I am just getting into the support structure behind the baffle; the basic materials will be #12 metal clamps on the engine mount, small transition angle brackets, small aluminum angle struts, and fasteners. I have angle bracket and platenut tie points already installed on the back of the baffling. Many configuration details have to be worked out as I make the struts. I will have what amounts to in inverted scoop from the center baffling up into the area above the exhaust pipes to eliminate the firewall lip from the exit air flow path and provide a normal cruise flight vent for everything behind the baffling like the cockpit heater air when the heater is off and the oil cooler exhaust air. For the AirVenture Cup Race I plan to block off the cockpit heater air port in the plenum, remove all of the hoses, remove the heat muff and cover the holes in the baffle for the heater air hoses. Before the first flight I will install red baffle seal material at all of the baffle penetrations for wires, tubes, etc. for both sealing and protection. After everything is finalized in a workable configuration I will seal the seals with Dow 736 RTV. Oh yes, I have to install the red seal around the outer edge of the baffle for the baffle to cowl interface.

I installed seven support struts today before I ran out of 2024-T3, 1/2"x1/2"x1/16" angle stock today. 

6/7/06
The side seals are riveted in place. The aluminum angle has arrived from Aircraft Spruce so the support structure can be finished today. The most serious task remaining is the closure of all the holes in the baffling around wires, tubes, hoses, etc. That work should start today as well but we have a lot of distractions today. Received a message from Paul Lipps offering encouragement, photo and other information.
 

6/8/06
Well, I put the last cross brace in the baffle support structure today and all that is left is closing all the penetration holes.  I have to say this looks like a dirty stinking job that is near impossible to do with style and grace but that begins tomorrow - carefully.

6/10
Well it may not look like much but today's effort was very satisfying. There is a spider web of stuff down below the engine between the engine and the firewall. I can't tell you how much I dreaded trying to develop baffling for this area. As is often the case the actual work is not as bad as the worrying about it. I am confident I will get good closure and air flow. It may not end up any faster but it won't be because the baffling is no good.

6/13
Attached is a photo of the three pieces of the lower cowl baffle that I made today. Things are going very well as far as the mod implementation is concerned. There are no drawings. I have a concept that is directing my efforts and it is if I can get the cooling air out of the cowl smoothly without right angle contact with the firewall, etc. the plane will go faster. I will know within a couple of weeks.

There is a center baffle component (this is actually a very complex assembly consisting of left and right sides that are eventually joined at a point several inches below the engine) that extends down from the rear of the engine to the bottom of the firewall. I use cut up file folders and masking tape to construct baffle piece patterns. It starts straight down then curves back to the lip formed by the bottom of the firewall and the lower fuselage skin. I have read that this lip is a surprisingly draggy feature (may or may not be true).

Then I have two attached side baffles that extend forward and outward from the center component to the cowl side and bottom (inside surface of course). I cut red rubber sheet from Aircraft Spruce to follow the contour of the side baffles and the cowl in this way I figured I could avoid the tension, compression and rolling tendency of the precut 3" wide material when formed around a curved surface (even with notches) - it is expensive but it worked beautifully. These side baffles are expected to deflect the air inward to the width of the cowl outlet and the center baffle component will deflect it downward to the cowl outlet.

The baffling is supported with 1/2"x1/2"x1/16" 2024-T3 aluminum angle struts, AN742-D12 clamps, and homemade platenut clips. The support is good for vertical and horizontal positioning. The baffling must not make metal to metal contact with the engine, etc. but you don't want air leaks either. I will get into that work tomorrow. I'm planning to use Dow 736 RTV and red silicone baffle seal material to close the penetration openings against air leaks and keep the baffle from cutting anything (support and chaffing protection).

This is a one of a kind experiment for me and it may result in no speed gain and the potential for careless work to create operational problems is high. Be real careful if you try something like this. I will share my results.

Bob Axsom

6/14
I installed all of the sheet metal with screws this evening and verified everything looks fine. I want to put one more plate nut in the upper left part of the center section but otherwise the sheet metal is done. I really hope to get into the hole closure activity tomorrow but before that I want to take a clipboard out there and do a thorough inspection with formally documented squawks (an old fashion shakedown). I will organize my work after that based on what I find. So far I have spent 108 hours on this mod development.

6/15
Well the metal work is done and it came out very well. I am well into closing the holes in the baffles with high temp rubber (silicone) strips but the first was a bit crude. Have to back off and work through this more carefully. Just the same the holes closed up very well.

6/19
I went out to the airport last night after the air show was over and looked at my oil cooler installation. The nice neat flow conduit with gentle curves, a rubber shock absorbing link and an aft slanting port into the narrow channel in the area of the cowl outlet that I had dreamed up is very appealing but space limitations and engine mount obstructions make it virtually impossible. The options I see are:

1 - try to do it anyway with efficiency robbing accommodations for the obstructions
2 - do nothing and see if the air will spill over the outer baffles (these deflection baffles do not close off the chamber behind the engine up to the upper cowl) and rejoin the cooling air leaving the engine
3 - add a scoop to the inside of the baffle to provide a deflecting surface for the engine cooling air to keep it flowing in the right direction and allow the air behind the baffle (especially the oil cooler exit air) to join the flow out of the area between the baffle and the firewall
4 - put spacers between the baffle panels and the bottom of the fuselage (they overlap to avoid an obstructing seam in the cowl outlet) to provide an outlet for both the engine cooling air and the oil cooler air at the cowl out let.
5 - construct a deflector for the rear of the oil cooler to turn the air in the best (?) direction
6 - change the oil cooler location from the plenum baffle above and behind the #4 cylinder to the inlet air ramp in front of the #2 cylinder
This is not a one time all or nothing situation like firing a rocket so I can test and see and modify until I get it right ... well at least functionally acceptable. I'll let you know how it comes out.

Made some good progress today closing the area around the forward heat muff hose.

6/21
This simple little rubber installation took me around 7 hours to get right.  I have started on the other side but it will probably take a couple of days to finish that part.

6/25
I photographed the "rubbered up" right side lower cowl baffling today and did an installation of those pieces with flight hardware. I'm going back out to the hangar tonight to work on sealing off the area directly under/behind the engine.

 

6/27
I have one place up on the left side of the engine to add some rubber and I will be done with the "riveting on rubber" iteration of the development process. That will be completed today. I have two central bottom pieces of baffling that overlap the bottom fuselage skin that have to be modified to provide a vent for the heated air dump (from the heat muff when the cockpit heater is not in use) and the oil cooler. Basically, I plan to separate the baffle trailing edge from the skin by 1/2" and close in the sides. Then I will be ready to start the Dow 736 RTV sealing iteration. I will let that cure over night before I fly. I may be ready to test fly this experiment on 6-28-06 - or 6-29-06 for sure.

6/28
I finished the lower Cowl baffle - speed mod tonight. Started 4-28-06 finished on 6-28-06. 180.8 Bob Axsom hours consumed. It is all buttoned up and I will fly it in the morning. I put a small vent in the cowl outlet space for the oil cooler and hot air dump.

6/29
You can't see much but the small vent I talked about earlier is the 1/4" x 7" space at the top center of the cowl outlet. This may not be necessary because there is also a route from the area between the baffle and the firewall to the area in front of the baffling over the side baffles outboard of the plenum. Even so I will be watching the oil temperature closely during tomorrow's test.

6/29 (more)
It is a warm day here in Arkansas, with clear skies and light winds - perfect for a speed mod test. I used the test method provided by U.S. Air Race, Inc. which I am told is not perfect but it is the same method I used last year to handicap the plane for their races. The resulting speed from that test was 170.67 kts. A copy of the form (including the procedure) is attached. I took off with everything the same as last year same two man crew, same weights, vents closed and climbed to 6,000ft. The OAT was 19C so we flew the test at 4,400ft this was coordinated with the tower and he provided RADAR tracking/traffic advisories (there was none). We got very consistent speeds at 20 second intervals reflecting the calm conditions. I trimmed the plane for hands off level flight and flew the 360 degree leg first with my Tru Trak altitude hold and directional control. The speeds were 178, 177, 177, 178, 178. Then I turned to 120 degrees and flew that leg to get farther from the airport. It took seven readings before we got five sequential within 1kt. They were 172, 172, 170, 170, 171, 171, 170 - so the last five were used. Then I turned to 240 degrees and the speeds recorded were 157, 158, 158, 158, 158. The total of the fifteen numbers used is 2529 so the average speed is 168.6 kts or 2 kts less that the results I came up with last year. It is possible that the airplane is 2 kts slower than it was last year without the mod but it is clear to me that the mod did not increase the speed of N710BJ and it may have slowed it down. I would not recommend implementing a mod of this type as a method of reducing cooling drag and increasing speed.

I need to add that the performance otherwise seemed no different than before. The oil temperature was around 200 which is it's standard spot after 223 hours of flight time on our RV-6A. So far this year we have flown to St. Augustine, FL; Asheville & Winston Salem, NC; Alamogordo, NM (Trinity site); and Santa Ynez, CA on four separate trips. Retirement is good though I sometimes miss the work but not the people.

7/1
As I'm told the English say "In for a penny, in for a pound", I'm not ready to cut my losses just yet. The attachment shows the pattern for the closure for the right side of the lower cowl. I have actually cut the piece out of .025 2024-T3 and riveted on mounting clips that will be attached to the engine by the baffle mounting screws in front of cylinders 1 and 3. The actual piece is smaller than the pattern because of a 3/8" set back from the cowling. Baffle sealing material will be riveted to the outboard edge and it will form the closure interface with the cowl. The edge in contact with the engine will be sealed with Dow 736 red RTV. Next I will do the left side in a similar fashion. OK, the intuitive approach here is to isolate the lower cowl cooling air flow path from the area behind the previously added baffling. Why do this? Well why not? It is relatively easy to do and it seems reasonable to eliminate some of the variables and see if there is a performance effect. Two things stand out with this change; 1) the volume of the after cylinder passage chamber is greatly reduced and 2 the variability of the flow is also reduced. Will it have a performance effect - I will see and then let you know. There are risks and my instrumentation is minimal.

7/2
The basic metal part of the two new closure baffles are complete. Tomorrow if all goes well I will install the perimeter rubber and start working on the interface with the two vertical side baffles in the lower cowl. Not difficult but it does require some thinking to end up with a clean and effective interface.

7/4
I have the right new baffle installed. I'm going back tonight and see what I can get done on the left one. I received a very good message from a Lancair 360 pilot named Chris Zavatson explaining that my mod probably increased the mass flow which cools better but makes the plane go slower (exactly what we saw in the test). Reducing the exit area should bring the temperatures back to the before mod levels and yield the speed increase I'm looking for. The little bit tough thing I have to do is buy CHT instrumentation and get it installed. Get it done in time for the AirVenture Cup Race is going to be tough.

7/7
I finished all of the aluminum and sheet rubber work for phase two of the evolving modification to the lower cowl on my RV-6A. The first phase just put baffling in the lower cowl to direct the air coming through the cylinder fins, etc. to the lower cowl outlet. That effort appears to have lost speed rather than gained speed. This phase attempts to seal area under the engine from the area between the new baffling and the firewall. This is purely experimental and I have no assurance that the performance will be any better or different than phase one. I received an IE C4 CHT system from Aircraft Spruce today and I plan to install it before the next flight. One the data are collected for this phase I plan to incorporate a cowl outlet restriction if the CHTs are not high. The plan is to tune the outlet so that a safe but high CHT is achieved in an attempt to reduce cooling drag based on a suggestion from Chris Zavatson. I want to stay away from the inlet mod until after the AirVenture Cup Race on the 23rd of this month.

7/8
I am exhausted, my head is throbbing, my eyes are blurry but the new baffling is sealed and the first generation baffling is sealed better. I still have to hook up two power wires and two ground wires for the CHT system and get the cowl back on but we intend to fly the second iteration test tomorrow. Barry West will be recording data. This time we will have CHT data but it will have no history to be compared with. We will see what tomorrow brings. By the way, getting the cowl on may not be a trivial task.

7/9
Jeanine and I went out to the hangar and installed the lower cowl this morning with little difficulty. It is definitely a two person job now however. Yesterday I got the 4 cylinder CHT system installed but the power and ground connections still have to be made. It had been a while since I had used a 2 1/4" hole saw. I made a couple of panel scars below the instrument hole because I forgot how uncontrollable the saws are after the hole is cut and that you can't go back in with it to clean up without the pilot drill alignment function. Photos of the CHT installation and the sealed lower cowl are attached. I hope to fly this afternoon but the clouds may prevent us from going to 6,000 ft density altitude for a comparable test. I got my copy of "Speed With Economy" by Kent Paser out based on a posting I saw recently. A great straight shooter book on the subject bit I had not read it in years. Some of the information around page 65 corroborated the information received from Chris Zavatson and I am convinced that we are going the right way with this mod.

7/9
Fight test of second iteration lower cowl mod RV-6A
It took most of the day to get the cowl back on the airplane and get the Electronics International C4 cylinder head temperature system hooked up in the cockpit. The C4 system from EI has four power system wires for power and ground but one of the power wires is for back lighting so I connected downstream of the Nav light switch/CB. The display is very clearly visible in daylight and the backlighting should not be on all the time - I guess they get around telling one that by stating "If you are not an FAA Certified Aircraft Mechanic familiar with the issues of installing aircraft EGT, CHT, Carb Temp and/or OAT instruments. Do Not attempt to install this instrument." Anyway, I caught it and mine is hooked up correctly but I'd be willing to bet that there are systems installed with the backlighting on all the time because of the casual treatment in the installation instructions.

At 17:30 Central Daylight Time Barry West and I were ready to repeat our test with the extra baffling installed to isolate the area under the engine from the area above that is outboard of the plenum and its adjoined area behind the new baffling (between it and the firewall). So in this latest configuration there are three areas inside the cowling 1) the standard plenum, 2) the area under the engine leading to the outlet, and 3) the remaining limited dynamic area with a small vent at the top of the trailing end of the cowl outlet for the oil cooler and the heat dump when the cockpit heater is not in use.

Before start-up the cylinders all read 88F or 89F - ambient temperature. We climbed to 6,000 ft to get the temperature at that altitude in accordance with the U. S. Air Race Inc. handicap procedure included earlier. As we passed 4,000 ft in the climb the hottest cylinder was 388F the cylinder is not recorded on the data sheet. At 5000 ft. the cylinder temperatures were: #1 = 290F, #2 = 367F, #3 = 384F and #4 = 373F. At 6,000 ft the outside air temperature was 22C so based on the U.S. Air Race, Inc. handicap procedure we subtracted 3C and determined that we had to descend to 4,100ft for our 6,000ft density altitude speed check. The winds were light and out of the east (ATIS was broadcasting 4 at 070 on the surface).

Our first run was on a heading of 360 degrees (full throttle, max RPM, leaned for max power, trimmed for hands off level flight, etc. before engaging the autopilot to hold altitude and heading. The procedure requires 5 consecutive readings 20 seconds apart with a speed variation of no more than one knot. The second run was at a heading of 120 degrees and the final run was on a heading of 240 degrees. The speeds were:

360 = 173, 173, 174, 174, 174
120 = 171, 172, 172, 172, 171
240 = 177, 178, 178, 178, 177
The cylinder head temperatures at the end of the run were #1 = 322, #2 = 369, #3 = 375 and #4 = 343. The Manifold Pressure during the runs was 26" the oil temp was ~200, the oil pressure was over 75 PSI.

The average speed was 174.3 kts or 200.58 MPH. This is approximately 6 kts faster that the test preceding the latest mod (the three zone cowl) and 4 kts faster than the unmodified configuration. Based on the cylinder head temperatures it appears to me that there is some head room for cutting down the outlet cross section and gaining some more speed. I'm running out of time before the race and at this very moment I'm a little timid about pushing it further - we shall see. The outcome at this point in my opinion is that the modifications are worth the trouble if you are willing to press on beyond the initial two zone cowl configuration.

Bob Axsom
Race 71

Bob Axsom - jeaninebob 'at' cox.net

7/18/06
This is what I ended up with.

This is probably going to take a long time to complete but I am starting to work on another baffle from the front of the engine down and back under the exhaust pipes and above the air box.  The notches at the bottom of the first pattern piece are left over from something else and are not a real part of the pattern.  The baffle is going to have to be three pieces wide because of the alternator and starter.  It looks interesting.  the attributes are going to be reduction of volume, director for smoothly turning the air and reduction of cross sectional area on the way to the outlet.

8/13/06
I had time to get back to the hangar today and continue feeling my way through the pattern making for a new baffle in the lower cowl forward and below the cylinders. This is my most dreaded part of the task. It seems like you are playing with something meaningless. When I start cutting metal it is all worthwhile. Hopefully I will do just that tomorrow.


8/14/06 I removed the pattern today, subdivided it and cut out the first two panels. I have to slow down now and think through the best way to mount everything. When I held the two panels in place they look very right. I have high hopes for this mod.

I spent 4 hours on the new baffling today. The right side consists of two pieces and I made the small forward piece today. This was a better day. The initial crude work is always a struggle but it is essential. Today the work is the refined fitting around items to final tolerances and working with the installation attach points. Part of that last item is figuring out how and where to attach the baffle. When I look at the contours etc. I can't help but like what I'm seeing. Heat could be a problem but I think I'm careful enough to avoid conditions associated with conflicting hardware during operations. I'll get some more photos tomorrow.

8/16/06  I worked 5 hours on the right side of the new forward lower cowl baffling today. Attached are three photos that show the current status. The upper part is complete and everything looks fine. I made the initial two clips out of .016 2024-T3 for the flexibility and the tight bend radius but it was too thin for the platenut rivets. I re-did them with .025 stock and they are good. I tried to keep the inner surface of the baffle smooth but there are just going to have to be some bumps in there. The next baffle support was going to be a one inch wide .025 aluminum strap about 10 inches long wrapped around the 1.75" exhaust pipe (crossover) in one spiral with platenuts attached to the outstretched ends but I was unable to come up with a way to drill the holes in the baffle and locate the platenuts in such a way the strap would be drawn tight and secure when screws were inserted. After thinking about it tonight I came up with the idea of riveting the spiraled "strap clamp" to a rectangular plate that will stabilize that part of the installation and give me a big target for the drilling operation. Once I have the properly aligned holes in the plate and the baffle I will install the platenuts on the plate and secure the baffle to the plate and thus the spiral clamp and the crossover pipe that way. I have got to think a little more about the alternator and the carb heat pipe but I think I have worked out how to deal with them. I have to take care of some home chores tomorrow and probably will not get back to this mod development until Thursday.

8/17/06  I looked over the photos I took yesterday and they do show some different views of the baffle that I worked on and perhaps give a more comprehensive idea of the final product. Things like the contour of the baffle that will be extended across the entire cowl when this work is done. Intuitively it will deflect the air early and provide a more gently transition from down to aft and it will prevent any forward displacement of the air (not optimum by any means because the baffle starts several inches forward of the most forward opening in the lower plenum baffle openings - below cylinders #1 and #2. If you look toward the rear end you can see that the exit path cross section is reduced - a lot. That cross section reduction could cause and increase in exit air velocity and it should reduce the air mass flowing in the system. I would expect the CHTs to rise but they are fairly cool now even in high ambient temperature conditions. Information received from people that have done similar outlet path cross section restrictions with different methods indicate that this is an important factor in the reduction of cooling drag. There is also a significant reduction in the volume of the flow chamber below the cylinders - which I suspect will also reduce the air mass flowing through the system. At this point I envision the trailing edge of this baffle being approximately 0.5" forward of the two fwd NLG support braces and slightly above the inverted pyramid weldment (which I have re-capped as shown in the airbox section of this mod).

The alternator has many air vents that must not be obstructed but I have debated about sealing the baffle around it using the riveted rubber approach - it's a circulation vs leakage (drag) argument. I think I have rationalized sealing it and using the battery to get home if it can't survive. The carb heat pipe is another pain. I have never used it and with a similar induction system I only needed it twice that I can recall in 22 years but most of it was in California - not a strong argument. Right now I plan to leave it in and seal around it without blocking the rear vent. The hot air coming out here will go into the cooling airflow exit path and that coming out the bottom onto the airbox will have an exit path under the baffle to the cowl air outlet. I my close that off later to eliminate back flow into this area if the current mod works - I could end up pulling the pipe and depending of the airbox by-pass door to compensate for induction blockage and the engine design to compensate for carb ice. Still thinking about this.

I'm taking Jeanine to see The World Trade Center movie today (us retirees do these things) and then the weekly grocery shopping is on the agenda. Bottom line - no mod work today.

8/18/06  I spent 7 hours out in the heat today working on the new right front baffle experiment in the lower cowl. I made good progress even though it probably is not apparent. I had to take some things apart and do them over to make them practical but when you know it is getting better it seems like constant progress. I implemented the spiral clamp that I thought would solve my mid baffle support problem but decided to abandon it in favor of a simple strap clamp which seems like a much better solution. The spiral strap and plate would work but it added complexity and was simply not as good a solution as the simple strap clamp approach.

Once I worked through the clamp decision I re-thought my approach to interfacing the two baffle panels on the right side. I had planned to keep the air facing side of the panel as smooth as possible with flat head screws installed from the inside with the two panels partially installed - ugly and awkward. The decision to eat some small drag in favor of practicality necessitated drilling out a lot of rivets, making a new splice plate, etc. When the two major panels were put back together their fit was no good and I had to trim a long skinny triangular piece off of the aft panel to gat a perfect butt fit of the panels in front of the splice plate. After that was done I did all of the #8 screw dimpling, platenut installations. etc and was able to install the panels with flight hardware for the first time. The right panel installation is complete except for two things as far as it alone is concerned. I have to stabilize and support the trailing edge in the right position and I have to verify a good fit with ~3/8" set back from the cowl side and the rear side baffle from the original lower cowl baffle mod, then add the sealing strip along the outboard edge. I am going to add the trailing edge support tomorrow but I'm going to delay the seal strip installation until the left side baffle is installed and the two sides are interconnected. There is just too much possibility for change driven by the left side development to do the sealing now. There will be a need to add splice plates across the left to right side panels to produce a "air tight" joint And I need the rivet rubber seals around the alternator, the carb heat tube and the starter. I should be ready to test fly it next week sometime.

I am starting to feel a little pressure since I promised Jeanine I would fly her to Memphis the following week. I'm not sure if we should try to get a tour of the FedEx super hub or Graceland - oh never mind, Graceland it is. Unbeknownst to either of us we both saw Elvis at the same concert at Kiel Auditorium in St. Louis in January, 1956 - it was a big deal back then. Note: I went to the concert because Hank Snow and Web Pierce were performing there.

8/19
A little over five hours in the more than 100 degree F temperature was not stimulating but I did get the left side up to the same level of completion as the right today. Tomorrow I will work on a long "T" support for the trailing edge of the baffle halves. I'm truncating the baffle about 1/2" in front of the NLG structure currently but I may change that later depending on the outcome of this experiment.



8/20
I spent 5.6 hours at the hangar today which was another hot one here in Northwest Arkansas. The work today concentrated on the center split between the left and right side baffles. The most difficult was the forward most section because it also had to provide closure for the space between the alternator, the starter and the engine. There are three sections that needed to be closed, the area forward of the carb heat tube, the area between the carb and the carb heat tube and the area aft of the carb. The latter two are simple rectangles but they are surprisingly wide for practical reasons I wanted to install them without requiring removal of the airbox for drilling access. Because of the size of the two aft gap cover plates I am fairly sure I will cut out the excess overlapping part of the main panels. As a side benefit, some access will be available with the main panels in place. All of the installation holes were drilled, deburred dimpled or countersunk for the platenuts and attach points. Only one platenut was installed before I ran out of time.

I still have to rivet the rubber around the baffle penetrations by the Alternator, carb heat tube and carburetor and around the outer edge of the baffle to provide a seal at the cowl and rear side baffle interface (not sure about that last one yet). I am still going over the handling of the trailing edge of the baffle. Right now it is free standing and truncated just forward of the NLG structure. I may test fly it this way but I am concerned about lift motion (it is restrained by the cowl to prevent downward motion). Another concern involves the air flow - if I rounded it down to the bottom of the cowl outlet coincident with the trailing edge some venturi function should be gained and the back flow under the baffle would be prevented.

Work continues tomorrow, I am expecting a test flight early next week.

...more

I spent another 5.6 hours at the hangar today finalizing the metal work on this latest mod. I need to put one more strap clamp on the right side to support the trailing edge. The left side is solid and the right side will be after this last clamp is installed. It put the whole thing together and found that it fits nicely in the cowl as is but I have to add the sealing strips and when it is on the plane things change slightly. I'll finish the last clamp and start installing the rubber tomorrow.

Note: You may notice that the center closure strip in the uninstalled photograph is about 12 inches wide but in the installed photograph it is gone. Well there is a reason for that. I verified that I could get a stubby screwdriver between the airbox and the baffle in this area for service. With everything stabilized by the 12" closure strip (and the other two), I was able to make the smaller strip installation off the airplane where I had good access to both sides of the baffle. The ~3.5" strip is installed in the photo showing the baffle installed on the airplane. Those platenuts are valuable so the next time the baffle is off, they are coming off as well.

8/21
We had some serious weather come through this afternoon and we lost power at the airport. Luckily I saw it coming and closed the hangar door before power was lost. I made a checklist of the things I still have to do when I arrived at the airport:

1 - Install the strap clamp around the heat muff
2 - Install interlocking clips on the rear of baffle and the gap closure strip
3 - Install hole reduction plate on right side of carburetor
4 - Cut hole larger on left side of carburetor
5 - Seal area around carb heat tube
6 - Seal area around alternator
7 - Seal area around starter
8 - Seal area around carburetor
9 - Seal edges of new baffle
10 - Make and install something to eliminate baffle sag at rear center
11 - Remove unused platenutes and close holes.

I completed 1, 4 and 11 today.

8/22
I reported faithfully to the hangar at 10:31 this morning fully prepared to finish the latest lower cowl baffle addition. Well, it didn't happen. I worked items 3, 2 and 10 from my completion checklist in that order. Things are nearing completion and everything not done with care will impact something else. I drilled out rivets and re-did the installation of the filler plate in the baffle by the carburetor more than once before I got it right. The clips at the rear of the baffle that secure the gap closure plate where I can't get to it with a screw driver look great but they don't work properly so I'm going to have to revise that tomorrow. The sag limiter for the rear of the baffle came out OK and I did photograph it.

Tomorrow I will try to get the clips fixed and start working on the rubber seals. I am not at all sure how this is going to work out so the rubber seals are going to be riveted to the baffle rather than using RTV (DOW 736) for the seals. That will keep it looking good if I have to pull out this latest mod - even though I don't think that is going to happen.

8/23
I spent over 5 hours out at the hangar today trying to get everything to fit nice and clean (no force fit). I ended up redoing the rear end closure sheetmetal. I seems very good now but that's all I got done.

8/24  All of the basic metal work is done but I don't think it is going to fit in the cowl as is. Jeanine is going out to the hangar with me after dinner and we are going to attempt to put the cowl on. I do not have the rubber on it yet but if I do that before I trial fit the cowl and then I find that it doesn't fit ... well you know - drill out all of those rivets, etc. etc. The trial fit will just take a few minutes then when I verify it doesn't go on I have to work out a method to determine where the interference is and how to fix it. I expect to have to put in some kind of width reduction where the baffle makes the bend to go back to the outlet. I'm visualizing something like the profile of an hour glass.

8/25
The baffle hits where I thought it would down at the bottom where the baffle turns back and the cowl turns in but I will not have to trim much off. I'll probably take it back an inch at the point of greatest interference then taper back to the original edge for a start. If I'm careful it should come out OK.

8/26  I think I am ready to start putting on the rubber seals. I spent another 5.6 hour day at the hangar working with the interference between the baffle and the cowl where the forward lower cowl baffle turns back toward the cowl outlet. I a laser on a tripod a template of the inside of the cowl and several other approaches but it ultimately came down to sketch what I thought would work and cut and try. I think the fourth iteration got it right. I will try the cowl on again in the morning to verify that then start working on the baffle border seal strips. First I will use the fiber lined high temperature rubber strip (3" x 9') from Aircraft Spruce. I will have to overlap strips to cover the curves but if it doesn't come out right I'll buy the high temperature sheet rubber and do it over. More tomorrow.

Friday 8-25-06
I think I am ready to start putting on the rubber seals. I spent another 5.6 hour day at the hangar working with the interference between the baffle and the cowl where the forward lower cowl baffle turns back toward the cowl outlet. I used a laser on a tripod a template of the inside of the cowl and several other approaches but it ultimately came down to sketch what I thought would work and cut and try. I think the fourth iteration got it right. I will try the cowl on again in the morning to verify that then start working on the baffle border seal strips. First I will use the fiber lined high temperature rubber strip (3" x 9') from Aircraft Spruce. I will have to overlap strips to cover the curves but if it doesn't come out right I'll buy the high temperature sheet rubber and do it over. More tomorrow.

Saturday 8-26-07
Well the verification didn't go well in my 5.5 hours of sweat and strain today. I installed the lower cowl and there was interference in three places involving both sides. Fortunately, I learned that if I do not insert the horizontal baffle seals (the ones on the baffle by the valve covers) down inside the cowl and use a flashlight I can see how the the new baffle fits. The first time I had one place on the right side and two on the left side that were hitting the cowl even though I could get it on. Charlie Heathco happened to be out at the airport and the extra set of hands were helpful because of the interference issues, especially with regard to aligning the induction port with the airbox. I trimmed 1/2" off of the parts of the baffle that were hitting the cowl and tried again. This time it was hitting at one place on the left side. I trimmed off 1/2" in one of the same areas I had trimmed before and reinstalled the cowl. This time it was hitting a little farther back on the left side. I trinned off 1/2" more in the same spot and reinstalled the cowl. It was still hitting in the same spot so I cut off another 3/4" in a fairly long cut going most of the way to the rear of the left side of the baffle and reinstalled the cowl. Finally, it wasn't actually hitting but it was very close in two places on the left side and one place on the right side. Off came the cowl again and 1/4" was taken off of the right side in one location and 3/8" were taken off of the two left side locations. I still didn't want to start installing the rubber without a visual check so the cowl went back on again. There were two non-critical items that I wasn't satisfied with so when I took the cowl off and trimmed them I did not reinstall the cowl. So a major milestone in the development was reached, the installed baffle fits inside the cowl.

I removed all of the baffle components and started making the metal strips that will be riveted to the outboard edge of the two baffle sides, sandwiching the rubber seals in position.
Sunday 8-27-06

6.5 hours at the hangar today but all I have to show for it sofar is the rubber on the outboard edge of the left half of the baffle. I am going back out there after dinner to work on it some more. I expect to get the outboard rubber on the other hafl of the baffle and that's about it. I plan on flying us to Memphis Tuesday so the pressure is really on.It is doubtful that I will get the very important (to me at least) sealing done around the baffle penetrations (carb, heat tube, starter and alternator). It doesn't sound like much but working out the implementation details takes a lot of time.
 

8-27-07
I installed the rubber on the right side of the new baffling so all that is left is sealing the four baffle penetrations.  I will attempt to reinstall the baffling and the cowl tomorrow morning then test fly it.  Because of the gaping holes around the carburetor, starter, alternator and carb heat pipe I don't expect to see any improvement but it will be another data point.  It will show how much effect the sealing has later on of course but the main thing I want to determine is the safety of the airplane configuration for a trip.  If all goes well I will fly to Memphis this way a little later in the week.  If not I will pull the baffle and make the trip in the old configuration.

8-28-07
I put everything together today and test flew the plane without the baffle penetrations sealed. It was stormy, cloudy and I had to deal with a Class C airspace (avoid it in other words) but I did complete a test flight. I could not maintain the variability standard required by the US Air Race handicap standard and I was solo but some info is better than nothing.

Outside air temperature at 6,000 ft was 19C so the test was conducted at 4,400 ft. The speeds in the three directions were as follows:

000 - 170, 170, 171, 171, 172
240 - 168, 168, 169, 169, 167
120 - 187, 188, 187, 186, 185

Average = 175.2 kts

CHTs after 45 minutes; 1 = 318, 2 = 353, 3 = 358, 4 = 329
Oil Temp = 190-200
Oil Pressure = well over 75 psi
RPM = 2690
EGT = Cylinder 4 only = Well rich of peak (~120 ROP)

I hope to pick up a little when the holes are plugged but the only way I would recommend something like this is if the person were really into racing and can't afford the big engine and fastest prop. It's hard, it's frustrating, the sacrifices are great and the rewards are pitifully small. It does help develop humility.

8-29-07
"When I fill in the gaps created by the baffle penetrations I suspect the speed will increase a bit."

9/10/-6   
  "Following leads given me by Alan Adamson and John Huft I have a 12"x60"x1/4" felt and red (with mat) rubber to seal the holes in the last baffle I added. I ordered a sheet of 0.025"x48"x24" 2024T3 aluminum from Aircraft Spruce and some more dimpled platenuts from Van's as well. When I work on the mods I build up a lot of home chore debt and I am working through that currently. For the last couple of days I have been spraying a satin white enamel paint on the 40+ year old bedroom furniture from our spare bedroom. Tomorrow we will mask and hand paint the gold trim. By mid-week I should get back to the airplane mods. If all goes according to plan I will test fly it to get a speed check after the holes are closed. Then I have an idea for bringing the truncated end of the latest baffle down to the bottom of the cowl air outlet. As with everything else this has to be done with care. Isolation from rigid contact with the engine and everything attached to it are required for this baffle extension. I'm thinking that I will mount it with metal clamps to the nose gear support structure (engine mount) and use red rubber or felt on the edges to interface with the cowl and the truncated end of the latest baffle. After that is done I will run another speed test to measure the effect.
  We are leaving on a trip to Europe on the 22nd of September so things will probably not flow smoothly to completion."

9/21/06
  "I removed the cowl today and inspected the latest baffle modification. I found one location below cylinder #2 where the metal strip used to secure the sealing rubber was contacting the thermal blanket inside the cowl. I added a 3/32" thick red silicone rubber with mat strip to the metal with red Dow 736 RTV to isolate the metal from direct contact with the blanket. I also found the aft clamp/strap on the left side was broken. This may be related to the metal to blanket contact problem. A photograph of the broken clamp is attached. The metal is 0.025" 2024T3 aluminum and it was under tension and torsion and in direct contact with the exhaust pipe. I plan to go to 0.032" aluminum, form a less stressful fitting clamp and line the inside with thermal blanket."

9/22/06   Attached are four photographs I took today.  The first one shows the rubber abrasion protection strip I added to the latest baffle yesterday.  The forward shorter strip is duplicated by one on the right side of the baffle and they were installed before the first flight with this baffle.  The second photo shows the area where the metal was rubbing on the thermal blanket inside the cowl.  The third photo show how I have use a section of a thick stainless steel butter knife blade to protect the cowl from even the small sized alternator pulley. The fourth photo shows how I use file folders to protect the paint on the cowl, spinner and nose landing gear strut during cowl installation and removal.

10/2/06   Yesterday I was caught up enough on my home chores to get back to the hangar. I started replacing the failed vacuum pump but could not gain access to the bottom inboard nut without removing the oil filter. I decided with 25 hours on this oil I would change it and install a new filter after the vacuum pump is replaced. To gain access to the oil drain I had to remove the main lower left forward baffle. The baffle came off with no difficulty and the oil was drained in the normal manner. I consider this a engine serviceability test of the new baffle and it passed. This also provided an opportunity to examine carefully the condition of the baffle itself after 6.1 hours of both local and cross country flying.

Four photos labeled "Outside.baffle", "inside.baffle", "before.trim" and After.trim are attached. First it is important to note that the aluminum strip used to attach the sealing rubber is set back quite a bit from the baffle edge. The idea was to allow the edge of the baffle to hold the rubber seal against the cowl without interference from the securing strip and avoid metal contact with the cowl. I rubber strip was bonded to the securing strip to assure a positive separation of the metal from the cowl.
As you know from my earlier comments and photos, I found evidence of contact with the cowl anyway and bonded on a second rubber strip that has not been used in operation.

Looking at the Outside.baffle photo you can see the original and the added pieces of rubber bonder to the seal attachment strip. Before removal of the baffle for the oil change I pushed on the edge of the baffle where the original rubber "bumper" is attached, trying to form it away from its current mild cowl interference shape. It moves with great pressure but springs back upon release. Looking at the inside.baffle photo you can see two black lines where the crossover pipe makes the turn aft and a little further back in the straight part of the run. You can also see if you look closely permanent little bends perpendicular to the edge of the baffle where
the greatest pressure exists in the operational state. Today I used a
nibbler, file and deburring tool to reduce the width of the baffle back 3/8"
from the original shape so that the outer edge of the baffle and the seal attachment strip are no longer staggered at this point. The before.trim and after.trim photos show this modification.

Studying the situation I recognize that this situation is not acceptable.
What needs to be done is to remove the two centermost seal strips and all of their attachment hardware and reinstall them on the inside of the baffle (top side is another way of thinking about it) and bond a rubber strip directly to the baffle in this area. This will reduce the material height in this tight area between the cowl and the baffle from 0.2125" down to 0.09375". This reduces the interference further and provides the same chafing protection. If the problem is not eliminated by this mod I will take it apart, trim the baffle edge back farther, reassemble and inspect until I am comfortable with the fit.

The exhaust pipe marks on the inside of the baffle need to be dealt with I think and there are a couple of ways that seem reasonable. First on the curved line I think I will add a piece of thermal blanket as a buffer and not change the clamping arrangement. In the aft clamping location I hope to kill two birds with one stone by changing the clamp configuration. I will make a strap that wraps around the pipe and attaches to the two existing strap clamp mounting holes. This will remove the tension and torsion loads that I think caused the clamp to fail and at the same time it will provide a standoff between the pipe and the baffle in this location.
 
Bob Axsom

10/11/06   
The attached photos show the Vacuum Pump shroud bought from Aircraft Spruce in hopes of improving the Vacuum Pump life. A new 1.125" hole has been added to the rear right baffle behind cylinder #3. Tomorrow I should be able to get back on the cowl mod.

I test ran the engine to check for oil leaks after the oil and filter change and I checked the new vacuum pump function. Everything was fine so I safety wired the plug and filter and got back to the new baffle mod. I have a new 1.125" dia. hole in the baffling behind cylinder #3 for the cooling air to the vacuum pump shroud. That will have some effect on aircraft performance but I'm not going to test it as an isolated change so I will never know how much or in which direction. I drilled and punched out all of the rivets in the baffle assembly securing the two center most sealing strips. Then I used the flat dies in the squeezed to remove the dimples in the rubber attachment strip holes and then used the dimple dies to dimple the baffle attachment holes for the rubber. The mfg. flat head will now be on the baffle side and the shop head will be on the attachment strip side.
Tomorrow is a full home chore day so no more progress will be made until Wednesday. I'm still not back in peak form following my Russian experience so my efficiency suffers.

10/12/06   
I completed the repositioning of the center most edge seals of the left lower forward baffle to the inside for cowl clearance. The baffle edge was cut back 3/8" in this area. I added a thermal blanket to the inside of the baffle where the turning crossover exhaust pipe is forced hard against the baffle. Two rubber strips were bonded to the edge of the baffle on the outside as a buffer against hard contact with the cowl. A wrap around clamp was made up to replace the broken straight clamp at the rear part of the left exhaust pipe. Platenuts still have to be installed on this clamp. The baffle was reinstalled and after the clamp is finished I will be ready to work on the hole closures.

Couple days later....
Not much progress today. I installed the platenuts on the wrap clamp that I made yesterday but I was not satisfied with the results. I made a new straight strap clap from some 0.032 6061 T? given to me by my friend Barry West. That looks good but I lined the mounting flanges and the inside of the clamp with thermal blanket and I will add a piece of thermal blanket to the baffle under the exhaust pipe. More tomorrow. Hope all is well at LOE.

10/15/06   
I put thermal blanket pads on the new baffles everywhere the exhaust pipes were robbing on them (four pads). I also padded the left rear clamp. I have riveted a rubber seal around the starter and solenoid. Across the top of the left baffle I have started to rivet a felt pad/seal. This is breaking the ice on the hole closure effort. This is very tedious work that evolves as I go through the seal development process. This is going to take a while but now that I'm rolling, I feel pretty good about it. Photos attached.

Bob Axsom

10/16/06  
 completed the front end of the left side baffle as far as the felt along the leading edge is concerned. Then I started working on the hole around the Alternator. I have the underside rubber installed and the attachment strip for the top rubber fabricated. We have to do quite a bit of work around the house for before Jeanine's brother comes to visit this Friday so things will stall for a little while. Hopefully, I will get one more day in sometime this week. The hole closure work looks good to me. You never know exactly how to deal with these things and there are many opportunities for mistakes. Sometimes as you work the problem good solutions reveal
themselves that were not even considered in the beginning. This is one of
those times.

10/19
I cut a rubber seal for the baffle segment that extends over the alternator. The holes were drilled, deburred and dimpled in the aluminum and punched in the rubber. Everything is clecoed together but I ran out of time before I could rivet everything together. That is strictly routine from this point though. I will probably have to trim the rubber to get a good seal without too much excess. The last thing I have to do up at the front end is fit rubber to the fairly small center closure plate.

One sealing method does not work for every thing and I will be using RVT (DOW 736) in some little nooks and crannies between the starter and solenoid, etc.

For the hole around the carb heat source tube I'm thinking that I will use rubber riveted to the baffle halves one on the front and one on the rear so they can overlap and provide a good seal.

The last opening for the current baffle configuration is around the carburetor and it is a big one. I haven't finalized this yet but I am mentally evaluating removing the airbox and mounting plate, making a rectangular plate with a 360 degree one piece seal cut from a 36" wide sheet of red baffle material with mat, riveting it together and reconfiguring the back end so that the rear baffle assembly platenuts are on this new closure plate instead of the current locations on the baffle halves. Some rivets will have to be turned around to get a smooth interface but at this point it looks like a good configuration.

10/24
I was able to spend 7.7 hours at the hangar today. This work is very tedious. the rubber seals exert pressure and things get loaded differently than they were before the seals are added. It requires a lot of thought and work to retain the seals and eliminate the undesirable conditions. Any way the upper and lower seals around the alternator are complete. In addition to completing the front end seals I got most of one half of the seal around the carb heat source tube riveted. If I can get this work done and tested by Saturday, I will fly it to California Sunday. If not I will remove the new forward baffling for this trip and continue after we get back.

10/25-26/06
I was able to spend a little time at the hangar this morning and I riveted the rubber on for sealing one side of the carb hole in the new baffle. I am going to glue in thermal blanket pads on all the clamps tonight. Tomorrow I should finish up.

(more) I brought the clamps and thermal blanket material home with me last night and glued the thermal blanket material to the inside and the mounting feet of the clamps. Today I am going out to the hangar again and I will complete the last of this baffle installation.

(more) I spent over 5 hours at the airport today I finished the last piece of rubber on the main baffles of this latest mod and started fitting everything back together. The big task is cutting rubber out of the rubber webs for the carb heat source tube and the carburetor itself to allow the baffles to come together in the middle like they did before the rubber was added while maintaining the seal. It is very tricky and perhaps critical to any speed improvement. With the front and middle joined with the splice plates I still have 1/4" separation at the rear end. I should finish and test fly it tomorrow.
 

10/28 and Performance Changes
Well Doug, it is time to judge the effectiveness of the latest modification in the lower cowl. Before I started these mods, the baseline speed established using the U. S. Air Race Handicap procedure, was 170.67 kts.

I've attached three photos to summarize the current configuration. Dvc00015 and Dvc00013 show the free standing configuration. Dvc00001 shows the seal of the top of the lower cooling air flow path to the installed lower cowl.  "rightcowl" is a photo taken of the right side of the cowl after I reassembled everything for flight test. You recall that my first mod channeled the cooling air that had passed the engine to the cowl outlet much like a funnel, isolating it from the firewall and all of the hardware between the rear of the plenum and the firewall. Flight testing this mod yielded a decrease in speed to 168.6 kts. Then I added the horizontal seals between the cylinders and the lower cowl which separated the cowl cavity into three chambers. Flight testing this mod showed an increase in speed to 174.27 kts. Then I added a cover to the recess in the top of the airbox and lost 1 kt. Then on July 3, 2006 I started working on this latest mod to provide an earlier turning surface than the cowl for starting the air flowing rearward, creating a fourth cowl chamber, reducing the post cylinder chamber volume and providing a flow path isolated from the hardware in the bottom of the cowl. I completed the basic sheet metal and rubber of the initial configuration and made a test flight on August 28, 2006 and the winds were too rough to allow me to conform to the consistency standards of the U. S. Air Race Handicap procedure but I reported that the data seem to indicate a speed increase to 175.2 kts. I flew the plane to Memphis and back for an operational test (and to have a little fun) and found several unacceptable discrepancies requiring attention. Many configuration changes were made to correct problems and all of the baffle penetrations were sealed. Today (10-28-06) was the first flight after completing that work.

Reinstalling the cowl did not present any new problems. I noted an interesting phenomenon while I was working under the plane - when I was laying directly behind the cowl outlet I could feel a soft breeze on my face that went away when I moved to the side and it returned again when I got behind the cowl outlet again. It was positive and consistent. The was with the hangar door closed. I had to think venturi effect but I do not know where the initial force to move the air came from. No explanation but it is not something to dismiss.

When I started the engine all functions were normal and the engine sounded very good, both solid and robust. Climb out was over 1,000 ft/min up to 6,000ft pressure altitude (29.92 altimeter setting). At this altitude the OAT was 12C. So I descended to 5,100 ft per procedure and flew the first leg on a magnetic heading of 360 (000)deg. Once again I was unable to maintain the consistency standards of the U. S Air Race Handicap procedure (no more than one kt. variation for five consecutive readings 20 seconds apart. I gave up and turned to 120 deg. for the second leg but had no luck there either. Finally I turned to 240 degrees for the last leg just to complete the required three leg run. There were winds and vertical air movement over the Ozarks east of Fayetteville, Arkansas where I conducted the flight test so the numbers are not valid but here is what I got:

360(000) deg. = 154, 152, 152, 157, 158, 156, 156, 155, 154, 154, 158, 154 = 155kt average 120 deg. = 196, 191, 192, 194, 190, 192, 193, 196, 192, 192, 191, 193, 192, 193, 194, 193 = 192.75kt average 240 deg. = 166, 164, 164, 163, 166, 166, 166 = 165 kt average Overall average for the three headings = 170.916 kts.

Very disappointing but this kind of result is not as baffling (pun not intended) as it would have been a year ago. Here are some additional data collected during the flight:

CHT at the end of the test (1 through 4 respectively) = 315 367, 371, 333 Oil temp = 170 Oil pressure = 80 MP = 25 RPM = 2690

As was pointed out to me after the first setback, if you clean up the air flow path, more air may flow through the system resulting in better cooling but increased cooling drag. I think I am going to have to work on airflow restriction - either at the inlet or outlet. At this time I am leaning toward the outlet because it is easier to implement.

Bob Axsom"

130.4 man hours of work.

Final test on the latest cooling air configuration
...from Bob Axsom

I got up early (for a retired guy) this morning to test the top speed of my airplane in its current cooling mod state (the added baffle below and in front of the engine to turn the air toward the cowl outlet). Instead of one run I made two, one at 2450 RPM and one at 2700 PPM. The surface winds were calm but there was some smooth air flow aloft. I am going to modify further and will let you know the results.

[U]2700 RPM[/U] (Heading - Speed in Kts) 360 deg. - 177, 177, 177, 177, 178 120 deg. - 172, 172, 172, 172, 173 240 deg. - 165, 166, 165, 165, 166

Average Speed = 171.6

MAP = 25.5
Oil Temp = 200
Oil Pressure = 80
Mixture = 100 ROP
CHT (1,2,3,4) = 324, 379, 377, 343

The OAT at 6,000 ft Pressure Altitude was 21C so the test was flown at 4,200 ft pressure altitude (which was approximately 200 ft lower than the corrected altimeter MSL reading). The OAT at 4,400ft was 26C.

1/8/07
...I clumped the last several days of Bob's entries here in one day.

5.5 hours at the hangar today produced the two major pieces of the baffle extension.  There will be some trimming before they are finalized but they are looking OK. When the process gets this far along it becomes its own motivation source. I need to take off all of the lower forward baffling and complete the platenut/screw interface. Then the baffling needs to be reinstalled along with the lower cowl to establish a proper gap by trimming the metal from the new baffle parts. Once I have a uniform gap, I will install rubber and RTV to seal and vibration isolate everything. I will be back out there tomorrow.

I expect to pull the two lower forward baffles and mate the extensions to them with screws and platenuts.

They are marked so I can properly align them for assembly in their in-aircraft configuration. I will progressively drill the screw holes, clecoing them as I go then install the platenuts on the top side of the forward baffles. This will place them in the airflow path but that is the only easy way to make the installation serviceable for oil changes etc.

There are so many obstructions in there already, especially the crossover exhaust pipes, carburetor and nosegear support structure, that these are relatively invisible. Then I will reinstall them in the airplane and make clamps to support the new extension baffles from the exhaust pipes. These will be installed several inches forward of the trailing edge of the baffles. Then reinstall the lower cowl (UGH!), curl the trailing edge of the baffle extensions down to the bottom of the cowl outlet opening and mark it for final trimming. The lower cowl will then be removed and the trimming completed. Then I will rivet matted hi-temp rubber around the edges at the sides, the trailing edge and the nose gear support structure holes.

Somewhere in the process I have to fabricate and install a diagonal rib to stiffen the outboard corners of the trailing edge. I may also may add a straight rib or two. These ribs will be under the baffle and out of the airflow path. I plan to anchor the inboard trailing edge by installing a screw through the inboard rubber seals and the center cowl support structure in the center of the outlet. After that I'll be ready to reassemble everything and start testing.

As you can see there is a lot to do yet.

Only 2.1 hours on the mod today. I removed the two baffles that were final installed at the end of October. Those were test flown on the 28th of October and the total flight time on them in that configuration is 32.2 hours including one trip to Southern California. I mention this because this is the first time I have had the baffles off since late October 2006.

You may recall that in the process of developing that baffle mod one strap clamp broke and there was chafing of the baffle surface where it contacted the exhaust pipes. My experience at that time with the replacement of the broken clamp with a similar one lines with the expensive aluminized thermal barrier cloth from Aircraft Spruce was very good so I added it to the other three strap clamp inner surfaces and the surface of the baffles at the clamp locations. Well that approach was not adequate. The aluminization delaminated from the cloth, the cloth scorched and the cloth wore though where there was hard contact. That is not a good use of the product. See attachments Dvc00002,jpg and Dvc00003.jpg. As a temporary fix I coated the affected areas with Dow 736 (see attachment Dvc00004.jpg). I am thinking about the problem and one possible long term fix is to buy some of the exhaust pipe wrap from Aircraft Spruce and apply it only to the exhaust pipes in the four problem areas and bond some hi-temp rubber sheet to the baffles in those areas before clamping it back up in the hopefully final configuration. This fix would address the chafing and thermal problems.

OK back to the mod with the baffles removed the extensions were placed on them to verify the alignment. Everything looks good (see attachment Baffles.and.extension.jpg). Tomorrow I will add the plate nuts and hopefully, install all four baffle components on the airplane.

I came up with a better idea for the heat & chaffing problem. I will add a bottom element to the four clamps then add metal standoffs between the clamps and baffle. This will allow air to circulate under the exhaust pipes at the mounting points and separate the pipes from direct contact with the baffles.

I completed the right baffle extension interface today.

I completed the interface of the extensions to the main part of the previously developed lower forward baffle. Two photos are attached. I also made the standoff bases for the four clamp attachments of the existing baffle to the exhaust pipes. These are to deal with the abrasion and thermal problems in that area. I will take everything back to the hangar tomorrow, install the baffles in the airplane and trim for final fit.

Misc shots from the last batch:

01/14/07

I took everything back to the hangar today and installed the lower forward baffles and the new extensions today. I also installed the baffle support clamps on the exhaust pipes with standoffs. The installation went well but the sequential assembly requirement is very inflexible. Some things that I assembled before I found the proper sequence had to be disassembled to allow the process to go to completion. Everything did go together with no interference and stress as far as the baffling was concerned. When I installed the lower cowl I found that there was hard interference between the sides of the new baffle extensions and the cowl transition down into the inlet/airbox/outlet tunnel on the bottom of the cowl and the left baffle extension was 1/2" too long. I measured marked and cut the baffles to provide clearance but I did not have time to reinstall the lower cowl and check the fit. Tomorrow I will continue with the task after we complete our grocery shopping.



First let me say that one of the chores I do not enjoy is installing and removing the cowl. I got to the hangar at 13:39 today and the first thing I had to do was install the baffle extensions that I trimmed and photographed yesterday, and the lower cowl to check for fit. Well it didn't of course, the left part was too wide and the right part was too narrow. I removed the cowl and the two baffle extension panels and went back to work. I removed 3/8" from the outboard edge of the left panel and riveted a 1/2" extension to the outboard edge of the right panel (photo 1). After all of that was done I reinstalled the panels and the lower cowl for a fit check. I still didn't fit right. The right panel was basically ok but the left panel was still interfering with the upper edge of the induction and outlet tunnel where the panel edge turned outboard. I removed the cowl and the panels again for further work. I trimmed the left panel where it was interfering with the cowl and rounded the rear outboard corners of both panels to match the cowl (photo 2). Then I reinstalled the panels and the cowl for the next fit check. The fit appears to be perfect. See attached photos.



No photos yet. I have been working with the next phase of the development of these baffle extensions. I want to hold a downward curving shape at the rear end of the two baffle extensions I am adding. I have thought about a few methods but I have settled on the roughly centered rib for the shape and I think I can use the rubber edge strips to stabilize the alignment in the cowl outlet. The first rib is about half way done. I loaded everything in our '85 Corvette to bring home because my small pickup is in the body shop for repair ... quite a squeeze and I hate to do it but progress must be made. I hope to get quite a bit more done this evening but you just never know what will turn up for me to do on the home front. More later.

After three different approaches I have one rib made. The quality is not good but it will do the job. I have one more to make and I may use another approach. If it is better I may redo this one.


The first photo is rib I finally came up with working most of yesterday afternoon and evening. At 2:30 this morning I got up and went out in the garage to make the second one and try a different design. Simple is sometimes better. The second photo is the latest version. It was so good and so simple that I made another one to replace the one I made last night.  I plan to assemble them onto the baffle extensions later today and start installing the rubber seals.

I have an idea for sealing all of that open area around the nose gear support structure that I think is going to work well. I think I can install a large piece of hi-temp matted rubber at the rear of the existing baffles just forward of the nose gear then pull it through the structure so that it lays flat and seals around the struts. The extreme ends of the rubber will be pulled out and secured to the underside of the baffle extensions using dimple washers, #8 screws and platenuts. That the idea for now anyway, it's funny how these perfect mental images fall apart when you get to look at the real situation.


I went to the airport this morning and checked the baffle extension rib clearance. I have about a one inch wide longitudinal window for the rib and I had to cut off part of the forward (big) end of the ribs to assure clearance of the filter air box . I also realized that I need to mount the sealing edge rubber before I mount the ribs. I have been working on that today. I have the rubber marked and cut for the right baffle extension and the most of the metal backing strips are completed for it as well. We have some nasty weather coming in to our area so I had to spend a few hours doing non-airplane things. They are forecasting bad weather until early next week so test flying is a few days off.

It has been a long day but the right hand baffle extension is complete and ready to install. I will work the left side tomorrow. The first photo shows the parts for attaching the main rubber component. The other two show it in its current form. The rubber will be trimmed as part of the
installation. One installation detail still has to be worked out. I plan on suspending the aft part of the baffles off of the exhaust pipes. I called Larry Vetterman's home today to see if I can but some clamps from him. A lady answered and told me that he was not home and basically I should call back Monday when the office is open ... I guess they want a life. I will do that but I may make up a couple of homemade clamps for a test flight.. I will need spacers to get the baffle trailing edges where I want them. That will also give me a means of adjusting (I'm hesitant to say tuning) the outlet cross section area.

As I said tomorrow will be dedicated to the left side baffle extension.

1/14/07 - 1/24/07

The pictures Bob sent with all this text

I have the "big rubber installed on the left baffle extension and the provisions for mounting the little one are complete. The rib is also ready to be riveted on. I should complete this in the early morning hours. I found a way to compensate for the difference in radii of the baffle and the rubber hold down strips at the rear of the baffle extension assemblies. It worked well on the left baffle so I plan to drill out and redo the strips on the right baffle to eliminate the "kinks" between rivets. Basically I sandwich rubber between the parts and bend them before drilling the holes, clecoing as I go. In the first attempt I bent the parts before I drilled ok but the difference in radius with the rubber between the parts is just to great to work well.

The rubber is complete on the left baffle extension. I'm having a little trouble with the rib installation so I'm calling it a night.

The baffle extensions are done and ready for installation. I decided not to replace the kinked rubber securing strips on the right baffle. I am tired, it is cold outside and there is a light snow falling but I am going to the airport and install the baffles. I called Larry Vetterman this morning and ordered four clamps ($8 each) to support the read of the baffles from the exhaust pipes. At the moment I plan to wait until they arrive but that could change. We had a very good conversation about RVs relative to speed.
He has tried many things in his RV-4, RV-7 and Harmon Rocket and I am anxious to learn. It was a conversation I will remember.

I loaded everything in the car and drove to the airport but the gate is not working. Will try to do the installation (partial) without the clamps tomorrow.

The gate was permanently open this morning at the airport so I got in fine.
Then The lock on my hangar was frozen. After around 5 minutes I was able to work the key all the way in and open the door I unloaded everything out of the car and held the right baffle extension in the general location where it will go and it looks very good. However, it is so cold I just can't work with the care and precision I need to cut the rubber for a perfect fit. I will wait for the clamps from Larry Vetterman and warmer weather before I try it again.

No work today. Went to a movie - Dream girls and went grocery shopping.
Tomorrow I am planning the big task. I thought of a procedure for installing the baffle mods that should give me a nice fit of the rubber around the NLG structure. Basically, trace the baffle details on file folder stock, cut it out and work with the fit until it is perfect, then trace the outline back on the baffles. It is supposed to be 7F tomorrow morning so it will be a challenge.

Well, I'm still not done. I made the patterns as I had discussed before I believe and I worked in the cold this afternoon trying to integrate the pattern with the airplane. The mating between the baffle extension and the Nose Gear Strut Support Structure is awkward. I only get one shot at cutting the rubber and I want it right. So far the right side is all I worked with and it is pretty close. One thing that bothers me is the closeness of the right front strut to the metal part of the baffle. I tried to be careful and provide ample clearance in the baffles around each strut element but when I transferred the working pattern to the actual baffle with a Sharpie you can see it is off center to the rear quite a bit (Dvc00006.jpg). With engine displacement, at startup and shut down especially, I would expect metal to metal contact. I looked through my earlier photos and I don't have a good one with the baffle extension hard mounted on the forward part of the baffle that shows the forward strut location relative to the baffle hole. If I do not get the hole where the strut is I will not get any kind of a seal and if the strut and metal are to close I'll get metal knocking on metal. I think I will make a small hole in the forward part of the target circle provided by the pattern and gradually enlarge it as required to make sure I get it in the right place and if that is too close to the baffle metal I'll just have to trim baffle metal.

Well, I got the holes cut in the rubber for the struts everything came out fine. There is a lot of finish detail work to do, especially rubber trimming, but you can see the configuration has a functional look about it.
I am hoping for some speed gain, maybe small but some!
I have to modify the clamps I bought from Larry Vetterman so they will hold the rear of the baffle extensions in the right location. I bought 4 of them even though I only need 2 so if I mess them up I will have another shot at it. He has an extra bent flange on them that I don't need but otherwise they look good.

Appeared today on the AirVenturecup.com web site:
"LATEST NEWS: 19 JAN, 2006 The EAA seems to have approved the 10th running of the Airventure Cup Race. WOW a whole decade! The applications and
paperwork were sent up to HQ a few weeks ago. We'll updates here as they
come."

I completed the provisions for clamping the aft end of the new baffle extensions to the exhaust pipes today. I went out to the hangar and worked first with positioning of the clamp. I did the right side first finding a location that was clear of obstructions with the baffle installed. I eyeballed it from the side and the back and marked the clamp mounting hole location with a blue Sharpie. Not seeing any criticality and the awkwardness of measuring the location I felt this would be adequate. I drilled it with a #30 drill as a pilot hole compatible with bronze Clecoes.
Upon reinstallation it looked OK. Then I went to the left side and found that I had to locate the clamp farther aft to clear a rubber mounting
flange. At this point I decided to start measuring in spite of the
awkwardness and see if I could work out symmetrical clamp locations. I could but it required moving the right clamp farther aft. I dimpled and rivet plugged the existing hole and drilled the new ones 2.5 inches in from the outer edge and 2.5 inches of the trailing edge of the metal portion of each baffle. This put the clamp back where the baffle curves down in a tightening radius. This location requires a 9/16" average length spacer between the surface of the baffle and the clamp mounting flange. I don't have a drill press, grinder or vise at the hangar so I had to go home to make four spacers with ground to and angle approximately conforming to the baffle surface. Two were needed on top between the baffles and the clamps and two were needed on the bottom between the baffles and the bolt heads. I said "I had to go home" but I was so glad to get out of that cold damp hangar that I was looking for an excuse. They came out well in that worker friendly environment.
The photos show today's work.

I installed the new baffle extensions with flight hardware for the first time today. See the attached photos. The right side unit is pressing the rubber hard against the center cowl support assembly and I don't like that much. I will loosen the screws and see if there is any float to be had to loosen the contact. Otherwise it looks good. I had some trouble installing the clamps but once I thought to use my nut driver I was able to push the
bolt through and tighten the nut. I should be able to install the cowl and
test it tomorrow if weather permits.

Well I didn't get to fly today because I ran out of time. We had to do some shopping for picture frames and something that we could put in a 14"x14"
photo album to hold tickets, diagrams, programs, etc. from out trip to Russia in October. Once I got that taken care of I was able to rough trim the mod rubber and install the cowl but not fly. I plan to get up early and
do it in the morning. One of the attached photos is with the lower cowl
installed without trimming the rubber. The other three are with rough rubber trimming completed.

When I was leaving the hangar late this afternoon I saw a low wing fixed gear airplane taxiing from the east side of Drake field toward the runway.
The sun was at a low angle and I just categorized it as a Columbia XXX as we are want to do. When it made the turn toward the end of Runway 16 I could see that it was an RV-10 but I know of no RV-10s at Fayetteville's airport.
I could make out the color scheme in profile and it dawned on me that this must be Jim Younkin's RV-10 that he has been flying for some time with its bare aluminum exposed to the world. How did I figure out that it was his?
It is now painted red and black. Anyone that saw his son Bobby fly his air show routine will understand. Jim is a passionate guy.

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

The subject doesn't say it all but that is the bottom line. The test flight this morning showed an average speed at 6,000 ft. density altitude of
170.67
Kts (according the U.S. Air Race Inc. handicap procedure
(www.us-airrace.org) a 3 Kt decrease in speed from the like test made just before starting the latest mod. The speed is precisely the same as the baseline speed before any mods were made on September 21, 2005. Making cooling drag mods to increase airspeed is not for one that discourages easily.

I went to the airport this morning and the winds were calm (verified by
ATIS) and according to my OAT instrument the surface temperature at Drake Field (1,251 ft. MSL) was -3C. I started the engine at 0757 and departed to the east. At 6,000 ft the OAT was +6C(there was visible evidence of a temperature inversion layer somwhere around 4,000 ft.). According to the procedure I descended to 5,700 ft. trimmed for hands off level flight with wide open throttle, leaned to approx. 120 degrees rich of peak, 24.5 inches of manifold pressure, and 2,700 RPM, oil temperature 170, oil pressure in the acceptable range above 75 psi. One slight error was I left the heater and defroster on so air extracted from the plenum was being dumped into the cockpit instead of overboard through a SCAT hose. I don't know the effect of this change in configuration from previous tests. Once everything was
stable the autopilot track and altitude hold were activated. The three
leg
test was flown in three directions, 360, 120, 240 degrees. According to the procedure the speeds were recorded at 20 second intervals and accepted as valid after 5 consecutive recorded GPS speeds were within 1 kt. The recordings at listed below:

360 - 140, 142, 141, 141, 142, 142
120 - 187, 188, 187, 186, 186, 185, 185, 186 240 - 182, 185, 185, 185, 185, 184

The IAS was approximately 165 kts and the TAS was at the upper end of the calibration window approx. 180 kts. throughout the test flight.

After completing the run, the cylinder head temperatures were recorded at 2,700 RPM. They were:

#1 = 282
#2 = 329
#3 = 340
#4 = 321

During the test run before starting the latest mod I only looked at #3 CHT and noted at the random time of the observation that it was 337 degrees.

Observations:

The right exhaust pipe was higher (up near the bottom of the fuselage at the base of the firewall) than before the flight.

The gap between the bottom seal rubber at the rear of the baffle extension and the inside bottom of the cowl outlet was greater than before the flight.

The outboard seal rubber on both baffle extensions are not sealing against the inside of the cowl outlet tunnel because of pressure of the rubber against the outlet "shoulder" transition to the bottom of the cowl where the extension starts curving down into the outlet "tunnel".

When the engine is running there is a new low frequency rhythmic sound which could be from the rubber covered right baffle extension where it presses hard against the aft center cowl support, the right exhaust pipe hitting the bottom corner of the fuselage/firewall. I knew before the flight that I am going to have to drill out the rubber seal rivets cut back the baffle and reinstall the rubber to relieve the pressure in this area.

Conclusion:

More air is moving through the cooling path and the greater air mass flow is causing the plane to fly slower.

Action:

Fix the baffle/cowl support mechanical pressure problem.

Trim the outboard side seal rubber to get a uniform contact (no gaps).

Evaluate the bottom seal to determine if longer bolts and spacers are needed to reposition the trailing edge of the baffle extensions.

Re-fly the test after the corrections are made with the heater and defroster off.

If the Cylinder head temperatures are still low, start work on inboard plugs for the cooling air inlets.

Bob Axsom

 

1/28/07 - 2/8/07

pics sent in by Bob in these 23 emails

I'm calling the baffle extension mod finished. I took some photographs then removed the cowl and inspected everything. Except for the roughness of the rubber trimming it looked very good. Attachments Dvc00002 and ...4 show the outer baffle seals as they were in the test flight. You will notice the gaps between the rubber and the cowl outlet tunnel walls. I trimmed the rubber for a better fit. Dvc00009 and ...10 show them after trimming.
Dvc00008 is simply a shot right up the outlet. I put everything back together and I have the alarm set for 0630 to fly it again in the morning.
There may be some improvement in speed but I don't really expect any.
Regardless, I am satisfied with the quality of the project and so after 72.3 man-hours the baffle extension mod is done. I will tell you what kind of performance I get tomorrow.

Next I will work on the other end of the system, the inlets. Conceptually, I think I will develop restriction blocks to be held in place on the inboard surface with long screws. I would like to have an inner and outer block/plate for each inlet that will sandwich insert plates of perhaps 1/4" thick. The two parameters I will be looking at closely are the CHTs and the airplane speed. I expect to progressively add insert plates and record the effect on speed until the CHTs get too high.

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

I got up at 0622 this morning, anxious to re-fly the test. I thought that I would try to satisfy a few different test method preferences today so I flew tracks of 360, 090, 120, 180, 240 and 270. When I tookoff at 0730 the surface winds were calm again. In the process of conducting the test the winds were coming up a little. It became increasingly difficult to achieve the "no more than 1 kt variation over five consecutive measurements taken 20 seconds apart" standard required by the U.S. Air Race Inc. Handicap procedure. First I will list the other items then got to the speed measurements. OAT at 6,000ft = +8C, 8-3=5, 5x100=500, 6,000-500=5,500, 6000 ft density altitude = 5,500 ft MSL (per the handicap procedure). The speed test was run at 5,500ft. Mixture 120 rich of peak, manifold pressure 24.5, oil temperature 180, oil pressure 75+ psi, RPM 2710, CHT 1=309, 2=367, 3=362, 4=331 (at end of test). Speeds:

360 - 181,181, 182, 182, 180, 180, 180 179, 179 090 - 196, 194, 194, 194, 193, 194 120 - 186, 186, 185, 185, 184, 185, 184 180 - 163, 162, 162, 161, 162, 163, 164, 162, 162, 161, 163, 163, 164, 165, 164, 164, 164 240 - 149, 148, 150, 151, 148, 150, 151, 149, 150, 149, 147, 149, 148, 150, 148, 148, 151, 150, 148, 149, 149, 147, 148, 147, 148, 148 270 - 152, 151, 153, 152, 150, 153, 152, 153, 153, 154, 153, 153

Using the U.S. Air Race 3-way method (360, 120, 240) I get 171kts Using the 4-way procedure (360, 090, 180, 270) = 172.7 kts Using 2-way (360, 180) = 171.9 Using 2-way (090, 270) = 173.5

Please recheck my math - I made several errors on the calculator that I did catch in a recheck so there may be some that I didn't catch. Using the baseline method that I have used throughout these test I will claim 171kts.

I observed after the flight that there was a 1/2" gap between the rear of the right baffle extension and the bottom of the cowl outlet that did not exist before the flight. This same condition was observed in the test on 1-24-07. I think there is a low pressure above the curve baffle surface and/or higher pressure under the right baffle that is causing this - I have to think about this as it has to be a source of drag.
--------------
The three-way speed for today should have been 170.6 - the same as on 1-24-07.
---------------
I've been thinking about the next step in the Cooling Drag Reduction process which is to restrict the air mass flowing through the system. The brute force, straight forward, intuitively obvious approach is to restrict the size of the inlets. However, I am told the this can be done effectively by restricting the size of the outlet. I can understand that but I have a hard time accepting that it can be as efficient as never letting air into the system. Restricting the outlet with no change in inlet should result in increased internal pressure and it is the back pressure at the inlet that would restrict the amount of air entering the system. The big advantage of the "block the outlet" approach is implementation simplicity. I can't ignore that feature so I am going to try the outlet plug approach first.

Since I have an "A" model RV-6, I have a ready-made mount in the nose gear strut and the center cowl support. I envision an airfoil shape, perhaps favoring the right side to offset some of the need for right rudder, that extends to a flat sided point several inches aft of the end of the cowl.
Metal, wood and fiberglass materials are being considered.
----------------
The response to my earlier message was so convincing that I was taking the wrong approach, that I have decided to go back to the front end of the system and work there. It will take longer to implement but I think it is the right thing to do.
---------------
I am going to attach a tension element from the underside of the baffle extensions to a point forward to hold the extension rubber against the bottom of the cowl outlet. This is to eliminate the separation that occurred between the baffle and the cowl outlet on both test flights.

On the air mass flow restriction I received several inputs that disagreed with my decision to work on the outlet first but one message was especially insightful and it would be a terrible waste not to share it with you. It is from Paul Lipps:

Subject: Re: Change of Plans on Air Mass Flow Restriction
> Hi, Bob!
> I tend to agree with you. When you block the outlet, you decrease
> the inlet flow, but form a large bubble of slowed down air in front of
> the
flat
> inlets, which affects the flow into the root of the prop. When you
> block
the
> inlet with a streamlined shape, the bubble is smaller. Block the
> inlets until the temps are ok, THEN reduce the outlet aperture! Or so it seems to
> me! Paul

----------------
First of all Thanks for the great technical feedback Chris. Below your
input I have copied, in reverse chronological order, the rest of the feedback I have received from Paul Lipps and from two of my fellow EAA Chapter 732 members Charlie Heathco and Steve Chambers, on the next step in my effort to reduce cooling air drag on my RV-6A (photo attached). Steve and Charlie are proponents of the inlet size reduction based on observations, experience and intuition. I think everyone is in at least high level agreement but you and Paul have actually done the work on your planes and have accomplished what I hope to achieve. I feel I must carefully examine what you have to say and focus the scope of my air mass flow restriction on the most probable success approach.

Both of you are aware of the bubble of slower air in front of the inlets. I knew there would be some back pressure resistance to the incoming air but I never thought of its configuration. In thinking of it this way it is easier to visualize some things. For example, I see this bubble that changes in size with changes in to changes to inlet size or outlet size. Intuitively, it seems to me that this bubble size has an external effect on aircraft speed as this bubble is being pushed through ambient air. If I reduce the inlet size in a way that replaces the open hole of the inlet with streamlined structure as proposed by Paul the bubble being pushed would become smaller and the external drag would be reduced. It also seems reasonable the with the reduction of the bubble opposing inlet flow, the inrush flow velocity would increase. It is hard for me to fully accept that the flow velocity would increase to the point that there is no change in air mass flowing through the system. If that were completely true the air mass flow would remain the same as the inlet shrunk to full closure. However, I don't think that was your intent. The important thing for me to grasp is the concept of increased internal air flow velocity with the reduction of inlet size. Given that the cooling is closely related to the air mass flowing through the system, it seems reasonable to look at CHT as the measure of adequate air mass flow. I am not prepared to go into measuring the air flow velocity at this time but I can work on that in the future if it is necessary to look at that parameter directly in order to tune the system for maximum safe airplane speed. I know of no other easy way for the experimenter in the field to do it. This is how I interpret what Paul recommended. Now, if this is done over a series of flights with no change in outlet size then the outlet flow capacity will eventually exceed the airflow mass in the system - like flying around with the cowl flaps wide open. Now I should be able to essentially "close the cowl flaps" (with a thicker plug) to a size that just starts to restrict the airflow mass because the increase in outlet air velocity can no longer handle the air mass capability of the inlets. I think this should be recognized by a rise in CHT. This idea is consistent with the jet inlet outlet work mentioned by Steve. The design of the RV "A" models is such that there are four 0.75" and one 1.75" cylindrical structure members in the middle of the outlet plus a sheet metal assembly to support the rear of the cowl where it has a long notch to clear the nose gear strut that are still exposed in my lower cowl mod, though much less so than the stock design. My thought was to enclose the 1.75" strut and the sheet metal support in an airfoil or streamlined shape to cut turbulence and at the same time reduce the cross sectional area of the outlet. I plan on adding "teardrop" streamline fairings to the 0.75" struts in this just because it seems like the right thing to do and this shape will still allow me to remove baffle components for servicing the engine etc.

I am still working my way through conceptual methods for working on both ends. My general thought on the inlet involves a growth from the inboard edge into the opening using extended screws where three are used to tie the upper and lower cowl halves together around the prop shaft. These need to be adjustable the measure and identify the "best" size. I think a rounded ridge or a flat plate flow fence is needed to keep the inboard air from spilling into the inlet opening but I'm still thinking that one. On the rear end I am still thinking of a vertical streamlined or airfoil shape but I'm now thinking that the trailing edge should be consistent with the trailing edge of the sheet metal cowl support assembly perhaps with a reflex in the shape after the strut is accommodated.

Thanks again for your input, I really appreciate being enlightened about this.
--------------
The approach that I am starting with is to make 1/4" wide intake plugs of balsa wood and attach them to the cowl cooling air intakes with the three cowl splice screws on each side of the prop shaft. I made two sandwich aluminum patterns today and the first actual balsa test plug. I call it a test plug for a couple of reasons but mainly because I intend to test fly with a small plug, record the results, add another test plug to the stack, test fly, record the results etc. The two main parameters I will be looking at are CHT, and airplane average speed at 6,000 ft density altitude, wide open throttle, ~2700 RPM, leaned to ~120 deg. rich of peak EGT on cylinder #4 (I have the 4 probe system but I haven't had time to install it yet).  When I find the best opening I may just glass and finish the stacks as racing intake plugs or I may use them as molds for some other implementation method.

-----------
I wish the leading edge of the plug were not so blunt. When I went to the hangar yesterday and made the template I measured about 1/2" from the front of the inlet to the front of the cowl face behind the spinner. That was the basis for the shape of the leading edge of the plug. The question is, is the drag of the plug of this shape less than the cooling hole? We are making our big weekly trip to the local Wal-Mart for groceries this morning and I have Jeanine's agreement to stop by the hangar first to check the balsa test plug plate in the cowl. If there is any room I will consider making new templates to extend the nose of the plug out more. One thing I do not want to do is force air in behind the spinner, however. Another thing I am thinking about in my test configuration is to use an aluminum flow barrier plate derived from the template with an oversized front end place at the outer end of the stacks in each inlet. I have to make sure that when the prop goes to high pitch that it does not contact the plate. Once I verify that the plug shape is right I can complete all of the rest of the duplicate test plug plates in a day or so and go right into testing.

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

We got out of our house yesterday for the first time after the snow and I was able to stop by the hangar to try to capture the hole patterns and positions in preparation for drilling the mounting holes in the test plug plates. I drew paper patterns on tracing paper, removed the center cowl screws, held the paper in place and tried marking and actually punching holes in the paper to get hole positioning but in the end I knew the alignment was no good. The contours in the inlets just don't allow easy consistent alignment with a 2-dimensional pattern. I came home and decided that I need to produce 3-dimensional mates, locate the holes in them and then transfer the hole pattern to two 2-dimensional patterns for drilling all of the rest of the test plug plates. I glued two sets of three test plug plates together in preparation for the 3-dimensional mate (see attached). There are a lot of details to be worked out yet for locating the holes but I believe I'm going in the right direction. The 3-dimensional mates will also become the first test plugs, so the first test will reduce the inlet area by 5.25 sq. in. After that the reductions will be 1.75 sq. in. if the incremental changes are made to both inlets or half that if only one inlet is changed. The three plug start should save me $60 or more in fuel cost for the tests.


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

Well Doug, things are moving forward again. I went to the hangar and started working on the right inlet plug. The hanging point had been getting the cowl screw hole locations transferred accurately to the test plug plates and the test plug outboard securing plate. I bought some #60 sandpaper and a 5/8" dowel from Lowe's and used them to shape the inboard side of the three layer balsa wood plug for the right cooling inlet. As I sanded I kept trial fitting the plug in the right cooling inlet. The interface is curved from top to bottom and it tapers out from front to back. The coarse sandpaper on the round stick worked quite well to shape the plug. After I had a good fit, I reinstalled the three screws that tie the upper and lower cowl halves together in the inboard surface of the cooling air inlet. I then stuck the plug back in the inlet and pressed it hard against the inboard edge of the inlet. Then I removed the plug I could see the impression of the first two screws well enough but I was not sure about the rear screw location. I put blue Sharpie ink on the screw heads and repeated the process which resulted in three very clearly and accurately marked screw locations. I went home and used the drill press to drill three parallel holes, perpendicular to the flat outboard edge of the plug, directly through the marked spots. Then I clamped one of the aluminum patterns used to make the 24 test plug plates to the outboard side of the plug and drilled a hole in it, in line with one of the holes in the balsa wood plug. I clecoed the plug and the aluminum plate together at the hole just drilled then drilled a hole in the aluminum plate through one of the other holes in the plug.
Finally I drilled the third mounting hole following the same process. The aluminum plate can now be used to drill the mounting holes in the rest of the test plug plates for the right inlet. I sanded the plug with finer (#220) sandpaper and it was ready to trial fit. The fit was perfect! The process must be repeated for the left cooling air inlet as the hole patterns are not identical.

To mount the plug I planned to use an undersized aluminum plate in the inlet outboard of the balsa wood plug that grew to oversize in front of the inlet to form a lateral flow barrier to keep the air from spilling into the cooling inlet. I used the right inlet aluminum pattern with the three holes described above to make the end securing/barrier plate. I drilled the mounting holes through the pattern into a new piece of aluminum clecoing as I went. With the two pieces of aluminum clecoed together I traced around the pattern with a sharpie and used a homemade tool to mark 1/8" inside the outline of the part there it was to go in the inlet and a similar tool to draw a line 1/4" outboard of the portion of the new part that would extend in front of the cooling air inlet. I cut out the part and finished the edges (deburred sharp edges, etc.). Finally, I installed the plug and the aluminum plated in the right inlet. Some trimming was required for the aluminum to prevent contact with the front of the cowl.The fit was then fine.

I am not ready to do the left side, however. I have two major concerns and one minor concern.

First, the balsa plug is stiff and retains its shape and position very well but the aluminum is not stiff at all. since the line of mounting hole is high on the inboard inlet wall the lower end is not well secured. If you look closely at the head-on photograph you will see a slight separation between the bottom of the aluminum and the plug. This is not acceptable. I need to preload the lower edge by bending the aluminum, add a fastener near the bottom or cut the aluminum plate off closer to the mounting screws.

Second, the forward extension of the aluminum is too close to the prop arc to be sure it will be well clear when the constant speed propeller blades rotate to a higher pitch in all of the propeller stations (diameter points) that will be used as more test plug plates are added in the test program. I can come up with a method to determine the clearance is adequate through the full pitch range and diameter range of concern or eliminate the lateral flow barrier.

At this point, using an aluminum end plate with an equal distribution above and below the mounting screws and no forward extending lateral flow barrier is winning out in my mind.

The minor concern is the mounting hardware. I have some 2 and 3 inch round head 8-32 screws that I can cut as required and use with AN960 washers for the shorter test plug plate stacks but nothing for the longer stacks and I would like to go to flat head screws and dimpled holes in the end plates.
This is not a big deal but it needs to be worked.

I should finish up the first flight configuration tomorrow but you never know...


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

I completed a basic test configuration and flew it this afternoon. I completed the left intake plug like the right one today. I gave up on the lateral flow barrier and I went to the two aluminum patterns as the end plates. I bent the periphery to load the edges when the screws were tightened rather than add additional fasteners. I learned some interesting things about the RV-6A design today. The thrust line is offset to the right and the distance from the front of the inlet on the right to the front of the cowl is much shorter than the same distance of the left side. My standard test plug plates extend beyond the front of the cowl on the right side but not the left. After working with the fit of the left basic plug (made of three standard test plug plates glued together I decided to fill a gap on the forward end between the lower cowl and the plug with black silicone. I cut off 2" screws bought from Aircraft Spruce to what I thought would be a good dimension to mount both plugs. They were OK for the front two mounting locations but the rear ones were too long and when tightened
they caused interference with the starter ring gear. I could not run the screws all the way down in the back where the basic plugs have been sanded much thinner and when I checked the freedom of the prop to rotate by hand - it wouldn't. When I backed out the rear screws I could rotate the prop by hand with no problem. I cut those screws off another 3/8". The rotation was unobstructed so I decided to make the flight but every test plug combination will require its own custom screw set and I am going to have to test install them with the upper cowl off to make sure there are no problems created behind the cowl by the screws. After the fit check the screws will be removed and reinstalled with the upper cowl reinstallation.

The plane flew normally but it was not calm and I could not meet the consistency requirement of the U. S. Air Race handicap standard of five consecutive GPS speeds, 20 seconds apart, within 1 nautical mile per hour over the set of five measurements. I did record a bunch of measurements but Razorback Approach was hinting that they were feeling some traffic pressure so I could not prolong the test. The OAT was 3C at 6,000 ft so the 6,000ft density altitude test was flown at 6,000 ft MSL. The manifold pressure was 24.5; the oil temperature was 200; the oil pressure was over 75 psi, the prop rpm was 2710; the CHTs at the end of the run were 1=299, 2=360, 3=367, 4=327; the mixture was leaned to approximately 120 degrees rich of peak EGT.
The speeds I recorded in the three tracks were:

360 = 148, 149, 144, 144, 141, 142, 142, 147, 148, 145 120 = 202, 204, 201, 206, 206 240 = 161, 160, 164, 162, 166, 171, 168, 164

Adding the speeds recorded in each track and dividing by the number of recordings in the track, then adding the three averages and dividing the sum by 3 gave an average speed of 171.1 kts. You can't simply add all the recorded speeds and divide by the number of recordings because the result will be skewed in the direction of the track with the most recordings. The indicated air speed and TAS looked about the same as the last few tests so a half kt increase in speed is possibly correct but I can't be sure since the requirements of the test method were not met. However, I do not plan to re-fly this configuration tomorrow because there is no enough information to be gained to make it worthwhile. I am going to add one test plug plate to each side and fly the next test in the morning if weather permits.

Three photos of the test configuration flown today are attached.


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

I got up at 0630 to do some testing in calm air. There were delays but it worked out well. When I got to the airport I looked at the outlet area and found that it looked good but there was a 1/4" gap between the inside bottom of the cowl and the rubber on the right outlet baffle extension - I still have to address that problem. I removed the upper cowl and inspected the area behind the test plug mounting screws. I saw faint marking on the top of the starter ring gear teeth where the rear test plug mounting screws rubbed against them when I manually rotated the prop - nothing else. The temperature was in the 20s but I had an electric heater and dual post mounted work lights to make it tolerable. Three new test plug mounting screws had to be cut but three were still long enough for an additional test plug plate. installed an additional 1/4" balsa wood plate in each inlet and 1 hr 45 min. after arrival I had everything back together. I had to top the tanks before the flight but the self serve fuel pump had failed. The good folks from Airport Operations & Maintenance came right over to fix it but after trying to get to work one more time they determined that it required a new solenoid and they called the folks from Millionair to come over and fuel me. At 09:54 I started the engine for the test. I was afraid that calm test period had expired but it was very good actually.

I climbed to 6,000 ft per the U.S. Air Race, Inc. handicap procedure
(http://www.us-airrace.org) and found that the temperature was 7C which meant that the test would be flown at 5,600 ft MSL. The manifold pressure was 24.5"; Oil Temperature was 180 degrees; Oil Pressure was over 75 psi; the throttle was wide open and the prop was max'd out at 2710 RPM. The GPS ground speeds in three directions were very steady as follows:

360 - 180, 179, 179, 179, 180
120 - 188, 189, 188, 188, 188
240 - 148, 147, 148, 147, 148

This gives an average speed of 171.733 kts. or 0.6 kt higher than yesterday - finally something that is positive. I'm taking today's results as validation of yesterdays test results as well

The Cylinder Head Temperatures were:

#1 - 316
#2 - 381
#3 - 392
#4 - 343

These are considerably higher that the CHTs reported on previous tests so it appears that I will run into limiting temperatures before I use all of my test plugs.

I'm going to tackle a nap then add another plate to each inlet. The plane will be ready to fly in the morning right away.


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

I went back to the airport after a short nap and installed the next pair of test plug plates. I had just enough time to squeeze in another test but that plan was done in by the tower. I taxied out to Runway 16 and sat while a tri-gear Maul, and a Bonanza landed and the same 172 did a couple of touch and goes. Oh well, tomorrow morning will probably be better anyway. In all of the taxiing and holding the CHT of #3 never reached 330 degrees. Two photos of the current configuration are attached.

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

170.533 kts. I don't understand 'em I just report the results. Started up at 0709 and shut down at 0751.
MP = 25
Oil Temp = 200
Oil Press. = 75+
RPM = 2710
360 - 174, 173, 173, 174, 174
120 - 198, 199, 198, 197, 197, 198, 197
240 - 141, 141, 140, 140, 141
Avg. Speed = 170.533 kts
CHT - 1=313, 2=377, 3=386, 4=344

I'm going to have breakfast then install the next two test plug plates and fly the next test later today.

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

I forgot to report that the OAT at 6000ft was 12C so the test was flown at 5100ft per the U.S. Air Race Inc. Handicap procedure. Ref:
http://www.us-airrace.org.

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

While contemplating this mornings test flight in the throne room I came up with an uncontrolled variable that I should take care of. I still have the magnetic latched filter bypass door (with my fix to keep it from falling apart again) and Van's documentation on the subsequent control wire revision says the door can open with the magnetic latch and reduce power. I already have a platenut mounted on the door and a "racing strap" that I mount there to keep it closed. I hate to delay the completion of the testing but I am going to have to pull the cowl, lock the door down and perhaps re-fly the current configuration. ARGH!!!

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

I got everything back together this afternoon and it is ready to repeat the test flight with the three piece plug bases and two additional test plug plates in each cooling air inlet. The air filter by-pass door strap down is shown in the attached photo.

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

ATIS was broadcasting 3,700 ft overcast. It didn't look that low but I was not able to reach 5,000 without starting to loose visibility. I'll let you know when it happens but they are forecasting clouds for days now.

Latest Updates
 

 

- Link to Excel Spreadsheet

2/13
Well Doug, I have done nothing for a week. I strapped down the filter by-pass door in the filter air box and I have been waiting for a ceiling over 6,000 ft. to conduct a test of the effect on speed with the basic plug and two test plug plates in each cooling air inlet. I might as well be living in Oregon. I am going to break good experimental test practice and make another change which of course will obscure the effect of the door strap down on the speed achieved in the previously tested configuration with the door held closed by the magnet latch only. It bothers me but not as much as just sitting around waiting for the weather to clear up.

The change I am going to make involves the inlet plug configuration. When I established the plug plate profile I measured the distance from the leading edge of the inlet to the front of the cowl to establish how far forward I could extend the leading edge of the plug and minimize the flatness of the plug as seen by the air being smashed through by the plug. I settled on 1/2" as the maximum forward extension of the plug which follows the curve of the cowl at the inlet pretty well.

Here's my concern. As the plug gets wider with the addition of test plug plates more air is impacting this rather blunt surface and there is no
directional control of the air flow from this impact. Some part of it goes
outboard and enters the inlets and part of it goes inboard behind the spinner and into the cowl. Most of the air entering the cowl behind the spinner will flow into the cavity between the lower baffle with the extensions to the outlet and the lower cowl. The baffle extensions are not held rigidly against the bottom of the outlet and after the last test flight where I improved the fit the right baffle extension rubber was raised approximately 3/16" from the bottom of the outlet. The trailing edge of the baffle extensions are supported of the exhaust pipes so this may be do to the movement of the turned down exhaust pipes sticking down in the 200 mph slip stream. The upper surface of the baffle extensions are smoothly curved downward to the bottom of the cowl outlet and it closes to approximately 2 inches of the oppositely curved upper (rear?) baffle which extends from the back of the engine to just below the bottom of the fuselage at the firewall.
I believe there is a lowering of pressure in this area and it is probable that the baffle moves upward as a result. The other force air entering the cowl in front of and below the lower (forward?) baffle which finds a somewhat linear way out by pushing the baffle extension up from the bottom of the cowl outlet.

I've been thinking of this for some time but I wanted to test fly the filter by-pass strap down with no other changes before I did anything about the plug configuration. What I intend to do is, take the plugs out and sand them at the inboard leading edge so that the air is allowed to go back behind the leading edge of the cowl and the outboard edge of the plug. The bad part of this is there will be no unambiguous baseline for the test because two variables are being changed. The plug will compress the air between the inboard and outboard edges and at the point of maximum compression the leading edge will be more blunt than before so it is not certain that the plug shape change is a good idea. Strapping the filter bypass door closed will not reduce the speed of the airplane so if there is a reduction I will assume it is due to the plug shape change.

... and the test flight wait continues.

2/13
I did some conservative sanding on the front of the plugs today. They still have a rounded leading edge to help move through the air and the have a concave grove to thy to restrict it to a straight line. If the clouds ever go away I will test it. Before and after photos attached.

2/16
Finally after 10 straight days of overcast skies we have a glorious crystal clear day in Northwest Arkansas. I got up at 6 am but it was too cold at 7F to abusive to crank the cold engine and at 1:30 pm I promised Jeanine I would take her to a movie. So, with great concealed anxiety I let the
opportunity slip by. It is supposed to be bad again tomorrow but Sunday
looks good for now.

2/17
Well, it was windy and gusty but it did clear up this afternoon. The test results are not considered valid but then again, who knows? I flew and flew until I got 5 consecutive speeds 20 seconds apart that were within 1 kt of each other in three 120 degree separate directions per the U.S. Air Race Handicap procedure (http://www.us-airrace.org):

360 - 130, 131, 131, 133,132, 127, 128, 130, 133, 131,130,130, 131, 131, 131 240 - 175, 177, 174, 170, 174, 174, 173, 176, 172, 173, 175, 175, 178, 177, 177, 176, 176, 176 120 - 207, 201, 201, 201, 203, 204, 203, 203, 203

For an average of 168.066 kts.

I made a significant mistake in selecting the altitude to fly when I used the ATIS altimeter setting (30.07" Hg I believe) instead of the standard
29.92 when flying to 6,000 ft to get the temperature to determine the altitude to fly at for a density altitude of 6,000 ft. per the procedure.
That coupled with the wind makes the results for today unreliable. Today's flight was the same configuration as the previous test flight except the filter by-pass door was strapped down and the leading edge of the cooling air intake plugs were sanded to provide a vertical air flow control groove in the plugs.

I went through my records in a casual manner and put my findings in an Excel spreadsheet which is attached. There may be errors but they will be corrected over time as they are found. Other data from today's flight are:

Oil Temperature 160
Oil Pressure 80
Manifold Pressure 23.5
Flew at 6,600ft (with wrong altimeter setting)
CHT:
#1 - 318
#2 - 392
#3 - 395
#4 - 353

After landing I fitted the next to test plug plates and installed them so it is ready to fly tomorrow with approximately 5.25 sq. in. of the cooling air inlets blocked off. Hopefully things will go better.

2/18
I got up before dawn and went to the airport to get in a good test. The sky was clear and the winds were light down on the surface. Once again the speed was less than 170 kts but it was over a knot faster than yesterday at 169.848kts. What I read into this is the "grooved" front end of the plugs increased the drag over the original "straight line" front end plugs but as I reduce the inlet area the speed is increasing. I could not maintain the 1 kt variation limit of the US Air Race handicap procedure but in these $4+ Gallon days I am not going to quibble. I just took all for each of the three headings, averaged them and then averaged the averages for the three directions.

The Manifold pressure was low again at 23.5. The temperature at 6,000ft pressure altitude was 0.0C so 6,000 ft density altitude was at 6,300 (around 6,800ft MSL) and that is where the test was flown this morning. The CHTs were slightly lower at 1=314, 2=386, 3=390, 4=343. The oil temp was 180 and oil pressure was ~80. All of the speed readings were:

360 - 145, 145, 146, 146, 146
120 - 200, 200, 199, 199, 197, 197, 197, 196, 198, 199, 198, 196, 198, 1978, 199, 198, 199, 196 240 - 168, 168, 166, 164, 165, 165, 166, 166, 166

169.848 kts avg. (You cannot add up all the numbers and divide by the quantity of entries because of the skew resulting from the large number of recordings in one direction.)

The plan is the insert two more test plug plates and fly again, record evaluate and iterate. I expect the speed to increase as the inlet area gets smaller. Once I find the "right" area I will work on a nice pair of "hole fillers".

2/18
I added a test plug plate to each side this afternoon and test flew it after the EAA Chapter 732 monthly meeting. Since the plug material is 1/4" balsa wood and the basic plug on each side is made up of three of them and four additional have been added to each side the inboard 1.75" of each cooling air inlet is blocked off. The height of the inlet openings is 3.5" so a total of 12.25 sq. in. of cooling air inlet is now blocked. I think I have turned the corner with this mod and the speed is starting to creep back up.
This afternoon I reached a three-way average of 172.158 kts. The weather is supposed to be bad for the next couple of days so this is probably it until Wednesday.

2/19
As my nightly activity when I wake up in the middle of the night (every
night) I cleaned up the Excel file on testing the "speed mods" last night.
I referred to my Builder's Log which I still maintain to try to put the configuration detail in a better perspective. Some of the test flight information is only recorded in my Pilot's Log but using the dates I was able to correlate the flights with the configuration in the Builder's Log.
Most interesting are flights made with the identical configuration producing average speeds that varied over a 4 kt range. So much for accuracy. I may switch to the two way method proposed by Paul Lipps just to save gas. The family CFO (Jeanine) was a bit shocked at my test flight fuel bill so I have to be a little more careful and keep the cost under my weekly budget to avoid turmoil - we can both live with that. It will be a few days before I add another test plug plate to each inlet and make another flight.

2/20
I added a test plug plat to each side today and flew another test flight. I stuck with the 3-direction test method again and the resulting speed and CHT's were very similar to the previous test. 171.866 kts; CHT #1 = 311, #2 = 385, #3 = 387, #4 = 358. I will add another test plug plate to each side and try again later. See attached Excel spreadsheet and two photos for more details. The CHT on cylinder #3 reached 402F fairly early in the climb out and I flattened the climb. The temperature stayed at that level during the rest of the 500 FPM climb then dropped to a more comfortable level in cruise. It is definitely becoming something to watch in my opinion.

2/23
Up before dawn again to test the next plug plate addition to the cooling air inlets. The inlet plug configuration is shown in the two attached jpg files and the results are shown in row FT#20 of the attached Excel file. The speed went down to 170.6.

Thinking about this, it seems to me that it is probable that I have gone beyond the point where the increase in speed achieved by the reduction of the air mass entering the cooling system is being more than offset by the air pressure on the leading edge of the plugs. At one point in inserting the plugs I decided that it would be worth trying a recess in the plug leading edge between the inlet and the spinner. I thought it would be good to cause the air to exit this area above and below the cowl and not go into the cowl behind the spinner or into the inlet. A test before and after this change with the plug base and two test plug plates indicated that this was a mistake. In the otherwise identical configuration the grooved plug sample indicated a speed reduction from 170.5 to 168.1 kts. Because the base (three plates glued together and shaped to conform to the cowl) was greatly altered in this change and it was time consuming to replace, I decided to continue testing with the grooved leading edge plug configuration. The speed increased in the next two test flights to 169.8 and 172.2 kts with base plus
3 and base plus 4 configurations respectively. When I went to the plus 5 configuration the speed dropped off slightly to 171.9 kts and the plus 6 configuration continued the decrease in speed to 170.6 kts.

It seems reasonable to start making a plug that occupies the inboard 1.75"
of the inlet opening. The shape should be such that the air resistance is minimized. Forward space is limited by the prop location but the plugs could extend forward of the gap between the cowl and the prop. My first though is to build plugs that extend up past the back of the spinner as far as the front face of the prop back plate. Outboard the leading edge will taper back in a straight line to the inside of the inlet. Air seeking the least resistive path will enter the inlet as well as going over and under the cowl on the outside. None should enter the cowl behind the spinner as a result of plug leading edge pressure. It is thought that this configuration will reduce drag and provide a little extra cooling air. The height of the plug is still in question. I have to decide, given the short longitudinal space to work with, is it better to blend in with the upper and lower surfaces of the cowl and be rather blunt or to maintain a low profile for the plugs and tolerate an bump in the airflow where the plug and the cowl meet. Right now I'm leaning toward the low profile plug.

I have to start on my annual condition inspection soon so it will be a while before the next step in the plug development and test is taken.

2/25
as I indicated before I feel like I have reached the limit on fixed inlet size reduction. Even in the current relatively cool environment the CHT rises rapidly on the ground an in the climb out. Going back to the 1.75" reduction of inlet size on both sides of the spinner appears tolerable for these phases of operation and seemed to give the best speed as well.
After I got into maximum speed cruise the temperatures came way down even with greater inlet restrictions. It seems that cowl flaps to close off the inboard area of the outlet might improve performance in cruise.

2/27
I fixed the unit of measure on the chart for the AirVenture Cup speeds. I am adding a legend below:

Stock = no cooling drag mods

Plugged ramps = the ends of the cooling air inlet ramps are plugged with foam and fiberglass

Lower rear baffles = left, right and center baffles added from read of engine to outlet at bottom of fuselage

Horizontal baffle = baffle added just below cowl split line from left & right baffles to front of cowl

AB w/def & NLG covers = cover added to top of filter air box (with deflector at back) & nose landing gear structure cavity

Removed AB cover def = removed deflector from rear of filter air box cover

Lower fwd baffle /no seal = lower forward baffle added from front of engine to a point just forward of the nose landing gear

Sealed LFB/ hole for VP = sealed all openings in lower baffle and added a hole in the upper engine baffle for cooling vacuum pump

No change = repeat flight of same configuration

Lower Baffle Ext. = lower baffle extended to the bottom of the opening in the cowl outlet

Inlet plug base = three 1/4" balsa test plug plates glued together are fitted to inboard edges of cooling air inlet opening and installed

Base + 1TP = 1 test 1/4" balsa test plug plate added to the base in the cooling air inlet

The rest of the tests are conducted with the base and the number of test plug plates indicated. 2 TP = two test plug plates, etc.

3/5
All of the cooling air inlet test plugs were removed and I test flew the plane again today. Of the baffles are still in place with no other modifications. I have flown enough test flights and thought about the test method enough to know that the U.S. Air Race Handicap triangular approach with fixed headings 120 degrees apart always starting at 360 it is not the optimum way to go about this. Today I flew directly down wind then turned around 180 degrees and flew directly into the wind recording 5 stabilized GPS speeds 20 seconds apart added them and divided by 10. I determined consistency of track and wind direction when the digital readout for track and the CDI were the same and the speed down wind was maximum. At small relative angles the sine is near zero so the cross wind component is inconsequential - that's what I want. That is the test method I intend to use from now on. I want to eliminate the inefficiency of cross wind flying from the calculation. I do not believe you can say "well, the GPS Speed is the speed you are traveling through the air in the track direction so cross wind doesn't matter". I believe the fact is, the track speed is the vector speed resulting from the heading speed and the cross wind drift speed, and only when the cross wind drift is eliminated can you see a GPS speed that truly represents the absolute airplane speed. By flying in the opposite direction the and recording the GPS speeds in a similar manner the average wind component can be eliminated. Anyway that's how I am looking at the test method now.

In today's test the headings (and tracks) were 040 and 220 degrees. At a density altitude of 6,000 ft all of the parameters recorded were essentially the same as previous tests except for the speeds and the CHTs. They were:

040 - 182, 179, 179, 181, 182
220 - 163, 164, 163, 163, 163

Average = 171.9

CHTs 1=289, 2= 356, 3=363, 4=329

As I said earlier this is conditional inspection time so I am going to be working on that for most of March. I am however, thinking of the removable cooling air inlet plug design and a little farther down the line the outlet internal flap design.

More later.

3/15
The annual condition inspection is still underway and it will not be finished by Sunday so I will be driving up to the meeting at Rogers.

Removing the baffling in the lower cowl for the inspection gave me a chance to study the cowl flap possibilities more closely. I now have a concept that looks simple to implement so I am going to pursue it for now at least.
I am going to add vertical flow fences just inboard of the exhaust pipes.
They will be riveted to the lower baffle extensions. The cowl flaps will be attached to the upper aft baffle behind the nose landing gear structure using aircraft hinge stock that I have. The actuator will be and extra Mixture control cable that I have and it will be mounted on a plate atop my console below the instrument panel. With this configuration the flap will go down to eventual closure against the upper surface of the lower baffle extension. The control pushed all the way in will close the flap (downward
travel) and pulled back will open the flaps. Some details are to be developed using the cut and try approach. For example, it would be nice if I could mount the flaps at angles so they would wedge the air to the reduced outlet outboard of the vertical flow fences. More thought required as I continue the annual.

3/19
The annual condition inspection is complete and I'm back to mods. I have to put in the 4 probe EGT system that I bought last summer first. I flew with the single probe in the pipe from cylinder 4 for 300 hours and I suspect it is not the leanest cylinder. It will be interesting. The two attachments show the old installation before starting with the mod. You can see the Alcor probe extending through the rear baffle and I was worried about that.
It went quite well actually. I had to remove the outboard baffle on the left side but that was all. The EI probe is a larger diameter than the Alcor probe so drilling out the old hole to 13/64ths was a straight forward operation. I mounted the probe in the exhaust pipe from cylinder #2 today
as well. I have good access to everything. The right side requires some
thought as the lower forward baffle and the spark plug as well as access to the mounting clamp screw need to be carefully considered. Completion of the rear cowl flaps will be put off a bit to catch up with spring home chores.

3/22
Yesterday I completed the final replacement of my one probe analog Alcor EGT with an Electronics International digital 4 probe EGT except for the final hook-up behind the instrument itself. That means the old probe was removed, holes were drilled in the four exhaust pipes (#4 was increased in size to 13/64"), the probes were installed, the wires were routed and tied in the engine compartment, a hole was drilled in the firewall, the leads were routed through a grommet and the firewall hole, the grommet was installed in the firewall hole, the grommet and the wires were sealed into the firewall with Dow 736 high temp RTV, the old instrument was removed and the new one was installed in the panel. I still need to get an in-line 1 amp. fuse and connect the power, ground and temperature probe leads.

The installation was very time consuming to avoid mistakes and keep it neat.
This gave a lot of time to keep looking at the existing baffles and contemplate the cowl flap installation. I do not like what I have come up with so far and I am backing away from the cowl flap idea for now. I am currently planning to fabricate another baffle component that is fairly long. It will wrap around the carburetor and the nose landing gear structure and continue back to the outlet. The current attachment plan is to mount it with #8 screws to the lower baffle in the lower cowl half and seal it with edge rubber strips to the oil sump bottom, the upper aft baffle in the lower cowl and the center cowl support with edge rubber. The aft portion will twist outboard at the top to accommodate the NLG support struts.

3/24
I completed the installation today and it appears to work fine but that was not the case initially. Electronics International provides a ready made extension cable for the temperature probes with very clear instructions.
Each pair in the cable has a male and a female connector so the connection is idiot proof. There is no detail interface wiring diagram at the instrument end. In my set they had terminated one pair in reverse. The symptom was a high temperature on probe 4 (108 vs. 76 on 1, 2 &3). When heat was applied the temperature indication went down instead of up. I happened to have the correct replacement terminals available so I was able to fix it.

3/31
I flew the plane today after installing a new digital 4 probe EGT system with 1 degree resolution from Electronics International. I installed it in place of the original 1 probe analog EGT gauge I had installed from Alcor. It works fine after I corrected the factory mis-terminated Interface cable but... the cylinder that I selected for my single probe system (cylinder #4) was the leanest/hottest cylinder so I was already using the proper cylinder to set the mixture and by the way there wasn't any significant difference between cylinders. The leaning process is not as intuitive as a fast responding needle going to a peak and dropping off on the lean side. The probes have the thermal couple isolated behind a dome/shield and the response is delayed when compared with the Alcor system.
I will get used to the system but I do not believe it is worth the money. I think just the opposite about EI's 4 probe CHT system - VERY much worth the money.

4/7
Before we flew to Charleston, SC this week I spent some time changing the way I supported the rear of the lower baffle extensions at the cowl outlet.
You may recall I originally supported them with clamps attached to the exhaust pipes. That configuration applied an upward force to the exhaust pipes that I didn't like. This configuration eliminates that condition. In the 10.1 hours of flying to Charleston and back there were no problems, the high CHT was 367F and the true air speeds at 2450 RPM were between 170 and 180 Kts while flying at 4,500 to 7,000 feet (IFR out VFR back)

4/14
I am finally getting back to the cooling inlet plugs. Always in the back of my mind is the question, "What is the best way to proceed with this effort?"
The physical constraints are the most limiting part of the job mechanically the implementation possibilities are many.

I have not developed a reliable set of test results for my cooling drag work to date but I do not intend to repeat everything with my new test method. I just have to say, I think some speed gains have been made and the potential for more have been made possible by the changes to date. Focusing on the cooling inlet plug work, the initial test program was to determine how much I could close off the inlets to reduce the air mass flow through the system and thereby increase the true air speed. My testing indicated that a 2.25"
plug at the inboard edge of each inlet pushed the cruise CHTs right up against 400F even with fairly cool OATs. The testing also indicated that after 1.75" plug width the TAS started declining. I assume this is due to the increase in parasitic drag with the wider plugs overcoming any further decrease in cooling drag.

In this development cycle I intend to start out with the 1.75" width plugs with a reduced drag profile. The distance to extend the leading edge of the plug is very limited and the right side is more limited than the left because of the thrust line offset. When the prop is at maximum pitch it does contact the spinner back plate so that plane is the absolute limit of forward extension of the plugs. The prop blade shape outboard of the hub expands fairly quickly so at high pitch the trailing edge extends behind the plane concurrent with the front face of the spinner back plate so clearance plane is reduced to a small disc.

In attachment Dvc00005 you can see the rough extension glued to the old test plug baseplate. This and the other photos are enough to let you see the direction I'm going with this.

4/17
I roughly finished the inlet plugs today and made a short test flight. The skies were overcast etc. so I could not really test the performance change but it looks promising. The plugs have been extended forward and one of the things I wanted to verify today was the clearance in high pitch situations.
That come out well. As far as speed I did run it at wide open throttle max, RPM and approximately 100 F rich of peak mixture. The speed I saw on the GPS was meaningless but the significant observation was the indicated airspeed was above the TAS window which I haven't seen before. The CHT on my hottest cylinder was 402 F at the end of the climb which is higher than normal and the OAT at 6,000 ft was +11 C (the surface temp was in the 60s F range). Obviously this is something I have to watch. There is quite a bit of finish work to do on the plugs obviously but I need to get some thorough testing in before I go into that part of the build process.