Exhaust drag question

[hecilopter]

Still trying to figure out differences in 2 RV-7A's. Cooling problems are now under control, now to a speed issue. One plane is about 10 mph slower than the other. I have several other threads about our 2 planes, but for all practical purposes, they are identically cowled, baffled, engined, propped, faired and equipped except for the louvres on the bottom of the cowl and removing about 4" from the bottom of the cowl where the exhaust exits due to cooling issues.

Has anyone noticed or experimented with the exhaust pipes and the angle with which they exit the cowl? I noticed that one difference between our airplanes (both equipped with vetterman exhaust) is that my pipes exit the cowl and are parallel to the center of the plane. The pipes on the slower plane are angled pointing to the center of the plane and at each other.

Could this make a difference with drag from the angled pipes or maybe exhaust thrust that is not aimed inline with the airflow? We are going to experiment and twist the pipes to be parallel like mine and see what happens. Just wanted to check to see if anyone else has done this and what the results were.

Thanks!

 

[Bob Axsom]

No experiment but ...

It seems unlikely that the inboard canted run of the pipes would make any difference but the cowl modifications sound like a significant difference to me. John Huft went to some trouble to extend his lower cowl exit surface to gain speed (he has a VERY fast RV-8). Cutting it off and adding louvers would be expected to slow the plane I think because both tend to turbulate the flow and reduce directional control. Back on the exhaust pipes. I also have Vetterman crossover pipes and I ofter look at the severely turned down tips. I have thought about sawing them off either before the downward bend or in a straight line following the lower surface of the pipes. The latter would still offer some fuselage bottom protection but retain some possible performance loss.

Bob Axsom

 

[gmcjetpilot]

Its small but every little bit helps.

Two things in theory or practically speaking:

Drag: The more pipe you have hanging in the breeze the more drag. (no pun intended). Unless you have 3 feet sticking out we are talking about a fraction of 1 mph at top speed, say 1/4 mph. Now if the pipes are angled down into the breeze a lot, they will have more frontal area, and more drag. Regardless we are talking about 1/4 mph may be 1/2 mph in worst case scenario. Its not like a Cessna with that big down spout sticking straight out.

Jet thrust: Yes in theory pipes pointing in the direction of the free air stream (not turned down) would add some thrust (in theory). Much has been written on this [1]. The problem is the ideal nozzle is not typical or practical to put at the end of our pipes, so just pointing them in the general down stream direction is good enough. For all practical purposes, its nil or small.

(Augmentor tubes, Coanda effect variable geometry nozzles could accentuate the exhaust pipes "jet thrust" effect, but we are talking small or no gains depending on engine power, altitude and airspeed. Attempts to gain jet power from exhaust would add drag and weight. The no free lunch rule is in affect.)

Drag is drag. No one thing will contribute to all the drag. It's many little things. Yes pipes can be a little thing. It's more likely the pipes effect on engine HP affects aircraft performance the most, not the amount of pipe drag.

Do put the pipes as parallel to the belly as much as practical, not angled down or angled in or out, you will reduce drag. If you make 20 little improvements you will see it on the airspeed indicator.

[1] Speed with Economy" by Kent Paser

 

[bumblebee]

tips angled down contribute drag in 2 ways: 1) pipe frontal area 2) exhaust stream frontal area.

the exhaust flow acts like a tornado-shaped (draggy) extension of the pipe....the exhaust stream must be redirected (accelerated) into the slipstream. this is a very turbulent condition which can induce a large drag component in the lower cowl exit area.

i doubt this is worth 10mph but it will be a meaningful number. the other major factor is power....unless both engines have been on a dyno there can easily be a up to a 10hp differential between identical engines. at RV speeds, 10hp is worth about 5mph. combined with exhaust drag, minor airframe differences and pitot calibration, +/- 10mph is very conceivable between "identical" airplanes.

you didn't indicate whether speeds were IAS or GPS.....

 

[Norman CYYJ]

Is your plane actually 10mph slower or is there a difference in your instrumentation?

 

[rv6ejguy]

I'm with Bumblebee on this one, angled exhaust can offer significant plume drag. I repositioned my angled pipe from right cowl bottom at 45 degrees to inside the cowling air exit pointing almost straight back. Best as I could tell, this netted about 3 knots.

 

[hecilopter]

Same indicated speed

Thanks for the responses. Both airplanes equipped with AS indicator and Dynon displays. When flying side by side, both indicate the same speed within 1 mph. When he is flying at 25"/2500 rpm and I am side by side with him, we both indicate about 186 mph, but I am running at 23"/2350. If he goes throttle to the firewall, 2700 rpm, he will indicate 195 mph and when flying side by side with him, I am at about 25"/2400 rpm and indicating 195 mph. If I then go full throttle and 2700 rpm, I will accelerate to about 206 mph. We are down low, about 1500' and both showing about 27" of MAP wide open.

 

[db1yg]

My RV9a has the Vetterman cross over system and initially the pipes exited at about 45 degrees and extended down about 6 inches below the firewall. Being concerned with the drag of this configuration I called Larry Vetterman about placement that would provide a good compromise between low drag and low noise (horizontal pipes create heat and noise thru the floor). Larry said he had done a great deal of testing on this and found that, in his opinion the best compromise was down at an angle of 25 to 30 degrees and around 5 inches extending below the fw. I re-angled my pipes to 27 degrees and did not cut any off---if I gained any speed it was no more than 1 kt.

As always YMMV!!

Cheers,

db

 

[Bob Axsom]

1 Kt is awesome

About an hour ago I sawed 3.5" off of the left pipe and 3.125" off of the right pipe (they are now even with respect to the firewall). If I get 1 kt. gain out of this I will be celebrating. ANY gain is worth a shot. I recently heard a highly respected aviation guy loosely say that he made some change and only gained 3 or 4 kts. Those darn knots are very hard to come by and I think you have to work very hard and test very carefully to know whether you are making progress or not.

Bob Axsom

 

[Ironflight]

Those darn knots are very hard to come by and I think you have to work very hard and test very carefully to know whether you are making progress or not.

Bob Axsom

I think that in the Grumman world, we used to value a knot at the top end at about a thousand bucks....(certified world, STC's, etc...)!

Paul

 

[db1yg]

Bob,

Please let us know if you see any measurable change in speed with the shorter pipes. I did not cut off any of mine since the bottom scoop on my James cowl extends down 4+ inches below the fw--so only 1-2 inches was in direct airflow.

Let us know and good luck.

Cheers,

db

 

[Bryan Wood]

About an hour ago I sawed 3.5" off of the left pipe and 3.125" off of the right pipe (they are now even with respect to the firewall). If I get 1 kt. gain out of this I will be celebrating. ANY gain is worth a shot. I recently heard a highly respected aviation guy loosely say that he made some change and only gained 3 or 4 kts. Those darn knots are very hard to come by and I think you have to work very hard and test very carefully to know whether you are making progress or not.

Bob Axsom

Bob,

I did a similar thing a while back as the lure of more speed was just to tempting. Here's my results... Little or no measurable speed increase, but a very noticable pounding on the bottom of the plane that made the plane feel rougher. My wife noticed it right off on the first flight and thought the prop needed rebalancing or something. The quick and dirty fix was a pair of stainless exhaust extensions with a flair at one end that would slide over the original pipes. I bought these at a local auto supply store and then slotted the ends where they overlap the exhisting pipes with several cuts to allow clamps to cinch down and grip. I've got about 250 hours now with this temp fix and it is holding well. With these tips installed with no trimming my pipes are within a quarter of an inch or so from original. The plane is much nicer with it put back. I hope you have better results than I did with your shortened pipes.

Best,

 

[Kevin Horton]

Thanks for the responses. Both airplanes equipped with AS indicator and Dynon displays. When flying side by side, both indicate the same speed within 1 mph. When he is flying at 25"/2500 rpm and I am side by side with him, we both indicate about 186 mph, but I am running at 23"/2350. If he goes throttle to the firewall, 2700 rpm, he will indicate 195 mph and when flying side by side with him, I am at about 25"/2400 rpm and indicating 195 mph. If I then go full throttle and 2700 rpm, I will accelerate to about 206 mph. We are down low, about 1500' and both showing about 27" of MAP wide open.

I wonder how much of this difference is due to drag, and how much is due to power. It would be interesting to load both aircraft to the same take off weight, take off, then set up side by side at Vy. Both aircraft go to full throttle and climb at Vy. How much difference in rate of climb is there? Don't rely on the VSI indications, as they are often in error. Rely on stopwatch vs altimeter to get rate of climb. If we know the weight, and the difference in rate of climb between the two aircraft, we can calculate the difference in power produced.

 

[hecilopter]

Great idea Kevin

We will try that next. Top off, get side by side, then go to full power and hold 100 mph indicated. That should be a good power test. The odd thing is, he should have more power as he has 9:1 pistons and I have the standard 8.5:1.

 

[penguin]

Hey Rusty,

You didn't mention that one airplane has a carb and the other FI! What kind of fuel flow are you both seeing? If Tom is flowing less fuel then he is making less power. Are the temps all similar? Wide variations would indicate not enough fuel getting to one cylinder. Have you checked both tachos with an optical tach? Very common for mechanical tachos to be 100+ rpm out (either way). Paint can often make a few mph difference as well.

When I was flying with the end of the tail pipes cut off the vibration through the floor was really bad. I bought some 'piccolo pipes' as extentions, but they loaded up the ball joints, made them leak. I sent the pipes back to Larry Vetterman and got them extended back to regular length. Floor vibration was much less making flights > 1/2 hour much more comfortable, didn't notice any speed loss (but didn't do any scientific checks).

Peter

 

[hecilopter]

Don't know the flows

I don't know the fuel flow on my airplane, I have a carb and no flow meter. He has fuel injection and a flow meter, but it isn't working at the moment. All his EGT's are around 1200 on climb out as well as mine. We both lean to about 1300 except when doing WOT down low. His EGT readings are all within about 25 degrees of each other.

Is it possible to be WOT full power and have low EGT readings? Seems like at WOT if there wasn't enough fuel flow the EGT reading would be high and you wouldn't be able to lean at all.

 

[Bob Axsom]

Will do but it will be a while

Bob,

Please let us know if you see any measurable change in speed with the shorter pipes. I did not cut off any of mine since the bottom scoop on my James cowl extends down 4+ inches below the fw--so only 1-2 inches was in direct airflow.

Let us know and good luck.

Cheers,

db

I will let you know but it will be a while since I have to replace my starter solenoid before I can fly again - next week probably. Here is a photo record of the work:

Bob Axsom

 

[penguin]

Is it possible to be WOT full power and have low EGT readings? Seems like at WOT if there wasn't enough fuel flow the EGT reading would be high and you wouldn't be able to lean at all.

1200F EGTs at WOT seems reasonable, especially as they are all similar. Low fuel flow ought to produce higher EGTs as you would be running partially leaned off - would also produce high CHTs on climb out. I would check both of the tachos to make sure you are both turning 2700 rpm.

Peter

 

[Bob Axsom]

Test soon starter solenoid out for delivery to the house

Test soon starter solenoid out for delivery to the house. Lot of chores to do at home and reinstalling the Sky-tec starter after the solenoid is replaced but if the weather cooperates the speed test should happen tomorrow or Friday.

Bob Axsom

 

[elippse]

Hoerner in "Fluid Dynamic Drag", P.3-11, Figure 18, shows the lift and drag coefficients of circular cross-section cylinders vs their sweep angle alpha. CD is given as 1.1 sin^3 alpha + 0.02 at an exposed area of diameter time length, where alpha is the angle relative to the airflow. This would correspond to the angle of the pipes as they emerge from the cowling, alpha=90 straight down.

 

[Bob Axsom]

Thanks Paul Lippse

It is always comforting to get your great technical input.

Bob Axsom

 

[gmcjetpilot]

Parellel

key is getting the exhaust to flow with the free air stream. Rotate that little down turn 180 degrees and weld it back on, so the exhaust exit exit is parallel to the floor. Noise, vibration and exhaust stain affect? Subject to limited data, subjective comments and rumors.

 

[elippse]

When a pipe is cut off square on the end, opened or closed, and a harmonic-rich noise source is introduced into one end, the pipe will have a sharp resonance with those harmonics that resonate with the pipe length. Those harmonics will be strong in amplitude and narrow in frequency. Take a short length of 1/4" or 3/8" aluminum tubing (you do have some of that lying around, don't you?), open at each end, and blow over the end of it. Now slice one end at an angle with a length of three times the diameter and blow over the opposite end. Wrap a little piece of paper around the baloney slice to make the tube the same as before slicing it, and try it again. Go back and forth several times and compare the results of the tones that are present and the relative strength of each. With the one I did, there was a very strong tone that emerged on the blunt-end tube when I blew strongly, and a much weaker tone at a higher frequency when I blew softly. With the baloney slice, the lower frequency tone was somewhat weaker, and the higher frequency tone became stronger. Also, there was an instability where the tones would try to go back and forth. Generally, by slicing the end of the exhaust pipe at an angle, you reduce the peak amplitude of a resonance since no longer does all of the pressure wave reflect from a single point. As an experiment, you might try this on your exhaust with two clamp-on extensions, one cut square and the other baloney-sliced, to see if the pounding on the bottom of the fuselage is reduced. Another thing that happens with the baloney slice is that as the gas and the pressure wave begin to emerge from the pipe where the slice begins, they meet opposition from the outside air which slows them down. The pressure wave slows since it goes from a medium of higher sound velocity (you did know that the speed of sound is about 50% faster in your exhaust?) and so it will turn and go away from the line of the exhaust pipe. This pressure wave is usually what causing the pounding, not the gas flow. Somewhat the same takes place in the gas flow. That means that if the baloney slice is away from the bottom of the plane, the exhaust will magically turn away without having to have a turned pipe angled downward, which has more drag. 'Something to experiment with. This is EXPERIMENTAL aviation, non?

 

[gmcjetpilot]

Interesting

When a pipe is cut off square on the end, opened or closed, and a harmonic-rich noise source is introduced into one end, the pipe will have a sharp resonance with those harmonics that resonate with the pipe length..................That means that if the baloney slice is away from the bottom of the plane, the exhaust will magically turn away without having to have a turned pipe angled downward, which has more drag. 'Something to experiment with. This is EXPERIMENTAL aviation, non?

Interesting, no reason to not try, but you will lose any residual thrust you might expect from the pipe with the angled cut. Kent Paser "Speed with Econ" has shown 5 mph increase in speed playing with the pipes and "nozzles" at 7,000 feet. Worth a look. Any way speed or noise, you decide.

 

[elippse]

I have Kent's book, and he actually made reducing nozzles to fit on the end of the pipes to obtain higher exhaust velocity to obtain thrust. This increased velocity came at the expense of creating back pressure. He obtained the best results at reducing the exhaust diameter from 1 3/4" to 1 1/4" at 7000', but he said the back pressure could reduce his speed at lower altitudes. I wish he would have published his results as the percentage increase each mod gave. You can multiply the increase from each mod to get the overall increase you can obtain from them, but you can not add the mph increases. If one mod gives 3%, another 6%, and a third 10%, then multiplying 1.03X1.06X1.10 would result in 1.20 for a 20% speed increase.

 

[gmcjetpilot]

Yep that is right

I have Kent's book, and he actually made reducing nozzles to fit on the end of the pipes to obtain higher exhaust velocity to obtain thrust.

Yep that is correct did I say something different. My point is pointing pipes down at any angle would be counter productive to gaining or recovering any residual thrust. You would agree?

As far as your algebraic math of his data (X*Y*Z=), not sure where you are going, but regarding your complaint of his little book, I do agree data presentation could be better, but I say go do your own test.

You are missing one key point I think Mr. Paser made. The nozzle size that provides some "exhaust thrust" at say 7,000 feet is a detriment at say sea level, where it will cause loss of HP due to restrictive exhaust. He had the idea for a variable nozzle and tried it. However the shape was not ideal. He envisioned the overlapping peddles of a flower design like many jet fighters have but left it there. The design and actuation of such a nozzle would be extensive he conjectured.

Long ago I did some consulting for Pratt & Whitney. Variable nozzles for some "engines" (can't talk about) where being worked on to increase and vector thrust. The physics of acceleration and direction of gases is the same; for every action there is an equal and opposite reaction (Newton).

http://www.youtube.com/watch?v=rYpm5OA78tg

 

[Bob Axsom]

I made a test flight this evening

On 5-11-07 I made flew a speed test with the exhaust pipes pointed down and the speed was 170.4 kts this was at 6,000 ft density altitude, three directions using the http://www.us-airrace.org handicap procedure for the conduct of the flight and the NTPS Excel spread sheet with imbedded formulas to calculate out the wind effect. Tonight I repeated the test with the cut off exhaust pipes and the speed was 170.8 kts. I had no noticeable difference in exhaust noise or pounding - probably because of the insulation on the floor and the overlaying carpet.

Other data: RPM=2710, MAP=24.5, temp at 6,000MSL = 17C, flew at 4,600ft MSL for 6,000 ft density altitude per USAR Proc., oil temp. = 180, oil pres. = 80, CHT/EGT 1=374/1309, 2=377/1365, 3=377/1341, 4=338/1309.

Conclusion: Apparent increase in speed and no ill effects. This was worth doing for my airplane.

Bob Axsom

 

[Nuisance]

yes, it matters!

It took me a little bit to gather the pictures, but now, on my website, down about the middle of the page, is a note about this.

http://www.lazy8.net/speedmods.htm

 

[gmcjetpilot]

Told Ya, Bob so how much total speed have you gained

On 5-11-07 I made flew a speed test with the exhaust pipes pointed down and the speed was 170.4 kts this was at 6,000 ft density altitude, three directions using the http://www.us-airrace.org handicap procedure for the conduct of the flight

Conclusion: Apparent increase in speed and no ill effects. This was worth doing for my airplane. Bob Axsom

Told ya. Nobody listens to me. ......
Seriously where is this US Race "handicap procedure" you mentioned. Went to site and it did not pop out at me. Thanks.

P.S. you're getting real good at doing flight test. You noticed a 0.40 kt gain, cool. Sounds like you got it all down with pretty repeatable results. It would be good to do it over several days, you may have gained more. Bob, what is your TOTAL gain you got so far, for all your mods, since you started on the quest for speed?

 

[Jerry Cochran]

Still trying to figure out differences in 2 RV-7A's. Cooling problems are now under control, now to a speed issue. One plane is about 10 mph slower than the other. I have several other threads about our 2 planes, but for all practical purposes, they are identically cowled, baffled, engined, propped, faired and equipped except for the louvres on the bottom of the cowl and removing about 4" from the bottom of the cowl where the exhaust exits due to cooling issues.

Has anyone noticed or experimented with the exhaust pipes and the angle with which they exit the cowl? I noticed that one difference between our airplanes (both equipped with vetterman exhaust) is that my pipes exit the cowl and are parallel to the center of the plane. The pipes on the slower plane are angled pointing to the center of the plane and at each other.

Could this make a difference with drag from the angled pipes or maybe exhaust thrust that is not aimed inline with the airflow? We are going to experiment and twist the pipes to be parallel like mine and see what happens. Just wanted to check to see if anyone else has done this and what the results were.

Thanks!

Since I am currently in the throes of fitting gear leg and wheel fairings, and reading up on how a misalignment can cause lots of drag, I'm wondering if the "slow" -7a took lots of time and measuring addressing this.

A shot in the dark regarding this instance, but might be worth checking. I know of RV-ers who installed their fairings with the plane on the ground, then "eyeballed" the results. This method in my opinion would negate some of the benefits of the fairings.

This may be heresy, but I think Van's method gets them close, but not "dead nuts" as a machinist friend of mine likes to say... Searching the archives of the RV-list, Scott from Van's claims a perfect alignment of these will get you 25mph. The manual says 12, so go figure...

Just food for thought.

 

[rv6ejguy]

I gave up a long time ago trying to get repeatable GPS/TAS results during flight testing. I've found it best to use IAS and OAT and use the whiz wheel for TAS. wastes a lot less fuel and time and appears to work fine if you are comparing changes on your airframe/engine and not to others.

I've found it difficult to quantify changes of much less than 2 knots with any method.

 

[Bob Axsom]

The NTPS method works well

The NTPS method works well. You can't "not test" if you are seriously trying to increase speed by modifying your airplane. Some mods actually slow the plane down. The NTPS spread sheet takes out the wind mathematically and that is almost essential.

Bob Axsom

 

[gmcjetpilot]

Thanks Bob

The NTPS method works well. You can't "not test" if you are seriously trying to increase speed by modifying your airplane. Some mods actually slow the plane down. The NTPS spread sheet takes out the wind mathematically and that is almost essential.

Bob Axsom

OK Bob thanks I am aware of the spread sheet. Have used it and its great, but was wondering what US Race "handicap procedure" meant. No biggie. I agree you have to flight test.

I agree with Ross. GPS is not perfect, it can produce error itself, small but it's there. Most flight test has been done for decades with out GPS, Loran, INS or DME for a long time.

My personal flight test results are only good for about +/- 1 to 2 mph on any given day at best, a little less than about +/- 1%, with out any changes.

Using the pitot static instruments and the whizz wheel is still very valid method, especially with electronic flight instruments which are easier to read (e.g., not parallax error) and accurate. Besides when looking for a difference in a before/after modification, its relative.

The 3rd method I found useful is flying LOW over known ground references, track and distance. Using a stop watch you can get absolute speed. You do it in two directions and average each run separate. However wind can cause large errors, so you have to pick a no or low wind condition. Preferably the track should be right into the prevailing winds if there is any wind. A near by weather station (ATIS) or even your own ground weather observer helps greatly. The down side is finding a safe place to fly over low with all the desirable characteristics. The total distance has to be fairly long to mellow out reaction time starting and stopping the watch. With all that effort I found the results where very consistent and predictable. The airspeed indicator and GPS can be also used to compare all three.

The problem with flying high in three directions is wind can vary second to second and mile to mile. You have no idea up high. You have to assume the atmosphere is homogeneous.

The 4th method I find really obvious and easy Another is SIDE-BY-SIDE RV test. Get another RV on the field that has similar performance, a little faster or slower does not matter. Than every mod you can go fly side-by-side and see if you gain or not.

My suggestion or comment was you have to repeat the flight test over several different flights, over different days to get more confidence in any number. Small speed changes are hard to measure, not impossible but hard.


It's so hard to control nature and almost impossible to reproduce similar conditions. However the art of flight test and quality of flight test of RV'ers like Bob, Ross, Keith is top notch. I really have seen a great improvement in technique thanks to their input. I know I have learned a lot.

 

[elippse]

I have a Tru-Trak autopilot fed from my Garmin hand-held and an altitude hold autopilot. With those two aboard, it is very easy to see changes in TAS from various mods. The only time I see a difference is if there are up-or-down-drafts over the hills. Then I will note that my rpm has decreased or increased, and a look over at the level bubble on the side of my cockpit confirms that I am nose up climbing in a down-draft, or nose down in an up-draft. But the altitude stays constant. My prop has a gradient of about 8 mph/100rpm, so when speed changes, rpm does too, and it's usually the first indicator!

 

[gmcjetpilot]

Flight test and autopilot's = goodness

I have a Tru-Trak autopilot fed from my Garmin hand-held and an altitude hold autopilot. With those two aboard, it is very easy to see changes in TAS from various mods.

Yea autopilots are great tools for removing the pilot error. Good point with heading or track and altitude held for you, you have more time to read the data. When hand flying there is no doubt you add some error.