Cowl Mods

[Cyril Gibb]

I'm starting a new topic prompted by an append by Yetti on another thread.

His question about a simpler cowl closure has been on my mind also, but I hesitated to put thoughts on the forum lest I get blasted with my lack of aerodynamics/cooling/etc knowledge.

I have the guppy mouth closure that covers some of the cooling intake area because (I think) there is excessive area that can't be vented out the cowl flap exit, hence excessive flow that doesn't do any good other that increase cooling drag.  And on my F with the cowl closure, there's not a hint of overheating no matter how hot the day even in climb.

So what would be the issue of testing increasingly smaller front vents to see if there's a sweet spot that maximises the cooling flow without excessive cooling drag.  Is that iterative testing too simplistic?  What am I missing here?  It may not be as aesthetically pleasing, for me, speed increase per $ is what I want.  Especially now that the Canadian dollar is headed in the direction of the Peso. 

And here's a completely off the wall suggestion.  Has any aircraft been built or tested using variable cowl inlets?  That would seem to be a simpler and more effective solution.  Why have a fixed input that is excessive in airflow and drag for cruise and try to compensate by restricting the output by closing the cowl flaps?

[takair]

I like the thought of your iterative approach to shrinking the inlet.  As usual, the hard part is legality.  I suspect that the sweet spot for improvement is somewhere in between field approval and STC.  It might be easiest to put an experimental ticket on the plane to do the trial.  Regarding variable inlet, I have seen it on some radials, Russian vintage as I recall.  Have not seen it on an opposed, but it would be interesting to see how it works.  It does seem like a better way to control flow than cowl flaps, but I suspect that the geometry of making them work is the downside.  I too have been thinking of this, but always come up short with a simple, practical and aerodynamically beneficial variable inlet.  I am looking forward to seeing Sabremechs data on his cowl.  Would be great if he got something over 5kt, especially if he can price it right.

[carusoam]

Cyril,

have no fear in being blasted... Some times it comes accidently.

You have made some good points.  Go directly to the thread where Sabremech is building a new Cowl for the old birds and ask.  He has been incredibly thoughtful and open with the work he is doing.  In his case, he has limitations of function vs economics including CB economics.  There are many potential customers that have only a CB budget to work with.

From an engineering point of view, getting enough air flow to evenly pass over all four cylinders without increasing drag or over cooling in the winter is a challenge.  Putting variable restrictions to the flow has been done previously by closing cowl flaps. Closing the front end or the back end achieves the same or very similar flow adjusting goal. 

There are portable instruments that can measure the speed of air passing by a small electronic sensor. Trying to measure air speed going past the cooling fins would be greatly helpful when designing air flow channels for the cowl.  There are also fog generators that can help visualize how air flows around things.  There are tiny video cameras that get into tight spaces and don't interrupt the air flow they are trying to record.

really modern Mooneys have set cowl openings that are quite small for the six cylinders they are cooling.  There are no cowl flaps to close.  Somebody must have done some really interesting air flow studies.  Compare the front openings of the O with the larger round openings for the Acclaim.  The TN'd Acclaim will produce full power at altitudes where the air density has been more than cut in half.

It would make some sense to borrow these well proven ideas and apply them to the engine to be cowled.  The newer air intake for the O's and J's engines are also updated to better handle snow/ice before feeding the air to the engine.  It has really low restriction for not being the Ram Air type.  Leaving the alt air door open may have a measurable effect on MP.  It would be measured in decimal points on a JPI (or other) if it exists.

Modern manufacturing techniques with modern materials have come a long way.  The old aluminum dog house with cowl flaps worked pretty well for its day.  Today we can achieve a few things like better CHTs with better cowling design. Less drag than the fully open and reverse flowing Guppy mouth.

Going with Rob's idea of developing a variable closing inlet is done in the air conditioning industry.  They use something called an iris valve.  It is infinitely adjustable but usually designed to only handle air flowing by at 90fpm approximately 60mph...  Some modification may be required.

Additional challenges to this development, and the STC process, is what happens when it is a real plane in real flight with real range of angle of attacks throw different weather phenomena.  Air is also a compressible fluid and really like to misbehave.  Good news: Flow through a cowl is friendlier than flow through an intake system with various valves opening and closing.  

For additional air flow challenge stuff, look at various intake designs for various engines.  A Good range of LOPs requires both balanced FIs and balanced air intake tubes to avoid engine vibrations.  Six cylinders are more challenging than Four for this.  Clarence's Eight cylinder Comanche doesn't run high MP LOP does it?   

Taking it to the next level, see where the hardware gets mounted for TC and TN'd engines.  They have to include air cooling and space for a a pair of inter coolers.  These are large air to air cooling radiators of a sort.  More equipment to fit under the cowl with complex plumbing included with a pair of very hot snails.

How is that?  Am I even close?

I am only a PP, not a mechanic. This is for conversation purposes only....

Best regards,

-a-

[Hank]

I dunno, Anthony. You made it sound pretty easy. 

Let me see:

• From sea level to FL 180, air density lowers by half
• Modeling flow across the cylinders must take into account that air is compressible
• Air flow required varies proportionally with MP
• More air flow is required at higher altitudes due to lower density
• Cooling inlets/exits must be modeled based on volumetric flow
• Due to density variation with airspeed and altitude, cooling required must be modeled on mass flow
• I really hated my Mass Flow class in college

Cooling inlet design is more complicated than I want to play with right now . . . .

[carusoam]

If only AL had a computer guy back then, and a fiberglass guy, and some airflow sensors, and an IPhone...

and this guy to build some things for the really cool customers...

Best regards,

-a-

[Hank]

Considering he didn't, the results were pretty good. Kind of like Spock trying to build a trick rider with stone knives and a bearskin, except Al's project worked (almost) as intended. What's a little reverse flow out the air intake when the plane flies as great as it does? 

[Marauder]

I'm starting a new topic prompted by an append by Yetti on another thread.

His question about a simpler cowl closure has been on my mind also, but I hesitated to put thoughts on the forum lest I get blasted with my lack of aerodynamics/cooling/etc knowledge.

I have the guppy mouth closure that covers some of the cooling intake area because (I think) there is excessive area that can't be vented out the cowl flap exit, hence excessive flow that doesn't do any good other that increase cooling drag.  And on my F with the cowl closure, there's not a hint of overheating no matter how hot the day even in climb.

So what would be the issue of testing increasingly smaller front vents to see if there's a sweet spot that maximises the cooling flow without excessive cooling drag.  Is that iterative testing too simplistic?  What am I missing here?  It may not be as aesthetically pleasing, for me, speed increase per $ is what I want.  Especially now that the Canadian dollar is headed in the direction of the Peso. 

And here's a completely off the wall suggestion.  Has any aircraft been built or tested using variable cowl inlets?  That would seem to be a simpler and more effective solution.  Why have a fixed input that is excessive in airflow and drag for cruise and try to compensate by restricting the output by closing the cowl flaps?

Do you have the fixed middle cowl on your model F? Mine has it and I run cool. Sometimes too cool!

Sent from my iPad using Tapatalk

[Cyril Gibb]

15 minutes ago, Marauder said:

Do you have the fixed middle cowl on your model F? Mine has it and I run cool. Sometimes too cool!

Sent from my iPad using Tapatalk

Not sure what a fixed middle cowl is.  I have a 75F.  Temps are pretty well always 250-290 winter or summer at cruise.

 

[Marauder]

15 minutes ago, Marauder said: Do you have the fixed middle cowl on your model F? Mine has it and I run cool. Sometimes too cool!

Sent from my iPad using Tapatalk

Not sure what a fixed middle cowl is.  I have a 75F.  Temps are pretty well always 250-290 winter or summer at cruise.

 

If you look right in front of your nose wheel, the lower portion of the cowl is bent outwards. They began putting fixed cowl on at some point in the series. Not sure all of them have it.

Sent from my iPad using Tapatalk

[Marauder]

BTW - mine is a 1975 F as well.

Sent from my iPad using Tapatalk

[Sabremech]

Knowing full well what it takes to get an STC and working on my cowling STC right now, I would not want to go through the pain , financial and mental, to get a variable cowl inlet system through certification. One word, safety, will make this extremely expensive and very complicated to do.

It's an interesting idea that may best be done by an airplane manufacturer or someone in the experimental aircraft market first.

 David

[Yetti]

My thoughts.  You need an upper piece and a lower piece.  The upper J cowl is fiberglass and has issue with heat and delamination.   The F is all metal with trussing.   There is a nice big rounded surface in the guppy mouth to mount the pieces to.   There is a big fan right in front of the holes.  But the holes seem to be a the root of the blades so does air move from the propeller blades or the forward motion of the plane.   My 75 F made nov/dec of 74 has a long air gap in front of the firewall.   Non adjustable cowl flap?

[Yetti]

Google fu

 

It has the ARI cowl mod

 

[Yetti]

http://www.aeroresourcesinc.com/proj98-ci/WebContent/product_category/product_detail/1052

[carusoam]

David,

I think the variable inlet may have been for development purposes, to select a final size..?

Or, that's what I wanted to read. 

Best regards,

-a-

[Sabremech]

Hi Carusoam, 

I'm actually doing some different size inlet ring testing as I type this. I started out with 4.5 inch inlet rings and now I installed 4.7 inch rings for my next couple of flights. I'm interested to see if there's any impact on cooling with that little change. When I do my performance flights, I'll do the same testing to see if there's any noticeable changes in speed or cooling. 

Hi Yetti, 

I had the ARI enclosure on my airplane and it's a good minimal quick modification. I had some issues with mine, hence my reasoning for going ahead and doing my own. 

I'm hopeful for some good numbers with mine and with the decrease in cooling temps already seen, I'm halfway to my goal. Performance runs will be soon and I'll be able to disclose those numbers. 

David

[Yetti]

It seems like wind tunnel testing.  Would be the way to answer lots of my questions.  It was interesting to learn how much the Wright Brothers built wind tunnels and test rigs.   Should the inlets be inboard or farther out?   Is there just way more airflow than the engine needs to stay cool? Seems like to optimize it and not use a wind tunnel whoever mentioned a Ardiuno with some thermocouplers recording lots of temperatures and airflow in 4 places would be best.    Yarn pieces taped all over the cowl and running the engine would be a fun video for showing ground ops cooling.

[M20F-1968]

Sabermech,

Now that you are in FAA testing phase for your STC, you undoubtedly will be doing cooling tests with them.  Take advantage of this time to pretest the opening sizes since the FAA will certify your "best choice."  Since you have round openings, get some plastic or nylon type material, put it on a metal lathe and made a set of varying size inserts.  Who knows what you will find.

Hope all is well.

John Breda

[Yetti]

Oh Snap. I do my best work when asleep.   What is the first mod you do to a corvette?  Replace the mass airflow sensor.   So Ardiuno and mass air sensor  http://forum.arduino.cc/index.php?topic=3690.0

Why are the inlet holes centered between the cowls?   Why not put them in the upper cowl maybe more outboard?  Seems like it would be a more direct path to move the air in and  across the cylinders, then through the fins.

 

[Cyril Gibb]

2 hours ago, M20F-1968 said:

Sabermech,

Now that you are in FAA testing phase for your STC, you undoubtedly will be doing cooling tests with them.  Take advantage of this time to pretest the opening sizes since the FAA will certify your "best choice."  Since you have round openings, get some plastic or nylon type material, put it on a metal lathe and made a set of varying size inserts.  Who knows what you will find.

Hope all is well.

John Breda

My Cherokee many decades ago had an oil cooler blocking plate for winter.  Wouldn't it be neat for Sabermech's cowl to have snap-in openings of various sizes for summer/winter or perhaps to help some models that have more of a cooling problem.

[N201MKTurbo]

Everybody thinks the air just flows in the front and through the engine.

That is the wrong way of thinking about it. The upper cowl acts like a pitot tube, causing a dynamic pressure rise proportional to airspeed. The cowl flaps cause a little negative pressure on the bottom because of Bournelli. Either way there is a pressure differential across the engine. The pressure differential is what drives the air flow across the cylinders. The cowl flaps control the pressure in the lower cowl, therefore controlling the airflow across the engine.

[Cyril Gibb]

21 minutes ago, N201MKTurbo said:

Everybody thinks the air just flows in the front and through the engine.

That is the wrong way of thinking about it. The upper cowl acts like a pitot tube, causing a dynamic pressure rise proportional to airspeed. The cowl flaps cause a little negative pressure on the bottom because of Bournelli. Either way there is a pressure differential across the engine. The pressure differential is what drives the air flow across the cylinders. The cowl flaps control the pressure in the lower cowl, therefore controlling the airflow across the engine.

You are absolutely correct.  I realise the concept that high pressure air in the upper cowl and relatively lower pressure in the lower cowl creates a flow downward through the cooling fins.

However, from my experience with air cooled racing engines (motorcycles and go-karts), the last thing you want is drag-inducing, uneven and relatively ineffective cooling airflow perpendicular to the fins.  In our aircraft, the airflow comes in the front, turns 90 degrees downward through the fins with all manner of chaotic flow and then turns more or less 90 degrees again to get out the cowl flaps.  That's a whole lot of energy to redirect.  Air moving at that speed has a huge amount of inertia.  It has to slow down, turn, turn again and then speed up to exit.  Each of those takes energy away from forward thrust. The Bernoulli effect speeding up the air again using the partial vacuum exiting the lower cowl has additional drag all by itself. Bernoulli ain't free.

It just seems to me that the way we control cylinder temperature levels is analogous to keeping your bathtub at the correct level by turning on the faucets full and then removing the stopper periodically.

[N201MKTurbo]

There were cowl designs with louvers on the inlets back in the '20s. There are still some Russian planes that have these. I believe the current arrangement was determined to be the most efficient many decades ago. Roy Lopresti did a lot of work in the 90s to improve it further and he made some progress with the smaller round openings and trying to recover some energy from the propeller. But for the most part, no major changes.

[carusoam]

David posted some planning pictures of the internal parts for directing airflow.  Curved fiberglass pieces.  Could be an improvement compared to the flow through a leaky square dog house.

Best regards,

-a-

[mike_elliott]

On ‎2‎/‎12‎/‎2016 at 10:56 PM, Yetti said:

It seems like wind tunnel testing.  Would be the way to answer lots of my questions. 

Waaay too expensive. Target Chip Ganassi racing spends as much as a new Mooney ultra to build a 1/10 scale model of their cars for wind tunnel testing. Then there is the cost of the wind tunnel. Heck, Chip bought a mountain in PA that a road tunnel was in after a highway was rerouted just so he could plug up one end and make a full scale wind tunnel sort of out of it. Then the stinking sanctioning bodies banned that testing after one of his drivers shot his mouth off about it.