Basic Aerodynamics and Flight Mechanics

Hi All, I'm fairly new to the forum, building an RV-8QB, working on canopy skirt and FWF. I'm also a high-time sailplane pilot.

But more to the point, I am a Research Engineer at NASA with a PhD in applied aerodynamics.

I frequently see examples of minor to major misunderstandings of aerodynamics, and often a post will assert an argument based on "almost correct" understanding that leads to the wrong conclusion.

There is a current thread on the safety forum on turning back to the airport after engine failure, which is a really excellent thread by the way, with valid arguments on both sides. But some of the arguments are based on misunderstandings. So I thought I would really stick my neck out and try to clarify some points:

To keep the posts short, I will do this in three parts.

Part 1 On turning down wind:

There are three elements to this. The third element is the really important one:

a) In a UNIFORM wind, the airplane can not tell. As one poster said, its a reference frame issue. The airplane can not tell that the ground is drifting underneath it.
b) Near the ground, the pilot CAN sense the drift, and it often causes pilots to mis-coordinate the turn. When turning from downwind to upwind, there is a tendency to use too much rudder. This is no doubt a contributer to stall-spin accidents from turning near the ground. It takes a strong effort to resist the temptation.
c) In DESCENDING turns, the airplane is strongly affected by the wind, because it is not UNIFORM. There is a strong wind GRADIENT - as you get closer to the ground, the wind speed drops. The result is that descending through the wind gradient (whether turning or not) robs the airplane of energy - it has to accelerate to offset the loss in wind speed in order to maintain airspeed. The effect is most pronounced for airplanes that are heavy, and very noticeable in descending turns. One poster is a crop duster pilot who pointed this out. A pilot is likely to sense the higher sink rate and pull back on the stick. This also contributes to stall-spin accidents.

The take-away here is that the age-old debate about down-wind turns has valid observations on both sides that are not conflicting. Near the ground, making descending turns in strong wind is dangerous because of the pilot mis-queing and because energy is bled faster than normal. You must fly coordinated, and you must maintain airspeed.

Part 2 will be on the effect of descending turn on expected g-load and stall speed.

Basic Aerodynamics and Flight Mechanics, Part II

Hi again,

OK, Part 2 on effects of climbing, descending, turning, on g-loads and stall speeds.

One post on the "return to airport" thread seemed to think that because the airplane would be gliding, descending, that somehow the g load caused by the steep turn would be less than a level flight turn.

The key thing here is whether you are in steady, unaccelerated flight or not.
If you are in a steady climb at a constant climb rate, you are flying at 1 g.
If you are in a steady descent at a constant descent rate, you are flying at 1 g.
Normally, our climb and descent angles are small enough that it is accurate to assume that all of the force that balances gravity is lift force. For fighters and dive bombers, the angles get steep enough that a combination of thrust or drag , combined with a portion of the lift, is what balances gravity.

Now, if you are turning, you are in accelerated flight. By banking, a portion of the lift is directed inwards to the center of the turn, causing the lateral acceleration that makes the flight path curve. But if you have a STEADY turn, this acceleration is constant, and balanced by the inward component of the lift, while the vertical component of the lift is still balancing gravity. This combination leads to elevated g's and reduced stall speed.

Now the key point is this: if you are in a STEADY, DESCENDING turn, you still need the same vertical component to balance gravity, or else your descent rate would increase. And you still need the same lateral component of lift to keep turning. So you get the same g-load increase, and the same stall speed increase for a gliding, descending turn as for a level turn.

The only deviations from this occur from accelerated flight, that is, dynamic maneuvers that reduce or increase the g-loading because of the accelerations.

Part 3 will be on the effects of turning and speed change on drag and sink rate, and how the airplane design affects the optimum bank angle.

Basic Aerodynamics and Flight Mechanics, Part III

Hi again,

Here I want to address misunderstandings about the effects of turning and speed changes on drag.

You must understand that there are two main sources of drag.

First is skin friction and pressure drag. Skin friction is the obvious surface friction caused by the viscous drag from the air flow. Pressure drag comes from any areas of flow separation or pressure losses through cooling system, etc. The drag from these sources increases with the square of the flight speed. ( we use "dynamic pressure" which is the combination of air density and velocity squared, which accounts automatically for altitude effects as well as speed effects)

The second source of drag is "induced drag". Induced drag comes from the generation of lift from a finite wing. It is caused by the "downwash" influence on the wing from the trailing tip vortex system. You can think of the trailing vortex wake as wasted energy left behind in the flow, or you can think of the downwash as reducing the local onset flow on the wing, requiring a small increase in angle of attack to compensate. That small angle change rotates the aerodynamic force vector rearward so there is a drag component as well as the lift.

Induced drag is proportional to the square of the amount of lift, and inversely proportional to the square of the wing span, and inversely proportional to the square of the flight speed. All three of these things are important:
1) if you are heavier, or pulling g-s, the induced drag goes up FAST
2) if you fly slower, the induced drag goes up FAST
3) if you reduce the wing span, from lets say a sailplane to an RV-8, the induced drag goes up FAST.

RV's get away with such short wings because they spend almost all their time flying fast, where induced drag is negligible. You may have wondered why the speed for best climb is so fast for RV's compared to other GA airplanes which typically have higher aspect ratio wings - i.e. more wing span for the wing area and/or weight. This is why. An RV-8 has twice the induced drag as a C-172 at the same weight and flight speed. But if you increase the speed 40%, you offset the shorter span with the higher speed and the induced drag is the same.

Sailplanes obviously have very low induced drag because of the very high aspect ratio. Thats because they need to be especially efficient at gliding turns at slow speed to stay in thermals.

So - whats the best glide speed? The best glide speed is where the drag is minimum for a given weight. We already said that the viscous drag is proportional to speed-squared, so it is small at low speeds and increases quickly at high speed. Induced drag does the opposite - it is large at slow speed and decreases rapidly at high speed. So the combination of these makes a bucket of minimum drag.

How is that drag bucket changed when you turn? Well, you are pulling g's, which is just like an increase in weight. In a 2-g turn you have FOUR times the induced drag. You can compensate for this by speeding up - turning has the effect of shifting the drag bucket to higher speed. Thats OK, since the stall speed goes up 40% too, so you better fly faster! True, the viscous drag will increase some, but that is small compared to the reduction in induced drag.

Main point here is that a gliding turn in an RV is going to be like a brick. Adding speed helps. But that is hard for a pilot to do when it is so apparent that the ground is rushing at you. Combine this with a little wind gradient effect (from Part 2) and its no mystery why so many have pulled instead of pushed and ended badly.


So what would be the optimum bank angle for a descending, gliding turn? This depends on a lot of factors. And as I pointed out above, it depends on the speed as well as the bank angle. It does no good at all to say "45 degrees is optimal" unless you say what speed goes with it. There are many combinations of speed and bank that will give about the same turn rate and turn radius, but with very different sink rates. Without having done any analysis, my instinct is that 60 degrees is too steep - the induced drag is just going to eat you alive. Best thermalling bank angle in sailplanes is about 40-45 degrees, but depending on conditions, I may bank steeper or shallower than that. Most pilots think they are banking 45 when they bank 30, so be honest. 45 is when the screws on the instruments match the horizon.

Training and practice will teach you to make efficient gliding turns. Practicing at altitude does not expose you to the visual cues from the ground that will make you screw up. If you are going to practice this, start high and work down to low enough altitudes to where you begin to sense the effects of the wind and the ground-rush. You are not going to be able to sense the effects of wind gradient unless you are down around 200-300 ft, so that is pretty serious training.
Bear in mind, I'm not taking a position on whether anyone should actually contemplate a return to airport with engine out. But training and practice of gliding turns can only make you a better pilot - regardless of your opinion on that question.

I wouldn't worry about the differences between idle thrust and windmilling for this training. All the other factors are going to dominate. Certainly a CS prop windmilling is going to increase the sink rate a fair bit, but, IMHO, thats not the difference that is going to kill you. You fly too slow in the turn, you sense the ground rush, you pull back, and maybe at the same time, too much rudder because you perceive the wind drift - thats the combination that is going to kill you. You learn to make safe gliding turns and you will not depart controlled flight. Maybe you get 2/3 the way around the turn and see you are not going to make the runway, and roll out and flare and take what you get. You are probably no worse off than if you had gone straight ahead.

So I allowed myself to depart from the initial theme here, which was just addressing the aerodynamics.

In general, I am happy to answer questions about aerodynamics and flight mechanics any time, just send a PM.

Steve

RV-8QB FWF
NASA Research Engineer in applied aero.

Nice job Steve, thanks for the info!

"Nice job Steve, thanks for the info!" Ditto!
Thanks for your well-reasoned words of wisdom regarding the previous debate.
When I read Dr. Rogers article, I was struck by how well the problems inherent in turnbacks had been addressed, and how a possible solution had been researched, tried, and proven. Later, as the discussion deterioated, the signal-to-noise ratio increased beyond my comfort level. Thanks for validating my initial impression, and also the perils of mis-perceptions while manuvering in close proximity to the ground.

Joe

Steve,

Why does too much rudder in the turn aggravate flight and cause a stall? (I always thought the problem was too little rudder - cross control -- that causes problems.)

Also -- thanks for the extensive info -- i actually studied the whole thing!

Dave T.
Legacy RG

slips and skids

Hi,
Well, a well coordinated turn takes just the right amount of rudder. Too little is a slip, too much is a skid. Skids are generally worse than slips, because of the interaction of dihedral, aileron deflection, and yaw rate.

Certainly in the initial transition to a turn, too much rudder causes the inside wing (the wing toward the center of the turn) to fly slower ( it swings aft relative to the rest of the airplane). Slower means less lift, the wing drops which increases angle of attack, you react by adding aileron, which further aggravates the angle of attack, and boom, in extreme cases, you get a spin.

In a steady turn, too much rudder means the outside wing is yawed forward, so that dihedral makes the whole outboard wing see a higher angle of attack and wants to roll the airplane into the turn (that's what dihedral is for - it creates roll in the direction to try to stay coordinated). But if you fight that roll by "top" aileron pressure, you are balancing all that roll moment with just the part of the inboard wing that is affected by the aileron, and again, you get locally high angles of attack near the tip.

In a slipping turn, everything is reversed. The whole inside wing sees the angle of attack increase, and dihedral is trying to roll you out of the turn. You balance this with the ailerons again, but now it is the outside wing that is working hard. Since it is moving faster, it takes less aileron to balance, so there is less of a problem.

Generally, slipping turns are safer than skidding turns, which is partly why slipping makes good energy management. ( No one teaches skidding turns to increase sink rate while turning base to final!) In extreme cases, it is possible to stall the outside wing and spin "over the top", but that is much more difficult to do than a classic inside spin from a skidding turn.

Hope that all makes sense

Steve

RV-8QB FWF

Very interesting -- that makes a lot of sense -- the inside wing is slowing down! (I've studied many aero books and i've never heard this articulated this well -- thank you!)

I guess a take-away would be to make sure you're not pulling back on the stick on these slow turns -- and of course concentrate on flight coordination -- ie ball in the center. Maybe even make sure you're using forward stick a little when you increase your turn bank at high alpha (angle of attack.)

What aerodynamics book do you recommend for the non-Phd?

David T.
Legacy

begining aerodynamics text

There is a great book called "Fundamentals of Sailplane Design" by Fred Thomas. Although its focus is obviously sailplanes, it is remarkably good at explaining basic aero theory without going over your head. I recommend it often.

Steve

RV-8QB FWF

scsmith said:

There is a great book called "Fundamentals of Sailplane Design" by Fred Thomas...

Click to expand...

Speaking of sailplanes there was another fallacy repeated on that turning back thread. It is something glider pilots know well but seems to not be common knowledge among power pilots.

Greater weight does not adversely effect the best glide angle it just changes the speed at which the best glide occurs. There are secondary effects that cause some manufacturers to claim better glide angles at higher weights, but I don't think that has ever been well documented. Weight does effect minimum sink speed however.

Just remember those beautiful photos of gliders at the finish of a race. Those white streams coming out are water. No glider pilot would purposely load a bunch of excess weight and in fact sometimes cheat by loading to beyond maximum legal gross weight if it decreased the glide angle.

scsmith said:

...
One post on the "return to airport" thread seemed to think that because the airplane would be gliding, descending, that somehow the g load caused by the steep turn would be less than a level flight turn.

The key thing here is whether you are in steady, unaccelerated flight or not.
If you are in a steady climb at a constant climb rate, you are flying at 1 g.
If you are in a steady descent at a constant descent rate, you are flying at 1 g.
Normally, our climb and descent angles are small enough that it is accurate to assume that all of the force that balances gravity is lift force. For fighters and dive bombers, the angles get steep enough that a combination of thrust or drag , combined with a portion of the lift, is what balances gravity.

Now, if you are turning, you are in accelerated flight. By banking, a portion of the lift is directed inwards to the center of the turn, causing the lateral acceleration that makes the flight path curve. But if you have a STEADY turn, this acceleration is constant, and balanced by the inward component of the lift, while the vertical component of the lift is still balancing gravity. This combination leads to elevated g's and reduced stall speed.

Now the key point is this: if you are in a STEADY, DESCENDING turn, you still need the same vertical component to balance gravity, or else your descent rate would increase. And you still need the same lateral component of lift to keep turning. So you get the same g-load increase, and the same stall speed increase for a gliding, descending turn as for a level turn.

The only deviations from this occur from accelerated flight, that is, dynamic maneuvers that reduce or increase the g-loading because of the accelerations.

Part 3 will be on the effects of turning and speed change on drag and sink rate, and how the airplane design affects the optimum bank angle.

Click to expand...

Let's test this idea - you guys with g-meters. Bank the airplane to 45 degrees with no back pressure. Observe your g-meter. If you were holding level altitude at 45 degrees the g-meter should say 1.4. What does it show if you bank 45 degrees and allow the airplane to sink? How about 90 degrees? But it did turn, didn't it?

In a maneuver to reverse course while allowing loss of altitude I contend you can pull 45 or even 60 degrees, roll out and not pull greater than 1.0 g's. Remember that you must accelerate vertically to go from zero fpm to 1500 fpm vertically even if you eventually stabilize at 1500 fpm by which time you have turned a lot. That's the idea - that a steeper turn gets you around so much faster that the altitude penalty is proportionally less than with a shallow turn.

If you pull 1.4 g's at 45 degrees you will hold altitude and the wing force will be 70% vertical and 70% horizontal, but 70% of 1.4 is approx. 1.0 so there's enough lift to keep you up. Now if you pull 1.0, then you have less turn force and less vertical lift. You will turn at a given rate and you will sink at a given rate. For a while, both are increasing. You won't turn as fast as a 1.4 g turn but it's faster than a 30 degree sinking turn, etc.


Viewing this as a steady-state question is erroneous analysis IMHO. I'm not any kind of engineer, but I can fly a little and I can read a meter. Who will try this before the WX here gets better?

Good points Steve and Larry

I would like to add one regarding the downwind turn. When you take off and turn downwind, the pilot may misuse the increasing ground speed cue (visual cue) as an increasing airspeed. If the pilot sees an increasing ground speed, he may pull back and decrease the airspeed to a dangerous level.

Also, given the typical wind shear near the ground, the airspeed tends to decrease when climbing with a tailwind (or descending with a headwind as you noted in more detail).

Thus, the climbing downwind turn close to the ground has two problems. First, and most important I believe, is the misinterpreted visual cue (increasing ground speed being dominate over static or decreasing indicated airspeed). Second is the decreasing airspeed due to a downwind climb through a wind shear.

Of course, the solution is simple, pay close attention to the airspeed and suppress the urge to pull on the stick. So simple, but you have to make yourself learn it. And keep the ball (or yaw string) centered. Love you glider pilots.

Flutter?

Great! An excellent review of basic aero....now I have someone I can explore a few nooks and crannies with

I've become a bit interested in flutter with respect to the RVs. I talked to a presenter yesterday at the Golden West Flyin, Martin Hollman, owner of Aircraft Designs, Inc. Talking to him after his presentation, he said that the RVs have never had a proper flutter analysis done. To his knowledge Van did pay for an analysis, but it was a cheapo method known for inaccuracy.

Obviously in practice this is not a great problem because RVs do not have a reputation for coming to pieces in flight. Still, I would like to see a trustworthy analysis done of the several RV models using an accepted, modern numerical code. I suppose the time-consuming part would be inputting the structural elements. Perhaps the RV crowd could help with the grunt work?

Steve,

If you are going to do this. I would hope you would go all out and bring in Coefficient of Lift and angle of attack and all the other important variables that need to be known to be understood. It is all what the wing "sees" and how much lift is being generated that matters. As a pilot I want to see the hard actual math that makes it all happen. Thanks for diving into this. I am looking forward to asking many questions.

It's the Angle of Attack.........

....NOT the G's !!! A wing doesn't know what the bank angle is, nor does it care. Could be 90, could be inverted, doesn't matter. It's the angle of attack, which is DIRECTLY proportional to the stick position.

If you roll to 45 deg angle of bank with no additional back stick movement, you won't maintain altitude. You will be still pulling 1 G in the cockpit, but you will be descending.
To maintain altitude at 45 deg bank, you must move the stick aft a bit, which will feel like holding back stick pressure because you are likely trimmed for 1 G level flight. You could, if you chose to, add nose up trim so as to be trimmed to level flight in a 45 degree banked turn. There would be no back stick pressure, but the stick position would be aft of level flight position, and the wing would be flying at a higher angle of attack.

Stop thinking G, start thinking Angle of Attack. The wing stalls at the same angle of attack (and the same stick position, but not the same back pressure) no matter what the airspeed. The higher the airspeed, the higher the G at which the wing will stall, but the angle of attack is the same for all stalls.

[I don't have a degree in aerodynamics, but I did stay at a Holiday Inn Express recently, but not last night.] (Does it wear off with time?)

hevansrv7a said:

Let's test this idea - you guys with g-meters. Bank the airplane to 45 degrees with no back pressure. Observe your g-meter. If you were holding level altitude at 45 degrees the g-meter should say 1.4. What does it show if you bank 45 degrees and allow the airplane to sink? How about 90 degrees? But it did turn, didn't it?

In a maneuver to reverse course while allowing loss of altitude I contend you can pull 45 or even 60 degrees, roll out and not pull greater than 1.0 g's. Remember that you must accelerate vertically to go from zero fpm to 1500 fpm vertically even if you eventually stabilize at 1500 fpm by which time you have turned a lot. That's the idea - that a steeper turn gets you around so much faster that the altitude penalty is proportionally less than with a shallow turn.

If you pull 1.4 g's at 45 degrees you will hold altitude and the wing force will be 70% vertical and 70% horizontal, but 70% of 1.4 is approx. 1.0 so there's enough lift to keep you up. Now if you pull 1.0, then you have less turn force and less vertical lift. You will turn at a given rate and you will sink at a given rate. For a while, both are increasing. You won't turn as fast as a 1.4 g turn but it's faster than a 30 degree sinking turn, etc.


Viewing this as a steady-state question is erroneous analysis IMHO. I'm not any kind of engineer, but I can fly a little and I can read a meter. Who will try this before the WX here gets better?

Click to expand...

Steve Smith- excellent posts

Steve Smith: Thanks for your excellent aerodynamic exegesis. You've stated the facts better and more succinctly than any textbook I have read. I have avoided the turnback/downwind turn thread you referenced because that subject tends to generate more heat than light, but finally I cannot resist telling of a crop dusting experience I had as a young fellow. I think it illustrates some points you and others have made.

It sometimes happened that the air close to the ground was calm and smooth, while up two hundred feet, above a shear level, a pretty brisk wind would be blowing. This most often occurred near the ocean. I soon learned that pulling up (and decelerating) into an increasing tailwind created a very uneasy feeling in my stomach and an airplane that wallowed rather than flew, while pulling up into an increasing headwind was very reassuring. This was in the pre-microburst/windshear accident era and the mantra then was, "The airplane doesn't feel the wind". I knew from practical experience that that wasn't true but my understanding was incomplete. I asked questions at safety seminars and was laughed out of the room. Now attitudes are very different.

I also learned that turbulent air has a marked negative effect on aircraft performance. That is not often written about and is, I believe, underappreciated by many pilots.

Crop dusting is very educational if it doesn't kill you. My hat is off to those who make it a career. It is hard, dangerous work. I had a fortunate combination of luck and youthful enthusiasm and after a few seasons I escaped unscathed.

Steve, perhaps some refinements to make RVs fly faster will occur to you as you build. We are all eager students of aerodymanics and you may be the teacher we have been waiting for on the forums. Welcome, and please write more.

G's, Angle, etc.

Yes, it is the angle of attack that determines the stall. No ifs or ands. Period.

That said, it's the g's that are relevant here, but not the sink or non-sink. If the airplane experiences 1.0 g's then the stall speed does not change when the bank angle changes and it doesn't know that it's turning. We know it is turning and descending because the banked wing is providing less vertical lift. So if you don't change trim/aft stick position/back pressure and you do change roll command, you will turn and your stall speed will not change.

Now it also is true that for a given weight and angle of attack the indicated airspeed will be the same. When you are "engine-out", stick movement fore-aft changes aoa and thus airspeed and thus stall speed.

All of this is on a simple point: you can turn sharply while engine-out and not increase your stall speed, but you will increase your sink rate with increased bank or turn rate. The remaining important question is which angle of bank is best for this trade-off between sink and turn. I think 60 is the magic number, others say 45.

rfinch said:

Great! An excellent review of basic aero....now I have someone I can explore a few nooks and crannies with

I've become a bit interested in flutter with respect to the RVs. I talked to a presenter yesterday at the Golden West Flyin, Martin Hollman, owner of Aircraft Designs, Inc. Talking to him after his presentation, he said that the RVs have never had a proper flutter analysis done. To his knowledge Van did pay for an analysis, but it was a cheapo method known for inaccuracy.

Obviously in practice this is not a great problem because RVs do not have a reputation for coming to pieces in flight. Still, I would like to see a trustworthy analysis done of the several RV models using an accepted, modern numerical code. I suppose the time-consuming part would be inputting the structural elements. Perhaps the RV crowd could help with the grunt work?

Click to expand...

IMO Martin Hollman only has one agenda and that is to sell his book and services. His really only claim to fame is some work he did on the Lancair 4. Remember the Stallion? Total flop. I wouldn't believe anything he says.
cheers
tm

I think we can all agree that in the turn-back situation we aren't looking to maintain level flight. That said, steady-state is a good explanation, but steady-state at 1G. How many people have gone out and tried turnbacks? I've done them pretty extensively, mostly in Cessna's (150/152, 172, 177) And a little for my Commercial (PA28R-200). I can promise you in practical application, you are holding ~1G in steady state descent for almost the entire maneuver.

Now with keeping the 1G steady descent, we aren't trying to maintain a pitch attitude, we are shooting for best glide (theoretically, lowest Lift/Drag ratio). Thus, in the 60? bank turn the sink rate is astronomical, because the majority of the lift is going to the turn, not keeping the airplane flying. Coming out of the turn, point the nose at the numbers, nail the best glide and keep it coordinated.

The increased weight does help the glide, and keeps you floating longer in the round-out/flare, momentum being a factor here I assume)

If you haven't done turnbacks, don't try them for the first time when your engine dies, just land straight ahead.

A couple of pointers, an offset when you hit the end of the runway (if allowed per noise abatement) of about 20? will greatly help if you have to come back, allowing enough room to make the turn, and almost line right up on final. Second, find someone who understands the theory here, and can teach it if you're not comfortable, but PRACTICE THESE, the split second decision from practice could be the difference between making it back and not. I can make the turn-back in the Cardinal at about 400' with practice, but I limit myself to start it at 600' in the actual event due to the 3-5 second delay of recognition.

scsmith said:

In extreme cases, it is possible to stall the outside wing and spin "over the top", but that is much more difficult to do than a classic inside spin from a skidding turn.

Hope that all makes sense

Steve

RV-8QB FWF

Click to expand...

Oh my, how well I do remember this from my student days. My instructor just told me to bank 45 degrees, and pull back until it stalled. Said we were going to practice "accelerated stalls", but no warning of what/where/how the plane would react

"E" ticket ride for sure.

I also started in gliders, one of the very first thing they beat into your head is "push" when your brain is screaming "pull". Airspeed equals safety, and maneuvering energy. It is your friend.

Effect of weight addition

n5lp said:

Speaking of sailplanes there was another fallacy repeated on that turning back thread. It is something glider pilots know well but seems to not be common knowledge among power pilots.

Greater weight does not adversely effect the best glide angle it just changes the speed at which the best glide occurs. There are secondary effects that cause some manufacturers to claim better glide angles at higher weights, but I don't think that has ever been well documented. Weight does effect minimum sink speed however.

Just remember those beautiful photos of gliders at the finish of a race. Those white streams coming out are water. No glider pilot would purposely load a bunch of excess weight and in fact sometimes cheat by loading to beyond maximum legal gross weight if it decreased the glide angle.

Click to expand...

Yes, this is all true. In a sailplane, gravity is the engine. I describe the effect of adding ballast as being like coasting down hill on a skateboard. If you add weight, you will coast down the same slope at a higher speed. The increased drag is balanced by the increase in the "thrust" from the weight component along the flight path.

An interesting thing happens, however, when the flight path goes to horizontal flight. If you add weight to a powerplane, its level-flight top speed will go down. (because of higher induced drag).

Flaw in the 1-g turning description

hevansrv7a said:

Let's test this idea - you guys with g-meters. Bank the airplane to 45 degrees with no back pressure. Observe your g-meter. If you were holding level altitude at 45 degrees the g-meter should say 1.4. What does it show if you bank 45 degrees and allow the airplane to sink? How about 90 degrees? But it did turn, didn't it?

In a maneuver to reverse course while allowing loss of altitude I contend you can pull 45 or even 60 degrees, roll out and not pull greater than 1.0 g's. Remember that you must accelerate vertically to go from zero fpm to 1500 fpm vertically even if you eventually stabilize at 1500 fpm by which time you have turned a lot. That's the idea - that a steeper turn gets you around so much faster that the altitude penalty is proportionally less than with a shallow turn.

If you pull 1.4 g's at 45 degrees you will hold altitude and the wing force will be 70% vertical and 70% horizontal, but 70% of 1.4 is approx. 1.0 so there's enough lift to keep you up. Now if you pull 1.0, then you have less turn force and less vertical lift. You will turn at a given rate and you will sink at a given rate. For a while, both are increasing. You won't turn as fast as a 1.4 g turn but it's faster than a 30 degree sinking turn, etc.


Viewing this as a steady-state question is erroneous analysis IMHO. I'm not any kind of engineer, but I can fly a little and I can read a meter. Who will try this before the WX here gets better?

Click to expand...

You (and other posters) are of course correct that angle-of-attack is the key parameter for stall.

The flaw in this 1-g turn argument is that the transient time when the airplane is accelerating downward is very short compared to the time to make the turn. Except for the initial push-over to maintain airspeed, and the initial roll into the turn, it really will be a fairly steady gliding turn with relatively constant descent rate.

YES, You can fly the maneuver you describe if you would like. If you keep trying to turn and maintain just one g, by the time you get turned around, you will be pointed almost at the ground. Then you will have some g-s to pull out to a steady glide. Yes, you can turn around this way, but it is not the turn with the minimum altitude loss.

For the purpose of a gliding 180-degree turn, the minimum altitude loss will be pretty close to a steady gliding turn flown at the angle of attack that corresponds to the angle of attack for minimum sink in level flight. It will be at a higher speed, and higher g's, but as you say, it will be at that angle of attack.

And remember folks, the real challenge in trying to do this, under stress, with perceptions altered by high sink rate and wind drift near the ground, is to make a COORDINATED gliding turn. As the other thread emphasized, it is a human factors issue.

Ah, lift coefficients....

CraigC said:

Steve,

If you are going to do this. I would hope you would go all out and bring in Coefficient of Lift and angle of attack and all the other important variables that need to be known to be understood. It is all what the wing "sees" and how much lift is being generated that matters. As a pilot I want to see the hard actual math that makes it all happen. Thanks for diving into this. I am looking forward to asking many questions.

Click to expand...

Hi,

In trying to describe this stuff to people, I try to avoid the "lift coefficient" type stuff. I'm afraid people get distracted or bogged down in unfamiliar vocabulary, and miss the meat of the ideas.
I can certainly do that if there is interest, but probably better in PM's.

Steve
RV-8QB FWF

Truly great explanation of the aerodynamics of flight related to the turn back. Best I have ever read.

Some experience from sailplanes that may be helpful: As airplane pilots, we consider wind shear as a problem, and it can be. But wind gradient is so common that we hardly consider it. The 15 knot headwind at 50' becomes the 10 knot at 10', but we are used to that and expect it to the point of not even noticing it. In the downwind landing we are theoretically attempting, it is the opposite, and in this case it is a plus. I certainly hope that any attempt at turn back happens so that the wings are level at some reasonable point above the runway. From there, the 15 knot tailwind at 50' becomes a 10 knot tailwind at 10', easing the descent instead of hastening it. We have in effect gained 5 knots airspeed instead of having lost it, given the same groundspeed.

While this is almost like understanding theory, it does point out that practice at altitude ISN'T ENOUGH! Perceptions are different and the actual conditions are different. I have done very few downwind landings, but I plan to do some practice approaches, just to help understand these things. I do know that 60 degrees of bank gets me around in the least altitude loss. Now, to nail down exactly how much that is and determine how much "cushion" I want, just to be safe.

Great thread!

Bob Kelly

You've experienced 'dynamic soaring'

Stephen Lindberg said:

Steve Smith: Thanks for your excellent aerodynamic exegesis. You've stated the facts better and more succinctly than any textbook I have read. I have avoided the turnback/downwind turn thread you referenced because that subject tends to generate more heat than light, but finally I cannot resist telling of a crop dusting experience I had as a young fellow. I think it illustrates some points you and others have made.

It sometimes happened that the air close to the ground was calm and smooth, while up two hundred feet, above a shear level, a pretty brisk wind would be blowing. This most often occurred near the ocean. I soon learned that pulling up (and decelerating) into an increasing tailwind created a very uneasy feeling in my stomach and an airplane that wallowed rather than flew, while pulling up into an increasing headwind was very reassuring. This was in the pre-microburst/windshear accident era and the mantra then was, "The airplane doesn't feel the wind". I knew from practical experience that that wasn't true but my understanding was incomplete. I asked questions at safety seminars and was laughed out of the room. Now attitudes are very different.

I also learned that turbulent air has a marked negative effect on aircraft performance. That is not often written about and is, I believe, underappreciated by many pilots.

Crop dusting is very educational if it doesn't kill you. My hat is off to those who make it a career. It is hard, dangerous work. I had a fortunate combination of luck and youthful enthusiasm and after a few seasons I escaped unscathed.

Steve, perhaps some refinements to make RVs fly faster will occur to you as you build. We are all eager students of aerodymanics and you may be the teacher we have been waiting for on the forums. Welcome, and please write more.

Click to expand...

Hi,
yes, what you said is exactly right. Again, it is the wind gradient with altitude that causes the effect. Albatross and Frigate Birds are well known for exploiting this, its called "dynamic soaring". There has been some recent active research on applying this to UAV's, notably from Stanford U.

Steve

RV-8QB FWF

thanks for the practical experience insight

osxuser said:

I think we can all agree that in the turn-back situation we aren't looking to maintain level flight. That said, steady-state is a good explanation, but steady-state at 1G. How many people have gone out and tried turnbacks? I've done them pretty extensively, mostly in Cessna's (150/152, 172, 177) And a little for my Commercial (PA28R-200). I can promise you in practical application, you are holding ~1G in steady state descent for almost the entire maneuver.

Now with keeping the 1G steady descent, we aren't trying to maintain a pitch attitude, we are shooting for best glide (theoretically, lowest Lift/Drag ratio). Thus, in the 60? bank turn the sink rate is astronomical, because the majority of the lift is going to the turn, not keeping the airplane flying. Coming out of the turn, point the nose at the numbers, nail the best glide and keep it coordinated.

The increased weight does help the glide, and keeps you floating longer in the round-out/flare, momentum being a factor here I assume)

If you haven't done turnbacks, don't try them for the first time when your engine dies, just land straight ahead.

A couple of pointers, an offset when you hit the end of the runway (if allowed per noise abatement) of about 20? will greatly help if you have to come back, allowing enough room to make the turn, and almost line right up on final. Second, find someone who understands the theory here, and can teach it if you're not comfortable, but PRACTICE THESE, the split second decision from practice could be the difference between making it back and not. I can make the turn-back in the Cardinal at about 400' with practice, but I limit myself to start it at 600' in the actual event due to the 3-5 second delay of recognition.

Click to expand...

Thanks for adding your practical experience. My bet is, if I instrument your airplane with an accurate data-logging g-meter, we would find that you are pulling 1.2 or 1.3 g's, and it seems like ~1-g to you. As you roll out and establish best glide, I bet you pull a little too. Its a pretty dynamic maneuver, and amid everything else, it would be hard to really tell. That said, your main point is that it is probably not 2 g's. You are keeping the nose down as you turn and letting it accelerate a little.

Anyway, I think its great that you chime in as someone who is actually training and practicing this. We should be wary of what "B25Flyer" calls "internet flight instruction". Another poster who does low-level in supercubs just added that he went out and tried it in his RV, and it took 800 ft for him, experienced and skilled in doing wing-overs at the edges of fields.

Steve
RV-8QB FWF

Flutter

tin man said:

IMO Martin Hollman only has one agenda and that is to sell his book and services. His really only claim to fame is some work he did on the Lancair 4. Remember the Stallion? Total flop. I wouldn't believe anything he says.

Click to expand...

No, I don't know anything about "the Stallion". Regardless of Hollman's agenda, what about formal flutter analysis of the RV models? I would think this would be of considerable interest to all of us.

flutter analysis extremely expensive

A simple, approximate flutter analysis can be done easily, but has very little value because it ignores the kinds of details that can make a significant difference.

A real, full-blown flutter analysis like would be done by a big airframe company, would be extremely expensive. Like several millions of dollars. First a high-fidelity Finite Element Model (FEM), or a low-fidelity FEM tuned by experimental measurements; these are used for a detailed dynamic analysis, then a ground vibration test that validates the dynamics, and identifies additional effects of counterbalance, control pivot slop, cable stretch, that kind of thing.

There are companies that do this. There used to be a professor at U. Texas-Austin, I think his name was Ron Stearman, that would use a team of grad students to do it as a learning exercise, at a somewhat reduced price compared to a company. I don't know if he still does it or not.

For wings of such low aspect ratio, flutter speed is probably incredibly high, maybe even transonic. The flutter modes that might be dominant are probably related to control surface counterweighting and control slop, but even those modes are pretty high because our structures are so stiff.

Steve.
RV-8QB FWF

Low AR helps rigidity; hurts aero efficiency

Steve, Thanks for taking the time to post all of this. Being a practicing engineer myself ( and ex GD flight test), I wanted to throw in an observation:

The low aspect ratio, rectangular planform of the RV wing does help in terms of torsional rigidity, and therefore would help control structural flutter (not necessarily control surface flutter though) But it does not help in terms of aerodynamic efficiency.

I might point out that although the RV's do spend most of their time in high speed cruise, the rectangle planform still extracts a price in terms of design efficiency.

A tapered planform could produce the same low speed lift with less area, given other factors being equal. That area reduction helps to reduce both parasite and induced drag.

Not to mention a tapered planform reduces roll damping, which leads to smaller ailerons which leads to larger flaps, which further reduces stall speed and/or wing area.

Allowing for a redesign of the flaps, a tapered planform can out perform the RV wing on both ends of the speed spectrum with much less area. Add a newer airfoil, and the results get pretty impressive...

scsmith said:

A simple, approximate flutter analysis can be done easily, but has very little value because it ignores the kinds of details that can make a significant difference.

Click to expand...

Thanks. I guess, according to Martin, that's what was done before.

A real, full-blown flutter analysis like would be done by a big airframe company, would be extremely expensive. ...

Click to expand...

OK, but could a reasonably close approximation be done cheaply by:
- using a good code on a PC. Doing FDM and FVM codes in water resources I know that off-the-shelf PCs can handle numerical tasks that required a half-million dollar machine a decade or two ago.
- good quality input of structural detail by interested VAF'ers.
- bare minimum ground-truthing (emperical measurements) which surely some interested homebuilders would be interested in doing.

For wings of such low aspect ratio, flutter speed is probably incredibly high, maybe even transonic. The flutter modes that might be dominant are probably related to control surface counterweighting and control slop, but even those modes are pretty high because our structures are so stiff.

Click to expand...

That sounds good about the wings. What about people adding or extending fuel tanks? Wings would still be good without analysis? The other concern I had was painting the control surfaces. I always hear for certificated aircraft how when you paint you have to be very careful to counterbalance again. What should we be concerned with wrt the RVs and painting of control surfaces?

another good read on this topic

Excellent thread! another book worth reading is "Stick and Rudder" it explains much if this in layman terms.

pat said:

Excellent thread! another book worth reading is "Stick and Rudder" it explains much if this in layman terms.

Click to expand...

Interesting that a book published in 1944 is still getting recommendations and I agree that it is a real good book.

I find his advice to pull the nose up and slow down if you are too high on approach works particularly well in RVs.

rfinch said:

Great! An excellent review of basic aero....now I have someone I can explore a few nooks and crannies with

I've become a bit interested in flutter with respect to the RVs. I talked to a presenter yesterday at the Golden West Flyin, Martin Hollman, owner of Aircraft Designs, Inc. Talking to him after his presentation, he said that the RVs have never had a proper flutter analysis done. To his knowledge Van did pay for an analysis, but it was a cheapo method known for inaccuracy.

Obviously in practice this is not a great problem because RVs do not have a reputation for coming to pieces in flight. Still, I would like to see a trustworthy analysis done of the several RV models using an accepted, modern numerical code. I suppose the time-consuming part would be inputting the structural elements. Perhaps the RV crowd could help with the grunt work?

Click to expand...

Of all the gurus in this business, Mr. Hollman is not my favorite.

I remember him being somewhat of a pain in the side years ago when guys were building and flying the canard airplanes created by Burt Rutan. He had criticism of that effort. Now he has taken on another successful effort, how come? Seems like writing books on aviation design is a good idea but questioning the integrity of the most successful home built design ever won't make me go out and buy one of his books. I much prefer "Stick and Rudder" from 1944.

As you say, "....this is not a great problem because RVs do not have a reputation for coming to pieces in flight" which I presume means it is not a problem. So why go to all the trouble and expense of crunching so many numbers to come up with a conclusion it is not a problem?

I've read all your stuff here, Steve, and it is good technical information most of which I understand and agree with.

It is curious though in that with all the training I've been subjected to since 1959, there never was discussion or procedure about turning back to the runway with an engine failure. Maybe there's a reason it was not on the agenda - an operational reason.

When we took off in basic training we were cleared very close behind the guy in front of us, like when he broke ground it was go. You could see 2 or 3 or 4 airplanes ahead in the pattern as you rolled day and night. Wouldn't it have been interesting if one of them turned around and came back head on about the time you were ready to break ground?

Same can be said about a departure from a busy place like OSH during the convention. A turn around at that place at that time could be a disaster.

Technically the procedure is doable in any airplane if the pilot knows how it will perform. But I am really hung up on the variables of the maneuver. I'd hate to get half way through the turn seeing nothing but dirt to maintain flight and also seeing the airplane is not going to make it to the runway. Seems like going straight ahead would present more options of where the flight will end and for that reason will stick with that plan.

David-aviator said:

As you say, "....this is not a great problem because RVs do not have a reputation for coming to pieces in flight" which I presume means it is not a problem. So why go to all the trouble and expense of crunching so many numbers to come up with a conclusion it is not a problem?

Click to expand...

Fair question.

I'm building a -9A which recommends a max engine power of 160 hp....because of potential flutter problems at high cruise speed. That article got me interested in this flutter business. Not that I'm interested in putting a larger engine in mine, but limits always interest me. Why is the limit at that point? How good was the analysis?

Second, some builders are departing from the kit and plans. This is experimental, after all. In particular, for the -9, it's not uncommon to find builders extending or adding fuel tanks in the wings. Most consider the additional static load, but none consider changes to flutter.

Third, I'd like more practical information about control surface flutter and paint. I have a vague notion, from reading about control surface balancing of certificated aircraft after repainting, that this is important. Yet I can't recall any concern about (re-)balancing control surfaces for the RVs. What should we be looking for? How close are we to flutter? Will the flutter come on gradually, with warning, or suddenly to destruction?

Finally, numerical models are my career and always of interest to me. Very powerful codes are available these days for common PCs which can do quite sophisticated calculations in many fields; collecting data and inputting it is more often the bottleneck.

I have been waiting for years for someone to explain why: the Pitts S1S from at or near Vne will fly the length of a 300" runway in a perfect 90 degree bank, and then climb 500 feet, still in a 90 degree bank. It will also make a very rapid 90 degree turn on rudder only with the wings perfectly level. The wings level turn bleeds off speed MUCH more rapidly than the "knife edge". Knife edge is a zero G maneuver. I suspect if the G meter were rotated 90 degrees, it would read 1 G in level knife edge flight. I suspect that the Pitts is using every vertical or near vertical surface for lift in knife edge. Fuselage, wing struts, landing gear and wheel pants. The spring gear Pitts does not knife edge nearly as well as the standard V gear.

Leighton Collins started the magazine Air Facts around 1939. Wolfgang Langwiesche wrote for Air Facts, Stick and Rudder was allegedly inspired by the acceptance of Wolfgang's articles in Air Facts. Collins and Langwiesche talked about wind gradient, aileron spins and many similar issues for years before others caught up. Ask any group of pilots today "who was Wolfgang Langwiesche" and you will get a lot of blank looks.

knife-edge lift

Hi,
about the comments on knife-edge flight, I believe the poster is correct, the airplane must be getting 1'g worth of lift from all the side area on the airplane. The induced drag to do this is tremendous, because the effective wingspan is so short, but with enough hp, it works.

I also wanted to note that I think "stick and rudder" is indeed a great book. I provided a recommendation for an aerodynamics text, but totally true that you can learn a lot about flight mechanics by reading 'stick and rudder' and it language that will make sense.

Steve

RV-8QB FWF

Hey, no fair peeking in my garage!

Bill Wightman said:

Steve, Thanks for taking the time to post all of this. Being a practicing engineer myself ( and ex GD flight test), I wanted to throw in an observation:

The low aspect ratio, rectangular planform of the RV wing does help in terms of torsional rigidity, and therefore would help control structural flutter (not necessarily control surface flutter though) But it does not help in terms of aerodynamic efficiency.

I might point out that although the RV's do spend most of their time in high speed cruise, the rectangle planform still extracts a price in terms of design efficiency.

A tapered planform could produce the same low speed lift with less area, given other factors being equal. That area reduction helps to reduce both parasite and induced drag.

Not to mention a tapered planform reduces roll damping, which leads to smaller ailerons which leads to larger flaps, which further reduces stall speed and/or wing area.

Allowing for a redesign of the flaps, a tapered planform can out perform the RV wing on both ends of the speed spectrum with much less area. Add a newer airfoil, and the results get pretty impressive...

Click to expand...

Hi,

All your ideas on improved wing performance are right on. Remember Van's basic premise of all-around sport-flying performance and a safe-handling airplane. So we can't disagree with his decision to use a rectangular wing with lots of area. Also, in the induced drag, rectangular wings actually are not as bad as linear theory would indicate, because of some interesting second-order effects of the large tip side edge at angle of attack (read my thesis!)

However, you are right that if you want to move the design toward more efficient long-range cruise, perhaps at the expense of some 'sportyness', a tapered wing with a bit more span, a fair bit less area, maybe a mild laminar flow airfoil, slotted flaps like the Mooney and RV-10, would be really nice.

But like I said, no fair peaking in my garage!
25 ft span, 98 sq ft, taper ratio =0.6 NLF-0114 section. In principle your comment about improving at both ends of the envelop are right. In design practice, I'm struggling to match stall speed with the reduced area, even with the slotted flap. Partly its because the NLF section doesn't have the CL-max of the 5-digit section. I'll come out of the closet more once my RV-8 is flying with the stock wings first.http://www.vansairforce.com/community/images/icons/icon12.gif

Steve.
RV-8QB FWF

What a tease!

Great, now I can't concentrate on fixing the dishwasher, thinking about the speed and handling improvements to be had with a "modern" airfoil, tapered planform, and slotted flaps. More power to you for advancing the state of RVdom. Have you started cutting metal yet? No? Drawn plans, then? I am sure many of us have an intense curiosity about such a project, I know I do.

As a new RV4 owner I have noticed a few things. First, it has very reassuring slow flight characteristics with good aileron control through the stall, with adequate (but not great) aerodynamic stall warning, and a slow stall speed due to adequate wing area. That means I would likely survive an off airport landing in the event of an engine failure. I would not want to own a homebuilt that didn't have these attributes. I know a Glassair III builder who sold it because he felt it lacked this very thing. So question one is: can you keep the stall speed low using slotted flaps on a small area wing? Without excessive weight? I didn't mention complexity or expense because we homebuilders can absorb that gladly.

Another thing about the four I have noticed is that the generous wing area makes for a rough ride at high speed in turbulent air, enough so that I frequently have to slow down. Saves gas, though, so maybe it is all good. I am willing to suffer this loss of speed for the low end handling. I think Van made this choice deliberately. For those who want to go faster on a smaller wing, we eagerly await the results of your project.

Care to comment on the Rocket EVO wing? If I can read between the lines on what I have read about the EVO, it may be a long run for a short slide.

Won't the tail have to be sized upward for the tapered wing? At least on the four the tail size seems to be adequate but not large enough for standard category handling characteristics. In my limited experience, that is the thing that sets the RV4 apart from production airplanes, and the RV4 flying characteristic that an inexperienced pilot will have the most to learn about.

It seems to me that the key to flying fast on the RV4 wing is to keep the induced drag low by keeping it light, and thereby flying at the least angle of attack. I think Van has said as much. Many times.

Sorry if this is a thread hijack. Good luck on your project and please keep all of us informed. Now, back under the sink...

Uhh, what's that again?

scsmith said:

Also, in the induced drag, rectangular wings actually are not as bad as linear theory would indicate, because of some interesting second-order effects of the large tip side edge at angle of attack (read my thesis!)

Click to expand...

Please, some further explication. Large tip side edge at angle of attack? Now I am really curious. There is so much to know.

New wings in the works...

Steve, funny thing. We're within a couple percent of each other in terms of area, taper ratio and flap design.

And, yes I am cutting (and hydroforming) metal.

The rectangle wing will be left in the dust.