Plenum configuration

[Bob Axsom]

At this time at least I do not intend to go to the round inlet self contained plenum because it seems to me that a lot of attention to detail has been given to the standard design (seal at the front end behind the spinner for example) and my seems to be working very well for all of the dependent systems. One area that has been giving me a little thought is the upper "ramps" that run from the aft end of the upper edge of the air inlets up to the inside of the upper cowl. I thought of closing in the inboard gap from the ramp to the cowl but I never did. I still wonder what the effect would have been. It would reduce the plenum volume and may do nothing the reduce cooling drag - on the other hand... Well that's why just wonder rather than doing something. What are the thermo and aero thoughts on closing off the area under the inlet "venturi" ramps?

Bob Axsom

[mark manda]

can you post a pict or link to a pict of your inside cowl? I have some pinhole thoughts. the numbers on pinholes on the inner and outer upper cowl in 2 zillion and something. (did I do mine the way you're thinking? thought everybody did it THIS way. except the holy cowls? )

[Bob Axsom]

What your photo shows is exactly what I was/am considering. I would be very interested in your results especially if you are running a O-360-A1A with a non-blended airfoil Hartzell constant speed prop. At 75% power I typically see a true airspeed in the 171 - 173 kt range. When I flew back from Bartlesville Saturday I pushed it up to 2500 RPM at 5,500 OAT 23C and I was seeing a TAS of 176 kts. As you can see my plane is flirting with 200mph and I would like to teak it to reliably hit it. Driving up to Bella Vista (sounds like California but it really is a beautiful area in Northwest Arkansas) to the Chapter 732 meeting yesterday with my friend Barry West (ME from University of Arkansas and Masters specializing in Thermal from SMU) I asked him about the idea. He said it probably will not help anything and it will introduce a lot of unknowns that you will never understand without proper instrumentation. He said in a true plenum the velocity of the air is zero and changes to volume will affect the function. He felt there are probably performance gains to be made but they will require some formal development work to optimize performance. He went into the round inlets being the most efficient shape, etc. etc. which we have all heard before but one thing caught my ear was the statement the blending the side of the ramps to the cowl over a curved surface would probably be a step in the right direction. Your edge has a sharp break just like I was thinking about and in spite of what Barry says I would really like to see how yours performs.

Bob Axsom

[avpro56]

Bob:

I have the configuration that you describe on my RV-8. I'm running a O-360 A1A, with balanced components and Airflow Fuel Injection. I am also equipped with a Hartzell blended airfoil 74" propeller. I do not have a self contained plenum, but I went to a good amount of effort and paid close attention to baffles and the inlet ramps. I did close off the inside edges of the inlet ramps as in the photo posted, however I used a radius about the size of a quarter.

My whole theory was to come as close in performance to using a separate plenum as possible while using the stock cowl setup by paying close atention to detail. I cannot compare a before and after with the closed off inlet ramps because I built it this way from the start. However, during the AirVenture Cup Race 2005 while running wide open at 2675 RPM my cylindwer head temps never got above 360 degrees. OAT was 37 degarres C, boy it was hot in that cockpit! My limiting factor was oil temerature, probably because I have a remote mounted oil cooler on the engine mount right behind and below the #4 cylinder. It is connected via a rather short 3" scat duct. When things get hot I can see 225-230 degree oil temps. The probelm seems to be the ineffecient scat duct, and the valoume of air not being ample to cool. I am planning making a custom duct to supply more air the the Stewart Warner 8406R oil cooler.

I conclude that I have plenty of cooling air, so a larger oil cooler duct will not be a problem. It's my thought that if the cooling inlet ramps had not been sealed off then it's likely that cylinder head temperatures would not be so manageable.

Several cooling drag seminars at OSH have suggested that the popular round air inlets commonly seen with most plenum istallattions are the result of the original design software being limited to just two dimensions. So I'm not quite sure that the round inlet is best for the cowling inlet; it's just the most common. But, that's a whole other topic for another post.

Anyway, just my .02c.

Sorry I did not get a chance to meet and speak with you during the race or at OSH this year.

Best Regards,
Jon Ross
Race 27 RV-8 N207RV
2005 First Place Formula RV Blue @203.26 MPH

[Bob Axsom]

Jon,

This is great feedback! During the race I made the stupid climb to 8,500 ft and really blew my speed. My oil cooler is hanging right on the rear baffel behing cylinder #4 - standard/easiest installation location. It is well braced so hopefully it will hold up. My oil temps were stable at approximately 200. During that race I was only running about 2550 RPM because some little testing I ran showed my speed peaked then "appeared" to drop off when I pushed it on up over 2600 RPM. Probably bad testing but if the weather holds up we are flying to Mesquite, Texas for the Cowhand 300 air race and I can get some more data. Your experience is very stimulating to "cut and try" the modification. Thanks for this very meaningful input.

Bob Axsom

[gmcjetpilot]

Originally Posted by avpro56

Several cooling drag seminars at OSH have suggested that the popular round air inlets commonly seen with most plenum installations are the result of the original design software being limited to just two dimensions. So I'm not quite sure that the round inlet is best for the cowling inlet; it's just the most common. But, that's a whole other topic for another post. Jon Ross

I disagree with you. If you ask Dave Anders-RV-4, Tracy Saylor RV-6, Dick Martin-RV-8 and Jankowski-RV-6 I think they will tell you the difference. NASA sponsored research, conducted by Mississippi State in the late 70's early 80's did flight test using a Piper Aztec. All the data was from flight-test not computers. Several inlets where tested. LoPresti was the first to market and used this data with his aftermarket cowl inlet kits.

(Ref.: An experimental investigation of the aerodynamics and cooling of a horizontally-opposed air-cooled aircraft engine installation, Miley, S. J.;

Practical flight test method for determining reciprocating engine cooling requirements - WARD D. T.;MILEY, S. J., Texas A and M University, College Station, Texas;

Effect of Propeller on Engine Cooling System Drag and Performance - Joseph Katz, Victor R. Corsiglia and Philip R. Barlow)

You are correct it is 3 dimensional flow and the rectangular shape is not great. If fact if you tuft your Vans cowl inlet you will see air flowing out the inboard edge of the cowl inlet. The prop has a large affect and next to the spinner the prop hub is a blunt club and beats the air to death. Further out towards the tip you get better pressures (Cp). Bottom line round is a pretty good shape for getting the required area as far outboard (away from the spinner) and staying in the cowl shape. Also the secret is not at the round ring, but behind in the "diffuser". With Vans cowl and baffle system you have large discontinuities in areas and abrupt edges. With a SJ cowl, being round it is easy to join the inlet ring to the plenum with a smooth flex duct clamped to the cowl inlet rings and plenum.

Plenum sealing is also a result of the NASA study. If you look at the transition from the cowl to the baffle and all the soft seals you will many places for leaks. A round inlet with a duct (diffuser) is completely sealed with clamps and is much more efficient. You have know idea how much you soft seal baffles leak at 200 mph. These are just a few reasons the inlet area of a round Sam James style cowl has inlet areas 30% to 35% less. Speed increase has been documented to be 6-10 mph on a RV. When Randy Lervold did a prop test and Van tested the same prop on the factory RV-8 with a 180HP engine. Randys RV was about 15mph faster under the same conditions.

I am using stock cowl with round inlet rings and solid / sealed plenum. So if you are going racing, by all means use the stock cowl and soft seal baffles so I can outrun you. G

PS Congrats on your first place finish; look forward to racing against you next year. Also Dick Martins RV-8, with SJ cowl, did 232 mph. Even though he is running a IO-390 and 210 HP(?), the extra HP did not make up the whole 20 MPH difference. What is interesting is a O-320 tailwind went 214mph. I guess winds where a factor, so it is hard to draw any conclusion about airframes, cowls or engines. Those who went low won. In the open Blue class a Glasair I went 242mph!

[avpro56]

I disagree with you. If you ask Dave Anders-RV-4, Tracy Saylor RV-6, Dick Martin-RV-8 and Jankowski-RV-6 I think they will tell you the difference. NASA sponsored research, conducted by Mississippi State in the late 70's early 80's did flight test using a Piper Aztec. All the data was from flight-test not computers. Several inlets where tested. LoPresti was the first to market and used this data with his aftermarket cowl inlet kits.

(Ref.: An experimental investigation of the aerodynamics and cooling of a horizontally-opposed air-cooled aircraft engine installation, Miley, S. J.;

Practical flight test method for determining reciprocating engine cooling requirements - WARD D. T.;MILEY, S. J., Texas A and M University, College Station, Texas;

Effect of Propeller on Engine Cooling System Drag and Performance - Joseph Katz, Victor R. Corsiglia and Philip R. Barlow)

Controvesey is the spice of life. (or forum posts) what I said was that is was suggested, not what is. I was very caerful with my words. I will review the sources that you cite and comment.

You are correct it is 3 dimensional flow and the rectangular shape is not great. If fact if you tuft your Vans cowl inlet you will see air flowing out the inboard edge of the cowl inlet. The prop has a large affect and next to the spinner the prop hub is a blunt club and beats the air to death. Further out towards the tip you get better pressures (Cp). Bottom line round is a pretty good shape for getting the required area as far outboard (away from the spinner) and staying in the cowl shape. Also the secret is not at the round ring, but behind in the "diffuser". With Vans cowl and baffle system you have large discontinuities in areas and abrupt edges. With a SJ cowl, being round it is easy to join the inlet ring to the plenum with a smooth flex duct clamped to the cowl inlet rings and plenum.

I am familar with this subject area. However, it is true that the original rings were designed with 2D software. I also agree that a plenum is a more efficient way to cool an engine, and will reult in less drag. Less cooling air required = less drag. No doubt that soft seal baffles are leaky and cause a loss of efficiency and result in higher drag.

If you have photos or quantitative data regarding tufting in the area you cite, please post it heere so all can see and learn. I know that the GAMI guys have video of tufted Bonanza cowling areas in flight. I have not seen them, but would like to.

Plenum sealing is also a result of the NASA study. If you look at the transition from the cowl to the baffle and all the soft seals you will many places for leaks. A round inlet with a duct (diffuser) is completely sealed with clamps and is much more efficient. You have know idea how much you soft seal baffles leak at 200 mph. These are just a few reasons the inlet area of a round Sam James style cowl has inlet areas 30% to 35% less. Speed increase has been documented to be 6-10 mph on a RV. When Randy Lervold did a prop test and Van tested the same prop on the factory RV-8 with a 180HP engine. Randys RV was about 15mph faster under the same conditions.

I would like to see the comparative data on stock VS plenum configurations. Randy Lervold is a good friend, and to my knowledge he never raced his plane, and does not claim the speed increase that you cite as an average due the the plenum. However, I am very familiar with the propeller testing data that he did accomplish.

I am using stock cowl with round inlet rings and solid / sealed plenum. So if you are going racing, by all means use the stock cowl and soft seal baffles so I can outrun you. G

Where were you in 05? You might wish to review the data on the race. John Huft in his RV-8 came in behind me with an identical engine and prop but was equipped with a plenum and Sam James Cowl. You might want to ask John what he thinks of the Sam James Cowl, he has mixed feelings. Having said that, I feel his airplane is better than mine.

PS Congrats on your first place finish; look forward to racing against you next year. Also Dick Martins RV-8, with SJ cowl, did 232 mph. Even though he is running a IO-390 and 210 HP(?), the extra HP did not make up the whole 20 MPH difference. What is interesting is a O-320 tailwind went 214mph. I guess winds where a factor, so it is hard to draw any conclusion about airframes, cowls or engines. Those who went low won. In the open Blue class a Glasair I went 242mph!

You leave out the fact the Dick Martin ran out of gas and had to stop at the finish line. I know, I was there; Dick was unable to recover in Fond Du Lac because of fuel issues. Dick was put into another class because of the displacement of his engine.

Red Hamilton's Tailwind did very well at 214 MPH. He did run his O-320 at 3100 RPM, which made it put out considerably more power. The Tailwind is still a very, very good design that can be built on a budget.

The AirVenture Cup race is about camaraderie and skill as much as it is about fast airplanes. I?ve made a few new friends, and it was a great experience to take part in the race with Red Hamilton, Dick Martin, John Huft and many others.

Jon Ross

[Bob Axsom]

I have the stock baffle plenum configuration well sealed with the good red RTV from Aircraft Spruce (everywhere there is a metal to metal interface is sealed - EVERYWHERE!). The rubber seals are the standard seal material supplied by vans. I have a big hole in the aft baffel over cylinder #3 for the heat/defrost system and the oil cooler hole is similarly located over cylinder #4 - there are no other unsealed baffle penetrations. The engine is a Lycoming O-360-A1A and the prop is the non-blended airfoil Hartzell C/S. I have a custom thin aluminum lap cover over the leading edge of the cowl covering the split and taking the front end separation load off of the hinges (dimpled for screws - two on the outside attaching the upper and lower cowl together with this aluminum cover and one inside the inlet for both the upper and lower cowl halfs.)

Today Barry West and I had to fly under U.S. Air Race rules to establish our handicap for the race on Sunday. The rules state:

- Set altimeter to barometric setting of 29.92 (pressure altitude). Climb to 6,000' pressure altitude. Observe the actual outside air temperature (OAT). On a standard day it should be 3C/38F. Calculate the difference between the actual OAT in Centigrade and 3C. Descend 100 feet for each degree C that the OAT is above 3C (60 feet for each degree above 38F) or, climb the same amount if the OAT is below standard. Summer example: OAT at 6,000' is 27C, once your altimeter is set to 29.92, you will fly at an indicated pressure altitude of 3,600' in order to achieve flying at a density altitude od 6,000'. Equation is 6000'-((27C-3C)x100=3600'. [we read 22C so our handicap flights were at 4,100ft]
- Flight must be made at full throttle and full RPF for constant speed prop. Lean engine to BEST POWER per the aircraft's POH. Flaps must be full up, cowl closed, airconditioner in "off" position, ram air open, and anything protruding into the airstream retracted (vents on some aircraft).
- Aircraft must be trimmed to "hands off" level flight. Autopilot altitude/heading hold may now be used.
- Maintain level flightand fly 360 degrees until 5 consistent GPS readings are obtained. Readings should be taken every 20 seconds. The readings should not vary more than one knot. Record the readings.
Good examples: 171, 170, 171, 170, 170, 170 or 173, 172, 171, 171, 171, 170, 171 (last five are good)
Bad examples: 171, 173, 175, 173, 171 or 170, 172, 173, 170, 168
- Turn to 240 degrees and repeat the process.
- Turn to 120 degrees and repeat the process
- Add the 15 good readings together and divide by 15 to two decimal places. This is your OPTIMAL SPEED and is your HANDICAP.

Our results today complying with these rules gave us an optimal speed of 170.66kts or 196.39 MPH.

That could be used as a plain vanila cowl & plenum 6,000 ft density altitude speed. I would be greatly surprized if running a hose nose into a small hard plenum gives the magnitude of speed gains expected with no other power enhancing mods. Not only would I be suprised - I would probably modify my cowl and plenum. I am inclined to go with the rounded closure of the upper cowl inlet ramp.

Bob Axsom

[gmcjetpilot]

Originally Posted by Bob Axsom

That could be used as a plain vanila cowl & plenum 6,000 ft density altitude speed. I would be greatly surprized if running a hose nose into a small hard plenum gives the magnitude of speed gains expected with no other power enhancing mods. Not only would I be suprised - I would probably modify my cowl and plenum. I am inclined to go with the rounded closure of the upper cowl inlet ramp. Bob Axsom

Be surprised. As far as leaks get a mirror and look just inside the cowl lip, at the upper corners, inboard and outboard. You will see gaps. I know this gets a little touchy because it sounds like I am saying Van's cowl is bad. Look at the performance numbers LoPresti gets from his Mooney, Grumman Tiger, Piper mods, about 5 MPH. That is a pretty conservative number. Also remember the RV is already going fast so there is more cooling drag to be saved.

The speeds you mention are good but are in line with a stock RV according to Van's numbers (200-204mph at 75% pwr @ 8,000'). There is room for improvement with the cowl and baffles. You can add HP or lower drag. Cooling drag is the one big-ticket item there to be had. There is real flight test data showing the increased efficiency of moving the inlets outboard, making the duct (diffuser) smooth. If you took cross sections of vans cowl and baffle system you would see large transitions in shape and areas. This causes internal turbulence and drag. Think of your cowl inlet and exit as an inside out wing. You can be in denial but there is not argument that reducing drag will add speed. Van's cowl does not have as good of details as a SJ style cowl. Further you are reducing the inlet area by a third and still cooling the engine and therefore..... it is not rocket science you go faster. Would you leave you wheel pants off and loose 5 mph? Bob if you put a hose nose on, as you say and put a sealed hard top plenum on, I know you will gain speed, but it is your plane. If you are into racing you will have a disadvantage to other RV's with this NASA inspired cooling system. You talk of Handicap racing. They don't call it Handicap for nothing. If you want all out performance and heads up racing cooling mods do reduce drag. G

[gmcjetpilot]

avpro56 wrote: I am familiar with this subject area. However, it is true that the original rings were designed with 2D software.
(Well yes the inlet rings are based on the laminar flow A-10 series which is a 2D shape. However when talking engineering of round ducts and fluid dynamics closed form solutions have been known of decades. This is not new stuff. Also the flight tested it with instrumentation galore. Plus they used solid well know basic aerodynamic fluid flow theories.)

avpro56 wrote: Less cooling air required = less drag. No doubt that soft seal baffles are leaky and cause a loss of efficiency and result in higher drag.
(Proven by the NASA study at Miss State and a Piper Aztec, you don't have to guess if you look at the data)


avpro56 wrote: If you have photos or quantitative data regarding tufting in the area you cite, please post it here so all can see and learn. I know that the GAMI guys have video of tufted Bonanza cowling areas in flight. I have not seen them, but would like to.
(Geeee you want everything. I was an aerospace consultant and made about$128/hr before I started flying for a living. I will be glad to give you all my race secrets for a fee. No really it was published in the RVator many years ago. Look you have a stagnation point just inside the inlet. The prop blast is so great on the out board edge of the horizontally orientated rectangular inlet, verses the inboard edge, which has a negative Cp. Meaning air will flow to the low pressure and right back out the cowl. This is well known stuff. I think you all think I make this up. Look at the NASA report. I can?t do all the work. Having an engineering degree and working in the Aerospace industry helps me be able to read these documents, but it is not hard if you try. Before I committed to the idea of modifying my cowl for my RV, I researched it. I also did my own test on my old RV-4 with a stock cowl. I blocked off the inner edge of my cowl with shaped foam blocks and speed tape, reducing the inlet area. The result was almost no cooling change and may be 1-2 mph. It is too hard to measure small speed changes between flights and conditions, but it showed the inner edge was not doing much cooling. If fact as I say air flows out because of the props pressure gradient (coefficient of pressure Cp). This reverse flow causes plume drag as it interferes with the free air stream. The idea was not to show speed increase but affect on cooling. Again the transition from Van?s cowl inlet in to the cooling plenum are too abrupt, diverging angles too great and sharp corners to give smooth non turbulent, low drag airflow. Again thing of the are flowing thru you cowl like air flowing over you wing. All this is well known, but cooling systems where neglected. In the 40's mass research was done on cooling of WWII fighter's piston engines (radials and inline), but it stopped when jets came along. It was not till some 35 years later more work on cooling drag was done on horizontal opposed engines such as we have. Even still more work could be done. The LoPresti/SJ cowl are reflections of what was learned. Not that these are perfect examples of the concepts and theories, they are better than what was done in the past.)


avpro56 wrote: I would like to see the comparative data on stock VS plenum configurations. Randy Lervold is a good friend, and to my knowledge he never raced his plane, and does not claim the speed increase that you cite as an average due the plenum. However, I am very familiar with the propeller testing data that he did accomplish.
(Randy as you know has prop tests on his site. Van worked with Randy in testing props. As a scale factor van used the speeds for the props that where tested on both Randy's and Van's prototype to compare relative speeds and scale results on other props. Randy's plane was much faster than vans with the same HP and prop. This was published in the RVator a year ago. Look at the speed randy got and what Van published. You will see this handicap if you will. Randy has a SJ cowl. How much of the 14mph can be contributed? Randy has electronic ignition but 14 mph? Is Van's RV-8 slow? No the absolute speed was at target or specification nominal speed or more for a 180hp RV-8. My conjecture is the only major difference is the cowl, thus may account for some of this speed advantage. Even of it was half of the 14 mph it would be significant.)


avpro56 wrote: Where were you in 05? You might wish to review the data on the race. John Huft in his RV-8 came in behind me with an identical engine and prop but was equipped with a plenum and Sam James Cowl. You might want to ask John what he thinks of the Sam James Cowl, he has mixed feelings. Having said that, I feel his airplane is better than mine.
(I was building and working. John who? I understand the winds where a factor during a race. Pilot skill comes in to play. It is racing after all. Some guys get lost. My 150hp RV-4 use to clean up 160 hp and a few 180 hp RV's. No one is claiming the SJ cowl and plenum will make your RV go 250 mph. It might make it go 6-10 mph faster. Now why are some RV's slow? Weight, fit and finish, alignment rigging? who knows.)


avpro56 wrote: You leave out the fact the Dick Martin ran out of gas and had to stop at the finish line. I know, I was there; Dick was unable to recover in Fond Du Lac because of fuel issues. Dick was put into another class because of the displacement of his engine.
(Yea I know about the large engine but he is going much faster than a 200 hp RV-8, at least in other races. It is not all engine making him go faster. I estimated the cowl was worth about 6 mph on his plane, which is reasonable, gas burn a side, which has nothing to do with the cowl. You forgot to mention he beat you by 20 mph. )

avpro56 wrote: Red Hamilton's Tailwind did very well at 214 MPH. He did run his O-320 at 3100 RPM, which made it put out considerably more power. The Tailwind is still a very, very good design that can be built on a budget.
(Yea, still he whooped ***. We can turn 3100 also. **** 2700-3100 is only 20 hp. They still would be going +205 mph on a stock O-320, 160 hp. All is fair in love, war and racing. I guess handicap racing will eliminate this but why bother. In the end he went 11 mph faster than the fastest RV with 20 cu-in more engine displacement. Racing is racing, and you can't draw all kinds of conclusions except one, someone finished faster or slower.)

In conclusion do not use a SJ, LoPresti style cowl on your RV so I can win some more races. G