Cowling exit lip

[Steve W]

Some of Murphy’s products were being constructed with a high speed engine cowl which didn’t provide a decent inlet/outlet ratio at lower speeds when fitted with floats. Builders with a MC-3 cowl and a tight baffle had good results. I did with the one I built with low 300’s cylinder temps and solid 180 oil temps. The high speed cowl shouldn’t have been used on any of Murphy’s aircraft on floats. A lot of builders tried re inventing the wheel when it came to their cooling dilemmas and most didn’t solve the problem.

[Glenn Patterson]

Agreed the racing cowl is not good for a Bearhawk but there are principles that apply to all cooling. That is exactly what we did. Modified the MC-3A to work with our engine, went to the nth degree to get the baffles sealed to the engine to prevent air leakage to maintain pressure. The cowl at the cooling outlet is in the same line as the rest of the cowl. It may be that having the cowl extend over the tunnel helps with increasing the depth of the outlet. We also put the aluminum tear drop profile around the air exit on the firewall corners. The rounded corners make for more a laminar flow of the exit air. Sharp edges at the firewall exit creates turbulence that impedes air flow. The exhaust pipes are close to the outside to leave the centre wide open for air outlet. It does makes no sense to put a racing cowl on a Bearhawk. Use the best principles for cooling by retaining pressure over the engine, rounded firewall outlet etc... We have a clean cowl underside so the air from the prop may help with inducing exhaust air flow. We had to make the inlet flow less efficient to get good uniform temperatures across all cylinders. It may be that the Continental IO360 configuration allowed for the efficient cooling with the way the cowl was constructed or just out right good luck.

[schu]

I want unrestricted climb and I'm willing to give up some speed for it. Constantly monitoring cylinder head temps and flying the airplane according to what they say doesn't really excite me much.

I looked at the MC-3A and the price and fabwork don't phase me a bit, but it's really ugly. I was thinking about using a Stinson 108 nosebowl on my bearhawk. The inlets seem a bit bigger, and the chin is slightly longer which will give me more space to put in horizontal fuel injection. I even know where to order a CF variant.

Any thoughts on some other nosebowl that won't have CHT limitations? I see so many bearhawks with all kinds of cooling mods on them and I'd rather skip past that and put on a different nose bowl, have a generous outlet, tight baffling and be done with it.

[Nev]

I was reading about checking and setting the minimum fuel flow from the fuel servo. Apparently this makes quite a difference to the CHT on extended climbs. Effectively it means making sure that it’s running sufficiently rich to cool the cylinders, without being too rich such that the engine falters. Then it can be leaned according to preference.

[zkelley2]

Not great for a climb, but lean of peak ops have wildly cooler CHTs than rich.

[PaulSA]

"Not great for the climb ....." .Too true! - don't try LOP at high powers

If you don't need the performance, climb at a higher speed.

[Bcone1381]

Any thoughts on some other nosebowl that won't have CHT limitations? I see so many bearhawks with all kinds of cooling mods on them and I'd rather skip past that and put on a different nose bowl, have a generous outlet, tight baffling and be done with it.

All theory I've read (Nasa studies) says treat the Exit size as a cooling throttle. So that plus the tightness of the baffling primarily controls the CHT. Nosebowls not so much, but a poor design will not flow well.

I will echo and validate what Glenn and another Patrol Builder advised me... A bottom cowl that has some extra room at the exit tunnel for airflow out of the cowl had sufficient cooling with no lip. The Patrol was based in the south. I wish I had some dimensions. Lycoming wants a pressure differential of 5" of water across the cylinders. Measuring the pressure drop is pretty simple, tools are fairly inexpensive, and it seems that very few do it.

Glenn, what is the measurement of the opening between the top of the tunnel and the bottom cowl in your Four Place?

[schu]

From what I've been able to gather, speed affects inlet pressure, cowl inlet size affects available air (which there is plenty), outlet size affects the restriction unto the point of having negative pressure (negative restriction), the position of the inlets affect how much of the air comes from dynamic air pressure and how much of it comes from the prop. The current design has very little inlet in front of the prop, most of the inlet is behind the root where there isn't any prop twist. I think a cowl that moves the inlet to the outside edges will probably work better as it can take advantage of the prop induced pressure instead of just the dynamic pressure of whatever speed you are going.

Also, patrol data isn't very useful for us 4 place guys as we have another row of cylinders to cool, and we really aren't going much faster, especially with bushwheels.

[Battson]

The current design has very little inlet in front of the prop, most of the inlet is behind the root where there isn't any prop twist. I think a cowl that moves the inlet to the outside edges will probably work better as it can take advantage of the prop induced pressure instead of just the dynamic pressure of whatever speed you are going.

That is not resonating with me from an engineering perspective, if you are using carbon prop - they have a much better shape factor near the root. Metal prop - I would tend to agree that a different cowl could possibly improve things...

Still, I would assume the prop is creating lower air pressure in the cowl at low airspeeds - it's just not possible in a free stream to create very high pressure in the cowl without good airspeed.

I can climb unrestricted in our plane, but only if I am prepared to accept higher engine times, well within the design limits, but above my personal limits.
Cylinder head temps stabilise at about 210*C (410*F) on the hottest cylinder during a max power climb on a cold day (near freezing), on a hot day (30*C / 86*F) the engine gets much hotter in a full power climb e.g. 215*C (420*F) which is getting too close to the manufacturer redline for my personal comfort 240*C (460*F). This is cowl flaps open in all cases.

[svyolo]

A lot of the stuff I have read seems to be applicable to higher speed aircraft where ram air can be significant for delta P, just like Jonathon said above. I am not sure how well it applies to our application. The only non-stock thing I am going to do is have a exit ramp from the lower engine mount to the lip of the tunnel. A couple of RV guys claim it knocked 40 degrees off their CHT's. For me it also acts as a radiant heat shield for the EFI fuel pumps, which I mounted low on the north side of the firewall.

I bought enough Ti for the firewall and tunnel, but ended up using the SS tunnel so I could use the Ti for the ramp/heat shield.

[schu]

The RV guys is where I got most of this information: https://vansairforce.net/community/s...d.php?t=196881
https://vansairforce.net/community/s...d.php?t=196881 Post 6 says a lot.

Also, have pictures of a Bearhawk with Hartzell 3 blade trailblazer. With the stock nose bowl, very little of the actual blade crosses the inlet, it's all root. In my case, I have the standard Hartzell 2 blade metal prop. I plan on opening my cowling up a little bit.

As far as unrestricted climb, I desire less than 400F while climbing full power at Vy on a standard day. The cessna 180's I've been around have no trouble with this, I want the same from my BH and I'm willing to engineer to get there.

Given the number of people with a lip, huge cowl flaps, louvers, and everything else, I think it may be easier to just start with or modify the nose bowl to move the inlet 2-3" further to the top/side to collect some of that prop induced pressure.

[Bcone1381]

Dan H.... he's great!

[zkelley2]

If you look at certified airplanes with cowls that have been designed in the last 20 years, you'll notice they put a pretty big emphasis of laminar flow of the inlet air. Big square holes don't do that. I'd probably start with that in mind if you were making your own cowl, or looking for one to adapt.

[schu]

If you look at certified airplanes with cowls that have been designed in the last 20 years, you'll notice they put a pretty big emphasis of laminar flow of the inlet air. Big square holes don't do that. I'd probably start with that in mind if you were making your own cowl, or looking for one to adapt.

Can you shoot me some examples of cowls that have this?

[whee]

Mooney Acclaim, Cirrus, even the Carbon Cub to an extent. I can’t find the info I collected last year when I was thinking about a nosebowl change but the gist was to place the inlets as far part and high as possible with smooth transitions.

Done right and the engine runs significantly cooler and the reduced cooling drag can result in a nice speed increase.

[Glenn Patterson]

All theory I've read (Nasa studies) says treat the Exit size as a cooling throttle. So that plus the tightness of the baffling primarily controls the CHT. Nosebowls not so much, but a poor design will not flow well.

I will echo and validate what Glenn and another Patrol Builder advised me... A bottom cowl that has some extra room at the exit tunnel for airflow out of the cowl had sufficient cooling with no lip. The Patrol was based in the south. I wish I had some dimensions. Lycoming wants a pressure differential of 5" of water across the cylinders. Measuring the pressure drop is pretty simple, tools are fairly inexpensive, and it seems that very few do it.

Glenn, what is the measurement of the opening between the top of the tunnel and the bottom cowl in your Four Place?

-----------------------------------------

Hi Brooke,

Missed this as I have not been looking much lately. The tunnel was made to the drawings. The drawings and airplane are a half hour away so someone may be able to give the numbers quicker.

The cowl at the tunnel follows the natural flat curve of what the firewall would be. The back edge overlap of the cowl is the same for the full perimeter of the firewall. The exhaust pipes are close to the sides of the exit and are never below the cowl line. It could be that the cowl overlap on the fire wall may have an eductor effect in that the air passes clean over the exit opening. The air going over the cowl at the exit could cause a low pressure to draw cooling air out. We concentrated on making the metal baffles fit tight to the engine and then seal the aluminum baffles to the cowl and nose bowl with the rubber gap seals. The rubber is tight and we have to be careful to tuck it in when we put the cowl nose bowl halves on.

Whee hit it on the head with the Continental engines. The Cont IO-360 has all the air intake and fuel injection systems on top of the engine. These systems are not hung under the engine to frustrate cooling air flow. The open air channel down the middle underside is impressive from the nose bowl to the exit. The taper of the underside of the cowl expands front to back so as the cooling air volume increases going back to the firewall there may be no pressure restrictions to the air passing through all the cylinders so the underside pressure may be uniform. The underside of the cowl tapers quick and is relatively flat with a very gentle curve across the under cowl. There is no air intake on the lower cowl to alter the air passing so it is a clean air flow down the cowl. My partner wanted to emulate the underside profile of the Mountain Goat aircraft as he thought the shape did a lot for take off lift. ( http://avstop.com/news/goat.html ). This is pure conjecture. It may be that our cowl has the large relatively flat for the air to accelerate down the length of the cowl to create a back draft or vacuum effect as it passes over the boot cowl opening. The under cowl air flow is at a down angle projecting away from the airplane so it could leave a low pressure in the boot cowl opening. We softened the boot cowl exit with the tear drop profile wrap on the firewall corners to aid laminar flow.

I read all the information I could find on cooling as well as the Bingelis books. It is not a science based design but more an instinctive design of using principles that would work with our cowl set up, to seal the engine top side to perfection, promote as much upper engine pressure possible, maintain as clear an exit channel down the underside as possible, have the exhaust tips with a slight elbow but not hanging in the air stream, and modify/fab the MC3A nose bowl for cooling to facilitate the steep under cowl angle. Cross fingers, collect lucky horse shoes and run it as we built it then modify the cowl if need be to add cooling. My partner wanted to supplement cooling but I wanted to hold off and not solve an issue where there may be none. So somewhere between the Continental engine and amazing luck the cooling all worked out. Have to win one once in a while.

Looking at Nev's photo that started this thread. His cowl looks so good up to the point that the cowl was cut out so that the air flow out of the cowl is interrupted and becomes inefficient. The altered cowl removes the cowl's ability to create negative pressure in the boot cowl outlet. The lip further frustrates exit air . Nev's exhaust goes straight back to look very good to me as it leaves a lot of boot cowl open. It may be that if the lower cowl overlapped the boot cowl the same as the rest of the cowl that the engine may be close to efficient cooling. Could possibly even put a slight extension on the lower cowl over the boot cowl to artificially expand the opening area and create more eductor or vacuum effect without affecting overall aerodynamics. It is all dependent on the engine characteristics, how well Nev's upper engine is sealed and the attention to laminar flow and clean flow in the cowl including the firewall.

[Nev]

Looking at Nev's photo that started this thread. His cowl looks so good up to the point that the cowl was cut out so that the air flow out of the cowl is interrupted and becomes inefficient. The altered cowl removes the cowl's ability to create negative pressure in the boot cowl outlet. The lip further frustrates exit air .

Very interesting Glenn. I plan to put nutplates along the lower cowl lip so I can attach different size outlets as required. I’ve got the tunnel sitting in place currently so I’ll get some photos later today and post here, I’m very interested in your ideas,

[Nev]

Here’s the lower cowl exit as it currently stands (a work in progress). The outlet area is approximately 120 sq/in.

If I left it like this and flew with it, what are the effects likely to be ?
How does the exit normally create a negative pressure area ? I understand how it is created by a lip, but I can’t understand how it is created in what would normally be a high pressure area under the aircraft.

Would I be better to run the cowling back to past the firewall, with a “bulge” below the exhausts to increase the outlet area rather than the current cutout ?

Thanks.

39702A66-C77E-4409-81E4-302A8B5918EA.jpeg

[Glenn Patterson]

Hi Nev.

I am not deep into the aerodynamics but it would seem that the air passing down the lower cowl would be drawn upwards to perhaps hit the firewall by tripping over the edge of the cooling opening. Our lower cowl overlaps the firewall opening so the air going down the underside would create a back draft or negative pressure. The intent was to create induced air flow out of the boot cowl with a clean cowl shape. Cowl flaps make the cooling outlet opening variable and they create a bigger negative pressure to assist the cooling. Another form of induced flow. Once the cowl flaps are retracted then the cowl is streamlined with a clean air flow. With the nut plates you have options. I saw where a builder installed a shallow 60's or 70's muscle car like scoop to increase the cooling exit area. The rear edge of the scoop was inline with the rest of the cowl and looked very slippery. There are examples of reversed scoops for aircraft cowl cooling. The reversed scoop up would increase the negative pressure behind the firewall opening to draw out cooling air. Could form a scoop out of aluminum. Alternatively shape a piece of foam to make a streamlined reverse scoop then wax it and lay up fiberglass on it to form the final shape that you may like to try. There is also another aspect to our cooling in that the engine is a Continental IO-360 6 cylinder engine that is not generating the heat of a Lycoming 540 or a Continental 470.

[Battson]

Here’s the lower cowl exit as it currently stands (a work in progress). The outlet area is approximately 120 sq/in.

If I left it like this and flew with it, what are the effects likely to be ?
How does the exit normally create a negative pressure area ? I understand how it is created by a lip, but I can’t understand how it is created in what would normally be a high pressure area under the aircraft.

Would I be better to run the cowling back to past the firewall, with a “bulge” below the exhausts to increase the outlet area rather than the current cutout ?

Thanks.

The engine is still going to get some cooling. It's no different to flying with cowl flaps closed.
In fact, you would get about 50% more cooling than I have with cowl flaps shut.

That setup pictured should still provide adequate cooling to cruise lean of peak with a run-in engine.
You will have insufficient cooling to:
- operate rich of peak for extended periods
- break in a new engine
- climb at Vy for more than a couple of minutes
- cruise on very hot days

Of course it all depends how well your baffles are sealing, too. Any leakage, no matter how small, is unnecessary drag.

Without getting too deep into Bernoulli's principal and Newtonian laws.... Velocity of air outside the cowl is always higher than the velocity inside the cowl. The cowl outlet creates suction and pulls air through the cylinder cooling fins, combined with pressure above the cylinders forcing it through. The lip only enhances the effect.

Flying without a lip isn't going to result in air flowing up the firewall or anything like that.