Rear aero project - Civic notchback coupe
 

Current status of project - added roof extension panel. Also tapered the sides inward towards the rear. Did this by cutting away the lower edge, so the side panels can rest on rear qtr panel as they taper inward towards the trunk lid edges. I also added several aluminum bar stock stiffeners to the main lower center panel.

Have been driving with this setup for two weeks with pretty good FE numbers in fall weather.

If I'm very fortunate I'll be able to do some coastdown testing this weekend.

http://i188.photobucket.com/albums/z...02MoreBits.jpg

++++++++++++++++++++++++++++++++++++++++++++++++

Original posted message starts here:


With the rear of a notchback coupe contributing seriously to car's the aero drag, I've developed a plan to try to remedy it on my Civic.

This design is based on earlier discussions in this thread:
http://ecomodder.com/forum/showthrea...les-17796.html

The concept is to plant a long, flat spoiler with the rear edge at a point meeting the template curve. While not as effective as a proper shortened boat tail that would extend to the same point, the side panels will help sequester the air in the "chamber" above the panel, more or less "faking" the template as far as the air flow is concerned. And, you can see out the rear mirror and, with a few cuts and hinges, can still open the trunk lid.

Basically, it is a flat spoiler panel, who's rear edge meets the template curve, above the bumper's rearmost point. Panel's forward edge meets the rear glass. Two side panels extend the C-pillars rearward to a point even with the bumper's rearmost point, and support the flat spoiler panel.

The spoiler's forward edge is aimed at the rear view mirror, so the spoiler should appear as a thin edge from the mirror's viewpoint, and not block the driver's view out the back.

I also came up with a design for a bracket to attach the side panels to the edges of the rear glass. I expect to need additional attachment points but this has the benefit of being cleanly removable later if needed. Attaches to glass with silicone glue or double stick foam tape.

http://i188.photobucket.com/albums/z...sideviewv2.jpg
Photo above:
Tan = spoiler panel, aimed at rear view mirror.
White lines = template curve + vertical.
Yellow = side panel.

http://i188.photobucket.com/albums/z...aller02a-1.jpg
Photo above:
Heavy tan = spoiler panel, aimed at rear view mirror.
Thin tan = fold/hinge in spoiler, to allow trunk opening.
Yellow = side panels.
Thin red = cut/hinge in side panels to allow trunk opening.

http://i188.photobucket.com/albums/z...earglass4b.jpg
Photo above:
Mount bracket attaching side panel to rear glass.
To be silicone-glued to glass.

If I understand your image correctly, you have two side panels that match the shape of the template, but the central section actually extends from center of the rear glass and acts as a raised Bonneville style flat spoiler ?
If that's the case, then why not just have it extend from the roofline ? It could be mounted on a hinged edge so it could swing upwards with the trunk. It would them match the template curve without the additional drag of a flat spoiler design.
Of course it would have to be transparent.

I probably am not understanding you image . If that's the case, nevermind.

It might help if I actually read the post !
So yes it looks as though you intend to create a flat spoiler versus an actual Kammback.
Just curious why ? Why not go up a few inches and have it hinge from the roofline ?
Either way, I'd like to see your results with this setup .

I'm also curious to see what kind of vortices will form as the air spills over the sides and down into the depression.

We have had PM exchanges on this, but I just want to note that I'm really interested in wider response to this idea and your results. Eventually, I want to build a Kammback, and this mounting may be my prototype.

Though I don't plan to use the kind of plane you plan, I like very much your effort to resolve visibility with a panel that disappears into the line of sight from the inside rear-view mirror. Very clever. I experienced that with one of my abortive "raised trunklid" designs, and I can see it working-out really well. As long as the aero benefit is real and significant... that's the debate, eh? Build it. Test it. Good luck... I'm following too.

do you have any info on your wheel covers? they look nice!

Sharing pics from another thread...
 

(From - Aero mods for chumps thread)

Magical numbers of 10-12 degrees.
Here are just visual concepts. (quick tire change time is valid, could use quick release 1/4 turn camlocks or just delete)

Conceptually, there are rear vertical fences that holds the new lexan rear window/fastback at the best low drag angle. Ideally a true tapered boattail is the best but this might be the easiest to fabricate & even easier to test & adjust the back angles. You could also adjust trim edges depending on your track venue.

http://ecomodder.com/forum/attachmen...5&d=1296259422

http://ecomodder.com/forum/attachmen...6&d=1296259422

I love the second photo. Very cool look. It's a model for what I would hope to install one day (quite a way off right now). I would build it entirely clear for visibility.

http://ecomodder.com/forum/attachmen...6&d=1296259422

Brucey Babe...I was thinking of doing something similar to my 04 Civic, my plan was to design it so it had pressure points on the top of the class and rear or the trunk lid, then I was just going to put metal bracket hooks on a cinch strap and hook the front of the trunk lid on the outsides, then pull tight, so it would only rub on the car in 2 places, and the hooks would contact the bottom of the front of the trunk (bottom of the rear glass area) that nobody would ever see. It wouldn't move cause you are jamming a "V" into a "V" shaped hole essentially. I'll post an illustration later, on the road right now.

I think I get your idea. Pulling the spoiler assembly forwards against the glass and down against the trunk lid. I think you would need to take it off the car to open the trunk - but maybe you have a way around that.

Thanks, tru.

Have a look here:
http://ecomodder.com/forum/showthrea...tml#post250009

There's some info there on how I made the the discs. I used the acorn nut attachment method discussed in the beginning of that thread. You might want a different attachment based on how your wheels are designed. I definitely like some version of attaching to the lug nuts or to decorative caps that go on the lug nuts, it avoids drilling into nice alloy wheels.

Actually this is a very tempting thought. After all, the side panels are very close to the template curve so why not put the center panel on the template and really get it right?? The trouble is, all the neat solutions in the existing concept go out the window (nearly literally), or you might say out the roof.

For an mpg competition or a car show, I maybe could just make it out of cheap plastic and tape it on. Build it and use it, then take it off. But for long term use with full trunk access... ... ...

The toughest challenge I see in extending a panel from the roofline is that the roofline is bowed upward at the middle, when you view from the rear. This makes for problems in hinging the panel for trunk access. Couple that with the preference for a panel that curves downward towards rear per the template, and that means the panel would have a compound curve, not easy to build or even to fake.

The hinge for the horizontal panel - to enable opening the trunk - is what I'm having a very difficult time picturing, due to the bowed roofline. I think the widest possible spaced hinges would be two, about a foot apart, centered on the roof transverse line. The outer corners of the panel, where the roof meets the C-pillar, would have to be free to hinge the panel up to open the trunk. I could anchor those front corners with removable screws, and would have to remove them before releasing the trunk latch. Even so, I fear the plastic would eventually crack when lifting it up supported only by two hinges at the forward end.

To give it the compound curve, I could cut a lengthwise slit in the panel, and slightly overlap the edges. Or turn the slit into a V-cut and just bring the two edges together. Either way, the plastic panel is weakened by the cut, and would need fasteners down the center.

Cd, ya got any ideas on this stuff?

I hope Cd has answers for your questions, and I don't want to distract from your request for his thouhts: But I have a different suggestion. What if trunk access were significantly rethought. What if the trunk lid itself were modified so that the lid only lifted 6" and the aft portion of the lid, with the tail lights, swung down like a tail gate? The clear kammback would not need to move anymore.

Now you're thinking outside the box, um trunk!!

Subscribed!

MetroMPG (or anyone else with a good suggestion on this),

Any tips on how to attach the side panels to the C-pillar and the quarter panel? This is turning out to be important; the clever mount brackets I built won't pull the forward and rear corners of the side panels against the body.


I suggested rubber expansion plugs (aka well nuts) to a friend of mine.
http://i16.photobucket.com/albums/b1.../well-nut1.jpg
Flange side (right) is where screw goes in.

http://i16.photobucket.com/albums/b1.../well-nut2.jpg
The plug is inserted in opposite direction here; the flange is on left side under the washer.

My friend is concerned about:

A) There's additional folded-over steel layers at the top where the roof & C-pillar meet, put there for strength. Probably don't want to drill into there where it's supposed to have strength? But that's exactly where one of my attachment points needs to be. I haven't come up with a fix for that issue.

B) The expansion plugs will allow rust sooner or later. We've both seen it happen, but 10+ years after a roof rack was installed. I think I can delay rusting long enough by priming the drilled holes and maybe also some silicone glue smeared on the expansion plug before inserting it.

I still like expansion plugs but I'm open to other suggestions.

I'm going to go out on a limb here and say depending on how big the whole is there shouldn't be a problem. The "mohican line" is there not only for a little bit of strength, but to hide the body welds. I don't think it will cause the car to collapse upon itself but you might not want to roll it. :thumbup:

Priming the drill holes sounds like a fine idea, and silicone should hold up reasonably well. I'd use housing grade caulk that's made to withstand UV rays.

I think the fold down tailgate idea is a brilliant idea ! ( and reversible )
Just remove your stock trunk and find tailgate from a 92-95 Civic Hatch.
Everything bolts right up and it even has a lock to hold the new Kammback in place.
http://farm5.static.flickr.com/4038/...ddc4266e22.jpg

Hmmm ...or does it ? I think i posted too fast on that.

I'll do some measurements and post back.

The nice thing about using a stock tailgate is that is has a rugged construction and looks professional if you were to paint it to match the car ( good luck finding a silver 92-95 Civic hatch at the salvage yard )

Nice idea but mine has a trunk not a hatch. I can't imagine the hatch lid would fit where the trunk lid is now.

It's also an EK. I don't know if the trunks swapped between generations.

You could always fab something up but it would probably look bad and work badly (no offense to your fabrication skills- thinking about this from a designer's perspective).

Have you thought about mounting trunk levers up by the top of the C pillars? Some trunk lids now lift up as they open to clear the compound curves. You'll see this on most recliner foot rests too. Basically it would be two levers of different sizes on each side. Instead of opening parallel as matching levers do, it would tilt to the front. The longer the levers, the higher up the thing lifts to clear the roof. You could add a small strut or spring to ease opening.

That way, the entire thing would lift up like a liftback and might actually close and seal right if you do it well. Or, you could hinge the Kammback separately. This would probably be easier to do. You could line the bottom end of the thing with rubber tubing to seal it and attach some simple elastic hooks on the back to secure it.

Panel attachment ideas, anyone?

Alternatives to the rubber expansion plugs, or suggestions on using them without ill effects?

Have you considered how large that your blind-spots will be in the new design ?
I threw together a quick Gimp paint-over of your original side shot, and it seems like with your line of sight in the car, you would only need a small section of the Kammback in clear lexan. The stuff is expensive, so this is good.
The rest could be formed in whatever material you decide to use and it could be pieced together in sections to form the compound curves.
The problem that arises is that you would still have massive blind spot along the side of the car. You could form the entire piece in Lexan, but it would be horribly expensive.

I'm planning a near-horizontal "spoiler", as drawn on the first page. Pretty much what we've called a Bonneville spoiler but going back only as far as the rear bumper. Seeing out via the mirror should not be an issue.

The side panels will intrude on the driver's view but I can redo them in Lexan later. Maybe cut an opening in the side panels where appropriate and put in a Lexan insert. Doing the center panel as a Kamm back going down from the roofline is not my plan (sorry).

Panel attachment ideas, anyone?

Rare Earth magnets?

Building it.

http://i188.photobucket.com/albums/z...nked_plate.jpg
Panel resting on supports

http://i188.photobucket.com/albums/z...e/100_0551.jpg
After trimming to match defrost wires

http://i188.photobucket.com/albums/z...e/100_0554.jpg
Measuring for length of side panel. I cut the panel 55" long.

http://i188.photobucket.com/albums/z...e/100_0555.jpg
Side panel taped into position. Will be trimmed approximately as in original drawing (below), the yellow outline.

http://i188.photobucket.com/albums/z...sideviewv2.jpg
Original drawing. Yellow = side panel, brown/tan = spoiler panel.

The photo with tape measure rule shows the approximate position of the spoiler plane. It will be close to horizontal, likely not sloped as steep as the drawing. Rear edge position determined by template. Front edge position determined by view out of the mirror. Panel is cut oversize on sides and at rear to allow for adjustment before finalizing.

http://i188.photobucket.com/albums/z...cropped_02.jpg
Rear view of side panel before trimming.

Nice. I just realized two stylistic things I like in your drawing above: the curves in leading edge and the bottom aft edge of the side panels. These mimic the style of the body itself. The rear curve imitating the bumper below it, the leading edge curve tracing the tumble-home pretty gracefully. Well made it could look way-cool.

Assembled it today, coroplast version
 

http://i188.photobucket.com/albums/z...72_smaller.jpg

Assembled this today.

Question to aerohead:
Do you think I should trim down the "fins" so they end at the height of the spoiler's rear edge?

Here's view from the other side. You can see the artwork printed on the inner surface of the side panel, I got that one at a reduced price!
http://i188.photobucket.com/albums/z...573smaller.jpg

Did some AB testing. Results were definitely not as good as I'd hoped. Also, seeing a cop hanging out on my test route discouraged me from doing the final "A" part. The car has no cruise control so I just had to do my best to start each coast at the same speed. I attempted to pass a particular speed limit sign at 40 mph, simultaneously putting the clutch in and going into a coast. I recorded the speed at start of coast at the same location each time, and recorded the end speed as I later crossed the same road each time, after about a half mile coast.

Northbound test runs, with spoiler:
START - - - END - - DELTA (MPH)
41.3 - - - - 33.8 - - 7.5
40.0 - - - - 32.4 - - 7.6

Northbound test runs, without spoiler:
START - - - END - - DELTA (MPH)
38.3 - - - - 32.9 - - 5.4
40.7 - - - - 35.5 - - 5.2
40.9 - - - - 34.8 - - 6.1

These tests don't show the results I'd hoped for. Speed loss was more pronounced WITH the spoiler, not without!

Next time I have an opportunity to test the spoiler I'll attach it with 3M's no-residue heavy duty duct tape and take it on the interstate. I found two places where I should be able to do a decent high speed coast.

I had the same problem with my sedan kamm experiment:
http://ecomodder.com/forum/showthrea...dan-14330.html

I think what is happening is that now your car is poking a larger hole (wake) behind it! The sedan makes us think that the wake is as big as the rear profile of the car. (rear roofline,c-pillars and all the way down perimeter)
However that must not be the case even on my boxy 94 protege!

If the spoiler/kamm thing would extend futher back to the point where the cutoff point will provide a smaller cross section than the original cross section of the trunk/tail area,AND it did that while flow is still attached,then it would help reducing drag but now I'm talking about a boattail.

I think the sides and top of the kamm does not let air to spill in behind the c-pillars/rear widshield and it does not continue long enough to get the small tail cross section to occur like a partial boattail would, therefore increasing the size of the wake.
All I'm saying is that the aerodynamics on sedans (especially your car) is probably a lot better than we think.
A Kamm works on cars with the station wagon style chopped off rear but apparently not on sedans. Sorry man,I kinda went through the same thing.

That's disappointing alright, but I don't think it tells you much, at this speed, with only an "A" and a "B" run. Still, maybe you are making eddies worse right now. Here is what the tail of our car looks like with tufts pasted into a light mist of water at abt 45mph:

http://ecomodder.com/forum/member-ca...sting-tail.jpg

I'm thinking you could taper the sides more to reduce the eddies and smooth-out the down-wash coming from the roof and c-pillars. I wonder what of the aerocivic's dimensions you could just imitate:

http://aerocivic.com/imgs/rear-z.jpg

I think you need more complete testing too. Get as many runs as possible, not just ABA at higher speeds.

I feel you on the officer, and on the frustrated disappointment. I experienced the same process, as you no doubt recall. You'll improve my failure quite a lot if you get a cleaner and better designed shape that I managed.

3dplane,
Thanks for posting that.

I'm not sure yet what went wrong with my spoiler. I suspect one or both of these are causing problems:

1) The side fins are causing turbulence, more than the intended benefit. They are supposed to help keep the air in the cavity above the spoiler panel. Very possibly their turbulence is defeating that gain, if they pressurize the cavity at all.

2) Even if I trim the fins down or eliminate them completely, it will still be a project with sharp angled edges. Supposedly not a good thing. The designs I've seen mostly start up front with a rounded rectangular cross section, and as the boat tail extends rearward the cross section becomes more like a circle or oval.

And thanks to California8Civic!!

I'm not giving up YET. As hinted earlier, I plan to test this version on the highway with tougher tape holding it on. And if the results still aren't good, I can trim away stuff to see if I can improve it.

First, trim the fins so they don't exist at the tail end. Just a straight line from the current top forward corners, going down to the back to meet the spoiler rear edge.

If "no joy" from that, then I can remove every part of the side panels that is forward of the spoiler's front edge, and every part that is above the spoiler itself. So the spoiler would look like this from the rear:

_______
/______\

What if you brought the upper, most aft, ends of the side fins inwards, so that they "pinch in" over your trunk lid. *shrug* :confused:

^^ What he said, why not bring the ends of the sides inwards to taper at the aft end.
It might help the air merge together farther back causing less of a wake from the sides.
I can't wait to pick up the 95 civic HB/CX I just went to see for the second time today!
I took some garbage bags with me to cover up the broken pass window and engine because it has no hood, waiting on lien sale papers 92k miles clean title! =D
I was thinking of doing something similar to my 90 corolla coupe with some plexi-glass and heat gun once I get a proper shape.

TangoCharlie,

You mean, to bend the fin tops down at the rear so then lay on the spoiler panel?

Or if you mean that the side panels can be angled, they are angled now. This photo shows the angle pretty well, but the panel hasn't been cut to shape yet. It's at about this same angle now:
http://i188.photobucket.com/albums/z...cropped_02.jpg

Angle them inwards more towards the end of trunk? and cut the center panel into more of this( \_/ ) shape so the ends of the side panels curve inwards and lie down more or do your panels have to stay straight/flat?

Yes, the side panels are closer together at the rear than they are up by the glass. I think that's what you're suggesting. Another way of saying it is, the spoiler panel is tapered - it's about 5" narrower at the back than it is up front where it meets the glass.

5" narrower doesn't sound like much, but that's with the spoiler angled slightly up at the front. If it were dead level, the measured taper would be more. Because I'm measuring the front width higher up, where the spoiler meets the side panels.

I got the rear narrowing rate, and also the side slope of the side panels, by matching and following the contour of the body. I'm just trusting that the engineers did a pretty good job. So far it looks like that's the case - my first attempt likely increased the drag.

Yea, so taper it more? That's what I was suggesting but what are your plans now? extension or different angle? The civic has a pretty nice rear end already and slope from the roof to rear glass to trunk.
I think you may have to build a kamma-back for the trunk which should help more because in my mind, that's where all the original turbulence is. When you made this style it shows you made the drag greater because the air leaving is trailing off a larger/higher edge, I just have these ideas since I have been looking at many different cars for good rear aero since it seems where most of the improvements go. Your car is already a kammaback up to the end of the trunk, I think kamma backs are left for the hatch backs or non fastback style cars. You could use Vortex Generators on the rear of the roof before the glass and tape up the gap between the trunk and glass to get the air to stick more.
Since I am getting a 95 Hatchback I would totally go Kamma back or tapering tail but for a coupe the trunk is the best spot to build it off of imo. you may see more improvements there but I would love to see a Civic Coupe Boat tail!

Yeah, that's what I was thinking. Only get progressively more aggressive toward the rear. Tuft testing would tell you if you go too far.
Just a suggestion.

Very interested to see further testing/results! :thumbup:

My bet, looking at the aerocivic is that this tail shape will be the winner even on our car, if any work well at all.

Following their lead might suggest not having the fins at all. Seems possible that your fins are creating more turbulence, given how some of the air seems to travel down the back (my tuft images above). Some of the flow is spilling over and around and twisting, out back of the car, off those fins I suspect. In my tuft image, someone pointed out to me that you can see evidence of airflow sliding off the trunk lid at the sides, airflow that is probably spinning off the backend of the corners as a result. A kamm, a spoiler/kamm like this one, needs to eliminate that. I think the fins accomplish the opposite goal.

The kamm/spoiler wins if it reduces the turbulence off the c-pillar, and the subsequent eddies off the rear corners. Basjoos has the shape. Question is whether your spoiler version can accomplish it too. If you are keeping the fins, I like the suggestion of trimming them so they don't exist anylonger when you get to the end of the spoiler/kamm.

Just my untrained, hobbyist, opinion.

Thanks, all.

Well - - -
It's not a Kamm back, at least not now. You could call it a spoiler, or a false, lifted trunk lid, or maybe a short Bonneville spoiler. The reason for the design is I want to be able to open the trunk and see out the back. If this basic design would give aero improvement, I have a scheme to make 2 cuts across the flat panel, that will allow the trunk lid to open. I'll spare you the details.

I'll try lowering the horizontal panel's trailing edge. I can try a flat panel over the whole assembly, making a Kamm that goes to where the horizontal panel currently ends - though I think that will kill the scheme to open the trunk. I would trim down the rear of the fins for that. But I will want to test that concept before I start cutting the fins away completely; the forward part of the fins are part of that Kamm design.

I can also reduce and then eliminate the fins. That would make it pretty much as if it were made with a taller trunk, with the trunk trailing edge at the template point.

Oh yes - I have another idea, completely different. A span of clear flexible plastic, the heavy duty stuff for temporary storm windows, stretched across the whole area from roof rear to the current spoiler rear edge. Maybe some bow strips going across to keep it from sagging in the wrong places. Detachable at the sides (velcro??) so you can lift the trunk. But as I said, that's a completely different concept.