Updraught cooling

[Old Koreelah]

Why are most aircraft engines cooled by forcing air down thru the fins? Warming air naturally wants to rise, so updraught cooling seems more sensible. I believe it's been used successfully in Rutan-style pusher configurations where the exhausting air can exit from the top side of the engine. My baby exits its cooling air along the midline, into the low-pressure area above the wing, so it lends itself to updraught cooling. It should work better on full-power climbs, when engines get hottest because conventional cooling is hampered by the angle of attack. 

 

Before I rip out the existing cowl ducts, does anyone see a problem?

 

 

[turboplanner]

Why are most aircraft engines cooled by forcing air down thru the fins? Warming air naturally wants to rise, so updraught cooling seems more sensible. I believe it's been used successfully in Rutan-style pusher configurations where the exhausting air can exit from the top side of the engine. My baby exits its cooling air along the midline, into the low-pressure area above the wing, so it lends itself to updraught cooling. It should work better on full-power climbs, when engines get hottest because conventional cooling is hampered by the angle of attack. 

Before I rip out the existing cowl ducts, does anyone see a problem?

I suspect you'll find that the heating from the fins in still air would be a tiny percentage of the controlled airflow through the ducting at 100 kts cruise. I played around with it on race engines, bringing in cool air low down and extracting it from the top but it didn't make any difference to engine life compared to cooling the combustion chamber with fuel.

 

 

[Downunder]

Any oil or fuel leak into the airstream may cover the windshield.

 

Perhaps flames, smoke and heat in the case of fire also....

 

 

[IBob]

First, it's not that warm air wants to rise. I'm sorry to sound picky about this, but generations of teachers have been telling generations of students that heat rises, as though it has some magical ability to move itself around.

 

What actually happens is that where you have warm air and cold air, the cold is heavier, and so will tend to sink, displacing the warm air upwards. The warm air (or hot air ballon, if you like) is actually floating on the denser cold air beneath it.

 

Second, the forces involved in this, while very large in weather systems, are actually very small under the the hood of your flying machine. So, as Turboplanner discovered with his cars, introducing up/down or any other direction of airflow makes no practical difference once the fan is going.

 

 

[IBob]

As for for improved cooling, you may already have identified a possibility when you mention that the present ducting exits into the low pressure area above the wing.

 

I believe the Savannah cowl is deliberately shaped under the belly to create a low pressure area. While I am not suggesting you copy that, perhaps the focus could be on artificially creating a better low pressure area to vent into?

 

 

[Old Koreelah]

Any oil or fuel leak into the airstream may cover the windshield.

Perhaps flames, smoke and heat in the case of fire also....

A valid point Downunder, but my cooling air doesn't vent in front of the screen; it goes out the sides into the low pressure area above the wing.

 

... the forces involved in this, while very large in weather systems, are actually very small under the the hood of your flying machine...

That's interesting, iBob. If indeed the natural updraught forces are small, that would remove one perceived advantage of the configuration. The only remaining advantage would be increased airflow because of higher angle of attack during climb.

 

...perhaps the focus could be on artificially creating a better low pressure area to vent into?

I doubt any sort of draggy exit underneath could improve on my current outlet; it works well.

 

What I would like to improve is how the air gets into the engine, especially on steep climbs.

 

 

[IBob]

I'm sure you know a whole lot more about how the air moves through your engine than I can, Old Koreelah.

 

My point was that how much air you get through a given duct is a function of the pressure difference between the front and the back.

 

Or put another way, the flow will increase if you increase the pressure at the front OR lower the pressure at the back.

 

But I've thought before that's a particularly nice looking Jodel, and I can well understand that you don't want to be adding any 'draggy exits'.

 

Maybe post some pics showing how the air goes through?

 

 

[facthunter]

You don't want oil drops on your windshield and the convection forces are of a low order. It's easier to pressure the "cleaner , no openings, pipes etc" top of the engine and the higher the intakes the less muck goes there. IF you are exiting to the low pressure area on the top of the wing you would be losing lift by doing that. Considering the prop blade shape for your cowl openings and inclining it down would help in the climb. Cowl gills or shutters would enable you to better cope with temp extremes and improve cooling drag/efficiency in varying conditions of operation . The heat from the engine should be able to be utilized for a bit of extra thrust if done well. Nev

 

 

[Downunder]

I believe "rule of thumb" is the exit should be 3 times bigger in area than the air entry area.

 

 

[IBob]

I don't have any experience in aircooling of aero engines. Reading comments here and elsewhere, I do understand that it can be very challenging.

 

I do have a long background in industrial blast freezers and high temperature timber kilns: environments where we are trying to freeze or heat a central item or product as rapidly and efficiently as possible by moving cold or hot air around.

 

The rising price of electricity, together with more exacting export requirements has resulted in huge improvements in design and performance.

 

But the principals involved have remained very simple:

 

1. Design the space so that the air is going through the product, not round it.

 

Boxed meat is stacked on stillages, timber is filleted, both have air gaps between each layer. However, it takes far less energy for the air to pass through gaps around the product, than through the gaps in the product. Tests we ran in a well stacked timber kiln, but with ragged stack ends, showed that 45% of the air was passing over, under, or around the load, rather than passing through it. That's 45% of our air doing nothing. (Older blast freezers would have been as bad or worse. And the fix was to design a tighter space that could still be loaded/unloaded by forklift without damage to the structure.)

 

Reducing this leakage has another important effect too: with leakage, the product adjacent to gaps is heated/cooled far quicker that the rest of the load. Closing the leakage off results in far more even heating/cooling.

 

2. Help the air round the corners.

 

In the case of a blast freezer, the air is making what amounts to four 90deg turns, from fan back to fan.

 

Unless these turns are swept, and with sufficient room before and after, the result is turbulence, which effectively blocks flow. In a poorly designed space, it is actually possible to fit a bigger fan and get less flow due to some major shift in this turbulence. (In blast freezers, the fix is to add vanes to guide the air round the corners, and to ensure sufficent gap at the front and back of the load where the air is turning.)

 

3. Fit enough temperature gear so you can see what is going on.

 

I have been involved in fitting or retrofitting temperature monitoring gear to numerous spaces. And without exception, every time we increase the monitoring, we learn something we did not know about how the space is behaving.

 

If I was trying to cool an aero engine, the least I would have (in addition to the standard instruments) would be one temperature probe that I could move around between flights, in order to build up a picture of what was actually going on under the hood.

 

 

[facthunter]

 All of that is relevant. You must direct the air to use it  efficiently or you waste energy. All but the most basic engines use baffles and ducts and blast tubes. This may increase the pressure drop but will reduce the  total volume of air used .  Air has viscosity that responds to temperature . Contact with surfaces by moving air is essential to transport the heat away, If it's stagnant not much is happening . Long fins in a  (relatively) cold place  on a motor don't do much. It has to travel through the material by conduction which requires a temperature gradient to work as heat flows only from hot to cold(er) and the bigger the distance the more temperature  difference.

 

  There's a practical limit to what Horsepower you can get from a piston motor of a given displacement and aircool it effectively. Unless it's by forced fan and very exotic, I would suggest it's south of 100 HP / litre.. Nev

 

 

[Old Koreelah]

...Maybe post some pics showing how the air goes through?

Good point iBob. I'm away from home at present, but from archives here are three pix:

 

1. the ruff-as-guts cowl duct that forces incoming air down thru the fins

 

2. an external sketch of airflow (cowl flaps closed)

 

3. the left cowl flap fully open.

 

As you can see, the cowl flaps are huge (larger than those on a 300hp Cessna) and cool effectively on climb. They also no doubt cause a lot of drag, but during cruise I can close them down to about 30mm. CHT's are fairly even, but I suspect they'd be more even if all the air came from underneath

 

I could install ducts under the engine to direct air up thru the fins, cooling the hottest parts first. 

 

 

 

 

 

[Old Koreelah]

...IF you are exiting to the low pressure area on the top of the wing you would be losing lift by doing that...

That's what I originally thought, Nev, but smarter men than me suggested the opposite. Accelerating air over the wing can only improve lift. As you say "The heat from the engine should be able to be utilized for a bit of extra thrust if done well."

 

My first Jab installation used exhaust augmentors to do just that, but the noise was atrocious, so I removed them.

 

 

[IBob]

There must be a fair body of knowledge by now as to which way is best to run the air past this sort of engine/exhaust layout?

 

I'm guessing from your pics that some of the air comes in either side of the spinner and into the cowl duct? And other air comes in lower down, does not go through the cowl duct, but blows through, cooling the exhausts???

 

 

[IBob]

And I'm guessing again: but if you blew air up from below, wouldn't you be carrying heat from the hottest parts up onto the engine? Or put the other way round, blowing downward as it does, isn't it carrying heat from the hottest parts directly away from the engine?

 

Question: what is the problem with the existing arrangement?

 

 

[Old Koreelah]

There must be a fair body of knowledge by now as to which way is best to run the air past this sort of engine/exhaust layout?...

Conventional wisdom is usually right, but sometimes can be improved upon.

 

...I'm guessing from your pics that some of the air comes in either side of the spinner and into the cowl duct? And other air comes in lower down, does not go through the cowl duct, but blows through, cooling the exhausts???

Yes, but the inlet below the spinner feeds air thru the oil cooler, which then passes the hot pipes and joins air that's come down thru the fins.

 

...if you blew air up from below, wouldn't you be carrying heat from the hottest parts up onto the engine?...

Good point, but the exhaust area always needs cooling well. Perhaps engine temps should be "evened out" more.

 

...Question: what is the problem with the existing arrangement?

 Not a great deal, but I've always liked the idea of updraught cooling. After a decade of modifying the little baby, it's working pretty well. I just want to make sure I've made all the improvement I can before I tidy it up and give it a decent paint job.

 

 

[turboplanner]

Question: what is the problem with the existing arrangement?

Good question, and interesting to see the answers.

 

Someone mentioned steep climbs; in GA aircraft there's normally a climb out speed after achieving 500' on full throttle and the climb angle to maintain that speed will not cause the engine to overheat. RA aircraft are much less developed and much more individual, so the solution could be as simple as adopting a lower climb angle.

 

Power demand makes a big difference to combustion chamber temperature (as against exhaust gas temperature);  have a look at the heat transfer ability of aluminium vs steel for cylinders and fins, and you will see the stunning ability of aluminium to shake off heat. Add to that a flow of cooler air through the fins and you'll increase that by a percentage depending on the stage of flight; quite a lot at cruise speed, less at descent speed, less again taxying, and less again stationary for pre-flight checks and holding.  The aircraft has to be designed to keep the operating temperature within limits in all of these situations. At the extremes you need to have good high speed airflow at cruise, and you need at least 20 minutes of in-envelope temperature grade on the ground in case you need to take off from a City airfield where in some cases it may take that long from start up to clear for takeoff.

 

The multi-million dollar aircraft development programmes will use probe temperature check for each little change, because sometimes you gain in some areas and make others worse, 

 

 

[IBob]

Right now...and considering the number of variables you are juggling in all of the scenarios Turboplanner has outlined...my every instinct (as an automation guy who likes to go home evenings and weekends) says if it works, don't fix it!

 

About the only thought that did occur to me (and this is an increasingly rare event) is that you have effectively a split in put, and a common outlet.

 

So, assuming the outlet is contributing to the flow by being low pressure, if you blanked off some of the flow to the lower engine (for instance), more would go through the upper. I'M NOT SUGGESTING THAT. Sorry, didn't mean to shout.

 

What I'm suggesting is there may be room to fool around with some minor baffling at the bottom, for some minor gains in cooling at the top.

 

Or maybe not.

 

Is it almost time to go home????

 

 

[IBob]

And do you inject the  Berocca before starting...or is that for hard climbs???

 

 

[Old Koreelah]

Thanks for the responses, fellas. All worth considering.

 

There is an awful lot of valuable experience on this forum and I’ve sure tapped into some of it. Never realised how much there is to learn from industrial heating and cooling, but iBob’s analogies are very relevant.

 

The Berocca containers are excellent for ducting; in this case getting cool air to the coils.

 

i have so many half-finished projects at home that this big change to my plane may have to await another day... might give my feeble brain time to assimliate all the advice.

 

 

[IBob]

Well, and thank you for raising the subject, Old Kareelah.

 

Between posts, I have been wrestling the cowling onto my first build, and I have enjoyed thinking about all this.

 

 

[Old Koreelah]

Speaking of cowls: a mate has been rebuilding a well-known LightWing and building a totally new engine cowl.

 

Given the high temps, particularly after shutdown, I advised him not to use epoxy resin. Vinyl ester is much more heat-resistant.

 

 

[IBob]

I know another member here who always opens his oil inspection hatch in the top of the cowling after shutdown: seems like a good idea.

 

 

[facthunter]

 The "Proven system" is baffles and ducting with  effective cowl seals all of which is designed to achieve what IBob outlined in post#10. This controls and directs air in close contact and OVER all parts needing cooling and reduces the amount of stray air that just waffles through the cowl area doing not much at all but requiring bigger inlets and outlets and hence more drag..  Most FLAT engines are much the same in this respect. Other configurations are more complex..On climb or a hot day you need more cooling air available than on descent or operating on a cold day. A "fixed " aperture/flow system can't adjust  for all this and taxi after flight. and cool the engine before you shut it down. Engine warm ups take too long on a frosty morning unless you can block off some of the unwanted cooling air. Engines run too cool on descents and pistons get too loose and barrell face the rings..  Many problems with aircooled engines get back to bad  design of cowling.. You can "cook" your engine in minutes if you run it with the cowl off IF it's designed to have one.   Nev

 

 

[Old Koreelah]

All true Nev. This is my third set of cowl ducts and they work okey. I thought the previous ones were good until I did some thorough testing and found tiny air leaks make a big difference.

 

My current cowl flaps have a huge range and keep the CHT's to acceptable levels, even idling on hot days.

 

it's amazing how fast the heads cool when you back off the power for descent.

 

I'm sometimes a bit too busy to always adjust them to different situations; a foolproof automatic system would be nice...

 

A cooling engine doesn't retain much heat to combat carby icing; I presume Jab's new allow barrels will cool down much faster than the old steel ones.