I was thinking the same as Jeroen, some holes or slots at the very back of the aluminum...

With the back of the aluminum totally covering the exhaust as well, not only might air from the fan be prevented from flowing properly but the exhaust heat can not radiate away and may be somewhat trapped adding to the problem.

Hmmmmmmmmm

How about you reverse the fan? Instead of pulling air from above and pushing through the engine to below, you pull from below and exhaust out the top? Would that work? Do you need a 917 style vertical fan? If you geared the fan to spin it even faster or enlarged it could you not get the cooling AND get some suction effect thus to a degree turning the diffuser into a mini-venturi?

Final fitting and pictures:

http://forums.pelicanparts.com/uploa...1112290827.jpg

http://forums.pelicanparts.com/uploa...1112290844.jpg

http://forums.pelicanparts.com/uploa...1112290794.jpg

That is so cool looking! If there are any cooling issues.... you MUST resolve them in a manner that permits the diffuser to stay. Even if it results in no net gain in lap times or anything, its just too cool looking not to use it!

That WILL be functional. Nice job.
One experiment you might try is this:

Bring the engine up to 5k+ with someone's help.
Have your car on the ground
Feel under the engine to where the air is coming out from - get a general idea of what (engine fan) air flow is ocurring wihout/with the diffuser in place.
Be careful doing this -

I need something like that! Not for aero reasons, but to collect the oil drips!

Nice job!

Continuing with that theme, there is a useful test that you can do which involves putting some dots of oil along the leading edge of the defuser prior to going out for a session. When you come back the streaks that are generated will be able to tell you quite a bit about the air flow at the surface. The results might look something like the following examples that I've pulled from Simon McBeath's excellent "Competition Car Downforce".

http://forums.pelicanparts.com/uploa...1112294449.jpg

My concern with your design (in addition to the cooling issues described earlier) is that the angle of the defuser it too abrupt thus causing the defuser to stall. If this is the case you might be picking up more drag then you gain in downforce. The oil-drop test will help to clarify if in fact the defuser is stalling or not.

Jack - Maybe my original question was too general. Did you happen to make a template of the aluminum cuts? I'm very interested in your project, I'm just not the best at "fabing" however I think I can figure out the assembly.

Bigrubberjeep: Unfortunately, I made it up as I went along. But my particular configuration -- early car license panel/bumper, Flowmaster muffler, engine moved 1.5 inches forward, cut-and-rewelded B&B headers -- probably wouldn't match any other car.

If I can show meaningful gains (which I'm still a little doubtful of), then maybe we can put Widebody911 to work on it with his CF skills.

John, the oil drop test sounds like a good idea. If it seems to help (based on lap time, 'seat of the pants' and DL1 data) the first time out, I'll definitely do some more specific tests like that.

You could solve the potential flow separation problem by extending the diffuser into the rear bumper. Unfortunately, that would detract from the looks of the car, and you run the rist of adding a RiceFactor

I see an added benefit here - there is a lot of air off of the engine; this would join your smoothed air underneath and provide even more air volume for the diffuser to 'lift'

I'll never forget the wind storm of air behind my 911 as it ran on the dyno...

HUGE!

You could run holes in the upper rear portion for hot air exit with no loss of 'suction' -

Jack - I think you really are on to something here.

From pictures that I have seen, here is my 2 cents.

Maybe a streight line out like the ferrari works better vs. the curved.

Also, maybe a gap in the middle section of the diffuser where it meets the bumper, like an inch or two just to allow "hot" air from the motor to exit.

I'll try to draw my idea and post later.

Ideas....????http://forums.pelicanparts.com/uploa...1112304092.jpg http://forums.pelicanparts.com/uploa...1112304102.jpg http://forums.pelicanparts.com/uploa...1112304109.jpg

Jack - what top speeds are you going at on the track?

Also, a small camera can help with the oil drops - and use a colored oil or red ATF on a white or silver painted surface.

From my "bling" POV, I would have to say some of these diffusers are more for looks than effect.

Here's some interesting pics.

Sherwood

Depth, arch and proximity to the ground are the keys to making these work.
Jack's looks pretty damn good!

How about if you run your engine fan backwards and create a sucker car :)

Just kidding of course, but you have to figure Porsche tried it at some point.

"Depth, arch and proximity to the ground are the keys to making these work."

I'm also thinking a large volume of air would make this more effective for ground effects, but from where? Is there that much air sneaking in from under the perimeter?

Or maybe I'm thinking something else. Is the diffuser for ground effects or to allow under-vehicle air to escape more easily?

Sherwood

I'm really concerned about the engine cooling. The engine fans push about 1000 to 1500 liters per second at higher engine speeds. That's a LOT of air. Where is all of that air going to go with that current configuration? Even adding some slots isn't going to provide enough capacity. More importantly, when you're tracking the car at high rev's the engine is generating an awful lot of heat. If the air has no place to go, the fan will stall and your engine temperatures will go up really fast.

135 or so. My average speed on a lap is 99-100 mph.

Well, the openings are bigger than the factory engine cover. But are they better designed? Again, the monkey-making-a-laptop image comes to mind. ;) But before I do any testing with them, I'll have a CHT gauge and the oil temp gauge to watch for rising temperatures.

http://forums.pelicanparts.com/uploa...1112319696.jpg

If you were running just headers and straight pipes or megaphones, I would suggest having them ceramic coated inside and out as that will massively cut down on the heat the exhaust can radiate.

Then if the exhaust is not causing heat issues..... duct around it and reverse that fan!

Seriously.... how much air could the fan pull through running in reverse? I am sure the air channeling tin would need to be reversed at the cylinders and some means of ensuring the air when toward the cylinders would be required. But has anyone ever attempted to reverse the air flow over the cylinders?

How about a set of reversed NACA ducts at the end of the diffuser, to suck air out?

I think the one of the problems with a reversed fan is that it would then suck all kinds of crap up from the road and Jack's occasional Off Track Excursions into the engine which besides making the engine compartment very nasty would also probably be detrimental to cooling.

reversed fan... apart from mechanical difficulties of building something like that, sucking up hot exhaust air over the cilinders to the intake doesn't sound like a very smart plan to me...

There's much more as to why "pushing" air will be easier than "pulling" air....

Not just for all the reasons mentioned, but any fluid ( as "air" is) has a certain compressability which it can tolerate as it's forced to move...but lacks "strength" ( if you will) and "pulls apart" when you try to "suck" air from above....

This isn't really what happens...but it serves to create a mental picture of what then goes on. For those of you who might be designers of industrial cooling towers or finned, air cooled heat exchangers ( ACHX's), this is similar to "induced-draft" fan configuration over the tower or ACHX.... as opposed to "forced-draft". Notice the big difference in air flow HP required to get induced flow to work...

Porsche, I believe also tested and rejected this configuration...but I don't recall all the specifics of the reason for rejection...

Wil

Any reason NACA ducts couldn't be used to help cooling flow down and out thru the diffuser? Maybe use them in the existing slots?

good points made on the coating for the headers- would help reduce almost 100 degrees in a semi closed space. But absolutely not on the reversed air flow! despite the chapparal like ideas. The factory horizontal fan sucked up close to 35 hp on the 935's so your n/a 3.6 can't support that idea either (plus the cost close to 10k for a used unit!)

Based on that speed, I will go out on a limb and predict small improvements only....

You might think about boring some big holes in the pretty rear panel...

Jack, your holes should certainly help to resolve the cooling issue, but now where are your venturi??? All that you have left are the vertical fences. The good news is that I suspect that your engine cooling will be more effective given the diffuser behind the holes. Given the limited surface area of the diffuser, I doubt that you'll get much in the way of downforce -- but who knows without testing it.

My other thought is ... Where is the air flow going to come from to flow through the diffuser? Your car has a full width air-dam at the front which will do a good job pushing air around and over the car. If you don't channel air under the car (check out the front of a 956/962), you won't have any flow through the tunnel which will make the diffuser meaningless. This is why the F1 designers always spend so much effort getting the front of the car right, because it then determines effectiveness of the entire rest of the car's aerodynamics.

On a different thought -- have you considered using "turbine" style wheels (a la the 935-78 and 956's) to pull air out via the wheels? Given the way that you have the rest of the car sealed, it might just do the trick to develop a slight vacuum under the entire car's undersurface -- which would be a significant improvement.

What a great project, this thread really grabs my imagination.

All I'd like to add is a link to a rear engined aircraft/pusher. It has some interesting carbon fiber NACA intake scoops. I don't know if something like this could be all that useful under the car though.

http://www.rguerra.com/velocity/finalStructure/finalStructure.html


A view from the rear firewall looking towards the front of the craft.
http://www.rguerra.com/velocity/fina...20radiused.jpg



Link heaven:
http://www.eesc.usp.br/biblioteca/naca_nasa

That's the rub, isn't it? For this little project to be effective, I need to reduce air pressure (and velocity) at the end of the underbody chain. To cool a 911 motor, I need to pump air into that same area where I'm trying to make the low-pressure magic happen. I think you're probably right about this current setup not hurting my engine temps (of course, that's just a guess) -- but simultaneously not providing classic diffuser effects, either.

Starting out with zero expectations for the piece, I'll be pleased if it does anything at all. At best, it's going to be giving me a small percentage of what a diffuser does on a front-engine car. At worst, I'll be a monkey trying to find the Pelican BBS on my suitcase filled with old circuit boards.

(Actually, the worst-case scenario is big, sharp pieces of aluminum flying off of the car at high speed after they come into contact with the track. :eek: )

It also adds weight where you don't want it

I do hope it works out though

That's true about the weight, although it's all low down.

To see what we're talking about, I just took the postal scale outside.

The stock 964 engine cover weighs 11 lb 0 oz

My front splitter piece (with attached front skirts) weighs 4 lb 9 oz

My side skirts weigh (combined) 5 lb 1 oz

The flat panel between my rear swaybar and the diffuser weighs 2 lb 6 oz

The rear diffuser weighs 4 lb 8 oz

So, the full aero package (less the ABS panel between the nose and the rear swaybar) is 16 lb 2 oz

My rear diffuser alone is 6 lb 8 oz lighter than the stock 964 piece.

My full aluminum aero package (front, sides, bottom and rear) is 5 lb 2 oz more than the stock 964 piece alone -- which is less than the weight of one gallon of gas.

But it is still additional weight. (Not considering the 964 cover piece at all, I'm adding 16 lb 2 oz to the car.)

We'll see which pieces can justify their weight cost.

Someday someone is going to pull the cork on car aerodynamics and do every crazy/ugly thing they can to a car. Lord only knows what it would look like, bet it would take some getting used to.

Someone already posted this strake on a car:
http://www.mulsannescorner.com/newsmarch03.html


From a previous Pelican threads.......................

Riblets are used on aircraft wings:
http://aerodyn.org/Drag/riblets.html

Shark Fins....vortex generators used on the new Mitsu:
http://forums.pelicanparts.com/show...ight=shark+fins

Lex Fence on the F-18 Hornet:
http://www.nas.nasa.gov/Main/Redire.../Pubs/TechSums/

...............and some more plane stuff:
http://www.aerospaceweb.org/question/planes/q0176.shtml

Dimples on a golf ball:
http://www.aerospaceweb.org/questio...ics/q0215.shtml

Don't forget the fun & easy books on fluid flow I posted...