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

So are we seeing in Picture C a cross section of the firewall? The blue dotted line that turns into a black dotted line is some structure that does not exist on a RV-6(A) but if it did should make the cooling air flow smoother?

Note that the blue dotted line is a different shape than the similar rounded piece on a RV-8(A). Is that shape critical?
 
Final fit

Final fit tonight. Ready for flight testing. (Yes I am going to remove protective plastic. )

IMG_20120314_194556.jpg


IMG_20120314_194658.jpg


IMG_20120314_194709.jpg
 
Exhaust collector extension

Thanks for posting my photos Bob Mills. I don't have rights to post photos. Anyone know why and how I get that privelage?

The 4-1 exhaust system was chosen because data from CAFE showed it had the best performance of the many types tested. I also saw the value of one pipe taking up room in the cowl outlet area vs. more smaller pipes. My thought was this would provide the least restriction and I would be able to trim down the outlet area sometime in the future.

The original 4-1 short collector pointed aft and into the RV8 belly outlet ramp. Wiithin 40 hours I found the belly skin cracked at a few rivets. Repaired the belly skin, added foam to reduce vibration and then installed another short collector with an outlet that curved down. I did not like this because it was in the slipstream but I did not like a cracked belly skin even more. After reading the CAFE reports about exhaust scavenging and the Powerflow patent both stated a longer collector provides better scavenging and more power. Four seperate pipes provides zero scavenging and takes up the most area in the outlet. The lower the rpm the longer your pipes need to be for exhaust pressure pulse timing. 2700 rpm is pretty low when compared to most other engines these days. My collector is 18 inches long with a slight downturn at the end. No drumming noise on the belly and no exhaust stain. I have been flying with this extension for about 100 hours and it has not cracked and fallen off yet so next step is to extend the cowl outlet fairing. I was thinking about trying to incorporate an adjustable flap but I will probably just make up a metal extension and see what happens.

I have no idea if my current exhaust setup is optimized for 2700 rpm at low altitude but it seems to work OK.
 
Really looking forward to your test results Dan

Your setup looks like what I have to deal with. My metal from Aircraft Spruce arrived today.

Bob Axsom
 
Here is the picture Mike is talking about:

cowl.JPG

Source: http://cafefoundation.org/v2/research_reports.php > Local Flow II (they must have a direct link policy on their files)



I'm also glad Ed was able to post that...tried over and over to download the .pdf(s) and it drove my computer (and its owner) to fits! :rolleyes:

I was only able to view it on my iPhone (go figure!), and it was hard to correleate the verbiage and pictures that way. But it did lead to some questions, as did Dan's post below...



Note the identified features.

"External flow provides suction".....which translates as "the external flow provides the energy to accelerate the cooling outflow". Put another way, suction = drag. In return you get increased mass flow, i.e. improved cooling.

"internal flow guide, green".....simple enough, no square corners please.

and

"rounded door leading edge"....a nice radius on the bottom of the firewall, again the no square corners rule.

"aft hinged door...to smooth the exit flow".....aka an exit throttle. This feature does accelerate exit mass and reduce both drag and turbulence.

The reference does not attribute anything to the shape of the body shell beyond the suction due to its airfoil shape (the front half). There's no mention of the slashed form afterbody. Seems like if you expect goodness you'll need to incorporate the referenced features.

I'm just sayin'

The diagram still stymies me in a number of ways:

- The light blue leading edge door equating to the firewall radius makes sense.

- I can see equating the dotted extension of that light blue line aft as the upper exit ramp on an RV-8, though I think it is meant to be the "aft hinged door" in the open position, while the hashed line below it is where that door would be in the fully closed position (flush against the bluf body (which I think is the solid black line). Not sure about all that though.

- The green line (internal flow guide), seems to be more like the airfoil shape that Paeser was espousing in Speed with economy. Dan, I know you said "no square corners please", and while I agree, it seems more than that. That seems to be the part that was described in the .pdf as an airfoil shape, designed to accelerate flow towards the exit (the part they described as having its X% chord point right at the exit.

Looking at that diagram, it seems like its using opposing airfoils (light blue and green), just like Paeser, as a convergent nozzle.

I think that the extensions of Gary, Brian, Dan T and Bob are smoothing the exit flow by making it run along the belly a ways, instead of just letting it gush out in all directions at the firewall exit...good stuff, and results are positive.

However, (and this is meant to be constructive) it seems that the convergence piece is missing in those SBS exit extensions. Dan H and Ken do have some convergence. Ken's extension definitely decreases in cross section as it goes aft. If you add the descending exit ramp that the 8 has, his has definite convergence. Dan H, you squeeze the heck out of the space with your smaller and smaller exits, thus there is convergence there. I think you may be the only one to be approaching that 4:1 to 6:1 exit-length-to-exit-diameter ratio that is in the CAFE report. Seems if you extend your exhaust pipe farther back along the ramp, and take your smallest exit back aft of your firewall, you will get even more convergence with the exit ramp (and from an earlier post, I think that may be up your sleeve!).

Dan H, Is that what you meant by your comment about incorporating the referenced features (afterbody, etc)?

So how does a 6 or 7 guy, with no sloping exit ramp on the belly, get some convergence in that exit? Make the exit extension decrease in cross-section as it goes aft? Add an airfoil shape to the bottom of the exit, so it both squeezes and accelerates the exit air? Both of the above? I think that is what Bob is getting at.

I've also considered putting a radius at the bottom of the firewall, and extending it back along the belly a ways with a bit of a bulge to make it a downward facing airfoil, opposite the extended exit floor. The sides of that would have to mate with the sides of the exit extension.

The raked-upwards trailing edges of the sides of the 6 and 7 extensions (those in place and in work) look great (and are required by our turned down pipes being in the way), but I think they actually increase the opening cross section more than we want, due to the angle of the aft edge. My current exit is actually like that a bit too...the bottom is actually forward of the firewall (I need to fix that!)

I could be way out in left on this, but just trying to learn. Thanks for the great discussion...tidbits are sneaking through the thick skull little by little!

Oh, and Gary...sorry to send you into action...but would you rather I told you after he showed the results in a race! ;)

Cheers,
Bob
 
Last edited:
- The green line (internal flow guide), seems to be more like the airfoil shape that Paeser was espousing in Speed with economy. Dan, I know you said "no square corners please", and while I agree, it seems more than that.

It's not.

That seems to be the part that was described in the .pdf as an airfoil shape, designed to accelerate flow towards the exit (the part they described as having its X% chord point right at the exit.

No. The illustrations and text describe bluff bodies, a concept nothing like the subject of this thread. External flow over an airfoil shape generates a low pressure, the maximum low being at 18-25% of chord, and that's where you place the exit opening (a hole in the airfoil, rectangular in the photos) so the low can extract the cowl contents.

281tpgx.jpg


Point is, the exit fairings discussed here have little in common with the bluff body concept in the illustration......but some of the other features illustrated can be applied to a wide range of outlets.

Which is not to say the exit fairing won't have some benefit. I'm pretty sure it will be wake reduction, not cooling drag reduction. More later.
 
I used the bluff body concept outlined in the CAFE reports and discussed here by Dan Horton for the side cowling outlet for my oil cooler. I made a a few versions of what I call a reverse scoop and taped them in different locations on the cowling until I had the smallest one, least parasitic drag, that would keep oil temperatures where I wanted them. I was limited with the rocket on what I could do with ducting the oil cooler inside the cowling. The RV4 style foot well and larger engine just does not leave a lot of room in the cowling.
Basically what it meant was that my although I could reduce my main cowling exit air to quite a small size I did have an extra outlet area on the side of the cowling for the oil cooler.
Ideally I would like to have a dedicated inlet and outlet for the oil cooler but that will likely have to wait for the next aircraft.
 
It is beginning to make more sense to me now (thanks Dan) but I still don't see the "aft hinged door".

My design does achieve the light green line airfoil shape (although not optimized) but your right Bob, we have no convergence. I am thinking about it and I think I have a method. This method may also allow incorporation of the external airfoil shape to some extent. Stay tuned.

Note on Dan T's third photo in the right corner of his outlet along the belly. See that little curved oil trail that indicates turbulent flow?? That is the turbulent area of "reverse flow" that I was trying to smooth that started this all. With the fairing, that will go away and you will see streamlined flow straight aft in this area, and improved flow all around the exit in general.

Good luck with it.
 
Last edited:
Has anyone tried to add afterbody fairing to the down-turned exhaust tips?

If so, did you stop them at the edge of the exhaust pipe or continue them down a bit to attempt to fair the plume, too?

Dave
 
I think that if the fairing and other cleanups work to smooth your exit airflow, you will find that your exit area will be more than adequate even with the four pipes. I am trying to think now of an efficient way to reduce my exit area without chopping up the cowl.

This was from a couple pages back, I was addressing the implied question in his last sentence.

Airfoil shaped (flat on top, curved on bottom) after fairing on the bottom of the fuse-----fit it between the sides of your cowl extensions, depending on size and how far forward, you get to choke off the outlet, but at the same time give it a smooth ramp for the air to follow. ]

So how does a 6 or 7 guy, with no sloping exit ramp on the belly, get some convergence in that exit? Make the exit extension decrease in cross-section as it goes aft? Add an airfoil shape to the bottom of the exit, so it both squeezes and accelerates the exit air? Both of the above? I think that is what Bob is getting at.

I've also considered putting a radius at the bottom of the firewall, and extending it back along the belly a ways with a bit of a bulge to make it a downward facing airfoil, opposite the extended exit floor. The sides of that would have to mate with the sides of the exit extension.


Cheers,
Bob

Bob, yes, this is what I was talking about in my prior post. (quoted above)

I have no actual experience with making/doing this, but I have a decent memory and remember what others who know a lot more than I do have said in the past.

The info in the CAFE reports and other tech info they have seems to dovetail well with the current efforts of you, Dan, Greg, and others.

Hopefully I was able to contribute something useful to this discussion:confused:
 
Jason sent me the files this morning (thanks brudda), so I had the chance to read them in the full scale, easy reading ;) version.

Right you are Dan. The chordwise location of the exit on the airfoil shape and the length vs diameter discussion was on the external surface of the bluff body. Concur with you that the bluff body concept is not in play to a large degree in our outlets...with the exception perhaps of your smallest outlet? Looks like one from the back...did you shape the front-to-rear profile in a bluff-like manner at all?

I've seen Tom Martin's exit bluffs as well, and Mark Frederick has also incorporated them on his Rocket. However, those exits are all on the sides of the cowlings, so perhaps not applicable to the exits on the RVs...though its very much value-added conversation Tom! Makes one think about how to incorporate the concept on the belly...perhaps!

Concur with you Dan, that the flow smoothers (light blue and green) are merely curved surfaces (perhaps, or likely, coanda curves).

The RV's have fairly smooth lines on the bottom of the cowl, so the radius at the bottom of the firewall (or at the leading edge of a shield like Tom showed us) seems like the first step in flow smoothing (for SBS guys). That would be adding the light blue line.

I would think the next step is how to close down flow (throttle the exit), perhaps with an upwards curved surface on the bottom of the lower cowl exit area (adding the light green line), perhaps in conjunction with a flow smoothing extension aft of the firewall (Gary's work) that decreases in cross section moving aft. Lots of possibilities there.

It does appear that proper placement of those curved surfaces will generate both flow smoothing and acceleration (convergence)...if the airfoil shapes are placed correctly and are not too steep (to prevent a "stall").

Gary, I'm visualizing the aft-hinged exit in the diagram as lowering of the light blue line (and its extended dark dotted line) down to meet the green line (thus closing off the exit). Dragging the top of the light blue radius down the face of the firewall, if you will. The exhaust pipe (not shown) would prevent complete closure, but that is how I'm visualizing it now. I don't think the aft-hinged door concept applies to us, as we must do this with fixed-position structures. Dan uses switchable fixed position shapes, and Ken has used a fixed shape with a door to re-open the exit area up to get more mass flow and cooling. Now its our turn to invent some ideas for the SBS, and you have a great start Gary!

On the afterbody concept, David has an interesting idea. Not sure how I'd impliment it, but in Mitchell, SD before Airventure Cup, Larry Vetterman and I were sitting under my airplane and talking exit shapes a bit. He had been working with that teardrop-shaped closed exit after-cowl mod for a while. That day he shared with me that he had played with a teardrop shaped bump on the belly where the air exited the lower cowl. Imagine the same teardrop footprint as that large closed exit fairing, but just a small (1/4" to 1/2" thick) build-up on the belly, with a curved surface at the trailing edge (that's how I pictured it). Larry said that gave him the same speed increase and smooth flow as the large after cowl mod. Seems to fit the concept of an afterbody fairing, and though I never saw any pics or heard more about that mod, it was the genesis of my thought of making a curved radius at the base of the firewall that becomes a shaped bump on the belly, aft of the firewall opening.

Just throwing stuff out there for discussion! Mike, I'm with you...hoping to add value to the conversation, and not just a bunch of blather as I "think out loud"! :eek: Dan, I know between RV-1 and truck stuff you are really busy now, so thanks for the comments earlier (way earlier! :eek:)

Off to the hangar!

Cheers,
Bob
 
Last edited:
Has anyone tried to add afterbody fairing to the down-turned exhaust tips?

If so, did you stop them at the edge of the exhaust pipe or continue them down a bit to attempt to fair the plume, too?

Dave

Not sure if this is what you are thinking, but if you look at some of the exhaust pipes on some of the WWII stuff (F4-F, F6-F) the pipe itself was actually formed into aerodynamic (teardrop) shapes.

Not sure what that would do to the exhaust's main function - scavenging combustion gases, so I won't go there.
 
Last edited:
Remember a video from last year where the drag of a round object was shown to be something like 7 times what an airfoil drag is??

For a down turned exhaust pipe, an airfoil shape via use of an after body should reduce the drag of the pipe section that is in the free stream air.
 
The RV's have fairly smooth lines on the bottom of the cowl, so the radius at the bottom of the firewall (or at the leading edge of a shield like Tom showed us) seems like the first step in flow smoothing (for SBS guys). That would be adding the light blue line.

I would think the next step is how to close down flow (throttle the exit), perhaps with an upwards curved surface on the bottom of the lower cowl exit area (adding the light green line), perhaps in conjunction with a flow smoothing extension aft of the firewall (Gary's work) that decreases in cross section moving aft. Lots of possibilities there.

It does appear that proper placement of those curved surfaces will generate both flow smoothing and acceleration (convergence)...if the airfoil shapes are placed correctly and are not too steep (to prevent a "stall").......

On the afterbody concept, David has an interesting idea. Not sure how I'd impliment it, but in Mitchell, SD before Airventure Cup, Larry Vetterman and I were sitting under my airplane and talking exit shapes a bit. He had been working with that teardrop-shaped closed exit after-cowl mod for a while. That day he shared with me that he had played with a teardrop shaped bump on the belly where the air exited the lower cowl. Imagine the same teardrop footprint as that large closed exit fairing, but just a small (1/4" to 1/2" thick) build-up on the belly, with a curved surface at the trailing edge (that's how I pictured it). Larry said that gave him the same speed increase and smooth flow as the large after cowl mod. Seems to fit the concept of an afterbody fairing, and though I never saw any pics or heard more about that mod, it was the genesis of my thought of making a curved radius at the base of the firewall that becomes a shaped bump on the belly, aft of the firewall opening.
Bob

We are thinking a lot alike here Bob. There are a lot of ways to attempt getting efficient convergence, a lot!! The after body teardrop bump is one of them.

The hard part is deciding on a concept that considers the realities (dealing with exhaust pipes being a main one), focusing on the objectives and then SWAGing on the concept that has the best chance of success. And then going for it.

I have some ideas and sketches and a reasonably good start with my current configuration. Where to go from here??? It'll come to me with more thought and doodling.

At least we don't have to deal with the NLG structure!
 
Bob,
What are the results from the alumimum 'round down' pieces you added in the photo below to break the 90deg firewall edge? Did you notice any difference in any regards?
I experimented with this after looking at some RV-8s on the field, but I tried to do the entire length of the exit area opening. With a nose dragger, it was too painful for what seemed to be little gain. (Meaning, the exit area is so cluttered with the NLG, that I didn't see much potential gain, so I never installed the piece). Your design seems much more user friendly and easier than trying to mold aluminum around the NLG, engine mount, exhaust pipes, etc.

Dan


IMG_5498.jpg


Bob Axsom
 
My aluminum round down

I'm going to have to scan some photos to avoid getting into rambling word games.

Before I forget it Super Bob, Hey Bob! That concept of a curved surface at the bottom of the outlet to pinch of the outflow like a field effect transister doesn't work because of our old friend coanda. I'm not that smart intuitively (or most other ways either) so I tried it - it slows the plane down because the air follows the curve down after it passes the highest point. You want an implementation that causes the air to follow the curve up toward the superwide deflector aft - the bottom of the fuselage.

This is going to take some time so I will go off line and do some work and return when I have it in more manageable form.

Bob Axsom
 
Well I guess I can point out something in the photo already here

That aluminum held off of the bottom of the fuselage by the firewall is about 1/4" by 7" and that is the outlet for all of the air exiting zone 3 (between the engine and the firewall) which is all of the air coming out of zone 1 aft baffle penetrations (oil cooler, blast tubes and heater air - when the heater is off). Before a race I install special plates over all but the oil cooler opening - yeah I'm thinking about it in a curious sort of way because the oil temperature is always around 180. Outboard of this vent it curved baffle between zone 2 and 3 is attached to the flange across the bottom of the firewall and the fuselage skin. The other end of that large baffle has a rubber baffle installed and it is jammed up against the bottom rear part of the engine. It is secured in place and held in contour by aluminum angle struts mounted on the engine mount struts with metal clamps.

I guess the short answer is it slowed the plane down 2 kts.

Bob Axsom
 
On the afterbody concept, David has an interesting idea. Not sure how I'd implement it....

On the exhaust afterbody idea, I don't know how to implement it either.

Question - does the exhaust plume cause extra drag by disrupting the free stream air if it's not parallel to the free stream? CAFE wouldn't be able to measure that with a static thrust test.

....he had played with a teardrop shaped bump on the belly where the air exited the lower cowl. Imagine the same teardrop footprint as that large closed exit fairing, but just a small (1/4" to 1/2" thick) build-up on the belly, with a curved surface at the trailing edge (that's how I pictured it).

If this is the same as the Vetterman belly fairing, I'd think that the rounded front part of it would have about the suction characteristics as a bluff body air extractor. This sort of fairing should have the additional advantage of causing the boundary layer of the exit air to be compressed as it joins the free stream, and help this mixed flow to fair to the belly, thanks to the Coanda effect. (In the Cafefoundation sketch, it would be the light blue dotted line.)

Note that the bluff body air extractor is merely a clean way to increase the suction at the exit. What happens after that, the turbulence in the exit flow, depends on other fairing design.

Dave
 
This was the original configuration that cost 2 kts

You can barely see the curved baffle in the first photo" of course ALL of the openings were closed with high temp RTV:

IMG_0007-3.jpg

IMG_0008-1.jpg

IMG_0009-1.jpg

IMG_0006-1.jpg


Bob Axsom
 
Then I added these and got the 2 kts back plus 4 more kts

IMG_0010.jpg

IMG_0011.jpg

These isolated the area behind the engine from the cooling flow path.

Bob Axsom
 
Last edited:
Further effort and the squeeze turn down failure

I first made this turn back baffle and I really don't remember and difference certainly no improvement.
IMG_0013.jpg


Then I thought if I put an extension on the back of the lower/forward baffle and bring it toward the original curved baffle then turn it down to seal off the area containing the filter air box (FAB) against back flow this could only be good. I did trim the seal for a nicer fit before I flew the test but the speed of the airplane was lowered. In reflection I realized that the exit air was following the nicely curved surface downward forcing a turbulent interface with the air below the airplane. Super Bob that is what I was refering to earlier.

IMG_0012-1.jpg


Bob Axsom
 
Bob,
Thanks for sharing the pics.
Everyone told me that you should see under the "HOOD".

WOW, I have a long way to go.
You have it figured out. I think.

Now I just have to make some mods and test my results.
Obviously all results will vary.;)

SuperBob,
Just put it back together and go fly.
See ya soon.
 
I have big hopes for that outlet faring and the under fuselage bump

I'm going after both of them. Any results yet Dan?

Bob Axsom
 
Not yet!

Sorry Bob,
I still have a few more hours of break in on my overhauled cylinder before i feel comfortable running a test run. I am making two long flights this weekend and should be good for a test run mid to late next week! I am curious myself as I have had this fairing prepared for a few weeks now!

Dan
 
Bluff body for single long exhaust pipe collector?

Since I have a single long exhaust pipe collector (photos in post #41) and nothing else in the way I am thinking about trimming and pulling the factory cowl outlet ramp up as close to the exhaust pipe as possible (similar to Dan Horton's smallest exit), extending it rearward and incorporating an actual bluff body shape. It would look similar to the Mooney cowl outlet photo 'G' in the CAFE local flow control document but located under my feet. One single bluff body placed over my exhaust maybe 6 inches rearward from the firewall.

The CAFE document states "The two bulges on the bottom of the cowling shown in photo A are air exits for the engine cooling system. These 'bluff bodies' are shaped to suck exiting air out of the cowl into the freestream using the locally negative pressure generated by their convex, cambered shape. The suction so generated can be quite strong and is additive to the working static pressure recovered by the inlet".

Many Lancair and Glasair Reno racers have a cowl outlet bluff body concentric with the exhaust collector.

It seems like reducing cowl outlet size (parasitic drag reduction), extending the outlet to rearward (directing cooling outlet airflow) and using a bluff body (help pull cooling air out of the lower cowling) would be the best combination. Any thoughts? Can I have my cake and eat it too or is this a pipe dream?
 
Last edited:

Will refer to this diagram below...

<snip> Super Bob that is what I was refering to earlier.

IMG_0012-1.jpg


Bob Axsom

Bob, as the others have said...man, have you done some work! If I'm following it all, I think you said the only internal baffle mod that added speed was the side baffles at the valve covers. Have you since removed the others, and if not, what is still in place? Couple thoughts for you (all FWIW, and none may be correct assumptions...just for discussion):

The aft curved wall to separate the firewall area looks like a nice flow director, but it does not appear to have a coanda-like curve at the exit that would make the descending flow curve aft and adhere to the belly aft of the exit (the way the -8's radiused curve at the exit is supposed to do). The intersection of your wall and the ramp may just be too angular still. I've heard a 2" diameter is a baseline on that radius, but I've also heard there are a couple guys in Oregon building Super 8s (not hotels! :p) that use an even larger radius, with good success. All hearsay at this point...digging into that rumor too!

Next, I think the ramps you put in the exit may have been far too steep. If you look at Dan's modified diagram from the CAFE report (above), the airfoil shape of the bluff body stops at 18-25% of airfoil chord (at the exit). That is near the thickest part of the airfoil, before the flow separates. If you flip that airfoil over, and put it on the bottom of the cowl exit, I think you'd want it to still be rising at the exit, versus starting back down, as yours did. My guess is you were actually causing the flow to separate ("stall") in the exit, and that's why you slowed down...very turbulent exit flow. Total conjecture on my part though.

So here's what I'm thinking at this point (and Gary, let's see if we are still in sync ;)...others too):

First, I want to show Larry Vetterman's subcowl, or aftercowl mods. These are reproduced from Larry's website: http://www.vettermanexhaust.com/

There are two versions, the first with the teardrop aftercowl, with louvers in the aft-center, and bluff bodies around the exhaust pipes:
vetterman%25201.jpg


This mod was said to increase TAS 4.5 kts.

The second version put the pipes closer to the teardrop aftercowl, and moved the louvers off the teardrop and up to the cowl bottom (where the bluff bodies were in the first version:
vetterman%25202.jpg


If I read the site correctly, verson 2 speed was very close to version 1 speed. In both cases, I think what we see is an extremely throttled exit, with louvers to re-gain lost cooling lost from the throttling, and the teardrop shape is there to straighten the flow (perhaps even make it converge towards the center of the fuselage a bit...or reduce outward "spanwise flow" along the belly...that may be a stretch of my imagination though).

Now, what Larry told me in Mitchell was that he removed that aftercowl, and built up a bump, or thickened skin area, in the same teardrop footprint as the attached area of that aftercowl. He said the speed with that bump remained the same as with the full aftercowl (same benefit as the aftercowl mod). I never did see that confirmed or discussed on his site, and the last discussion of the above pics was in 2009 on his site (a year before I met him in Mitchell).

Below I have two quicky diagrams of what I was discussing earlier. One is the addition of a radius at the firewall that extends along the belly, and then forms that teardrop bump (keeps the light blue line convention from the CAFE diagram). At the same time, an airfoil-shaped restrictor is placed in the lower cowl exit area to throttle the exit and accelerate the exit air (keeps the green line convention from the CAFE diagram). The airfoil does not descend at the aft end, but rather rises slowly all the way to the exit:
Exit%2520mod%25201%2520sm.JPG


This is pretty much a step-one type of mod, and shows a stock cowl exit, and stock exhaust pipes (or at least pipes that protrude down, like I have). Not sure how to resolve the placement and shape of the leading edge of that (green) lower airfoil, nor how to shape the aft edge of that green body at the exit, but the CAFE report did have some info on airfoil type, so more study needed there.

The next diagram is of the same concept, applied to an exit that has one of Gary's (and now others') exit extenders added (red dashed lines). Actually I show the extender as designed a bit differently, as it does not rake aft around the exhaust pipes, but has a decreasing cross-section, and is squared off at the aft opening (inspired by Ken's latest mod). To make the extension like that, one would need run the pipes closer to the belly, and I've tried to show that. I also show my concept of how Paul Lipps told me to cut the pipes to make the sound waves go down, but keep the exhaust plume closer to the belly (the dotted diagonal line...and the lower aft triangle would be removed...just don't know how to draw that!).

Exit%2520mod%25202%2520sm.JPG


Since the exit is extended, the lower airfoil would be moved aft, to get that 18-25% chord location at the exit edge. This version would seem to give more throttling than the first, and hopefully more smoothing and acceleration. Same caveats apply as on the first...not sure how to locate and shape the leading and trailing edges of the green airfoil, and not sure how to shape the aft edge of the teardrop shaped bump on the belly.

All conceptual right now...whaddya think fellas?

<snip> It seems like reducing cowl outlet size (parasitic drag reduction), extending the outlet to rearward (directing cooling outlet airflow) and using a bluff body (help pull cooling air out of the lower cowling) would be the best combination. Any thoughts? Can I have my cake and eat it too or is this a pipe dream?

I'd be very interested in what others think of this too. Dan mentioned earlier that the suction causes an increase in mass flow, and thus additional drag. Perhaps its a balance of getting just enough mass flow to cool, while throttling the exit and smoothing the flow to reduce drag. The bluff body extractor seems to do a bit of both, while maintaining a small wetted area. Might not be free, but perhaps the cake costs a little less to eat that way?

Snip> SuperBob, Just put it back together and go fly.
See ya soon.

Sho 'nuf boomer. Waiting on parts (prop, ram air gear, and reworked plenum) to get the last big projects done...so doing condition inspection prep stuff (new tires and brakes, pulling all the panels, etc, etc) while I wait for the delivery truck(s)!!

Cheers,
Bob
 
My Current Configuration

The curved baffle, the two slanted baffles and the two horizontal baffles are still in. Zone 2 and 3 are sealed away from each other. All the rest were installed, tested, failed and discarded. Everyone is focused on the outlet now as am I. My thought is there are 4 areas to work on. I plan to extend two flow fences on the sides that will probably extend back farther the Gary and Dan's for a reason. I think a constant chord inverted airfoil attached to the bottom of the fuselage and extending all the way to the two flow fences is a good way to go. Extending the bottom of the cowl outlet back will address the brute force turn the air idea. A central vane will help keep the outlet flow straight and separated once they separate atound the cowl support on the "A" model. This also provides a convenient location for installing a semetrical airfoil to reduce the cross section of the outlet. You must recognize that I am latching onto all of you folks good ideas and discarding ones that I don't agree with. It is pretty stimulating.

Bob Axsom
 
A central vane will help keep the outlet flow straight and separated once they separate atound the cowl support on the "A" model. This also provides a convenient location for installing a semetrical airfoil to reduce the cross section of the outlet.
Bob Axsom[/QUOTE]

Bob,
I am running with the idea of gear leg fairings on the supports on
the A. There has to be drag and turbulence coming off those supports.
I just remember watching the film on those round rods vs the faired ones.
I hope this goes in the GOOD pile.

Two small ones on the 45's and One large one on the center with a 12-18"
vane down the middle.

My .05
 
Would it help if people trying these mods added tufts and cameras?

Bob Axsom, you added side pieces inside the cowl. Was there also a piece(s) between the engine and firewall that kept cooling air from going up? Seems like you would have to have a complete barrier horizontally.
 
This is very interesting

It really sounds like we are all working on similar ideas with the central question being which configuration most efficiently achieves convergence while dealing with the exhaust system.

I have several ideas sketched out and will try to get them posted.

I am thinking the same combination as Jason. I just wonder if it is worth incorporating augmentation via bluff bodies if you already have enough mass flow rate. It seems to me that maybe you are going to do work (that must create drag - no free lunch) to augment a flow that is already good enough.

Seems to me the most efficient (least drag) way is to get the flow that you already have (assuming it is adequate) out of the cooling system with the least amount of work being done.

To me that sounds like a convergent duct and getting the exhaust out efficiently and smoothly, directed as closely aft as possible (to aline the thrust produced aft).

I have been thinking along the lines of "what does the most efficiently shaped exit duct look like without exhausts??" Probably a very simple convergent duct working with an underbelly airfoil (the Cafe Green line). Then next is, "how do I integrate the exhaust system with the least disruption to that ideal outlet duct??"

Larry took a couple of interesting stabs at it. My Mod 1 effort is another stab, but none of them integrate all of the best ideas. Ken's design does a great job for a 4 into 1 system, and I'm thinking something similar but with two pipes. KISS
 
Last edited:
I'd be very interested in what others think of this too. Dan mentioned earlier that the suction causes an increase in mass flow, and thus additional drag. Perhaps its a balance of getting just enough mass flow to cool, while throttling the exit and smoothing the flow to reduce drag. The bluff body extractor seems to do a bit of both, while maintaining a small wetted area. Might not be free, but perhaps the cake costs a little less to eat that way?

That is my understanding. A little more drag with the bluff body but net result being less drag because the outlet area would be much smaller and the flow smoothed. Many of the fast Reno guys in the Sport Class have this setup and their exhaust pipe points rearward but hangs out quite a bit.

I think a bluff body desgin would fit on my plane if the leading edge was located where my collector extension starts the bend to the rear. The bluff body would only work at a bend in the exhaust pipe, otherwise it would have to be installed completely below the pipe which means more stuff hanging out in the slipstream.
 
Last edited:
I plan to extend two flow fences on the sides that will probably extend back farther the Gary and Dan's for a reason. I think a constant chord inverted airfoil attached to the bottom of the fuselage and extending all the way to the two flow fences is a good way to go.

Bob Axsom

This is what I am envisioning also.
 
<snip> This also provides a convenient location for installing a semetrical airfoil to reduce the cross section of the outlet. You must recognize that I am latching onto all of you folks good ideas and discarding ones that I don't agree with. It is pretty stimulating.

Bob Axsom

So Bob, is the symmetrical airfoil you are speaking of oriented on its side, with its leading edge being the vertical support of the nose gear? Just trying to visualize it. Sounds like the afterbody concept at work!

And fully agree with your last two statements in that paragraph! :)

<snip> Bob, I am running with the idea of gear leg fairings on the supports on the A. There has to be drag and turbulence oming off those supports. I just remember watching the film on those round rods vs the faired ones. I hope this goes in the GOOD pile.

Two small ones on the 45's and One large one on the center with a 12-18"
vane down the middle.

My .05

Boomer, sounds like you are attacking the NLG struts in the same manner, but with three tiny airfoils, one oriented on its side, and 2 oriented on the 45. Is that a correct interpretation...3 afterbody fairings?

Cool stuff you guys are pondering!

It really sounds like we are all working on similar ideas with the central question being which configuration most efficiently achieves convergence while dealing with the exhaust system.

I have several ideas sketched out and will try to get them posted.

My Mod 1 effort is another stab, but none of them integrate all of the best ideas. Ken's design does a great job for a 4 into 1 system, and I'm thinking something similar but with two pipes. KISS

I do think we are running on the same road. Look forward to your sketches! Also wondering how much a 6 into 1 is going to cost me! :rolleyes:

That is my understanding. A little more drag with the bluff body but net result being less drag because the outlet area would be much smaller and the flow smoothed. Many of the fast Reno guys in the Sport Class have this setup and their exhaust pipe points rearward but hangs out quite a bit.

I think a bluff body desgin would fit on my plane if the leading edge was located where my collector extension starts the bend to the rear. The bluff body would only work at a bend in the exhaust pipe, otherwise it would have to be installed completely below the pipe which means more stuff hanging out in the slipstream.

Good stuff Jason. I keep coming back to Dan H's smallest outlet, but extended along the belly....nice low profile duct is what I'd be thinking. The one think I'm concerned about there is if there is a diameter to length optimization that needs to be considered when lengthening that set up, so you don't create unintended consequences in flow with a long, narrow exit duct. Not sure if there would be any...just noodling it. Us SBS guys have a different set of issues to overcome, but we're all fighting similar battles.

This is a fun discussion! :)

Cheers,
Bob
 
This is what I am envisioning also.

Is that inverted airfoil in Bob's post like the light blue line in the drawings I made up? I think I follow, and agree. Trying to visualize your description...can you draw a stick figure of it?

Cheers,
Bob
 
Is that inverted airfoil in Bob's post like the light blue line in the drawings I made up? I think I follow, and agree. Trying to visualize your description...can you draw a stick figure of it?

Cheers,
Bob

Bob, the one in your drawing gets flat to the bottom of the fuse after the firewall, not what I would call an airfoil.

Look at the photo below, there are three antennas just aft of the duct extensions, one of the antennas has a more-or-less airfoil outline.

A shape something like that, but forward into the duct side extensions, is what I am thinking of. Full width, closed off by actually contacting the side extensions.

IMG_20120314_194556.jpg
 
Now, what Larry told me in Mitchell was that he removed that aftercowl, and built up a bump, or thickened skin area, in the same teardrop footprint as the attached area of that aftercowl. He said the speed with that bump remained the same as with the full aftercowl (same benefit as the aftercowl mod). I never did see that confirmed or discussed on his site, and the last discussion of the above pics was in 2009 on his site (a year before I met him in Mitchell).

Bob,
The version 3 from Vetterman looks A LOT like and probably works similarly to the system being developed in this thread. It includes the extended tapered sides of Dan Thompsons work in post #52 and also the bulged floor depicted in image C of your post #55.

18313491464f6381fae06f6.jpg




He is ahead of us........:rolleyes:
 
Last edited:
Boomer, sounds like you are attacking the NLG struts in the same manner, but with three tiny airfoils, one oriented on its side, and 2 oriented on the 45. Is that a correct interpretation...3 afterbody fairings?

Cool stuff you guys are pondering!



Cheers,
Bob[/QUOTE]

Exactly---Sounds good and I hope it works the same.
Will find out if I ever get to the hangar and get some
stuff done.
 
Yes, that is the curved baffle I mention

Would it help if people trying these mods added tufts and cameras?

Bob Axsom, you added side pieces inside the cowl. Was there also a piece(s) between the engine and firewall that kept cooling air from going up? Seems like you would have to have a complete barrier horizontally.

Yes, that is the curved baffle I mention the side baffles trailing/inboard edge is cut in a curve to conform to the rear curved baffle between zone 2 and 3 in my cowl (I call them that - zone 2 is below the engine and zone 3 is between all the baffles and the firewall). It shows in the two photos in post #70 in this thread during the implementation phase before the openings were sealed with high temp rubber and RTV and before the side baffles were installed. I included 2 photos of it but edge on it is very difficul to pick out. I riveted .016" 2024T3 tabs onto the side baffles, attach them to the curved baffle with #8 flathead screws the seal the two joints with Dow 732 high temp RTV.

Bob Axsom
 
Last edited:
I wonder about the flow in the lower cowl

I wonder about the flow in the lower cowl. I visualize Zone 2 as a big accumulator with a big change in velocity near the outlet. I actually tried 4 configurations in the lower cowl and all of them reduced the volume in zone 2 below what I am running with now and they slowed down the airplane. Obviously I think something related to this is happening to cause the slow down. The fellows measuring pressure differentials might see the connection if it is back pressure related but maybe not if it is an internal air mass velocity thing.

Bob Axsom
 
Last edited:
....one of the antennas has a more-or-less airfoil outline.

A shape something like that, but forward into the duct side extensions, is what I am thinking of. Full width, closed off by actually contacting the side extensions....

It might be profitable to make this shape so it can be moved fore and aft on a test basis. There's probably an optimum position for it.

Dave
 
It might be profitable to make this shape so it can be moved fore and aft on a test basis. There's probably an optimum position for it.

Dave

I would bet money there is an optimum position, and a side bet it is speed dependent also.

Optimum size and shape too.

That is what theis expiremental thing is all about aint it??
 
Where's The Energy?

Let's think about the matter of using the exhaust for augmented cooling.

Without augmented cooling, the energy it takes to pump the cooling air through the engine comes from the ram air coming into the cowl. That air flow occurs only because the airplane's moving forward. To move forward, the engine turned the propeller and the prop converted that energy into thrust. The thrust is balanced by the total drag, some parasitic, some lift-induced and some cooling drag.

This means that the engine has already created the exhaust that's right now pulling the plane forward. There are engine efficiency losses and prop efficiency losses, and what's left, the useful work, is what you wanted.

The cooling air flow comes out of what's left. It's paid for with prop efficiency (roughly 80% more or less) and since I don't know the engine efficiency I won't give an estimate for that.

We know that the exhaust has useful work left in it. We know this because it's hot, it's moving fast, and it's noisy. Some of that energy can be recovered by converging exhaust nozzles, and if you had them, some could be recovered by using a blow-down turbine. This is what turbochargers use, another example of using the energy in the exhaust that's otherwise wasted.

This energy can be used to pump cooling air through the engine with an augmentation jet. It might be one of the simpler and lighter ways to recover a bit of that energy.

Any energy that's used for that purpose is energy that doesn't need to come from the prop pulling the plane forward. It seems that the plane must therefore go faster (reduced cooling air flow drag), once the inlet and outlet ducts are optimized and the cooling flow path is made reasonably efficient.

And there we are - don't just throw that exhaust energy away, use it!

Dave
 
Bob, could you just siameeze the existing dual pipes into one??

Probably not without messing with exhaust flow balance, though with my non-crossover 6 into 2, I have no idea if there is any balancing going on. Gotta fess up to knowing very little on that...just beginning to look at and study that piece. Open to your thoughts, as I know you are an engine guy!

Bob, the one in your drawing gets flat to the bottom of the fuse after the firewall, not what I would call an airfoil.

<snip> ...one of the antennas has a more-or-less airfoil outline.

A shape something like that, but forward into the duct side extensions, is what I am thinking of. Full width, closed off by actually contacting the side extensions.

I'm with ya Mike, and that is where I started as well. Then reading the CAFE report gave me pause on it (though it could be a poor interpretation too). I'm still trying to determine if the airfoil should go past that 18-25% chord point at all...in other words, should it "camber" up, then back down (or in the case of the inverted airfoil, "camber" down, then back up. The bluff bodies seem to not do that...just not sure if that "rule" would apply to the surfaces we are talking about here. My concern with it is tripping laminar flow, by having the airfoil too cambered. (Calling Steve Smith! :))

Bob,
The version 3 from Vetterman looks A LOT like and probably works similarly to the system being developed in this thread. It includes the extended tapered sides of Dan Thompsons work in post #52 and also the bulged floor depicted in image C of your post #55.

He is ahead of us........:rolleyes:

Well check that out! Larry's bump has walls! I hadn't pictured it that way, but it makes a lot of sense. So we have a shape that would appear to do what Gerry's trident does (the flow smoothing of the vertical walls) and has a faired afterbody shape to it as well (the teardrop). It does look like it has a cambered shape, but in looking at it, it may be the curve of the walls and the reflection that give that cambered appearance. Not sure...and to be honest, I'm not sure what effect a curved surface that far aft would provide...maybe some acceleration (if flow does not trip); but I don't think it would provide much venturi or nozzling effect on the exit there...more noodling to follow!

Bil, any idea what performance gains (if any) were noted with this version 3?

Stealing pictures from Ken's exit drag testing thread, for comparison and discussion:

Ken's:
kens%25201.jpg


Gerry's:
gerrys%25201.jpg


Larry's:
vetterman%25203%2520sm.jpg


Tom's:
toms%25201.jpg


Commonalities and differences:
- Ken, Gerry and Tom each have a coanda radius (Tom's is farther forward than the others). Not sure about Larry.
- All have vertical walls of sorts, to smooth flow. Tom and Ken's are part of a convergent duct, while Gerry's is not, and Larry's is convergent side to side, but not top to bottom. So exit throttling is perhaps more active for Ken and Tom...not sure. I'm wondering if Gerry's walls could be extended past the pipes, and closed off a bit at the aft edge in some manner, to mirror what Ken's looks like at the aft edge.. Then again, maybe Larry's performs as well with just that teardrop shape (really wondering how that version 3 performed!)

Fun to contemplate all that!

Is that a correct interpretation...3 afterbody fairings?

Exactly---Sounds good and I hope it works the same.
Will find out if I ever get to the hangar and get some
stuff done.

Pictures please when you get there...guess what I've been contemplating for my tailwheel spring! ;) May have been tried...

Let's think about the matter of using the exhaust for augmented cooling.

This energy can be used to pump cooling air through the engine with an augmentation jet. It might be one of the simpler and lighter ways to recover a bit of that energy.

And there we are - don't just throw that exhaust energy away, use it!

Dave

Dave, the exhaust-augmented exit seems like a good idea, and Paeser talked a lot about it in Speed With Economy (placing the exhaust opening within and between the top and bottom ramps of the exit, and shaping those ramps as a venturi). Then again, as Jason pointed out, CAFE seems to state that extending the exhaust pipe well beyond the exit yields positive results. I've been mulling over both...seems they are at odds with each other. Any thoughts on that (from you or others). Kinda scratchin' my head on that one right now.

Man...hope I'm not using too much multi-quote madness here...just so many good posts to discuss, and don't want to make a gazillion posts to touch on each!

Cheers,
Bob
 
I did some testing with exhaust augmentation a few years ago on my blue HRII. I used information from a 1942 study done on this issue. I forget where I got it from but it was government stuff done in Mississippi. They were looking for the best way to get the exhaust out of a radial engine. You can see the results of this work on many of the old radials, some have very short exhaust stubs enclosed in curved ducts. The piper comanche is an example of a "modern" aircraft that uses augmentation. The experiments tried all sorts of exhaust pipe lengths, augmentation tube lengths and shapes. Results were positive and I constructed an oval tube about 12" long 10" wide and 3" deep, which should have been optimum based on the data. This tube was mounted on the bottom of the fuselage immediately aft of the cowl outlet. I had some crude ducting inside the cowling, from previous experiments, which should have provided a good transition. The exhaust pipes were dumped directly into this tube, about an inch aft of the cowling.
On start up the sound was much different, like a six cylinder Harley, it was cool. A flight quick flight test showed no speed increase but climbing cylinder head temperatures. On landing I noted that the plastic ferrel on my com antennae, mounted a couple of feet back, was melted off. The tube is now located in the cupboard of chaos. I keep this stuff to remind me of what did not work!
Clearly exhaust augmentation works on some aircraft. Some of the really fast SX300s, Lancairs etc use this type of system. However it is usually internal in the cowling and surrounded by a smooth flowing stainless duct.
It is just possible that these planes are fast due to other factors and not augmentation. The piper comanche is certainly not fast!!!
 
Last edited:
Has anyone talked to Larry Vetterman?

From a casual observer's point of view it appears that Larry Vetterman has already done a lot of this... Has anyone talked to him about his results? I have had on occasion the opportunity to chat with him in the past about all things exhaust (and even oil cooling) and found him to be very willing to share his ideas .
 
Larry Vetterman Consultation

From a casual observer's point of view it appears that Larry Vetterman has already done a lot of this... Has anyone talked to him about his results? I have had on occasion the opportunity to chat with him in the past about all things exhaust (and even oil cooling) and found him to be very willing to share his ideas .

Good suggestion. Yes several of us have spoken to Larry Vetterman directly face to face or through various communication media. He is a valuable resource because of his long term experience in the exhaust business and because of his unique experiments and tests. He has shared some of that in this forum.

The SARL racing season for 2012 starts 2 weeks from today in Sherman, Texas and continues for more than 20 races all over the country. A few of the people posting in this thread are trying to come up with that little difference maker on our airplane that will up our position at the finish line. Larry has never flown his airplane in a SARL Race but this may be the year with a race scheduled up in his neck of the woods in Idaho. It would be good to see his latest version tested in competition. At www.sportairrace.org under Calendar of Events:

Pappy Boyington Memorial Cup
18 August, 2012
Coeur d'Alene, ID (KCOE)

Bob Axsom
 
Last edited:
Back
Top