What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Forming aluminum for a cooling outlet fairing

An old report repeat on antennas and current status

When Jeanine and I built our RV-6A 1996-2004 our intention was to race in whatever cross country air races that came up. Usually one per year until SARL was created, now we have over 20 per year (www.sportairrace.org). With that focus during the original build, I contacted the manufacturer of the best looking "in the wingtip" antennas. When I told him that I fly IFR he told me that the antennas were intended for sport flying applications and he would not recommend them for serious IFR flying. That was a NO GO decision driver for me.

I could make cover plates for all of the external antennas, two communication, transponder and marker beacon (I already take off the navigation elements by unscrewing them), remove them for racing on site and reinstall them before flying home. The 12" stock wing tips have been replaced by a 3" streamlined tips of my own design and the performance of the com antennas placed in them for turn and finish line calls would be further compromised. I would have to route a coax to at least one wing tip and change the RF I/O connection to the race only coax someway (change connections, coax switch, etc.) disable the transponder (you don't want transmission attempts into an unloaded output, maybe some FAA problems there as well - depending on race location perhaps). Still, if I run out of more promising and less compromising opportunities the passive drag of external antennas will be kept in mind - I look at them and wonder each time I wash the belly.

The speed differential between out RV-6A and the current RV Blue race leaders is from 3 to 9 kts and I think I am going to have to deal with the thrust member of the speed equation to overcome that. Two of them will be in the AirVenture Cup Race next Sunday and we will see what we have accomplished so far. I am hoping this engine cooling air outlet mod cuts into the margin.

I completed a trial "final installation" of the new one piece cover this morning at 05:40 and there are some problems.

IMG_6157.jpg


The 0.040" 6061 0 material I had available and used, is very soft. It is just a little harder to form than sheet lead so it shapes nicely but its softness causes mounting problems. I knew this from the 0.032" 2024-0 material used in the development cover. I planned to drill large mounting holes and use #8 hardware with dimpled washers to provide more distribution of the fastener head compression but I went too small at 1/4" dia and the washers are not all flat because of imperfect hole alignment. I plan on removing it today, increasing the hole size to 3/8" like I do on fiberglass and plastic parts installed with dimple washers and see if that works. If that is not satisfactory I will go to harder material for a new cover and cure the problem that way. Since I plan to leave for Mitchell, South Dakota for the race Friday morning there is some schedule pressure so I guess I had better eat, get cleaned up and head back to the hangar.

Bob Axsom
 
Last edited:
3/8" mounting holes worked fine

The one piece cover is reinstalled and the dimpled washers are sitting the way the should. I would have liked to see if there was any effect on speed from the cleanup but I'm out of time. I've got a lot of prep work to do before flying to Mitchell, SD Friday fot the 2012 AirVenture Cup race - one secret weapon - so no more effort on the cooling air outlet fairing. I may test it later just to get a new baseline but I think this one is done.

The secret weapon, well it is a strategy thing - tell you later Alan.

Bob Axsom


P.S. I'm calling this configuration speed 183Kts speed which is 1.4 kts less that I got one time in testing a while back so it is hard to claim an improvement in performance. I did two tests in this series (20 and 21) with no fences, no covers and no bumps which yielded 177.6 and 178. 8 knots respectively for an average of 178.2 KTAS which would indicate a 4.8 kt speed gain but the three long angle mounting rails were riveted to the fuselage already and were exposed to the air in tests 20 and 21. S-o-o-o-o, conservatively I'll take the pre-mod baseline speed as the average of the 184.4 and the 182.2 or 181.3 kts. That still yields a gain of 1.7 kts. I'm happy with that. 04:30 and I'm on my way to the hangar to do race prep.
 
Last edited:
7 Degree NACA Ramp

I have spent many an hour trying to locate the source study for the sugggested optimal NACA exit ramp slope to be 7 Degrees. I can't find it. Since it came up in this discussion I thought I'd ask here.

Does anyone here know the source and have a link to it?

Thanks in advance,
Ned



Some very thought provoking discussion! I know Bob is gearing up for the Hill Country Air Race tonight, but this will give him some great post-race reading material...to bounce his results off of!

Comments below:



Paeser postulated the best angle for exit air was at 10 degrees to the free stream...not far from the NACA 7 degrees...though it may be a slightly different concept at work (not sure).

There may be some acceleration of the air on the first half of the bumps, but I would be concerned about separation on the aft side (as you and others have mentioned). There may be some coanda effect, but I doubt laminar flow can hack that angle, and it sort of has a venturi-like look to it. Only down side is the back half of the venturi would expand and slow the exit air...opposite the goal. (more on that below).

Also, I think slowing the freestream air locally would be counterproductive...I know you're talking about makiing the flows more compatible, but it would seem slowing anything in this area would be drag-producing. Just a comment for discussion...this is all facinating stuff! :)



Dan, I think there is probably some area reduction...at least compared to Bob's previous configuration, without the bump. I do see your point about re-filling those cutouts (though I'm concerned about blocking the exhaust directly and causing odd flow patterns downstream), and I also wondered about exhaust stains on the bump. Looks like it is there, but the swirl pattern may indicate separation or turbulent flow.

My pea-brained effort at a similar concept to what your last paragraph says is pictured below (a mod of a picture I did earlier):
Exit%2520mods2sm.jpg


From discussions with Paul Lipps, the coanda radius at the base of the firewall might do the lion's share of redirecting and attaching the flow exiting the cowl. Then the continuation of that curve...a softened version of Bob's current bump, as others have suggested) might maintain laminar flow and continue that coanda effect further aft. Opposite that upper bump, on the lower surface, another bump that forms the front half of a venturi could add to the area reduction and exit flow acceleration. Tuning of the system would mean trying to direct the flow aft without reopening of the venturi prior to the end of the ramp (I think)...which I don't depict well in this picture. I do agree with you that decreasing the down-angle of the pipes might increase Bob's current coanda effect, and might allow the aft part of the bumps in my picture to be more aligned with the belly and the freestream.

My kluge of a bunch of stolen ideas, somewhat applied to what Bob is currently working...all FWIW! ;)

And great work Bob...as always! Good luck in the race!! :D

Cheers,
Bob
 
"A" Reference, NACA-TN-2888

I have remembered the 7 deg angle since college, but never asked for reference. I searched the nasa database ( http://ntrs.nasa.gov ) and found this report: Search for this title, "NACA-TN-2888". I cannot add the link as it will yield an error. I just scanned the report, but it appears to show experimental development of pressure recovery for a 2-d diffuser. The efficiency appears to peak right at 7 degrees. ref figure 15. The report is 85 pages long so only 30 min was spent evaluating. Happy reading.
 
The "NACA" Inlet

I was the one that mentioned the 7 degree angle much earlier in this thread.
It is for INLETS, not exits, but the principle I was discussing at the time was the same. My original post was concerning expansion or channels with increasing volume. Depending on conditions, there is a maximum expansion rate that can be done before the air separates from the structure. Previous experimentation by NACA can help.

A poster asked for a ref document. Using the search method above, look for NACA-RM-A7I30. That is the study I based my commentary on.
 
Bill and Bill,
Thank you very much for your responses. I've read both those papers before but neither are specifically defining the design of NACA exit ramps. I remain quite curious. I would like to find the study explaining and defining the design. I've seen posters quoting the 7 degree slope and that the NACA papers say a straight side shape for exits is better than the curved shape formulated for inlets. I've spent literally days reading the NACA studies to no avail. One fellow Mark Langford http://www.n56ml.com/nacaducts/index.html experimentor made his exit ramps based on the "earlier" NACA studies and later posted that the "later" NACA studies indicated that the inlet design is not optimum for the outlet. When I ask him for references he vaguely suggested to look in an old Kitplanes. I eventually found the article he remembered and the author did respond to me with about 40 references none of which had the info on the exits. I'm wondering if it's some kind of urban legend perpetuated by the Internet....
The author was Todd Parker and his article was posted in Sport Aviation March 2006

So if anyone knows where the exit ramp design development for the NACA ducts I sure would appreciate the info
Thanks
Nec
 
What is your understanding of the 7 deg. optimum outlet?

What is your understanding of the 7 deg. optimum outlet? If you understand the concept it would be possible to begin an experiment and test process based on that.

Bob Axsom
 
I'm wondering if it's some kind of urban legend perpetuated by the Internet....
The author was Todd Parker and his article was posted in Sport Aviation March 2006

So if anyone knows where the exit ramp design development for the NACA ducts I sure would appreciate the info
Thanks
Nec

There is plenty of exit geometry testing by NACA. I cannot remember seeing an instance of them testing a reverse of the curved sidewall inlet. Maybe they did, I just did not see it. It is not a sidewall shape I would automatically try for an exit, due to my belief that the sidewall geometry is directional for generation of edge vortices spilling into the inlet. I don't see that working in reverse. There are some BASICS that can be drawn from theory and testing. A converging ramp, such as the RV8 exit floor has a large range of functional angles because it is in a positive pressure gradient (my assumption by inspection, not data) which keeps the flow attached. The backside of Bob Axsoms coanda bump was an expansion area with negative pressure gradient. The exhaust/oil pattern showed separation, a small piece of data. The NACA INLET is designed to minimize separation in an expanding environment. The 7 degree data is VALID for this narrow application (expansion) as suggested. I had additionally suggested convergence of the cowl at 7 degrees (trying to minimise separation on the cowl exterior and matching flows to reduce the disruption at the rear edge of the cowl) in order to reduce the volume expansion past Bobs coanda bump inside the cowl. As Mills said, that could be counterproductive. I'm not sure. The 7 degree common aero knowledge is hardly an urban legend

EDIT REF: "It is empirically observed that the best divergence angle for a two dimensional subsonic diffuser of a given area ratio is about 7 degrees" Hill, Peterson - MECHANICS AND THERMODYNAMICS OF PROPULSION. Addison Wesley Pub 1965.
 
Last edited:
Page 3 of Dave Anders notes
http://sacrvators.com/Aircraft Efficiency N230A.pdf

3) shape - NACA studies
- straight sided rectangular shallow angle converging ramp - ramp as parallel to free stream flow as possible at end
- width to depth ratio of ramp
a) 7 to 1 for cooling systems
b) 1 to 1 for exhausts (augmenter

Evidently Anders knows about the NACA studies?

The RV8 outlet and the Grumman Tiger/Cheetah outlets have straight sided exit ramps. The Grummans have some of the ramp covered. It would be great to know how to design it.

If there is a study that NACA did then I'm thinking a whole lot of experimentation has already been done. Why repeat that? I want to experiment with the benefit of the science already concluded guiding my efforts
 
I'm curious why RV's have the "exit ramp" out in the slip stream. The Grummans picked up quite a bit of speed when the "exit ramp" was placed internally. The Grumman Traveler exit ramp looks like an RV 6 until 1975 when they placed it "inside" the cowl. It picked up 6 knts. Bob Axsom have you got room to move all that "stuff" inside the lower cowl skin and make the skin even with the fuselage skin, not hanging below?

Dave Anders believed whatever NACA data he had enough to put the ramp inside the cowl AND to get the ramp slope down to 7:1 he modified the bottom of the fuselage to continue the ramp way back.......

Dan Horton picked up more speed the more he removed the hanging down part......
 
Last edited:
The RV8's exit slot (after trimming the cowl) is about 7 to 1.

Version 4 of my own cowl exit will again reduce fixed area by extending the exit 4" aft of the firewall. However, it will also incorporate some variable exit area for low airspeed use.
 
Correctly assessing or quantifying improvements I have found to be very difficult. I discovered it is extremely hard to get repeatable results with flight testing for a particular mod. So I understand Bob Axsom's efforts in this thread. He is an extremely hard worker. And I don't want to take away from all he is doing and has done. But here are so many variables it makes it really hard to determine if what's done really is what causes the speed increase. For example, in this thread you one time mention that CHT's went up a little. To me this says cooling air went down. The logic flows that the stuff put at the exit may have caused lower cooling airflow which in turn may be contributing to the speed increase. So how much speed increase is do to that versus the effect of the changes with the air exit? Well that's the whole problem. And until you properly instrument the cowl you just won't know whether all your work actually increased the speed by itself or it just affected another variable that actually caused the speed increase.
 
I've been away a while

I worked in aerospace for 50 years from the lowest level to the project office and I recognize wisdom were I see it I think and you guys have it. I just flew down to Abilene, TX yesterday and worked most of the night with changing winds aloft forecasts and race course changes including one added turn that was published after most people went to bed with the wrong latitude value. What a great group - We worked together and and got by what could have been an awkward situation. I woke at 00:50 CDT this morning and saw the e-mail announcing the added turn and started working it into my race flight plan and altitude strategy (winds). I developed a crude Excel spread sheet last year for helping me plan what altitude I will fly in each leg. The race is a real race not one of those compromise things with handicaps or fuel burn estimates, etc. The fastest wins and that tends to stimulate speed modification action. MANY speed modifications do not work even though the experimenter's intuition says it has to work. I had 3.4 hours sleep last night (yes I maintain a sleep log) and a very stressful day including some serious scud run challenges through the Ozarks to get home here in northwest Arkansas so bear with my lack of clarity - I'm tired but I'm trying.

When you try to exploit a concept I believe you must work with it until you succeed or you are convinced that it will not work and have some rational thoughts about why it failed. If you learn something from the failure it was a good experiment. The cooling drag concern is popular with the technically elite so it must be worth looking at and trying to reduce it must be a good thing. I have a significant stack of aluminum, rubber and stainless steel from my failures. Not what I have read or been taught in the classroom but what I have fabricated installed and tested. As I have gone through this growing and satisfying experience I have accepted intellectual input on test methodology from some of our best contributors to this forum and from lessons learned when I used to be a useful member of the aerospace community but in every case I control my experiments and I do not bow to technical arrogance and I judge the results. If you don't do that you will be defeated early in you life as an experimenter.

My initial thought on cooling drag was the cooling air should have easy passage through the cowl to the fixed outlet. I made a curved aluminum panel from the bottom of the engine to the bottom of the firewall so the air would follow the curved surface to bottom of the fuselage. THIS IS NOT A SIMPLE THING TO CREATE! All of the things that transition from the airframe to the engine have be accommodated by this thing you are constructing from many pieces and it has to be serviceable and the penetrating wires, tubes engine mounts, etc. have to be accessible for maintenance. OK ... ? Anyway, I did that and as I was doing it I said to myself, if I just let the air follow this cowl wide curved panel to the bottom of the firewall/fuselage the air outboard of the outlet goes into those lower corners and gets trapped there or at least does not have a equal shot at the outlet. So, I decided to make vertical panels that slant from the inner side walls of the lower cowl inboard at an angle determined by the rearmost location of cylinders 3 and 4 (of course they are staggered so the angles are different) to the width of the outlet. When all of this work was done and tested I found the the top speed of the airplane had dropped below my base line speed of 170.67kts by 2 knots. I made many modifications and nothing worked until I sealed the lower cowl with two additional horizontal baffles below the cowl split line preventing a back flow path from the lower cowl up past the valve covers and back to the area between the engine and the firewall. When those baffles were installed the speed jumped up 6 kts for a 4 kt gain over the baseline for my specific airplane.

I have made many changes since that was achieved but NONE did anything but lower the speed and when the change was removed the speed came back. I believe that a near optimum combination of upper plenum and lower cowl configuration has been achieved for my specific airplane with significant additions of aluminum and rubber to my sacrifices to the God of Speed. Some new idea may arise in this area that would make the plane faster but I personally have tried enough that I do not believe they will.

After some time I thought the outlet configuration was not effective based on tuft testing by others that showed back flow. A fellow named Geer, from Pennsylvania I believe, offered in this forum and experiment that he had tried with the outlet that he reported increased the speed of his plane. I did not want to just copy his shroud design and gain limited knowledge - basically GO/NOGO. I saw several areas to work on but I'm getting tired and I suspect you have long since stopped reading so I will jump ahead to the final configuration.

The bumps are now stainless steel panels that are attached to the curved baffle from the engine to the bottom of the fuselage so now the panel extends from the engine past the bottom firewall/fuselage interface and gently curves back to the fuselage with a narrow gap at the trailing edge to pass the air that comes through the oil cooler, blast tubes and heater dump from zone 3. There is a center fin that keeps the exhaust pipe outputs separated, outer vertical panels to keep the the cooling air and the ambient air separated and a horizontal panel that does not let the outlet air travel vertically downward into the ambient air under the fuselage without starting to travel aft toward the tail. The vertical and horizontal panels evolved through many trials and are now actually one piece. This all made sense to me but the real breakthrough came when someone in this forum suggested that I should try parabolic like cutouts between the the horizontal panel and the two outboard panels. After I did that the speed jumped over 1.7 kts. The cutout is not unlike the plume pattern of the exhaust on earlier versions of this outlet shroud. The original wide coverage of the bottom of the fuselage with oil and exhaust products is now defined by straight bands the width of the outlet all the way to the tail - no evidence of turbulent flow at all. Refinements in the implementation can be made but I do not believe there is anymore speed there. Wow that was tiring - now the system will probably say I used too many words in a post.

If you don't try to test because you say it's is too difficult to get consistent results that is too bad. I wish you would try harder to control the testing otherwise you will always be stuck with hand me down knowledge which is good but not the best kind. You have to work at it to really know the facts on the best level.

Bob Axsom
 
Last edited:
Plenty of us are still reading Bob. I've seen the improvements you've made and continue to make. You are a true Experimenter. Keep up the good work
 
Im another one who reads every post you make Bob. I am not building a race plane, but I sure enjoy how you keep searching for that one more knot! I would love to see pictures of your modifications, but since you have spent so much effort making them, they are your "advantage" in your racing!

Keep it up Bob, many of us here and one upstairs, are surely following your progress and racing successes.
 
Thanks Bob. I will likely never get the racing bug but your test methodology has been incredibly valuable for me. I too read your technical posts carefully and completely.
 
Speed Accuracy

Hi Bob,

I too read every one of your posts even tho I can't fly right now because of a panel overhaul. This question has prob been asked before but a quick search gave no joy.

How do you assure accuracy of one speed test to another since you're looking for a knot or two at a time? I've tried to repeat testing using the same altitude/rpm/MAP even on the same flight using the 3 or 4 heading GPS method and have not been consistent to within a knot.

What am I doing wrong?

Thanks, no hurry with a reply...

Jerry
 
Maybe nothing you are doing is wrong Jerry

USARHP_zps6af057bc.jpg


The flight procedure I use is the 2005 USAR handicap procedure image above. Once at wide open throttle, max RPM, leaned for max power and trimmed for hands off level flight at 6000 ft density altitude for EVERY TEST OF SPEED I engage the autopilot heading and altitude hold and start recording GPS ground speed for the 360 deg. GPS ground track until I have 5 consecutive 20 second interval recordings that do not vary by more than one GPS ground speed knot then I turn 120 degrees and repeat for the next leg until I have completed 3 legs then I return to base with my data and plug the valid leg average GPS ground speed for the five recordings for each leg and the GPS ground track for each leg into the National Test Pilot School (NTPS) Excel spreadsheet that was provided in this forum. Embedded formulas compute the wind experienced during the test and the true air speed of the airplane.

NTPS_zps3c35bcd0.jpg


Originally I used the handicap procedure alone bur Kevin Horton convinced me that that procedure would always error on the side of lower speed. I verified that in actual use and have used it with the NTPS excel file ever since. John Huft also supported the use of the NTPS spreadsheet and recommended multiple runs before accepting a result as accurate. I verified again that the suggestion was correct for refining the estimate but there are practical time and money limits to my resources so I now fly two triangles, one right after the other, and average the two estimates. The estimates are always very close but this approach dampens out the variations.

I now fly my tests under a MOA east of the Fayetteville, Arkansas's Drake Field and 6,000 ft density altitude is often around 4,000 ft MSL and ~2000 ft AGL over the Ozarks there. Wind and thermals during tests in non calm conditions occasionally cause a speed excursion outside the acceptable range of the USARHP and I compromise my standard by recording the observation and continue with data collection until I have a five good consistent recordings around the one anomaly and strike it from the list when I get back on the ground.

A few more more important lessons learned:

Try to minimise variations in the aircraft under test that are not the item of interest in the test (Fuel load, aerodynamic fairing installations, tape, vent position, etc.).

Always record the date of the test on the raw data collection sheet.

Always record what is being tested on the raw data collection sheet

Always print the NTPS spreadsheet for the test and staple it to the raw data sheet for the test.

Always File the paperwork.

I hope that helps.

Bob Axsom
 
Last edited:
Back
Top