rjsa wrote:Water thermal condutivity is around 20 times bigger than air, so not bothering with the other factors involved, like speeds and Reynold's numbers inside the water piping and outside the wing, you will need something in the order of 20 times air contact area compared to water's. That's why radiators have a very thin grid to create greater air contact surface.
call me dumb or watever I did not get the thoughts leading to your statement.
the water still has to disperse the heat into free air ....so in effect the area needed is still the same as you need to get rid of the haet of the engine still..
The Bt46 was liquid cooled ,of course ,so for sure the surface area was way too small for the efficiency those surface cooling modules could provide.To me this is
only proof to Gordon Murray completely missing the boat with his calculations presumably based on wrong efficiency numbers.
A aircooled setup is for sure feasible. As far as I know nobody yet got very sophisticated in cooling ribs and duct design to match the designs in water cooler cores with louvered folded and what have you special bits to improve heat transfer assuming someone is willing to start pumping recources into such a project it would surely work ,at least in f1
I should have been clearer. I was referring to the idea of water cooling thru the wings. For sure it will dissipate heat, but I don't think it is worth it.
That's a good sketch and it brings up a very relevant issue: Is there enough room to get adequate cooling air mass flow to the cylinder liner area?
I would propose that there probably is. As I noted in my previous post, my "aircooled" engine is in fact air, oil and fuel cooled.
There is not much cooling required around the cylinder liner area of most N/A engines. The bore area exposed to combustion gas is very short, there is lots of intake flow overlap that provides internal heat transfer, and an oil spray could provide the balance of cooling required. The only limit to how hot the liner surface can operate is related to the flash temperature of the engine oil. A good engine oil won't flash below about 400degF, so the liner surface can safely run close to that temperature. Thus, not a whole lot of cooling fin area would be required around the liner.
As for the intake side of the cylinder head, not much cooling is required in this area either. The high fuel flow rates would give excellent heat rejection thru latent heat effects. If you look closely at a liquid cooled F1 engine you'll note that there is very little coolant jacketing around the intake side of the head.
As for the issue of adequate cooling of an aircooled engine at low speeds, it would be no more of an issue than you have with a radiator, since the basic heat exchange mechanism is similar.
Finally, with regards to risa's point about the large disparity in specific heat values of water vs. air, he is technically correct. But I think he fails to see the big picture. As someone noted, ultimately the heat load must be transferred from the liquid coolant circuit to the passing air. How effectively this occurs is a function of the deltaT between flows, surface area, relative mass flows, and the thermal conductivity across the interface material. The deltaT in a liquid cooled system is limited by F1 rules to about 230(coolant)-70(air)=160degF. An aircooled engine would have no such limit, and its deltaT would be 400(cooling fin)-70(air)=330degF. This much higher deltaT would mean that the final heat exchange mass airflow and/or total exchange surface area would be much less with the aircooled engine.
As I noted, my "aircooled" proposal was premised on the fact that you could manufacture a high-temp cylinder head structure made from modern wrought aluminum alloys, with very thin and closely pitched cooling fins.
I am enjoying the discussion.
Terry
"Q: How do you make a small fortune in racing?
A: Start with a large one!"
I can see three real big problems with adding liquid cooling to the rear wing (aside from the ones mentioned):
1. Fragility. A single bump to the rear wing could cause a leak that would immediately lead to retirement. Furthermore, sealing up all the pipe work that leads to the rear wing cooling fin would be incredibly challenging, and prone to failure. It would also be impossible to quickly change a damaged rear wing.
2. Running pressurized coolant through the airfoil's "empty space" would create tremendous force that would deform the rear wing. Radiators can handle the high pressure because they're made of narrow tubes smashed flat and covered in fins. The internal surface area of any section of pipe is relatively small, hence the force generated by the coolant pressure is fairly small. The "empty space" inside of an airfoil would have much more surface area per unit of length, and would have to be strong enough to handle the force that 3.75 bar creates when acting over such a large surface area. Granted you could run a bunch of tubes inside of the wing, but that would increase the complexity of the structure.
3. Long manufacturing lead times due to increased complexity. F1 teams update the wings almost at every race. The wings are already very complex and difficult to fabricate. Adding plumbing would only increase that complexity, increasing lead times and costs, which would result in fewer aero updates, reducing overall aerodynamic efficiency.
fragility concerns have not stopped all F1 teams from using Carbonfibre wishbones wings etc..
Pressure deforming the tubes? -more tubes willl increase surface area and cooling efficiency ,so ...
Then again putting the heat transfer solely on the underside of the the wing ,that should help downforce ,as it changes air density?
My understanging is that there is a big difficulty with multi cylinder air cooled engines, it is cooling between the cylinders. With water a small volume of water is effective to move the heat away (provided it is not stagnant). With air it becomes difficult to get the heat transferred & moved away. Air cooling fins like to be deep, thin & tapered - just what you are missing between the cylinders & heads.
I think the air cooling is just as good. As riffraff said the water will eventually have to exchange with the air. So if it is looked at like a resistance circuit. The metal to water to metal to air, is not really better than metal to metal to air.
aluminum conductivity = 250 W/mK
water = 0.58 W/mK
air = .024 W/mK
The advantage with water cooling is not solely the conductivity but the actual volume of the water which acts a remote reservoir to draw all the heat. It's virtually like increasing engine block volume to absorb the heat.
With saying that, this "increased" engine block volume will still have to exchange with another reservoir, the outside air, Just like any other air cooled engine, (only that the block is of lower conductivity since it's a metal and water combination)
ringo wrote:I think the air cooling is just as good. As riffraff said the water will eventually have to exchange with the air. So if it is looked at like a resistance circuit. The metal to water to metal to air, is not really better than metal to metal to air.
aluminum conductivity = 250 W/mK
water = 0.58 W/mK
air = .024 W/mK
The advantage with water cooling is not solely the conductivity but the actual volume of the water which acts a remote reservoir to draw all the heat. It's virtually like increasing engine block volume to absorb the heat.
With saying that, this "increased" engine block volume will still have to exchange with another reservoir, the outside air, Just like any other air cooled engine, (only that the block is of lower conductivity since it's a metal and water combination)
this was exactly my point .In my understanding the water system just acts as a heat sink ,but you should have the drawback of two additional heattransfers:first from the block to the water ,then back from the water to the radiator and from there to the air.....sounds not like the best ideas I heard of really...
ringo wrote:I think the air cooling is just as good. As riffraff said the water will eventually have to exchange with the air. So if it is looked at like a resistance circuit. The metal to water to metal to air, is not really better than metal to metal to air.
aluminum conductivity = 250 W/mK
water = 0.58 W/mK
air = .024 W/mK
The advantage with water cooling is not solely the conductivity but the actual volume of the water which acts a remote reservoir to draw all the heat. It's virtually like increasing engine block volume to absorb the heat.
With saying that, this "increased" engine block volume will still have to exchange with another reservoir, the outside air, Just like any other air cooled engine, (only that the block is of lower conductivity since it's a metal and water combination)
Remember you have to do all sorts of calculations with things like Nusselt number and that sort of craziness I don't remember to get the final heat transfer coefficient of the air.
I think the Temperature difference between the engine and the air must not be too large because our goal is not necessarily to draw as much power away from the engine as possible but to keep the engine at a certain temperature, right. But then again if there is a system that uses higher temperature difference we can always minimize the flow/amount of the fluid to minimize bring the heat transfer in the proper range.
The original benefit of water over air cooling was due to the specific heat of water versus air. Water could conduct heat energy away from the critical areas of the cylinder head (ie. between the exhaust valve seats and around the spark plug bore) more effectively than air cooling. But this was only necessary with a cast aluminum head that had a temperature limit of about 250degF before structural failure. Additionally, the cooling fin thickness, configuration, and pitch spacing is limited with a cast aluminum aircooled cylinder head.
My proposal is for a diffusion bonded, wrought aluminum alloy head structure. Wrought aluminum alloys can easily operate safely at temperatures up to 400degF, and have tensile strengths almost double that of cast aluminum alloys.
I'm still waiting for someone to punch a hole in my theory. Tell me why with similar cooling air mass flows, a liquid-to-air heat exchanger at a deltaT of about 160degF across the core would be more aero efficient than an aircooled finned head with a deltaT of 330degF across the head.
Anyone?
"Q: How do you make a small fortune in racing?
A: Start with a large one!"
I think ringo has a most valid point in that the important property is the medium's thermal conductivity, which is 20 times higher for water than for air, than the actual thermal capacity.
Equally obvious, is it that water is only a conveyor for heat to be finally dissipated to the passing air through the radiator, but I'm afraid that the decisive parameters here are Aluminium-thickness and -area.
If you look at the unit for thermal conductivity, W/m*K, it is really based on Power*Thickness/Area*Tempdifferential,
why you want to minimize Thickness and maximize Area for the Aluminium barrier, which is so much easier to do with a
finned radiator that a solid cylinder head.
"I spent most of my money on wine and women...I wasted the rest"
Here is a cnc machined engine.
Made in small batches in Australia. http://www.jabiru.net.au/
The kit includes the air cooling ducting to make it simple to install on a different airframe.
Made in Australia, eh? That figures...no in all honesty, gawd how I love machined Aluminium, what a beautiful piece!
A little strange configuration though, by the cylinder-staggering I suspect each con-rod is on it's own crank,
why this is more like a "spread-out" Straight-6, no?
Possibly to help cooling?
"I spent most of my money on wine and women...I wasted the rest"
I really like the aircooled idea .....riffraff has brought up a major point for a development area in my view
Looking back to 917/30 @1100hp with fan cooling ,I´m pretty sure this is very very feasible and poosibly more reliable then a watercooled layout.
I think a bit of extra oilcooling passages and a bit increase area for the oilcoolers could do the trick.
Porsche went to watercooled heads first ,if I remeber correctly when they introduced 4 valves per cylinder ,but i firmly believe thes issues can be resolved without water cooling.
I think what you want from the F1 engine is compactness, minimum weight is limited, which would be very difficult to achieve with the surface-area needed from the cooling-fins.
My previous post above was an attempt to xplain why liquid cooling gives you an advantage in that respect.
"I spent most of my money on wine and women...I wasted the rest"
