Optimizing fairings for local airflow

[Norman]

The track of the prop tip is misleading you in terms of visualizing the flow. The prop induces a slow roll of the flow field behind the prop. To see the real flow you would need to plot the velocity vectors in the flow field which are much more axial than helical.

You could just mount a little flag poll in front of the canopy.

 

[TFF]

My friend added fairings and pants; cleaned up his parasitic but the biggest change was induced drag by letting the wing fly a flatter AOA trimmed out. Most of the time you will not find 7 mph like he did. If you are cleaning up parasitic drag, the changes are going to be small; they will add up some. Rarely does it change the character of the plane.

 

[PatrickW]

I did some similar testing (of nose wheel pant alignment) with strips of yarn and a GoPro camera.

What clued me in on the fact that something was going on with the airflow was a change in rudder trim after I'd fixed a flat tire (which involved removing/replacing the wheelpant). Additional hints were lots of bug splatters on one side of the wheel pant, but not on the other side.

Later I did a fair bit of tuft testing all over the airframe (Zodiac 601XL), and eventually ended up with an airplane that now tops out 38mph faster than it did during my Phase 1 testing.

Some of these aero modifications make sense, while others don't, or the modifications will make sense only during certain times of flight (for example, does the aircraft fly the same way during cruise as it does on short final? Of course not. So where do you want to compromise?).

Some aero modifications (on my airplane) would have been beneficial from an efficiency standpoint, but not so much from a practical standpoint. For example, one modification I tested made sense from a drag reduction perspective, but would have rendered egress of the passenger very difficult. Other tests proved that areas that I "just knew were a problem" were actually not a problem. So there are suprises sometimes.

Tuft testing is a lot of fun.

There's also the "dirty oil trick", which is just smearing a line of dirty oil perpendicular to the airflow in the region that you're interested it, and then looking at the smear tracks after having flown the plane.

There's a lot of efficiency to be gained with minimal effort. I think a lot of guys could benefit from this sort of stuff. Also, just because an airplane is of the "low and slow" variety doesn't mean that gains can be found. Don't forget to examine areas like inside cowlings or behind radiators (cooling drag), or on the sides of fuselages, etc.

- Pat

 

[BJC]

Later I did a fair bit of tuft testing all over the airframe (Zodiac 601XL), and eventually ended up with an airplane that now tops out 38mph faster than it did during my Phase 1 testing.

That is impressive. Will you share what your mods are, and the benefits of each improvement?

Thanks,


BJC

 

[Orange4sky]

I’m also interested in making aero improvements. It can be very tricky but rewarding. You will run into a lot of naysayers but keep testing and experimenting. I did a lot of work on road cars and even at low speeds, some mods make a big difference.

When you are operating at max cruise, single small improvements in parasite drag may not be perceptible especially since airspeed indicators don’t have much resolution and conditions change. However, many small improvements add up and compound as well since drag is proportional to the square of speed.

I’m looking at reflexing my flaps and ailerons. Maules get several knots just with a few degrees negative flaps.

the biggest change was induced drag by letting the wing fly a flatter AOA trimmed out.

How was that accomplished?

 

[Sockmonkey]

How was that accomplished?

I think if the gear is draggy, it's going to make the plane want to nose down so you need more elevator to trim it out.

 

[dog]

I don't see how it's really any different than the way things are already done. The wing and tail incidence are set to be most efficient at a certain airspeed, likewise the AoA of fairings is ideally set to be 0 at cruise speed. It may well be that the performance gains are too small to justify the expense of doing the testing and fabrication on production aircraft, but I did propose a way to investigate it, and a fairly cheap and easy way at that.

I like your thinking,though this particular idea
seems to ask a lot from local flow around the
struts on a high wing tandem(cub type),the suggestion is also that the prop wash is putting
significant loads on the strut.
From your avitar pic ,its clear that drag reduction is something you know and think
about,so here is my thought in line with what
you started,that the induced drag from the prop
wash spiraling over the airframe is going to strongest at the point where it first encounters
the airframe ,so a small change there would have a larger effect than something further aft.

 

[Map]

How much effect a gear leg fairing will have depends on the drag of the rest of the airplane. If it is an otherwise very clean airframe, streamlining the gear legs can be measurable. If it has lots of other drag sources, the the gear leg fairing drag reduction percentage is too small to be measurable.
I added gear leg fairings on both my airplanes. They are aligned with the overall airflow. I can see from the dead bug pattern and oil flow pattern that the prop slipstream does not change the direction of the airflow locally.
I know that the difference of flying my Pulsar with or without wheel fairings is 4-5 kts at 120 kts.
Gear leg fairings may be worth 1-2 kts.

 

[speedracer]

In the 90's I finished building my O290 D2 powered first Long EZ. The top speed was 175 MPH..... then I got the racing bug. I spent approx. 1,000 hours doing mods to make it go faster. Usually I'd have to do 3 things (fillets, etc.) to see a measurable result. The biggies were adding wheel pants (10 MPH) and more horsepower 135 - to around 170 (GO 435 angle valve heads, 11-1 compression, bigger carb, porting, 1/4" longer 360 rods, etc.). I also found that waxing the entire airframe was worth 5 MPH. IIRC, the CAFE people had the same 5 MPH increase by waxing a Mooney. The top speed increased to 230 + MPH.

 

[Mad MAC]

I can't find the image but I have posted it before from some flight testing of flow angle on an engine cowl. The problem being that propeller related flow tends to be very power dependent (something like 30 degrees movement between taxi and full power).

 

[Old Koreelah]

Not everyone wants more speed; with the extra power I’ve given it, my little Baby already cruises close to VNE.
How far will it glide if the noise stops? That’s a better reason to streamline your aircraft.

I spent a couple of months building fairings around my fat 6" tyres. I never particularly liked the look of those ugly big wheel pants, but they work.
Testing with them on then without them showed a 9% reduction in fuel burn and 4% more speed. The biggest benefit of cleaning up the airflow was my glide rate increased from 7:1 to almost 10:1. One day that might save my bacon.

With range over 350nm, my bladder is now the major limitation!

 

[speedracer]

Not everyone wants more speed; with the extra power I’ve given it, my little Baby already cruises close to VNE.
How far will it glide if the noise stops? That’s a better reason to streamline your aircraft.

I spent a couple of months building fairings around my fat 6" tyres. I never particularly liked the look of those ugly big wheel pants, but they work.
Testing with them on then without them showed a 9% reduction in fuel burn and 4% more speed. The biggest benefit of cleaning up the airflow was my glide rate increased from 7:1 to almost 10:1. One day that might save my bacon.

With range over 350nm, my bladder is now the major limitation!

I can't believe the number of builders who don't incorporate a simple "pee tube" into their plane. Mine tucks under the thigh rest along with a funnel for access. It goes along the bottom of the fuselage, down inside the gear leg fairing and out the bottom of the wheel pant. I plug the tube into a funnel and piss whenever I feel like it. Actually there's two tubes. The one in the backseat sports a "ladies version" funnel. I've spent up to eight hours at a time in the cockpit nonstop (Hood River, OR. to OSH). Also, at altitude where it's usually below freezing I pour a small amount of alcohol into the funnel to mix with the urine to avoid freezing. I was looking at a Lancair with the seat bottom down in a hole well below the thigh rest. The owner mentioned peeing on the fly. I pointed to the seat and asked him how he did that. He looked left then right, lowered his voice and said "I tape the tube to my dick".

 

[stanislavz]

My two cents - Anything high wing and strutted will deny all gain of fairing modifications due to increased load on lift - more mass mor lift generated more vortex drag.

Proper wing tips (like Strojniks for W8 Tailwind) - will gain you much more on simple Harshey bar wing low speed aircraft.

 

[PatrickW]

That is impressive. Will you share what your mods are, and the benefits of each improvement?

Thanks,

BJC

Here's the changes that were positive in the order they were done, to the best of what I can remember or can track down from videos and notes.

Intial top speed from Phase 1 testing: 108mph. This was with the smallest engine (100hp), no aero cleanup.

First group of modifications:
Blocking the flap/fuselage gaps with a piece of L-angle attached to the fuselage, at the point of maximum "up deflection" of the flap (so it does not interfere with flap travel). This is in the area where the width of the fuselage starts to neck down (Chris Heinz himself gave me that idea).
Fairing over the foot step support tubes (these are round pieces of 4130 sticking into the wind under the wing/fuselage junction area).
Sealing gaps in the underside of the wing/fuselage junction, and fairing around the landing gear support structures (both are in the same area under the fuselage).
These all combined added +3mph, which brought the top speed up to 111mph.

No further progress for a couple years. I was just flying and having fun.

Second group of modifications:
Rebuilt the engine, going from .020 overbore to a .060 overbore (Note: I did NOT play with prop pitch during this time, which was an opportunity that I missed. I simply went with what was working for other guys and left it at that. In hindsight I wish I would have done more testing in that area, as I believe I left some performance on the table...)
Trimmed excessive length from bolts and protrusions sticking down under the fuselage into the wind, and taping over some (for example, bolts for the mounting brackets for things like batteries and ELT's that poke down into the wind from inside the fuselage).
Combined +3mph. Now at 114mph for top speed.

A few more years pass, then a lot of changes all at once:
Upgraded to a 130hp engine, changed from a 2-bladed 66" Warp Drive to a 3-bladed 68" Whirlwind prop, changed the cowling, and hung a radiator under the fuselage between the main gear legs (all this was done at the same time). This increased my total weight by 2.5 lbs, but it also moved the CG aft by 25mm.

Before flying my now-rebuilt airplane I tried some math w/respect to speed/horsepower, just to see if I could get a ballpark idea of what to expect. My real-world numbers turned out better that I was expecting (probably a mistake on my part. Here is the equation I used: Vf = Vi (HPf/HPi)^.333). There may be some combination of drag reduction and prop efficiency going on that I don't understand that magnified the impact of just adding 20% more horsepower. I don't think there is any instrument error involved because it was the same airplane, with the same airspeed indicator, and with the same pitot-static system.
Combined +12mph, bringing it to 126mph at wide open throttle.

Other modifications around this time:
I was running up against a maximum rpm limiter at WOT, so I knew I needed to adjust prop pitch. I started doing that in 1 degree increments and graphing the results, eventually arriving at 23 degrees as the best (24 degrees of pitch was just as good in cruise, but made the airplane a lot harder to land because at idle I was still getting enough pull so as to not be able to slow down like I wanted to. An in-flight adjustable prop would have been really fun to test).
Different spinner.
Sam James wheel pants.
Fairing under the fuselage, between the main gear and behind the radiator (this was the single biggest aero improvement of them all, seemed like the airplane ran into a drag wall without this).
Fairing inside the cowl, along the bottom, between the air intake and the front of the radiator to minimize under-cowl drag (this was another really good modification).
Played with the angle of the radiator to the relative wind, to make it easier for the air to "get out" (another good one).
All that combined added another 29mph, topping out at 155mph at wide open throttle.

Straight away I want to say that this is NOT a comfortable speed, and is faster than anybody has any reason to be flying this kind of airplane.

I charted the data I took during this time and posted it here (don't know if anybody other than Zenith builders will have access to this): Airspeed and RPM vs Prop Pitch


Other stuff:
Been flying this airplane for about 9 years now. I don't fly at WOT very often, except in climb. Real-world, flying at about 125 or so is more comfortable in this airplane. Flying "fast" is like driving a pinto at 95mph down the highway - not fun.

I played around with a Gurney Flap on my elevator for a while, but eventually took it off after moving my CG aft (after installing the new engine). I was trying to get more "flyability" in the airplane, as it was always in nose-high trim (can see some of that elevator deflection in the in-flight pictures in the Jan 2015 issue of Experimenter). If your airplane needs excessive trim, then something's off, and there's going to be drag associated with that.

I got lucky when I made a fiberglass fairing to cover the gap in my "split canopy" (my airplane is technically a 601XL, but I installed the canopy for a 650 on it). My intent was to just clean up an area that I thought looked sloppy. But later tuft testing of that area leads me to believe that this strip of fiberglass trips the airflow enough such that it stays attached all the way down the aft section of the canopy and onto the top-aft section of fuselage. This was one area that I expected to see a problem, but did not have one.
Blocking the NACA inlets and installing little "barn door" type vent inlets (no performance gained or lost, but much better interior cooling).
Tuft testing video of the inlet testing (video shows 3 different tests, showing aft canopy flow, as well as aft wing-root separation and a VG test) is here:

I made "wing root fairings". Airflow separation in the area where the fuselage starts to neck down (above the wing root/flap area) has always bugged me. These fairings seemed to make a difference for a short time - a very short time. During the single test of these, the airplane started shaking and I slowed way down. When I landed those fairings looked like a bear had clawed them. The aft edges of them were torn to pieces (fiberglass, probably not thick enough?). I never revisited that, but I know other guys have - and with greater success than I had.

I am certain that VG's are useful in tripping airflow ahead of areas of known flow separation, so as to keep the flow attached or at least delay flow separation. People often just think of VG's as simply being for the wings, but there's more you can do with them. You can see in that YouTube video where there is a line of VG's up the side of the fuselage above the leading edge of the wing in one test. More testing needed there...


More stuff:
Flying "ball centered", and not having to be excessively trimmed. If your airplane is squirrly then this kind of stuff is harder to do.
Expect something to always break, especially on takeoff. Know where that nearby field is that you can glide to.
Slowly go from "known" to "unknown".

While I was making fairings with modeling clay, a guy came over and asked why I wasn't using SolidWorks and a 3D printer. I downloaded a copy of EAA's SolidWorks and played around for it for 3 days, then went back to the modeling clay. Maybe I need to give SolidWorks another chance.


How I could have done better:
Use better instrumentation. I simply recorded my notes by speaking (and having a camera recording). I would like to have data that was automatically logged. I was very limited in what I could detect, and no doubt I missed stuff that could have maybe been a fraction of a mph. Then again, maybe I got lucky on some of that stuff, too. It all adds up.

Some modifications seem to "not help, but not hurt either", and if I had better data maybe I could be more solid on stuff like that. Somebody who has better instrumentation will be able to get better data than I did. I have another airplane that has better instrumentation, and my next homebuilt will have MUCH better instrumentation...


Bottom line:
This was a ton of fun, and I learned a LOT. At the end of the day I ended up with an airplane that can exceed the "fun" part of flying and get into the "scary" part if you're not paying attention. Vne on this one is 160mph, and it starts feeling different around 145mph. Getting anywhere near Vne means you (or "me", anyway) get hyper aware and are watching the ASI more than anything else. It takes the fun out of it, and I'm kind of at the stage of life where I just don't want to push my luck as much anymore. So I keep it reasonable now.

Knowing what I know now, there are some things I wouldn't do, and some things I'd do again. I have another homebuilt project in the works, and I'll be following the plans a little closer and using that airplane for it's mission, instead of trying to shoehorn it into something that serves a different mission. I'll be looking at aero improvements on that one from the perspective of improving fuel efficiency and focusing on passenger comfort more than speed.

Kinda goes without saying that I'm not a pro at this. Just a rank amature who is learning and having fun, probably like just about everyone else here is.

- Pat

 

[BJC]

Thank you Pat. Lots of excellent information there. I appreciate your taking the time to write that.

Tell us about your new project.


BJC

 

[Pops]

I gained 17 mph on my Falconar F-12 with gap sealing on the ailerons, flaps, rudder and elevator. Big surprise was getting to remove the aileron trim tab because of a slightly heavy wing. Then had to keep a close watch on the airspeed at all times. Just a little nose down resulted in a large increase in airspeed and getting close to VNE. ( VNE was 187 mph and it cruised at 150 mph with a climb prop. Have passed VNE when not paying attention.

 

[PatrickW]

Tell us about your new project.

I'm scratch building a Wittman Tailwind W-10. Wings/ailerons/flaps are done. Fuselage is tack welded. Clement mods for those who are familiar with those. It's slow going. Engine still TBD.

I'm looking really hard at building another Zenith (a 750 Cruzer this time) given that it fits a pleasant mission that I share with my wife and local friends, and I think I can build one of those fairly quickly (the kits have come a long ways in recent years with a lot of "matched hole" stuff, plus my prior Zenith experience).

Also have a Grumman AA-5B "Tiger". Super fun airplane. Just put a 204hp LyCon engine on that. Probably more accurate to say I "had" a LyCon engine put on it - I'm not an A&P, so am somewhat frustrated in not being able to legally do some things that are well within my skill sets (and tools). The Tiger has essentially the same mission as the Zodiac, just that the Tiger does everything better, so I'm thinking about cannibalizing the Zodiac for parts (like the engine!) that I can use on the 750.

- Pat