freakingfast

My Feedback: (3)

Would you guys find the following statement true?

"In aircraft even slow flying models, when you cool something ram-air/dynamically, generally you want the inlet smaller than the outlet or the airflow stalls and goes around. Usually the outlet should be a minimum 1 1/2 times the area of the inlet. I have seen this problem time and time again where engines or high discharge batteries overheat. It's a common misconception that a big scoop can force air fast out of a little exit."

speedracerntrixie

My Feedback: (29)

I would agree with it in general. It is practically impossible to push air where you want it to go, you must pull the air. I don't think there is an actual formula of intake VS exit area, just about every install is going to be slightly different. Best case scenario you create a partial vacuum around the exit that pulls the air in through some duct work that directs the air through the fins of the engine or across batteries.

iron eagel

True.

Andy Lennon in his book on model aircraft design stated that in order for the cooling system to work properly, the outlet needs to be sized at least 140% of the inlet. Another wrinkle you failed to mention is that as the air is heated it expands, yet another reason for the larger outlet vs inlet. The real trick is to size the intake so you have just enough airflow to cool the heat producing components to minimize drag. It's important that you direct the cooling air over the components and keep the airflow as close as possible to the components as well. You have to look at the entire cooling setup as a system, the higher power the system is the more critical cooling becomes. Many of the big 3D type of airplanes use fans on the power regulator to insure cooling even many high power helicopters are now using cooling fans for the power systems as well. The last two airplanes I have built have had very close attention paid to the cooling, to a point it influenced the final shape of the aircraft. Granted part of the reason I had to pay close attention to this detail is because I am actually pushing the motors themselves far beyond the recommended power levels to start with. My past experience in the electronics industry taught me that cooling of electronic power handling devices can make the difference between a good reliable system or one that goes up in smoke. I have one larger high power electric plane, that in order to insure the motor, batteries and electronics stay intact, uses a EDF setup to insure adequate airflow through the cooling system even at low flight speeds.

BMatthews

This was most certainly a true case in control line speed models that use an airfoil shaped cowl around the cylinder head to both reduce drag as well as better guide the air thru the cooling fins.

da Rock

The ratio of outlet to inlet is a complex one. Locating the outlet in a negative pressure area has a great effect. Notice how many fighters in WWII had cowl flaps. Amazing what negative pressure can do. Cowl flaps provided very strong negative pressure. Locating the outlet on a "negative slope" can do a lot as well.

The amount of intake can be reduced also if you do what designers have done for years and years. Force the incoming air to pass close to the things you want to cool and close off any routes that allow the incoming air to bypass the heated objects.

Hot air is harder to move than cool air. Incoming air will take the path of least resistance. The hot air then stagnates. Great big open cowls actually hurt more than help. Even placing cowl outlets directly behind the engine can fail to cool enough if there is a way for much incoming air to get around the warmer air sitting around the things that need cooling.

shonny

It's not only true, but vital, That is why you should baffle the engine if the air outlet is as little as only 1.5 times the area of the inlet.
If you look at full-sized engines you will find that they are baffled to very close to the fins. Air further away just 'clogs up' the outlet.

doxilia

My Feedback: (3)

SRT,

actually there is. The physics goes back to Mr. Boyle and Mr. Stokes although most of that work applies to incompressible fluid dynamics (i.e., liquids). More recently though, Mr. Ernst Mach (among orhers) extended this to compressible flows (gas dynamics) which is the reason we have turbines today.

The dynamic pressure, q, of an ideal gas (air is pretty "ideal") can be expressed as:

q = 1/2*gamma*p_s*M^2

gamma and M are dimensionless constants that relate to thermodynamic variables (heat capacity & the Mach number) and p_s is the static pressure.

A simple relation between entry & exit areas can be established for a desired q value. Conversely, given inlet and outlet dynamic pressures, an area can be computed from the q values.

More can be found on your friendly online encyclopedia:

http://en.m.wikipedia.org/wiki/Dynamic_pressure

David

Spastic

negative pressure! My physics professor would scream!

in 3D planes it is common practice to shoot for 1:2 "in to out" ratio, not as much speed most of the time so you need all the help you can get.
spaz

da Rock

That actually confuses the issue. The inlet to outlet ratio isn't the reason.

Block "excess" cool from entering that would be able to get to the outlet without cooling the hot spots.

Cool air that can go around the engine will do so and fill the outlet, making it even harder for the heated air to exit. Look at full-sized cowlings and you'll see baffling to direct air as mentioned, close to the parts to be cooled AND there are baffles that block off "excess" cool air. They're designed so that the heated air getting to the outlets has no competition from cool air.

Cool air is routed to the things that need cooling and no air is allowed to enter that would compete for outlet area.

da Rock

My physics professors were known to scream too.

However, in aerodynamics quite a few people talk about positive and negative pressures. Of course they're attempting to make one complicated discussion easier to understand by using "positive" and "negative" to suggest direction.

iron eagel

A couple of questions.
What would be the values (using air as the gas) for heat capacity which I would assume is the gamma value. And as far as the velocity or Mach # that would be a function of the airspeed do you use the max value for that? Given that the air is being heated wouldn't that also change things a bit. I also wonder as far as the velocity in that as it passes the devices that it is to cool drag is going to reduce the velocity as well. The other thing I wonder about as well is what you would use as a value for the static pressure, in my case I have the inlet right behind the prop. There is an area of high pressure that is developed behind the prop that is higher than the surrounding ambient air so there is already a negative pressure at the outlet in regard to the inlet. Now throw into the mix that it is my understanding air starts to become compressible in our applications when the aircraft exceeds speeds on the order of 220 mph not much so beforehand, this has got to change the dynamics of the entire thing pretty radically, because once you (hopefully in my case) get beyond the 220 Mph airspeed the entire dynamics are going to start to shift fairly quickly. Bottom line I don't see where it is that simple of a process to get a good handle on.

speedracerntrixie

My Feedback: (29)

Iron, this is one of the reasons that I said for our applications there is no real formula that will always give a predictable result. There are just too many variables and the speed of our models is just too low. The best thing that we can do is simply move the air past the cylinders. As long as the airplane has forward speed and some prop wash moving past the outlets this is going to happen. Provided there is some sort of duct work in the cowl. In all the years I had been flying IMAC I would simply seperate the cowl into top and bottom halves, the air would come in through the top half and be pulled to the bottom half. There would be a splitter plate so to speak separating the top and bottom that would direct the airflow through the fins. It has always worked well and I can honestly say that I have never had a heat related failure. I know it's fun to throw in full scale theory and math into the disussion, but it usually dosen't work out with our models.

doxilia

My Feedback: (3)

I guess a simple example would be the way to demonstrate how the dynamic pressure of a moving gas is affected by the frontal area through which it passes.

However, there is a simpler way as compressible fluid dynamics is an extension of incompressible fluid dynamics. The point I wanted to illustrate is that this sort of physics is well understood although technically still "rocket science" ;-)

Think of your common garden hose... Turn the water on. Note that if the velocity increases so does the q value. More importantly, decrease the frontal area of the flow (stick your finger in the outlet reducing the aperture) and the q value also increases. This is analogous to air passing into a cowl (the "tap") and exiting the back after cooling your ESC & LiPo's (for example). The same principle is also used in tuned pipes (so called Helmholtz resonators). The numbers are just different as are some of the important factors.

The main difference between the hose example and our "air duct flow" is that gasses are compressible by definition (note, not at a given velocity - its not like a switch) and so the fluid dynamics are affected by thermodynamics to a much greater extent than with liquids. Ever try to compress water? Very, very hard and the temperature doesn't rise much. Take air, shove it in a scuba tank and the sucker will keep compressing while the tank is quickly ready to cook eggs... That's why they're placed in water. It'll take on all the heat and then some (that's all related actually).

Its a little hard to elaborate on your questions typing on my phone but the answers are yes and no. Gamma and M are dimensionless variables which means they express ratios of physical quantities with the same units. As a simple example, the Mach number, M, is defined as:

M = v/a

Where,

v = speed of gas.
a = speed of sound in medium.

When v = a, M = 1 and the gas is moving at Mach 1 - the speed of sound.

Similar considerations apply for gamma and its related variables.

Check out the link I provided. It'll give you a better idea of what's involved and should answer some of the questions.

David

PS one of the most beautiful physics equations I have seen which probably took the better part of 20 years to understand is the Navier-Stokes equation. Mathematically it can be generalized into something even more elegant known as Stokes Theorem. It can be applied to the universe just as easily as it can be applied to your wife's breast... I admit, both come to mind when I think of it (well, my wife's goods... )

doxilia

My Feedback: (3)

The section on internal flows in the link below describes the "area question" being raised by the OP. It presents the simple relation I was alluding to in my earlier post. Here's the discussion generally related to compressible flows:

http://en.m.wikipedia.org/wiki/Compressible_flow

David

iron eagel

Some interesting information. Being nothing more than an hobbyist with only a basic background in physics all I can do is try something and look at the results from data I obtain. All I know for sure is that the outlet needs to be at least 140% of the inlet for it to cool electrics so far.
My interest in this question comes from my trying to keep a electric motor from cooking (at power levels well beyond the motor spec) while at the same time provide cooling for the ESC and battery.
Now the setup I am using to cool the system works like this...
I have a round cowl inlet opening of 7/8ths of an inch in diameter. This area is roughly about 1/4 of an inch deep and terminates at the motormount/firewall which has openings that direct airflow through the motor as well as around it, now this area in total is about 1/2 inch in diameter. Going through the motor it is getting slowed down by the coils and magnets while it is absorbing heat and (I would imagine expanding to some degree as well in the process of being heated). The air flow which goes around the case of the motor is kept fairly tight to both the motor and heatsink I would imagine being slowed in the process. The entire motor assembly is enclosed within a duct until it get to the compartment which contains the ESC and Battery which has both a diffuser and baffles to keep the airflow directed around these two components. Now this area also houses the rx and servos which has got to be causing some turbulence and slowing it yet more in the process. Now it gets a bit more complicated this area is vented into a expansion chamber through a 3/4 diameter nozzle. At the end of the expansion chamber there is yet another 3/4 inlet 1.125" outlet nozzle, which is the tail of the aircraft (I figured this would be a lowest pressure area). I can't for the life of me figure out how much airflow I am actually getting through this system, but thus far I have keep the case temp of the motor and ESC below 110 degrees f and figure it has to be working. Just to put this in some perspective while not super high power levels are involved we are talking a 1Kw power system so there is a fair amount of heat generated by it. While cooling is the main goal of this system I am trying to use some of the waste heat energy to generate some thrust (no mater how minimal) and have no idea of how to tune it to do what I want, other than adjusting the intake/outlet proportions but I know there has to be a lot of other dynamics involved.

sensei

I personally like creating a 2 to1 exhaust to inlet ratio on the stuff I build and fly. On 3D airplanes I try to setup for a 3 to 1 ratio because in many cases there is not very much forward speed to work with if any. Cowl flaps do help a great deal by accelerating incoming air on the OML (outer mold line) of the cowl when deployed creating a low pressure area on the IML (inner mold line) around the cowl flap area scavenging the super heated air from inside the cowling. This action happen from prop blast and/or forward speed of the vehicle.

Bob

shonny

Here is a paper on baffling of air-cooled engines which should be of interest to aeromodellers. And especially so now as we see steadily more petrol (gas) engines that will need more exacting cooling than our methanol-driven ones with lots of oil that helps cooling and lubrication. http://www.experimentalaircraft.info...stallation.php
Or have a look at very plainly explained version by Aussie: http://www.youtube.com/watch?v=bfSv0V7r4HA
Oh, and Aussies prefer in-out ratio minimum 1:3. Pretty warm down there!

BobFE

I'm suprised that no one has mentioned Bernoulli's Principle and its effect on RAM air inlets. If the inlet is divergent (opening) then the velocity of the RAM air will decrease and pressure will increase. IF the inlet is convergent (narrowing) then velocity will increase and pressure will decrease. This is how engineers prevent supersonic aircraft engines from loosing efficiency when the go supersonic. They slow the airflow to subsonic before it hits the first stages of the compressor. I was a P-3 Flight Engineer for 17 years and became an expert on that aircraft. We had 2 RAM air inlets for the AirCon and 4 inlets for the oil coolers. For the Air Conditioning system we ducted the RAM air to the heat exchangers and out the side of the plane. The inlet was much smaller than the exhaust. For the oil coolers we could control the size of the exhaust opening by moving a flap anywhere from full closed to full open. On the ground we had the flaps full open. In flight most of the time they were almost fully closed regardless of the temp. I'm not sure how this applies to RC, but it should give some fuel for thought.

shonny

We're still talking about model airplanes? Like these things we normally have in the air for 15 to 20 minutes?

iron eagel

Yes but as the line in the movie "The Flight Of the Phoenix" states, "The principles are the same."

speedracerntrixie

My Feedback: (29)

to a certain degree yes but in most cases when we start getting this complicated when trying to resolve an issue with a model airplane it turns into information that has little use.

sensei

I like the KISS principal myself, been flying gas engines exclusively for nearly 15 years without burning them up, so I see no reason to get over technical now days about cooling.


Bob

iron eagel

I agree with you for the most part, but the one thing that he mentioned about this full scale airframe that applies here is that the inlet size was smaller than the outlet, and that basically reinforces what the OP was saying.

iron eagel

"I like the KISS principal myself, been flying gas engines exclusively for nearly 15 years without burning them up, so I see no reason to get over technical now days about cooling."

Where the OP mentioned that this was occurring with high discharge batteries, I got the sense he was referring to electric powered aircraft. With electrics cooling is a very critical consideration. The solid state speed controller and motor themselves produce a fair amount of waste heat, and solid state electronics don't fair to well when overheated, and LI-pol batteries can actually blow up if they are allowed to get to hot. Over the years I have seen both gas and glow powered airplanes that have had issues when little or no consideration was given to cooling as well.

sensei

Read the thread starter's first post, He is making mention to both battery and engine overheat issues.The KISS principal has nothing to do with giving little or no consideration to an issue, in fact it means to consider everything and approach it with this thought in mind, (keep it simple stupid). Simple is generally the most direct, and lightest way to approach issues on our models.

Bob