cooling fans

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
816
this whole thread and many others is like a backhanded advertisement for turbo props
where there is one path for air and heat,all internal,no Hx,no fins ,fluids,coefficients(except the one for wallet contraction)
all the systems with there gizmos just gone
 

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,498
Location
Saline Michigan
Yes, ME-109 radiator occupies the full depth of the wing.
Is there any way to transmit torsional loads through the external radiator fairings?
The usual way is skin doublers around the perimeter of the opening and rib from top skin to bottom skin around the opening. Basic rule is to put as much material into the doubler as was removed by the opening. Yes, it requires a bit of analysis. The diffuser and nozzle are removable, and usually not structural to the wing skin.

This approach is taken because the diffuser and sometimes the nozzle are removed to gain access to the HX's and its plumbing. Also, HX size, and geometry of diffuser and nozzle are all things likely to tuned in the process of flying. See posts on Russell Sherwood's system - he has different diffusers for travel and for racing...

Billski
 

Monty

Well-Known Member
Joined
Jul 15, 2010
Messages
1,367
Location
Fayetteville, AR / USA
Billski,

Like I said, no argument about the cooling system design. Even if you copy the original design, most of the war birds had problems in hot climates, and would overheat on the ground. But the engines still don't operate at 100% duty cycle relative to rated power. Even the war birds didn't. If they did for any appreciable time they had to be rebuilt. Most of these replicas can't use the full rated power of the engine in level flight without exceeding VNE. It's only useful in climb. Even normal air-cooled aircraft will overheat on a hot day at Vx climb. I'll also say that verifying power output via fuel flow is "squishy" when the nut that connects the pilot seat to the yolk has control of the mixture.

Draggy airframes with low output engines C65 etc may operate at 75% DC if flown at sea level, but most of these auto conversions will be closer to 50% rated power DC. Just like a race car isn't operated with the pedal down all the time. Neither is the aircraft! Climb, yes, banner towing, yes, otherwise no. People use this argument to say auto engines can't work in AC, and it's a bad argument.
 

BJC

Well-Known Member
Supporting Member
Joined
Oct 7, 2013
Messages
15,917
Location
97FL, Florida, USA
A friend flew P-47’s in WW-II. First mission was a ground attack in a valley, flying wing on the Squadron commander, whom he followed through the valley.

Commander’s airplane came through clean, but my friend’s was full of holes. Commander explained that, out of a dive, the airplane was faster than the AA could track. Turned out that my friend had been trained to attack at full power rather than War Emergency Power. The commander explained, “Son, any time that they are shooting at you it is an emergency.”

WEP was only available for a few (5?) minutes. Even without using WEP, military engines back then didn’t last long.


BJC
 

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,498
Location
Saline Michigan
Billski,

Like I said, no argument about the cooling system design. Even if you copy the original design, most of the war birds had problems in hot climates, and would overheat on the ground. But the engines still don't operate at 100% duty cycle relative to rated power. Even the war birds didn't. If they did for any appreciable time they had to be rebuilt. Most of these replicas can't use the full rated power of the engine in level flight without exceeding VNE. It's only useful in climb. Even normal air-cooled aircraft will overheat on a hot day at Vx climb. I'll also say that verifying power output via fuel flow is "squishy" when the nut that connects the pilot seat to the yolk has control of the mixture.

Draggy airframes with low output engines C65 etc may operate at 75% DC if flown at sea level, but most of these auto conversions will be closer to 50% rated power DC. Just like a race car isn't operated with the pedal down all the time. Neither is the aircraft! Climb, yes, banner towing, yes, otherwise no. People use this argument to say auto engines can't work in AC, and it's a bad argument.
Let’s go back to the OP, the airplanes he is talking about are LS warbird replicas, they run constant speed props, taking off from many locations they can start their flights from sea level.

Analyzing these flights and airplanes, they can be counted upon to launch some flights with both oil and coolant on their thermostats - 220F and 192F. The airplanes can start at 100% power, and can be analyzed to take somewhere around five to six minutes to 8000 feet, which if the cooling system has any shortfalls, is plenty of time to boil coolant.

Then we look at power available. If the pilot keeps both knobs forward and starts at Port Angeles WA, it starts at 100% and smoothly drops to about 75% at 8000‘, atmospheric density also drops to about 75% of sea level, which means that if the cooling system is barely good enough for carrying away heat from 100% power at sea level, it is also only barely capable of carrying away heat from 75% power at 75% of sea level density. Vy is only approximately constant, it actually drops slowly with altitude, and density drops, so the mass flow drops with altitude. So if the owner is using all of the normally aspirated engine they can, they need cooling suitable for 100%. And they need that capability at whatever climb speed is used.

Now if the pilot instead goes to say 75% power by whatever method (too much prop pitch, throttle reduction, combinations), a few seconds after lifting wheels, and has a cooling system barely up to cooling the engine at 75% power, the airplane will hold temperature from power reduction initially. If the pilot keeps the power at 75%, as the pilot climbs to 8000 feet, the heat rejection shortfall starts at around zero and increases smoothly… That too will boil coolant and cook the oil too.

Let’s say we want to live with a cooling system capable of 75% and not boil coolant. You can start your climb at 75% power, but we have to reduce power smoothly as we go up until we are 75%*75% = 56% as you go to 8000 feet. And this continuing power reduction continues as you go higher. This is the major actor. There is also the fact that the air does drop in temperature as you go up, but IIRC, the effect is substantially smaller than the density effect… You still start by pulling power and continue to back off power as you go higher or you start boiling coolant… Does the builder really want an airplane that has all this power but has to keep pulling power and artificially lowering the ceiling of the bird?

In all of these cases, once you get to altitude, you can put the power higher because airflow goes up proportionally with airspeed and thus heat rejection goes up. Temps will likely drop towards the thermostat set temps, and you can throttle the output.

One other thing to think about is that turbo-normalized engines need even more cooling than normally aspirated engines. The engine is breathing air at higher density than it is being cooled with, and so higher airflows are needed for turbo normalized than for normally aspirated at altitude. Don’t believe it? Read up on the Malibu and how it is flown to altitude… because cooling decreases with altitude while power stays high until reaching critical altitude.

So, the airplane CAN be lived with even with a cooling system not capable of firewalling it. But does the guy who is scheming out the bird want to do that? It takes a downtown analysis starting with power profile with altitude and time, then look at cooling as altitude changes. Best to ask the builder and have them figure out which way to run it.

Billski
 
Last edited:

addaon

Well-Known Member
Supporting Member
Joined
Feb 24, 2008
Messages
2,342
Location
Kanab, UT
Everything Billski said, as usual.

I drive a turbocharged TO-360, with pretty typical cooling for a certified plane. If I take off at my home field at 5000' altitude at 100% power (42" @ 2550), things get hot quick. I pull back to about 80% once established in the climb (36" @ 2550), and best climb is pretty close to 80 kts, where I can get about 700 fpm at gross. Climbing at 80 kts, though, just ain't an option in summer... climbing below 100 kts gets toasty fast. At 100 kts I get about 400 fpm, and throw about 20% extra fuel out the tailpipes for cooling, and I'm still thermally limited.

At top of climb, I go down to 75% power (34" @ 2400) or 65 power (32" @ 2400), and at 135 KIAS cooling is no issue, even leaned for efficiency and cowl flaps closed. But boy would I love some additional cooling in the climb...
 
Joined
Aug 15, 2013
Messages
3,106
Location
Western US
...Here is a 400+hp machine and you are saying more than 12,000 cfm is going through those 2 little holes . . .
Thermodynamically speaking, Andy Findlay is cheating a bit. There are a bunch of smaller holes through which he squirts a gallon or two of water per minute onto the outsides of the cylinders. And there are a couple holes in the intake through which he squirts a water/methanol anti-detonation mix. All that liquid enters the engine compartment at ambient temperature and leaves it as steam.
 

Bellaire MK

Active Member
Joined
Oct 11, 2021
Messages
44
Anyone think a ~4000 CFM puller fan and matching large radiator could cool a 6L LS engine ok? With no intake scoop I mean, just cockpit air with open canopy and exiting through louvers in the belly behind the wing.

I'm building a scale replica and don't want to spoil the outline with a ventral belly scoop. Original (ME109) had rads in the wing but with a big gear well cutout ahead of spar, if I go cutting another big hole for a scoop right behind that, wouldn't the torsional strength of the wing skin be severely compromised? I believe so and there is so much room for a larger rad in the back.

Thanks for any thoughts, yay or nay!
Locate the radiator below the surface of the wing! Slight angle to the rad. simple!
 

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,498
Location
Saline Michigan
Locate the radiator below the surface of the wing! Slight angle to the rad. simple!

the far and away lowest drag way to run air through a HX is a small opening, run an expanding diffuser to lower airspeed and raise pressure to get air through the HX, then a contracting nozzle on the way out. The easy scheme is to make the nozzle adjustable.

This can all be run at modest efficiencies with inlet and nozzle below the wing, the HX filling a width from top skin to below the bottom skin, a short inlet diffuser, and a short outlet nozzle. We have usable guidelines on how much radiator area and volume we need on per hp basis plus ratios of inlet area to radiator face area. These guidelines apply to radiator faces perpendicular to the airplane travel direction. I bet it would look about right for the replica. It is not as efficient as if it had longer inlet and outlet shapes, but if it cools OK and looks right, I suspect the small performance deficit would be accepted.

A previous post seems to imply an oblique radiator with wedge diffuser and nozzle. These have been used, but have proven to be large drag compared to one’s with the face perpendicular to airplane travel. I highly doubt that we need to go to this extreme for 300 hp and cooling equipment that looks correct for the replica. Look up Ross’ guidelines for rad area and volume, and see how it fits the scaled wing and scoops.

Billski
 

Malish

Well-Known Member
Joined
Oct 11, 2013
Messages
839
Location
Russia. City of Volgograd
Here the drawings of radiator to use with GM LS-6 V-8 engine in aircraft setup that Len Bechtold sent to me few years ago. It's also show air duct layout. Air duct outlet area should be 150% of inlet area. What is air duct inlet area should be compare to a radiator area?

Radiator (Large).jpg
 

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
816
Here the drawings of radiator to use with GM LS-6 V-8 engine in aircraft setup that Len Bechtold sent to me few years ago. It's also show air duct layout. Air duct outlet area should be 150% of inlet area. What is air duct inlet area should be compare to a radiator area?

View attachment 129099

is the sketch representitive of the actual install?
putting the end tanks, inlets and outlets,in the duct seems to be a bit bodged and clumsy
there are any number of rad makers that
can custom build
and are there any HX cores that are optemised for
install in a duct,ie: tapered air pasages through the core to begin pressure recovery right away?
 

Malish

Well-Known Member
Joined
Oct 11, 2013
Messages
839
Location
Russia. City of Volgograd
is the sketch representitive of the actual install?
putting the end tanks, inlets and outlets,in the duct seems to be a bit bodged and clumsy
there are any number of rad makers that
can custom build
and are there any HX cores that are optemised for
install in a duct,ie: tapered air pasages through the core to begin pressure recovery right away?

On the sketch the ends tanks of radiator are from point of view and they are not in the air duct. I think Len is know that the radiator tanks shouldn't be in the air duct as he is building replica of P-51 Mustang with LS V-8 engine...
What I forgot, at what proportion air duct inlet area should be compere to radiator area?

Новое изображение 1 (Large).jpg PC100401.jpg
 

rv7charlie

Well-Known Member
Supporting Member
Joined
Nov 17, 2014
Messages
3,376
Location
Pocahontas MS
What I forgot, at what proportion air duct inlet area should be compere to radiator area?
There's no single answer to that. Kuchemann & Weber discuss ratios as wide as 25% inlet to core area, but diffuser (duct) design gets extremely critical the wider the ratio. K&W has long been considered the go-to resource for cooling system design, but even it has stuff that could well be improved on today. It was published in 1953, right at the end of the piston engine era in government/military research. Chapter 12 discusses diffusers & exits; chapter 4 discusses inlets. Note that a whole chapter is devoted to the shape of the inlet lip on the diffuser. Some critical factors: climb speed, cruise speed, internal vs external diffusion, turbulent vs freestream flow into the cooling inlet...

Find a copy (a university that will loan books is likely your best source; you have to really want a copy to purchase one), and dig through those two chapters, to start. Get back to us in a couple of years, after you digest those chapters, and maybe explain them to me. ;-)

edit: Oh, and that 'approximate duct shape' in the earlier drawing, based on my testing of wedge diffusers, might well just use about 1/3 of the core for cooling. Probably depends on whether there's so much volume in the diffuser that it can act like a plenum instead of a diffuser.
 
Last edited:
Top