Mustang,
The NACA duct (scoop) is for air entry only, not for exhausting air.

Well I guess I learn something every day, I always thought the naca duct was exclusively for intakes

Air inlet on a flat panel is one application, not the ONLY application. Here's an implementation of NACA duct for engine exhaust on a full scale aircraft - pretty much exactly what we are talking about here.

http://home.hiwaay.net/~langford/nacaducts/

SO this ducting increases the speed of the ambient air to help with scavaging exhaust?

Ken Rand 2- never saw that application for a NACA duct -

This is getting interesting ....
Learning more about cooling and air flow.

Just talked with the Revolution designer ... they are having problems cooling this engine in the mustang.
Their research seems to show that the crankcase is getting too hot (not the cylinder head), and the engine is experiencing vapor lock.
Planning on providing a cooling fan to help the flow.

I will be trying some of the ideas expressed in this forum in the next week or two.

I have studied these ducts a long time ago. They depend on edge vortex flow to draw the flow into the duct.
It does not work the other way around. Just to be sure, I [link=http://naca.larc.nasa.gov/index.cgi?method=search&limit=25&offset=0&mode=sim ple&order=DESC&keywords=naca+submerged+inlet]did a search on NACA/NASA[/link]. This yielded 16 publications, none of which were about the duct as air outlet.
I then did a search on air exit in race cars. Here the duct form is used in the bottom sheet of the car to keep the flow adhered to the curved up section at the rear, and thus help the car to cling to the road. Here again, the vortices work exactly as in the submerged intake duct. The narrow part of the duct always must be upstream, and not downstream as the experimental plane builder showed in his photo session.
I very much doubt that he had any gain in drag reduction by this idea of inversed naca scoops.

Unconventional does not equal non-functional. Burt Rutan designed many of his famous aircrafts precisely to challenge the conventional thinking in aviation design. I remember him saying that most of what we call "engineering design" is actually " engineering analysis" - not design.
Nope. What is at work here is precisely what the NACA duct is designed to do: varying the fluid pressure and flow velocity by varying the chamber volume. As the equation (assuming isothermal) describes:

P1V1=P2V2

when volume is increased, pressure is correspondingly reduced. At the same time, flow velocity also decreases. The decrease in pressure is what holds the vehicle tighter to the road. The differential pressure between the top of the car and the bottom of the car literally sucks it to the ground - increasing downforce on the tires. Those big lateral extensions on F1 cars house large scale versions of these "suckers". Of course, there is no free lunch. Such devices add drag to the vehicle, as it takes forward energy to produce the tremendous down force. Such tradeoffs are advantageous in a racing vehicle. Some production vehicles also employ such "ground effect" devices, notably the likes of street legal Ferraris. A few others actually use ground effects to correct body shapes, which otherwise would produce unacceptable lift at high speed. The early production Audi TT is an infamous example. Ferdinand Piech, VWAG boss at the time, was adamant about not varying from the flowing line drawn by TT's designer, Freeman Thomas. The result was that Audi engineers ended up having to use ground effects to produce sufficient negative lift at autobahn speeds.

Some literature describes vortices generated by the NACA duct sidewalls as contributing to the suction effect, when in fact, it is far from being the principal action. The vortices actually adds to the overall drag produced by the duct.

Well, why not? The equation works both ways. The NACA duct in reverse can produce roughly the opposite effects. Decreasing volume will increase static pressure and increase the fluid flow velocity. It is this increase flow velocity that Mark Langford counted on to help cool the exhaust pipes. Apparently you also did not understand why it is that Mark says it would also reduce drag. Here's the scoop... literally. The cooling air enters the fixed-size intakes on the front of the cowl at the cruising airspeed. The cooling air expands somewhat inside the cowl and slows down. If this air is purged out the cowl through a substantially larger opening at the bottom of the cowl, the slower exhaust air mass will enter the faster moving air stream and cause turbulence (i.e. drag). One way to minimize this drag is to reintroduce the cooling air (now hot exhaust air) back into the airstream at closer to the same velocity. Such is the reason for employing the NACA duct in reverse, which conditions the exhaust airflow by increasing it velocity steadily before re-entering the slipstream. Is there a price to pay? But of course! There always is! The increased static pressure build up before the NACA duct can potentially impede airflow out of the cowl. No doubt Mark calibrated the exhaust ducts to the inlet opening to ensure more than sufficient cooling air flow. He even said he added the center duct for good safety margin and can block it off if later dtermined to be not required.

It is a sound use of the NACA duct - one that shows good understanding of the underlying principals at work.

On that KR2- It looks like a simple cone section would do just as well.
Consider this .
The air enters the cowling because of the low pressure at the exit, permits it to enter.
It isn't blown into the cowl.
In your explanation, there is positive pressure in the cowling .
No/Yes?

Quote "Decreasing volume will increase static pressure and increase the fluid flow velocity"...sorry but thats wrong, it will increase fluid speed, whilst decreasing pressure

Gentlemen - Man your textbooks!

How does cooling air flow thru the cowling ? Is is blown in - (increasing pressure in the cowl -then looking for avenues of escape)
Or
Does a low pressure outlet decrease pressure in the cowl - allowing the inlet air to readily flow .
PS-Leave out referrences to the air cooled VW's where a fan in a box is used .

Volfy,
Burt Rutan is unconventional, yet had to abide the laws of nature in his design. He has a very good understanding of these laws. He is a very able bodied designer, making good use of all possibilities offered to him.
Not everything unconvential works by law however, so that is not a discussion item. Only ignorants walk that road, until they know better.

Please check your fluid laws:
Boyle; P/V=C (used for static gas, non-adiabatic, so T=C) (V=volume)
Boyle-GuyLussac; P/(VxT)=C (if we consider the flow in the cowl low enough to be static)
Bernoulli; P x v=C (used for dynamic flow, considered adiabatic) (v=speed) You should have used Bernoulli and Boyle-GuyLussac.
Using these, you also will notice, that heating the gas either increases pressure without slowing down the gas, or makes the gas move faster because of their increased Volume and constant pressure, like in cooler ducts or jet engines.
The heating up of the exiting gas by the exhaust pipes will increase the pressure inside the cowl, because the pressure where the cowl flow merges with the outside flow is the same for both, and the larger pressure difference between start and end of the duct.
Thus the KR2S design effectively reduces cooling flow through the cowling, because there is less pressure difference left to sustain the coolant flow.

I do know my fluid laws, having taken Thermodynamics, Fluid Mechanics and Energy systems, amongst other courses in college. Boyle's Law is Boyle's Law: P1V1/T1=P2V2/T2. only difference is that varying assumptions will lead to different terms dropping out.

You're mixing two processes. First, the flow through the NACA duct is assumed isothermal. So the T terms disappears, thus the equation becomes P1V1=P2V2. It is exactly the same as your PV=C (I assume your C means constant), which BTW is not the conventional way it is expressed in most engineering texts. Anyhow, what this means is that as the air enters the NACA duct, it is expanded through the widening path. This volume expansion decreases the static pressure, which is principally responsible for NACA duct's ability to draw air without incurring much drag. Flow velocity also decreased as the volume expands - it has to because the same amount of air is now flowing through a larger cross section. Bernoulli does not apply in this instance.

Second, you have the air entering and exiting the cowl. Strictly speaking, this process is non-adiabatic, since there is heat gained from the hot engine. Still, the temp differential is not great enough (unlike that which occurs in combustion and steam engines) so as to produce a significant difference in air entry and exit velocities through the cowl. Otherwise, the cowl would act like a jet engine. We all know that doesn't happen.

Since compressibility is not relevant unless the velocity approaches sonic speed., if the cowl intlet and outlet sizes are exactly the same, the air velocities will have to equal. Again, the same amount of air is going in and out - conservation of mass dictates that if the outlet size is much larger than inlet, the exit velocity must be lower than the inlet. Therein lies Mark Langsford's reasoning for employing the NACA duct for accelerating the exit air velocity back up to same as the slipstream.

As many of you have correctly pointed out, there is ram effect on the cowl inlet. As long as this positive pressure sufficiently overcome any added back pressure inserted by the reversed NACA ducts, continuous cooling airflow through the cowl should not be a problem.

I cant be assed getting the books out, but ill agree sucking out is eminantly more effective than (trying to) blowing in

Ram Air--(higher pressure)
Is at the inlet --
not thru the inlet.
engine cooling requires that the air flow thru the fins -not over the heat bubble that is generated by the engine.
The question that needs complete addressing is not resolved by restating formula
How , can the model builder , best cause this air to cool the engine?
Does the ram air do the cooling?
No
It is simply the supply point and as such the inlet can be small IF the path thru and out is correct (back to the exit low pressure placement and shape)
Inlet drag is significant problem on full scale stuff- not on our models
Creating a full duct for best flow is simply beyond the ability or desire or understanding (NOT practical) for most hobbiests.
The best setup is to try to create a path which allows the low pressure to pull air thru the fins
You certainly can't blow it over and down around them .

Pe, Dick, Crusty, Volfy

As a long time R/C flier (45years plus) but a non technical graduate I'm enjoying the discussion!
Pragmatism lives!!!

Not really certain how to express the 'score' at this point------but I'm strongly leaning toward the point of view as expressed by Pe, Dick and Crusty.

Keep at it gentlemen------ it is informative AND fun to observe.

Hey,
It's not a contest. Even Volfy is back on track, so we all agree, don't we?
Core in the discussion is, that theory and pragmatic approach show matching results, with the prime question being: "do you know how to do it?".

Look at Nature. Sharks for example. Their trial and error design has stood up against eons of time. The coolant inlet is in a high pressure zone. The outlet is through the gills in a low pressure region. The series arrangement of gills show, how effective this contraption is when moving at a certain speed. With the shark asleep and zero speed, he relies on jaw pumping action to keep that flow going.

Now look at the Suchoi 26 aerobatic plane design. Incorporated here is all the knowledge gathered in WWII radial fighter planes, and the testing for best performance that went with it. The aircooled radials were well baffled, and when stationary on a start lineup or a ship's deck purely relied on prow wash aerodynamics to keep the engine in a proper operating range. It is no good to start your mission with the engine overheated! Their gills (cowl exhaust baffles) could be regulated to adjust the flow, and hence the engine operating temperature.
Like Dick stated, the prop hardly contributes, because the centre sections were cuffed, and hardly were able to add to the cooling of the engine by providing ram air pressure. Yet there was sufficient pressure difference to sustain a good flow of air through the engine.

With our 3D model airplanes growing in size, and the lack of internal fuel cooling of gas engines, we find ourselves in nearly the same situation as the fighter planes waiting for that take-off signal. There seem to be no shortcuts. Baffling gains importance as sizes increase, because engine swept volume (cube) doubles with available cooling fin area (square) when engine size doubles. Lacking forward speed, we need the low pressure of the air outlet, as well as the generous dimensioning of that outlet to keep gas flow up to par, and resistance to a minimal value.

Emperical values (derived prom field observations) are a ratio of at least 1 for inlet to 1.5 for outlet area. Preferrably 2 for the outlet, where the inlet is in a high pressure region, and the outlet is in a low pressure region.

Don't know what you're trying to get at, but pressure and flow are interactive. There must be a pressure differential between the inlet and outlet, or else there will be no flow.

Creating an oversized outlet in a model engine cowl does help minimize backpressure, which is why you often hear the "twice the inlet size" recommendation. If one does not possess more in depth understanding of fluid mechanics, such rule of thumb is a fairly safe bet to go by. However, as we should all know by now, there is always a drawback. An overly large outlet will result in more drag (as I explained previously). Drag is a major concern for full-scale aviation. For most RC applications, however, drag is not a big deal. Heck, when you make the outlet as large as possible - that is, the size of the cowl - the airplane model will still fly just fine. How big to make the cutout depends on how technical you choose to be and what you consider esthetically acceptable - the latter of which is a non-technical decision.

Let me make it absolutely clear that I am in no way advocating doing to our RC models what Mark Langsford has done. He has his full-scale design concerns, we have ours. The two sets are not always the same. The engineering principles at work, nevertheless, apply universally.

I bought up Mark's example only to counter the incorrect assertion that NACA duct only applies to fluid inlet on flat panels.

NOW, For model considerations, the size of the cowl oulet is but one consideration. Equally important are the location and configuration of it. I have seen people hog out the bottom of a cowl, much of which faced forward. In general, it is far better to have the cowl outlet face rearward, such as with a oversized cowl that extends below the fuselage or a "tunnel" that recesses into the fuselage inside of the cowl.

Good save there, Pe. Though it is debatable whose hide you were trying to save. But I won't press the issue.

I do agree that this is not a popularity contest. As long as folks here benefit from the discussion and hopefully gains a higher level of appreciation for the issue at hand - and most importantly - becomes better at telling BS apart from sound advice, it is all for good and matters not who is perceived to "win".

As I stated in the very first post in this thread, I don't particularly care to get as fussy about this issue with my own airplane models. One thing a good engineers learns with experience is decidng what constitutes "good enough". That, in essense, separates us engineers from the pure scientists. We may spend 2months performing a FLUENT CFD two-phase flow erosion study on our oil field pressure chokes (and we have done just that), because it could mean the difference between using plain stainless steel, or stellite coating, or having to go with carbide inserts. The huge difference in the overall performance and associated expense justifies the engineering excercise.

OTOH, spending 1/2hour dremeling a hole that "looks right" to me on my 35% Giles 202 cowl satisfies my requirements plenty.

Sorry not convinced, the more I think about it that reverse naca outlet is very wrong
any supporting data?

Well -what I am getting at is the point of the discussion.
Maybe I don't know all there is to know about fluid dynamics -however it is enough knowledge to invent workable ,proven money making patents involving fluid flow and heat control.
Almost all the Leisure suits made by Levi Strauss in the 60's-were thermal set on the machines I designed and patented.
So much for my credentials.
Those and a buck will maybe get you a cup of coffee today
The discussion is about how does the cooling on model engine really work-
Not analogies- nor who's ---- is the biggest .
My comments on cooling -in models- is from only one viewpoint - actual, measured cooling on many different engine/cowl setups.
And I made some pretty good screwups along the way .
Still do -
Cooling the engines -is like moving a chain down a road - you can't push it along - put it will pull very nicely.

Throwing in a little frivolity ....

It's like the saying ... their is no such thing as gravity ... the earth sucks.

How does trees get the water to the top ... same as a straw works ... they suck.

The wind moves because of a low pressure region ... so the wind sucks ... it doesn't blow.

Because a horn has a bigger outlet than inlet ... a horn doesn't blow ... it sucks ! (A Stretch)

I think we're on to something !

Hey 'stang, tell me.
How can a tree grow taller than 10 meters? (analog to Dick's chain gang?)
I know this question sucks [&:] ROFLMAO

Does this have anything to do with fibervascular bundles or capillary action?
I once saw those terms in a learnin book --

preivers, D.Hanson

Scientist use big words to explain stuff ... I just get confused. [:-]

HOW DOES a tree get stuff up past 10 meters .... thats a lot of sucking !!! [X(]

I'm glad someone is watching

Going now ... to build a low pressure louver and ducting for my Mustang Cowl ...