions of my aero mods--mostly an undertray and front tire deflectors. I've gone a bit further with it over the last few months, adding pieces and refining what was there before.

This is a side-view schematic of my car, a 92 Civic 4D DX 5MT:
http://i656.photobucket.com/albums/u...eschematic.jpg
It's not a bad design for its age. The main problem areas as I see it:

• the steep angle of the rear window
• the rough underbody, with open engine bay and parachute rear bumper void
• the front bumper, which does not extend low enough, so that the stagnation point is relatively high and forces too much air underneath the car rather than over
• the rear body, which tapers gradually to prevent clean flow separation
• the fenders (front and rear), which are too large and do not fully cover the front profile of the tires.

Another thread posted this pic of tuft testing the front. The front of my car is identical in shape to the one pictured.
http://i656.photobucket.com/albums/u...nttufttest.jpg


So far I have mainly focused on improving underbody flow, since it has more room for improvement and it's easier to attach things there. I would greatly appreciate any analysis and suggestions!

This is the front undertray as it was when I posted the first time.
http://i656.photobucket.com/albums/u...wheeldefle.jpg

http://i656.photobucket.com/albums/u...dertraynew.jpg
It is essentially the same now, but I have since covered the strip in the center where you can see the exhaust tube, and also significantly closed up the wheel wells.


This was my first attempt at an aerodynamically shaped front tire deflector (as opposed to the flat deflectors seen on production vehicles).
http://i656.photobucket.com/albums/u...flectornew.jpg
I was not happy with these at all. They are far too small, and I have since learned 2 critical points: it is just as important to trip flow past the low-pressure void of the wheel wells as it is to deflect air around the front face of the tires, and on most cars the airflow approaches this area at an outward-facing direction, so the deflectors need to be placed and angled inward.

I just installed these new prototype deflectors:
http://i656.photobucket.com/albums/u...t/DSC04566.jpg

http://i656.photobucket.com/albums/u...t/DSC04567.jpg

http://i656.photobucket.com/albums/u...t/DSC04578.jpg
I'm reasonably happy with the angles, however I'd like to expand them further inward to "cover" the entire exposed wheel well.


This shows the mid underbody untouched. It's not the worst, but far from optimal.
http://i656.photobucket.com/albums/u...beforeback.jpg


I've since extended the undertray all the way back. These pics show the new section from the front doors forward.
http://i656.photobucket.com/albums/u...t/IMG_2259.jpg

http://i656.photobucket.com/albums/u...t/IMG_2262.jpg
The opening is wider in the view here since this is where I want to extract air from the engine bay. It narrows greatly further back. Unfortunately this zone is too far forward. It is so close to the wheel wells that for several inches the undertray reduces to 2 narrow strips in between, meaning that there is little attached flow at this point. In the future I may narrow the opening at the widest point.


Here is a newer pic showing where I have covered up the place where the exhaust downtube was exposed, and also the lip I added on the leading edges of the opening at the exhaust/transmission tunnel.
http://i656.photobucket.com/albums/u...t/DSC04573.jpg

Another pic of the lip:
http://i656.photobucket.com/albums/u...t/DSC04588.jpg


A view from the rear wheels forward.
http://i656.photobucket.com/albums/u...t/IMG_2249.jpg
Unfortunately the exhaust hangs down below the undertray by a couple inches at the point where cuts over to the passenger side. This will break up the attached flow on the passenger side. I may try to use metal flashing to cover it, trying to make it as flat as possible. I have left a small section of the gas tank uncovered as it is already quite flat.


A bit further back:
http://i656.photobucket.com/albums/u...t/IMG_2246.jpg


A closer view around the rear passenger-side wheel:
http://i656.photobucket.com/albums/u...t/IMG_2253.jpg
The trailing arms forward of the wheels take up quite a bit of space, and since they pivot down at full suspension travel I had to cut out the undertray underneath them. Further down is a pic showing a cover I made for the trailing arms/lower control arms.


The rear of the undertray at the passenger-side corner:
http://i656.photobucket.com/albums/u...t/IMG_2255.jpg
I got as close to the muffler and tailpipe as comfortable. The coroplast doesn't show any signs of melting or warping. I did my best to preserve smooth flow around the muffler area considering that it hangs rather low.


The full width of the rear undetray:
http://i656.photobucket.com/albums/u...t/IMG_2240.jpg
You can see that the trailing edge extends past the bumper, by about 4 inches. This keeps flow attached a bit longer and allows it to break off sharply and smoothly. The angle of the undertray from the rear wheels back is about 5 degrees, well within the range necessary to keep flow attached, but tapering up so that the wake height is reduced. It could be angled more, but there isn't an easy way to do it without cutting up the rear bumper cover.


Recently added are rear tire deflectors (just to have something there; they're from a failed front deflector design) and rear trailing arm/control arm covers, shown here:
http://i656.photobucket.com/albums/u...t/DSC04580.jpg
I'm going to redo both; these are just trial versions.


And my grill-to-radiator duct:
http://i656.photobucket.com/albums/u...t/DSC04598.jpg

http://i656.photobucket.com/albums/u...t/DSC04600.jpg

http://i656.photobucket.com/albums/u...t/DSC04602.jpg
It's very tight in there, so it looks like a hack job, but I think they're effective enough that the same amount of air is getting to the radiator as stock even with a 2/3 grill block.


I've also been thinking about how to reduce the drag created by the side mirror (there's only one on my car). While I can't really reduce the frontal area, I would like to reduce the degree to which the mirror housing breaks up airflow along the side of the body. Looking at newer cars with more sophisticated aero design, it seems that compared to earlier designs they are (1) reducing the frontal area of the arm that attaches the mirror to the body, (2) making the inner edge of the housing more parallel with the body, and (3) making the entire width of the front of the housing angle away from car instead of having the inner half angle toward the car. I used a few scraps to test how easy it is to approximate the latter 2 aspects:
http://i656.photobucket.com/albums/u...t/DSC04574.jpg

http://i656.photobucket.com/albums/u...t/DSC04576.jpg
When I redo this I will do before/after tuft testing.


Finally, here are the places I've tried to seal up the gaps in the hood:
http://i656.photobucket.com/albums/u...t/DSC04605.jpg

http://i656.photobucket.com/albums/u...t/DSC04606.jpg

http://i656.photobucket.com/albums/u...t/DSC04607.jpg


Planned mods to add soon:

• revised tire deflectors
• plastic strips on the front edges of the fenders to trip flow past the wheels (in lieu of full covers--too much work)
• revised smooth grill block cover
• Prius-style tapered extensions behind the rear wheels
• rear trunk lid extension (to lower reduce angle from the top of the rear window so flow is less detached and turbulent as it reaches the wake)

Apology: The more committed among you will be wondering where my coast-down test results are. I don't have them 'cause I haven't done any. I'm reasonably sure that the mods I've done have reduced aero drag measurably, though not nearly to the degree of someone like Basjoos. First, my HWY-only FE is significantly greater (>15-20%) than before the mods (yes, at the same speed and using hypermiling techniques). Further, there are several downhill freeway sections I frequent on which I notice that I can coast down and loose almost no speed, whereas before I would loose quite a bit. I use markers and constant speeds, but I haven't written results down. Alas, I love FE and aero but I'm just not that scientifically minded (and I don't typically have several-hour chunks of free time).

Looks good!

Undertrays are not as easy as some people think.

I admire your attention to detail.


The smaller your radiator opening, the smaller the opening at the rear of the engine bay has to be.

How many repair/maintanence jobs will require you to remove the engine bay tray? How many fasteners will have to be removed?

My engine bay tray was torn off at >100mph airspeed, and it was held on by two bolts up front, two at the back.

did you makes a hole to drain the oils?

nice work on the pans, im interested in your tire deflector design as this and skirts are my next mods,, i noticed that some factory tire spates are angled to defect air under the car, backing up your theory

I have small holes in the undertray for the oil drain plug and the front jack points. I have access to the oil filter and some other rear-engine components through the exhaust tunnel opening.

Unfortunately the front edge of the undertray is fastened by a long row of screws along the edge of the bumper. I have thought that if I have to get more access underneath I will simply cut out a section and reattach it later. I couldn't come up with a good way to make the undertray more modular because shaping and attaching it over quite uneven surfaces required very exacting measurement and the ability of the undertray to bend smoothly over some contours. If I have to remove it, I will probably use it as a template for cutting a new one, as I have a few areas that are patched up.

On another note, I have been extremely happy with the durability of the undertray. My suspension is extremely soft and I live in a city with lots of uneven pavement and steep driveways, so I have scraped bottom many times (I drive kind of hard over bumps too, so as not to lose speed). All the undertray has to show for this punishment is a couple small scrapes and punctures, far too small to make a difference to aero. It's at least as solid as stock.

Tire deflectors are something I've been studying even since I noticed how many cars have them. The are almost always positioned inward of the tire centerline, and for a long time I wondered why. I believe it was here I read in some discussion that the front underside is a high pressure zone, which means the air wants to divert outward along the sides of the car. That was a "duh" moment for me. There is some talk about the wheel wells being a major source of drag and a complex problem in this post of an interview with a BMW aerodynamicist. For more of my thoughts on deflectors and wheel wells, see this thread.

I've paid particular attention to BMW and Mercedes designs. The deflectors often wrap around and curve inward along the arc of the wheel well. Usually the height of the deflectors also tapers as you move inward. A good example is the E-class, which in coupe form has a cD of .24.
http://i656.photobucket.com/albums/u...tic_2010_1.jpg

http://i656.photobucket.com/albums/u...atic_201-1.jpg

My understanding is that they do this to both divert air around the front tire surface and across the void of the wheel well. The large difference in pressure fore/aft of the deflector also makes the wheel well a prime location to extract hot air from the engine bay. This makes more sense than extracting it at the center, where the slow exiting air breaks up the clean and fast flow underneath the otherwise smooth underbody. The airflow near the sides of the car is already somewhat dirty, especially rear of the front wheels, so extracting air there doesn't hurt anything and may even help by equalizing pressure.

im also going to do side skirts on back maybe the front if i can muster the courage to drive such an odd looking car, if the front wheels are skirted, like basjoos with rollers for tire contact do these front skirts change the tire deflector design, if so how?

I'd lower it about 1-1/2".
I have the same car a year newer with a B20B. The lowering on AGX's and sport springs made it handle better, and cut some drag. I see 33mpg in mixed driving with the 2.0L, where I was lucky to see 31mpg with the stock engine/stock car.
The lowering has increased negative camber which wears the tires on the inside a little, but I feel this reduced contact patch is helping mpg. When you throw the car into a turn, tire roll pitches in and you get full contact patch and grip when you need it. I've been running Nitto NT450's on 15" rims in the 195-50-15 size. They have a soft carcass and during cruise I'm on the inside edges, but in heavy braking and cornering the "squirm" makes them hook up well so the car still performs. I get about 30k miles out of a set, but have to have them swapped to the other side or the inner tread is done alot sooner.

I used to have a 91 SI with Eibach Sports and AGXs and performance tires. It was stolen, and I really miss driving it. This 92 DX doesn't handle worth a darn owing to its soft springs and old limp struts, but it's what I got. Yes, new suspension would improve aero and handling, but it's a really old car and I just want to get the most out of its remaining life for as little money as possible. The aero mods I've done so far are extremely cheap and reversible, which is why I've done them.

BTW, I plan to have newer pics up tomorrow, and I'm keen to get feedback on the prototype front deflectors.

what did you use as mounting points? i see you used zipties. where did u attach them though? in the front to be specific

this is all i have aero-wise
http://i183.photobucket.com/albums/x97/imokrs/2.jpg
flush
http://i183.photobucket.com/albums/x97/imokrs/1.jpg

The front edge of the undertray is attached with screws to the front bumper. There are also several places where it's screwed into the factory belly pan to keep it from sagging. I don't remember exactly where all the zipties are attached. I had to get creative as I didn't want to drill any holes. Basically I found holes in the chassis wherever I could and looped the zipties through them. Between the axles most of the support for the undertray comes from having the outer edge stuffed under the plastic rocker panel covers.

BTW, I updated the first post to include newer pics.

Some nice moods, a good inspiration for me - thanks for posting all those pictures

New front deflectors w/pics
 

So I finished my new front deflector design and took a few pics. I didn't get a good reading on my last HWY trip so I don't know if they measurably improved FE, but I've finally got the design right I think, as far as theory goes.

From the front:
http://i656.photobucket.com/albums/u...t/DSC04706.jpg

From the angle at which I imagine the airflow will approach the deflectors (notice the inner side strakes bordering the wheel wells):
http://i656.photobucket.com/albums/u...t/DSC04708.jpg

From the side:
http://i656.photobucket.com/albums/u...t/DSC04714.jpg

Another angle:
http://i656.photobucket.com/albums/u...t/DSC04720.jpg

This shows the little center section of the front undertray I covered up, and further back, the center engine bay vent which I removed the lip from:
http://i656.photobucket.com/albums/u...t/DSC04711.jpg

Next up for mods are a smaller engine bay vent, revised grill block flush with the bumper, revised rear suspension arm covers, revised rear tire deflectors, revised mirror mod, and a trunk lid extension.

i would definitely say look into vacuum forming, your deflector prototype looks great! if you vacuum formed some abs plastic to those contours it would be impeccably smooth

Very very nice !

The angle in the first picture seems to show a large area of the tire that is still exposed to oncoming air. Is this just because of the angle of the camera when you took the shot ?

One other thing that I wanted to mention was that I noticed that you were trying some aero mods with the mirrors.
You are of course aware that our mirrors fold back right ?

Yes, you're right, from directly in front the outer portion of the tire is exposed. That's intentional, and in line with what you see on OE deflectors. The underbody, especially the front, is a high-pressure zone, so that as air approaches it will want to escape out the sides, hence the direction of the airflow coming at the front tires is more like what you see in the second picture. Deflectors that extended to the outer edge of the tire profile would be diverting air further out than necessary, effectively increasing the frontal area of the car as a whole.

On production cars with deflectors you see a variation in how far to the sides the deflectors extend. This is almost certainly because every car will be different as far as the precise direction and pressure of the airflow as it approaches the front tires. It's pretty hard to tell w/o a windtunnel, though tuft testing might reveal something, but in my estimation my car will create more than average pressure in the front underbody, and therefore a greater angle at which air comes at the tires (say 35-40˚). My reason for suspecting this is that the lower edge of the bumper is quite high in comparison to the ground clearance, so the angle of my front undertray is quite steep. I wish there was an easy way to remedy this, but it would be challenging and impractical for a number of reasons, not least of which is the steep approach to many of the driveways I frequently use and my soft suspension--I need that ground clearance.

On the mirrors, yes, I know they fold back. I'm not sure it would help reduce drag. Yes, it reduces front area, but it might cause even more flow separation along the sides of the body. You'll see mirrors on newer (especially high-priced) cars that have a larger gap between the car body and the mirror housing, as well as a smaller mounting arm, I think because it disrupts flow along the length of the car less. Ideally, mirror housings would be as small and slippery as possible and mounted at least several inches from the car.

Now that I think of it, perhaps I should try some sort of extender, barring replacing the mirror with something a lot smaller.

http://farm3.static.flickr.com/2094/...ca1f61e6_o.jpg

I tufted my car with the mirror folded back. It seemed to help, but you can still see some effects from the mirror on the sides of the window.

http://farm4.static.flickr.com/3095/...7b8dc97fcb.jpg

Ah, that's your Civic being tufted. Thanks for posting those. Do you have anymore pics of that test (or know of any others), particularly further back? I am surprised to see that the folding the mirrors in helped, but I need my mirrors anyway.

I'm contemplating what can be done to improve aero at the rear of the car.


The rear undertray extending past the bumper is the best I can come up with for the underside.
http://i656.photobucket.com/albums/u...t/IMG_2240.jpg


Not much can be done with the sides, especially since the rear of the fender curves in so sharply that there is no way to get flow to reattach past the rear wheels (those little vertical strakes are worthless, I realize).
http://i656.photobucket.com/albums/u...t/IMG_2254.jpg
Perhaps I will try wheel covers in the summer, but that will test my fabrication skills quite a bit.


For now I am toying with some sort of coroplast lip at the rear, either from the roofline or trunklid. At the roofline I can keep flow fully attached for another foot or so while reducing the wake profile, but can't extend it further as it would block my sightline from the rearview mirror. Since the flow on the bottom stays attached several inches past the front bumper, the difference between when flow detaches at top and bottom is creating quite a swirl effect, I assume, which is very bad. A lip at the trunklid extending to at least the edge of the undertray can reduce this, but will not prevent flow from separating at the roofline. A full boattail is not an option, as I need my trunk and rear lights, and I'm no basjoos. Here are the two options, with the red dotted line showing the angle from the roofline to the edge of the trunklid lip, which is too steep to keep attached flow between these points, but neither will it break up quite as much as in the wake.
http://i656.photobucket.com/albums/u...ct/options.jpg


What say you?

spoilers
 

Daq,they might be worth a go.Mercedes did this sort of thing before and it made a showing in their wind tunnel.
End-plates would help prevent adulterated flow,in from the sides.They would also give some structural reinforcement.The captured vortex created,would help give you some 'phantom' aft-body.
You WOULD have some 'blind-spot' and you'd have to rely on your mirrors a bit more.
The extra length is always welcome as far as air is concerned.

How about a quickie duct-tape and coroplast test of the green line? You could add some vertical supports underneath the outer edges down to the bumper bulge. That and the curvature of the trunk should make it stiff enough.

EDIT: Aerohead beat me to the post but I think we had the same idea about the side pieces. He said it much more eloquently.

I agree with (both of) your comments. I am not familiar with the idea of a"phantom" body, but it makes sense if attached flow breaks cleanly. I will perhaps try the trunklid extension first, since it will be much easier to fabricate, and also I can do a before/after tuft test on the rear window to gauge whether its helping. aerohead seemed to be saying that the roof extension is the better option (is that what Mercedes tested?), and I agree if done correctly, but it will be harder to make and I would have to get someone to snap pics from the outside with tuft testing it. Side extensions would definitely add to support and make more effective.

Daq, If I understand it correctly, the "phantom body" would be similar to the idea of pick-ups having less drag with the tailgate up. It's not as perfect as an aerocap or full boattail but you still get some of the benefit. With the tailgate up, you have a captured bubble of air that the slipstream slides over. You may not see much difference tuft testing on the rear glass with a trunk extension because of that captured bubble above the trunk but you may have less drag. My suggestion refers to a couple of vertical extensions inside the tail lights at the outside edges of the trunk. The rear end of the extension would match the 7º(?) plan taper from the sides of the car and help blend the side airflow in addition to the airflow coming over the roof and the trunk.

You have me excited to try this concept on my hatchback. I have started on a cargo box style form that would attach to my receiver hitch to serve as a "phantom boattail". Hopefully, Aerohead will weigh in if I have this all wrong.

Why no skirts?

phantom
 

The 'phantom' tail dates to Koenig von Fachsenfeld/Kamm with the 'K-car',or Kamm-back ( but Fachsenfeld deserves equal credit with Kamm).
Kamm considered the 'utility' of a full boat tail car and decided only a 'shortened' version would be practical for parking lots,parallel parking,etc..
So he essentially just started sawing away the tail of his car,as you would with a loaf of home-baked bread.
He discovered that you could remove quite a bit of length without degrading the Cd terrifically,and when examined in the wind tunnel under smoke,the air behind the car skipped over an invisible 'phantom' cone of air which mimiced the original long tail.
Bomber designers noticed the same effect when they chopped a fuselage off to add a gun turret at the rear.
This theme has been re-quantified over the decades with the likes of Walter Korff at Lockheed and Dr. Alberto Morelli at Pininfarina's wind tunnel in Turin,Italy.
The important thing is to fashion everything up to the chop as if you were going to continue to the full tail.
The Aerodynamic Streamlining Template Part-C,2nd image provides a preliminary guide to design from.It's purely amateur,but it's grounded in very good science of road vehicles operating in ground-effect which don't have the benefit of the axisymetrical flow jet pumping action of a free-flying structure.
The template can be used for a roof or side of a car.You'll notice amlost imperceptible curvature at the onset,building into steeper curvature the further back,up to a maximum tangent angle of 22-degrees which can't be obtained in less than one body height length behind the point of max body camber,whether top or sides.

That tecnique is used on theese trains.

The rubber front capture a big bubble of air, and reduces resistance, and it's actually areodynamicly not that bad, as it actually has an invisible rounded nose made of air.

http://funini.com/train/denmark/imgs/ic3.jpg

hows this project goin?

Whoa, aerohead, thanks for the history lesson! Very interesting, and helpful for getting a grasp on what I'm trying to do. This rooftop extension will not be anything close to as long as a full boattail or even kammback, so there will not be that much curvature to speak of, though I will try to approximate the template as far as the profile of the roofline allows, which will mean the extension will have a very shallow angle.

edit-
I just found this quote from you:
"As far as the drag reduction goes,you can't cheat on the curvature,and your drag reduction will be an arithmetic function of the new,smaller wake area the extension creates as compared to the vehicles original wake.--------EXAMPLE: your extension allows for a 10% wake area reduction,that translates to a 10% drag reduction.Simple huh? . . . So using,say the template,you can engineer your extension on paper and know the results before you ever begin buying materials."
Does that mean my extension will only reduce drag to the degree that it reduces the area (mainly height) of the wake? That would make it seem pretty pointless unless it extended quite far--enough so that it could curve down significantly according to the template.
-end edit

From what I've read on the EV1 and others of similar designs, as the body tapers in from plan (?) view, the taper in profile can be greater than if there was no side taper. That would mean the sides of a rooftop extension could curve in slightly and therefore downward and at a greater angle than the middle (if it has side supports) correct? Or is the boatttail template meant to serve for side taper also?

BTW, do you think I can improve the rear undertray in any way?

Here is a link to my Flickr page with a few images that might be useful to someone. The car is a hatchback, so the airflow is quite different from your car.

Flickr: swbin73's Photostream

plan and profile
 

It was Kamm who narrowed the body concurrent with the drop in roofline,providing a progressive reduction of the body cross-section such that when he made his chop,it was at a point where the wake area was 50% of frontal area.
Kamm makes no provision for 'cheating' on the roof in light of the plan-taper.
On a notch-back car it looks like you can cheat the backlight angle when the pillar buttresses are rounded to allow flow carefully in from the sides.Then the whole soup recovers over the trunklid,finally separating at the rear edge.
For a constant,progressive-curve roof it appears that you must stay close to the template until you hit 22-degrees.That angle has been demonstrated to be the maximum for attached flow.
Stay as close to the template line as best you can.Those shallow curves will guarantee good flow.You could always add to it in the future if that struck you.
Reducing the wake area is the goal.
The VW Rabbit photos in the FLOW-IMAGES will show you what the air wants to do.Give it a look.

Cd, thanks for that! I was mainly interested in seeing front and side airflow, particularly aft of the wheels and mirror, so those are perfect.

Add me to the "didn't know what I was doing until aerohead bequeathed unto me his vast knowledge" club.