Liquid cooling

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Royal

Well-Known Member
Joined
Jun 8, 2020
Messages
68
I have had an idea for a while. While radiators have been around for quite some time and work they do create a lot of drag. I've seen people making certain contraptions to block air after takeoff to restrict the amount of air and reduce drag. The idea i have is making a copper radiator that doesn't have any fins. It dissipates heat SOOO much better.

The whole science of cooling with a radiator is not to shove as much air as possible but to let enough air through at the correct speed so the air can absorb as much heat as possible before becoming heat soaked. If it goes over to fast it will only pick a little and then you over heat.

Copper is expensive compared to aluminum and that mostly why they use that. Not sure if there would be weight savings. The copper one could be smaller but its a bit more dense. I'd have to run some tests.

What I'm proposing is have a NACA for inlet and an exit hole with a lip that can closed and the lip pushed flush when in flight reducing drag even father.

The air could be slowed down and not have to use heat sinks on the copper since it give up the heat so easily. Pretty much have a tunnel lined with the tubes and the air passes by. Whats everyone think?
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,787
Location
Saline Michigan
While radiators have been around for quite some time and work they do create a lot of drag.
Well designed cooling systems do get close to zero total drag. The keys are clean air inlet, an expanding duct to reduce air velocity before the radiator, a contracting duct to get the air back up to speed, and size the contracting duct to use the energy added from radiator waste heat. For low speed and high power operations you open the exhaust door, and as speed goes up you close down the opening more and more. THe Meridith effect describes all of this, and we have had whole threads on the topic. Just use the advanced search tool on the word Meredith.

I've seen people making certain contraptions to block air after takeoff to restrict the amount of air and reduce drag.
They have been around a long time. WWII liquid cooled airplanes made big use of inlet and/or outlet vanes.

The idea i have is making a copper radiator that doesn't have any fins. It dissipates heat SOOO much better.

The whole science of cooling with a radiator is not to shove as much air as possible but to let enough air through at the correct speed so the air can absorb as much heat as possible before becoming heat soaked. If it goes over to fast it will only pick a little and then you over heat.

Copper is expensive compared to aluminum and that mostly why they use that. Not sure if there would be weight savings. The copper one could be smaller but its a bit more dense. I'd have to run some tests.
Copper does have higher thermal conductivity than aluminum, but there is a lot more to it than that 0.020" thick walled tube in the system. In cars, aluminum is favored over copper for cost and weight. Nonetheless amazing performance radiators with aluminum cores exist.

Area exposed to the airstream is a huge part of how much radiator you need to move a unit of heat to the air, and finned radiators are a huge part of making more efficient and light radiators we have available today.

What I'm proposing is have a NACA for inlet and an exit hole with a lip that can closed and the lip pushed flush when in flight reducing drag even father.

The air could be slowed down and not have to use heat sinks on the copper since it give up the heat so easily. Pretty much have a tunnel lined with the tubes and the air passes by. Whats everyone think?
NACA scoops were all the rage about 50 years ago, but for continuous use inlets they are not the best. They are still great for an inlet that might be used some of the time and will be shut off at other times or where only modest pressures are needed - when shut off, it is as if they were not even there. For continuous use with higher flow resistence, like driving air through a radiator, they just do not move much air.

The variable round outlet is a complex thing to build. We do use them on jet engine exhaust nozzles. Most folks find that simple exit ramp works pretty darned well.

There are numerous threads on this subject. I suggest that you use the Advanced Search tool and search on "radiator" and for "rv6ejguy" in the author. Lots of great posts on how he did it, including his flow versus pressure drop testing, radiator effectiveness, and rules of thumb on radiator sizing.

The whole idea of making a tunnel and lining it with tubes seems pretty low yield. If the center of the tunnel does not have tubes in it, the air will largely stick to tubes along the wall and flow down the center. What flow you do achieve along the tubes will get warmer and warmer as the air moves along the tube, making for less and less heat transfer. The use of large relatively few layers of tubes is based upon all of the tubes seeing nearly the same high delta T between tube and air, raising the heat flux while having relatively thin radiator to keep pressure drop through the radiator within the bounds of the inlet air energy.

The metrics of interest: How much heat do you need to reject? How much drag does it cause? How much much does it weigh? Everything else is academic.

While this scheme could work, my engineering assessment is that for it to move as much heat to the airflow, it will both need more air and more weight, both of these things being negatives for good performance in airplanes and other vehicles.

Billski
 

Royal

Well-Known Member
Joined
Jun 8, 2020
Messages
68
I did do some research. I got to excited about what I found. Then I did more research on copper vs aluminum radiators. Aluminum beats copper because its stronger. It can be made thinner therefore transferring heat better Sooo um yeah. Lighter and thinner wins it.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
13,054
Location
Memphis, TN
Copper also work hardens. Old planes you use to anneal copper lines every so often to keep them from cracking.
 

trimtab

Well-Known Member
Joined
Apr 30, 2014
Messages
85
Location
rocky mountains, rocky, usa
I suggest Incropera Dewitt. If you are designing a beat exchanger even on a napkin, you'll find resources that will help get your thinking closer to reality. Although it's a rigorous treatment of heat exchangers, it has a lot of info to keep in mind when trying to put something together that is 1) effective for the real requirements of the mission and 2) fabricable.

Heat transfer hasn't changed for this type of application in 80 years. Choose an older edition for economy.
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,345
Copper also work hardens. Old planes you use to anneal copper lines every so often to keep them from cracking.
Copper is also more than three times the weight of aluminum and has less tensile strength.

There are good reasons why copper isn't widely seen in airplanes other than in wiring. Weight and weakness don't make for good airplanes.

I had a copper oil pressure line break in flight and dump most of the oil overboard before I landed. The airplane was built in 1946. They've been using much better materials for a long time now. If I bought an old airplane with a copper oil pressure line I'd replace it with a modern hose before I flew it.
 

Sockmonkey

Well-Known Member
Joined
Apr 24, 2014
Messages
1,880
Location
Flint, Mi, USA
The whole idea of making a tunnel and lining it with tubes seems pretty low yield. If the center of the tunnel does not have tubes in it, the air will largely stick to tubes along the wall and flow down the center. What flow you do achieve along the tubes will get warmer and warmer as the air moves along the tube, making for less and less heat transfer. The use of large relatively few layers of tubes is based upon all of the tubes seeing nearly the same high delta T between tube and air, raising the heat flux while having relatively thin radiator to keep pressure drop through the radiator within the bounds of the inlet air energy.
The way to make that work, (as much as it could work) would be to use counterflow. Have the coolant run forwards through the tube against the airflow so the hottest part of the tube is in contact with the hottest part of the airflow so you always get a net heat transfer of heat to the air. Fish use the same setup in their gills to maximize the maximum amount of O2-CO2 transfer.

The only build I can think of where you might want to do it like that is one where you want it to look like a jet because that system would have the external shape of a jet engine.
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
13,252
Location
Port Townsend WA
The whole science of cooling with a radiator is not to shove as much air as possible but to let enough air through at the correct speed so the air can absorb as much heat as possible before becoming heat soaked. If it goes over to fast it will only pick a little and then you over heat.
Huh?
Need faster airflow for increased heat transfer until air friction gets significant at supersonic speeds.
 

Royal

Well-Known Member
Joined
Jun 8, 2020
Messages
68
Huh?
Need faster airflow for increased heat transfer until air friction gets significant at supersonic speeds.
I think it actually causes a boundary layer of air on the fins keeping the air from exchanging heat.
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,076
Location
Rocky Mountains
I suggest Incropera << >>

Thanks for the link!
It looks like Incropera is to thermodynamics as Shigley, Bruhn, and Hartog are to their respective fields?

Free PDF of an older version can be found on the internet. At over 1000 pages I'd really prefere hard copy but the cost...........:(
 

BJC

Well-Known Member
HBA Supporter
Joined
Oct 7, 2013
Messages
11,251
Location
97FL, Florida, USA
Not necessarily.

Engine cooling via liquid coolant typically is determined by the engine designer. The heat rejection to the atmosphere is then determined by the airframe designer. Consideration there needs to be given to the typical lack of a radiator cooling fan during extended ground operations.

In the typical air-cooled engine in an HBA, cooling is determined by the airframe / FWF designer who typically is trying both to minimize drag and to adequately cool the engine. Mature designs, properly implemented, operate without “hot spots”.


BJC
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
3,943
Location
Warren, VT USA
Is liquid cooling a significant advantage to reducing hotspots?
Short answer Yes. Proper EFI and proper thermostatically controlled heat exchange in an aluminum engine should be more conservative and more idiot proof than air cooled at the cost of weight. Modern engines are really idiot proof through starting, initial warm up and most operating modes. This is why you don't find ANY air cooled car engines anymore. Even lightweight motorcycle engines are oil cooled with external heat exchangers and dry sump. Even super light weight motocross bikes now are water cooled and EFI.

That being said, aircraft air cooled engines are highly evolved and at the cost of complicated operating procedures and not being idiot proof and with the benefit of weight and somewhat less complexity mechanically they have a long history of success. I personally loath carburetors. But most of the certified stock in the world is still using manual mixture controls and pilot monitored critical engine heat management also manual. Seems crazy to me.

Really liking the progress on the Yamaha front. The more hours they build up the more we all win. To me a proper modern aircraft engine would be dry sump, all aluminum construction, head block, accessories, forged internals, excess cooling capacity (like german car engines designed for continuous WOT), dry sump, and with some sort of bullet proof PSRU. The PSRU should be able to be mounted upright sideways, below without issues for different installations with different thrust lines.

Just reading BJC post while I finished writing this: True bad design is always going to yield bad results but still water cooling in an aluminum engine is A LOT easier to get right and way more conservative in the end. Why does a hot spot exist? Cause either the mixture went bad (EFI) or the cooling system has a problem either by design or due to maintenance or installation issues. A mouse nest in an air cooled engine can be an expensive disaster.
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,345
That being said, aircraft air cooled engines are highly evolved and at the cost of complicated operating procedures and not being idiot proof and with the benefit of weight and somewhat less complexity mechanically they have a long history of success. I personally loath carburetors. But most of the certified stock in the world is still using manual mixture controls and pilot monitored critical engine heat management also manual. Seems crazy to me.
I know that things are becoming more automated and that idiot-proofing is part of that. Works well as long as the computers continue to operate and as long as they're less idiotic than the human operator. Automation has made a huge difference in automobiles, but that automation does something: it dumbs down the driver. He/she no longer feels the road and trusts the ABS and other stuff to keep him out of trouble; works fine until he encounters a road surface or a corner that is beyond the computer's ability to save him, and there's an accident. If the driver has to move a vehicle with a manual transmission and a clutch he's in trouble. If he wants to buy an antique with a manual choke and manual steering and brakes and shifting and no ABS or stability augmentation, he has to learn to drive all over again.

But there are very few old cars on the road. There are a lot of old airplanes around due to the cost of new machines, and they need certain pilot skills to operate. A computerized airplane will start itself and adjust everything as it needs to and the airplane will find its way to its destination and maybe even land itself and taxi to the fuel pumps or tiedowns. Great, but the pilot that learns to "fly" in that stuff is stuck with it. He has to buy one, for a million dollars and up. If he wants to buy an old Champ so he can afford to fly, he now has to learn to fly. Taildragger, automatic nothing, whiskey compass, adverse yaw and more. Learning something the second time is ten times harder than learning it first. The Learning Factor of Primacy is at work here: the stuff learned first is the most ingrained and is almost unshakeable. I could see a whole different level of license for automated airplanes. Nobody is going to come up with a retrofit automating kit for a Champ.

Automation is coming. Can't stop it. But it has its downsides. The numbers of modern vehicles in the ditch beside an icy road is testament to that. It's way worse than it was 25 or 30 years ago.
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
3,943
Location
Warren, VT USA
All modern airliners are getting more and more automated every year with every new model. GA is going the same way. Experimental is using old parts so not plus it's cost conscious by its nature. Cars are design in no small parts by liability and product life cycle officers as well as teams of lawyers that understand how much non-idiot proofed products cost them. I have been in product development my entire adult life, decades. Seen it. Ran the meeting where the lawyer won the day. You can thank them for wiping out the manual transmission in the USA. Idiotproofing is good and bad :)
 
Top