Static margin… does it really exist?

[Gonzalo]

Explain this will be kind of difficult to me because of my bad English but I’ll do my best.

I was reading a book about model aircraft aerodynamics, and it defines the static margin as the distance between the CG and the “neutral point”, being the “neutral point” the place where the combination of all lift and drag forces effectively act.

It also says that a plane in order to be stable must have its CG in front of the “neutral point” and this distance is called the static margin.

In my opinion if the CG is in front of the neutral point, it will produce a nose down moment and in order to avoid a dive “up elevator” should be applied, that will reduce the lift force of the tail (or even make it a negative lift).

This change on the tail lifting force without changing the main wing lifting force, will move the “neutral point” forward closer to the main wing aerodynamic center (or even in front of it if we get a negative lifting tail) until the “neutral point” is at the same longitudinal level of the CG, because this is the only way to achieve a level flight.

Bottom line there is nothing as a static margin because level flight requires “neutral point” longitudinally coincident with the CG.

am I wrong?

regards

Gonzalo.

[globemaster3c17]

If I'm reading you correctly, then you are saying that the center of lift and center of gravity would be the same spot. On the average plane, that is not correct. The CG is typically ahead of the CL, and the tail produces negative/down lift. With this setup, the wing actually produces more lift than the weight of the aircraft, the difference due to the down lift produced by the tail. A constant pitch attitude is held when the nose-down moment of the CG being ahead of the CL balances the tail-down moment produces by the horizontal tail surfaces.

[BMatthews]

In my opinion if the CG is in front of the neutral point, it will produce a nose down moment and in order to avoid a dive “up elevator” should be applied, that will reduce the lift force of the tail (or even make it a negative lift).

This change on the tail lifting force without changing the main wing lifting force, will move the “neutral point” forward closer to the main wing aerodynamic center (or even in front of it if we get a negative lifting tail) until the “neutral point” is at the same longitudinal level of the CG, because this is the only way to achieve a level flight.

You're right, the CG being in front of the neutral point (NP) does generate a nose down moment and the elevator needs to be trimmed so it's lift coefficient varies to balance that force. But you're wrong in saying that the NP shifts. If the NP shifted to coincide with the CG then the CG would not be generating a pitching moment any more. But that moment is always there. Besides, if you look at the calculations for finding the NP you'll see that it's all based on the design geometry and has nothing to do with tail downforce values or lift. So what's happening? How does it balance this moment of force?

The CG being in front of the NP produces a nose down moment acting on the NP just as you said. But the tail trim also produces a tail down moment acting on the NP to balance the CG moment. The common point is the NP, which remains static, and the rotational vectors are balanced around that point. The overall lift is always through the NP as the center of the whole airplane's lifting effort but the elevator trim produces a pitching moment to balance the effect of the forward CG. Since the elevator pitching moment is speed related we see the pitching moment change as the model's speed changes. And if the elevator trim stays constant the model will attempt to hold the same airspeed over a fairly large angle of attack change. If you throttle up the nose will come up as the model responds to the rise in airspeed and then the airspeed drops with the nose still high until the CG moment and the tail down moment are once again balanced. And the airspeed at that point will be very close to the same airspeed as it was in level flight. Similarly if we cut the throttle the model will slow, the nose will drop from that forward CG moment and the dive will make the airspeed rise again to be very close to the original level flight speed. The "very close" is used because as the model rotates to nose up or nose down the CG shifts to a new angle compared to the NP location and that slightly shortens the horizontal part of CG moment arm... trigonometry and all that you know....

[CloudyIFR]

I have a spreadsheet that's written in MS Excel that will do all the math calculations for you. All you need is a ruler that measures in inches.

Email me and I'll send it to you.

suterc@msn.com

An article will be written someday and I'll have someone host it for me.

Curtis
Montana

[Andy W]

Why not zip it, and post it here as an attachment?
As a side note, we (EZonemag) can certainly host the article for you ..
..a

[vintage1]

If the plane is flying in steady climb, level flight, or glide, then the CG MUST be at the point where all rhe other 'forces' average out to meet, otherwise the plane wouldn't BE in steady flight etc etc.

So if the neutal point is defined as the place at which all these other foces apart from weihght can be 'considered to act' then obviously it moves about all over the place as you vary the speed and the elevator setting etc.

Otherwise you would not be able to pull the plane OUT of a steady flight in any given direction.

A stable plane in pitch, has an NP that moves fowrard with increasing speed.

An unstable plane has one that moves back.

[adam_one]

vintage1,
You don't take into account the effect of the stabiliser...
You may increase or the decrease the lifting force at Neutral Point, (which will make the plane climb respectively dive) but the NP location depends on the wing's and stabiliser's AC locations, which remain in one place at all usable flight conditions.
During steady flight the stab continuously corrects the altitude in the same way as a zener diode stabilises the voltage by oscillating at its zener voltage with a frequency that is filtered by the circuit's capacitance. The stab's oscillation is dampened by the inertia.
By locating the CG in front of NP creates a negative feedback to the oscillation process, which gives a positive stability (self-correcting).
By locating the CG behind of NP creates a positive feedback to the oscillation process, which gives a negative stability (unstable).
By locating the CG coincident with NP there's no feedback and the system is considered neutral stable (not self-correcting).

- Positive stability - tends to return to original condition after a disturbance.
- Negative stability - tends to increase the disturbance.
- Neutral stability - remains at the new condition.

[CloudyIFR]

AndyW

Thanks for the offer and I will certainly take you up on it, however, I want to first write an article for it. I'm slow at doing such things but as soon as I get it done I'll certainly take you up on the offer.

Thanks
Curtis
Montana

[Ollie]

http://www.djaerotech.com/dj_askjd/dj_questions/hlgcg.html
http://ciurpita.tripod.com/rc/notes/neutralPt.html
http://www.engin.umich.edu/dept/aero/msgc/kids/gliderdesign.shtml

[BMatthews]

Semantics is at work here....

I can't find it at the moment but the calculation for the neutral point is pureley a geometrical equation that only considers the size and layout of the wing and stabilizer. No mention of speed enters into the equation. So the neutral point, UNDER THIS DEFINITION, is not a point that moves with the speed or attitude of the model.

To balance the weight and moment arm (or torque) of the CG being in front of this NP we have the torque of the stabilizer downloading. The resulting torques cancel each other out in level flight and we have a plane that is stable. I believe you'll find that this is how current aerodynamic engineering explains this matter.

If you want to consider a point where the lift all centers on so it's purely an upwards force lifting against the CG then Center of Lift would probably be more accurate than Neutral Point. But here again the CoF is always going to be located right at the CG for any time the airplane is in stable non-pitching flight. Forcing the CoF to move forward or back with elevator input produces a pitching force that rotates the airplane as desired.

The reason this CoF idea is not a good one is that you assume it is always lifting UPWARDS. This concept falls apart or gets very messy when you consider a model flying around a loop. It's far easier to use the proper Neutral Point with torque values as described in my previous post and the first paragraph here.

[markdrela]

Wow. There's a lot of incorrect definitions flying about here.

CP = center of pressure = force centroid of the wing+tail combination. This moves around during maneuvers, but in trimmed flight the CP sits at the CG location.

NP = neutral point = the location about which the pitching moment of the whole airplane is independent of the AoA. If you artifically constrain the airplane to pivot about NP, it will have no resistance to AoA changes, or be neutrally stable. Pivoting the airplane ahead of the NP will make the moment resist AoA changes. In reality the airplane wants to pivot about the CG, so placing the CG at the NP gives neutral stability. If the CG is ahead of the NP, the airplane is positively stable.

There is a common misconception that a stable airplane must have a downloaded tail. Not so. The direction of the tail load at any given stability margin depends mainly on:
1) Tail volume. Larger tail volume make the tail load more positive.
2) Wing airfoil pitching moment Cm. More negative Cm makes the tail load more negative.

A sufficiently large tail can carry an upload, and still give a stable airplane. Many FF airplanes have large uploaded tails.

If the wing airfoil has a zero Cm, even a tiny tail can carry some upload and still give a stable airplane.

[vintage1]

vintage1,
You don't take into account the effect of the stabiliser...
You may increase or the decrease the lifting force at Neutral Point, (which will make the plane climb respectively dive) but the NP location depends on the wing's and stabiliser's AC locations, which remain in one place at all usable flight conditions.



Semantics. You are defining the neutral point as somethimng that doesn't include the stabilizer. I am defining it as something that does. Or using the definition the OP used, which is the same.

Whether you say that omncreasd speed causes a tail down moment to be applied by the stab, or say that incraesed speed causes a rearward shift in the NP, doesn't amount to a hill of beans. Its the same thing described by two different sets of words.



During steady flight the stab continuously corrects the altitude in the same way as a zener diode stabilises the voltage by oscillating at its zener voltage with a frequency that is filtered by the circuit's capacitance. The stab's oscillation is dampened by the inertia.



What? Zener diodes don't oscillate!!.
I think I understand what you are trying to say, possibly better than you do, ....but actually the aerodymanics are an overdamped system, so there is no oscillation, unless you get into a stall/dive/stall/dive type of porpoising, which shows you have too much stability feedback and not enough damping....:D


By locating the CG in front of NP creates a negative feedback to the oscillation process, which gives a positive stability (self-correcting).
By locating the CG behind of NP creates a positive feedback to the oscillation process, which gives a negative stability (unstable).
By locating the CG coincident with NP there's no feedback and the system is considered neutral stable (not self-correcting).

- Positive stability - tends to return to original condition after a disturbance.
- Negative stability - tends to increase the disturbance.
- Neutral stability - remains at the new condition.


As I said, I suspect I understand control system theory and negative feedback better than you do. ;)

[vintage1]

Wow. There's a lot of incorrect definitions flying about here.

CP = center of pressure = force centroid of the wing+tail combination. This moves around during maneuvers, but in trimmed flight the CP sits at the CG location.

NP = neutral point = the location about which the pitching moment of the whole airplane is independent of the AoA. If you artifically constrain the airplane to pivot about NP, it will have no resistance to AoA changes, or be neutrally stable. Pivoting the airplane ahead of the NP will make the moment resist AoA changes. In reality the airplane wants to pivot about the CG, so placing the CG at the NP gives neutral stability. If the CG is ahead of the NP, the airplane is positively stable.

There is a common misconception that a stable airplane must have a downloaded tail. Not so. The direction of the tail load at any given stability margin depends mainly on:
1) Tail volume. Larger tail volume make the tail load more positive.
2) Wing airfoil pitching moment Cm. More negative Cm makes the tail load more negative.

A sufficiently large tail can carry an upload, and still give a stable airplane. Many FF airplanes have large uploaded tails.

If the wing airfoil has a zero Cm, even a tiny tail can carry some upload and still give a stable airplane.


Absolutely. What counts is the derivative with respect to speed of the pitching moment.

The tail can be a lifter, as long as the increase in speed generates MORE lift on the front wing than on the rear.

The extreme case is a canard. With rhe CG somewhere between the for and aft wings, and BOTH definitely doing some lifting.

Its allegedly more efficient because the stabiliser in long tail moment and big stabilser planes contributes to lift.

[HELModels]

....but actually the aerodymanics are an overdamped system, so there is no oscillation, unless you get into a stall/dive/stall/dive type of porpoising, which shows you have too much stability feedback and not enough damping....:D




;)

It is Futile, I mean Phugoidal.

The novice designer, such as myself, must decide alot of issues that determine the amount of dampening. The first thing is to design a crash resistant and adjustable airframe so that you can fiddle with the issues. Nothing worse than needing to completely rebuild to get the answer you seek. If you are fortunate enough to know the calculations needed, then crashabletestbed might not be necessary. I'm not that lucky. :o

[adam_one]

What? Zener diodes don't oscillate!!.
Yes it does, and you can see it as a noise by checking with an oscilloscope across any zener without a filtering capacitor. :p

....but actually the aerodymanics are an overdamped system, so there is no oscillation,
Wrong, in every control loop system there are oscillations with more or less amplitude and frequency depending on how the integral, proportional and derivative controls are adjusted to the actual system's inertia.
I really suspect you don't understand control system theory and negative feedback at all. :D

[Gonzalo]

Semantic problems… let see it, I’m basing my point and my definition of Neutral point on the following sources that I copy literally to avoid errors due to my poor English:

“RC Model Aircraft design” by Andy Lennon (2002 edition): Page 27: “REAR CG AND THE NEUTRAL POINT Modern aerodynamic analysis for assessing the stability of an airplane is based on the fact that a win and tail plane represent a pair of airfoils in tandem. Each has its own aerodynamic center, but the combination will also have a corresponding MAC equivalent to the point where the total lift (and drag) forces of the two airfoils effectively act. This MAC is called the Neutral Point”

“Model aircraft aerodynamics” by Martin Simons (2002 edition): Page 148 “12.15. The neutral point. As described previously, every wing or wing-like surface in an airstreams at a moderate angle of attack has an aerodynamic centre close to the quarter chord point. This applies to fins, tail planes, fore planes and such streamlined shapes as struts, wheel spats, nacelles, faired undercarriage axles etc. Even long, slender forms such as arrow shafts or fuselages have an aerodynamic center and this is normally close to the quarter length position for moderate angles of attack. If the structure of a model is fairly stiff, it may be treated as a fully rigid body. Then it is possible to regard the entire aircraft as one object which produces lift and drag at some fixed point equivalent to the aerodynamic center of the whole.”

Rc Model World magazine, October 2004 “Aerodynamic forum” by Alasdair page 22 “Stability. Where is the neutral point? My guess, for what is worth, starts from basic principles. Stability is very simple. We imagine the aero plane trimmed in level flight and then being given a slight nose up pitch, which causes a small lift increase on each flying surface. The neutral point is the point where the resultant of all these increases acts…”

To me all those definitions are in agreement and mean that the neutral point is the point where the combination of all forces except weight and power trust effectively acts, and this includes the tail plane. My point is that under that definition in order to have a level flight the neutral point must be the same as the CG and can not be behind of it because it will cause a nose down and a dive.

[kallend]

Wow. There's a lot of incorrect definitions flying about here.

CP = center of pressure = force centroid of the wing+tail combination. This moves around during maneuvers, but in trimmed flight the CP sits at the CG location.

NP = neutral point = the location about which the pitching moment of the whole airplane is independent of the AoA. If you artifically constrain the airplane to pivot about NP, it will have no resistance to AoA changes, or be neutrally stable. Pivoting the airplane ahead of the NP will make the moment resist AoA changes. In reality the airplane wants to pivot about the CG, so placing the CG at the NP gives neutral stability. If the CG is ahead of the NP, the airplane is positively stable.

There is a common misconception that a stable airplane must have a downloaded tail. Not so. The direction of the tail load at any given stability margin depends mainly on:
1) Tail volume. Larger tail volume make the tail load more positive.
2) Wing airfoil pitching moment Cm. More negative Cm makes the tail load more negative.

A sufficiently large tail can carry an upload, and still give a stable airplane. Many FF airplanes have large uploaded tails.

If the wing airfoil has a zero Cm, even a tiny tail can carry some upload and still give a stable airplane.


I have an AT Phoenix rocket glider (Mark, I think you know Bob Parks, the designer) that I have balanced way behind the recommended position... it has a very large tail and even though marginally stable (in fact, I think it's actually unstable), it diverges so slowly that it's still OK to fly.

Anyhow, thanks for clarifying the definitions, it always makes things easier if everyone agrees on what the words mean.

[BMatthews]

.....To me all those definitions are in agreement and mean that the neutral point is the point where the combination of all forces except weight and power trust effectively acts, and this includes the tail plane. My point is that under that definition in order to have a level flight the neutral point must be the same as the CG and can not be behind of it because it will cause a nose down and a dive.

By those definitions you're quite right and frankly I don't know what to say since these folks have a reputation for knowing what they are talking about. But their concept doesn't match mine or the way I've seen some aerodynamic research types describe it lately in threads on that "other" forum, RC Universe.

There IS a point on the model where if you balance it there the model will show neutral pitch stability. I've always seen that described as the neutral point, a usage that matches with the other pundits I'm refferring to above. "My" neutral point is fixed by the wing and tail areas and the moment arms and the neutral point is determined by the tail volume coefficient and other fixed charactaristics of the aircraft design. Flight loadings or elevator trim does not enter into it so it cannot move around to follow the CG or produce pitching forces when you use the elevator.

Do Simons and/or Lennon describe such a point in their books?

It's still an error in sematics but perhaps it's my error.... :o

[vintage1]

Semantic problems… let see it, I’m basing my point and my definition of Neutral point on the following sources that I copy literally to avoid errors due to my poor English:

“RC Model Aircraft design” by Andy Lennon (2002 edition): Page 27: “REAR CG AND THE NEUTRAL POINT Modern aerodynamic analysis for assessing the stability of an airplane is based on the fact that a win and tail plane represent a pair of airfoils in tandem. Each has its own aerodynamic center, but the combination will also have a corresponding MAC equivalent to the point where the total lift (and drag) forces of the two airfoils effectively act. This MAC is called the Neutral Point”

“Model aircraft aerodynamics” by Martin Simons (2002 edition): Page 148 “12.15. The neutral point. As described previously, every wing or wing-like surface in an airstreams at a moderate angle of attack has an aerodynamic centre close to the quarter chord point. This applies to fins, tail planes, fore planes and such streamlined shapes as struts, wheel spats, nacelles, faired undercarriage axles etc. Even long, slender forms such as arrow shafts or fuselages have an aerodynamic center and this is normally close to the quarter length position for moderate angles of attack. If the structure of a model is fairly stiff, it may be treated as a fully rigid body. Then it is possible to regard the entire aircraft as one object which produces lift and drag at some fixed point equivalent to the aerodynamic center of the whole.”

Rc Model World magazine, October 2004 “Aerodynamic forum” by Alasdair page 22 “Stability. Where is the neutral point? My guess, for what is worth, starts from basic principles. Stability is very simple. We imagine the aero plane trimmed in level flight and then being given a slight nose up pitch, which causes a small lift increase on each flying surface. The neutral point is the point where the resultant of all these increases acts…”

To me all those definitions are in agreement and mean that the neutral point is the point where the combination of all forces except weight and power trust effectively acts, and this includes the tail plane. My point is that under that definition in order to have a level flight the neutral point must be the same as the CG and can not be behind of it because it will cause a nose down and a dive.


If those are the definitions then I totally agree with your conclsions.

And the implied conclusion that as airspeed increases, if the NP does not move forward, the plane will stay in the dive until it crashes.

[vintage1]

By those definitions you're quite right and frankly I don't know what to say since these folks have a reputation for knowing what they are talking about. But their concept doesn't match mine or the way I've seen some aerodynamic research types describe it lately in threads on that "other" forum, RC Universe.

There IS a point on the model where if you balance it there the model will show neutral pitch stability. I've always seen that described as the neutral point, a usage that matches with the other pundits I'm refferring to above. "My" neutral point is fixed by the wing and tail areas and the moment arms and the neutral point is determined by the tail volume coefficient and other fixed charactaristics of the aircraft design. Flight loadings or elevator trim does not enter into it so it cannot move around to follow the CG or produce pitching forces when you use the elevator.

Do Simons and/or Lennon describe such a point in their books?

It's still an error in sematics but perhaps it's my error.... :o


Ah. Now I understand where the confusion is arising.

That's the CG position for neutral stability...

Neutral stability is when the neutral point doesn't move around as the speed changes.

[adam_one]

Originally Posted by Gonzalo
My point is that under that definition in order to have a level flight the neutral point must be the same as the CG and can not be behind of it because it will cause a nose down and a dive.
Yes, the primary tendency is to cause nose down, but the tail counteracts that motion, hence the name stabiliser.
Any surface such as wing or fuselage, which has its aerodynamic centre AC ahead of the CG tends to destabilise the model, whereas any surface aft of the CG tends to stabilise it.
So, the answer to the question above is: Yes, Static Margin does really exist, and it should be between 5% to 15% of MAC in order to get a stable flight.

[Ben Lanterman]

It depends on how detailed you want to get sometimes - I read Model Aircraft Aerodynamics by Simons. Page 149 shows 3 cases - I'll attach them- didn't get his permission but I'll plug his book, everyone should buy one, good stuff.

The one for Neutral stability shows the sum of aero forces or the ac (aerodynamic center) at the neutral point but that of course is the definition of Neutral stability. The other two figures show stable and unstable cases and apparently have been ignored by the guys reading the book. What is missing is that the detail that the Lift vector (which is really the aerodynamic center including wing and tail needed to trim effects) does move around a little with angle of attack and tail deflections. The Neutral Point doesn't move though.

Remember NP (in percent of chord) = dCm/DCL - static margin. (if I have the signs right). It is a fallout of being able to accurately measure the slope of the pitching moment curve, knowing the wind tunnel reference center and knowing the CG location.

In flight you can work backwards and find it also but it is easier to see and work with by using wind tunnel data (my first aero job was reading NP slopes from wind tunnel data, hundreds of them, arrrrg - bad memories)

I'll attach a graph I made of the way the aero stable case would plot up for a 0-0-0 angle airplane with a randomly selected reference point (again it is chosen at random but is close to the final guess of CG) for ease of reading the slopes of the curves.

The blue line is the stable case at zero tail deflection. As the airplane is displaced to the high angle of attack it finds itself with increased lift and a negative pitching moment about the chosen Reference center to restore it to the original trim condition. The wing increased lift and the tail increased lift in the up direction. Again note that the figure is a pure aero case. When we apply the data to a real airplane the CG gets into the figure and you can end up with an initial tail setting that is either down or up depending on the airplane layout just as Mark Drela said (by the way, when Mark mentions something do pay attention, he knows wherein he speaks, he is not like one of those professors I never liked when in college - a prof that flies models would have been great to have - I am always at the wrong place at the wrong time).

For the real airplane case with a CG at the Neutral Point then the ac is shifted by the required elevator deflection to trim.

Remember

1. The ac definition is where the sum of aero forces end up to have zero moment when not considering the real CG effects. It will change with tail deflection and airloads.

2. The Neutral Point does not change as a function of stabilizer or angle of attack. It's the slope of the dCm/dCL curve. As the elevator changes the slope of the curve doesn't change. The position of the curve on the plot does change however. But Neutral Point is about slopes.

Note that the changes in the Lift vector in Simons's figures - that to be perfectly accurate they should include a moment due to the tail. I don't know if he intends that to be the case or not, lets assume so. I would prefer the figures to show wing and tail vectors but the figures are close enough to get the general idea.

The NP is a function of the geometric areas and their relative aero efficiencies. The aero center is the geometric location of the final aero configuration and takes into account all the lift and pitching moments due to deflections, aero efficiences and angle of attack and varies everywhere depending on those items.

I'm not too sure what I have left out - its early for us retired guys.

Ben

[markdrela]

I have serious issues with Simmons's figures B,C. These show that the center of lift is behind or ahead of the CG in trimmed flight. That's just plain wrong. In trimmed flight, the center of lift (i.e. the CP) is always at the CG, whether if the airplane is stable or unstable.
The equivalent statement is that in trimmed flight, the net moment about the CG is zero.

Stability deals with how the moment changes with perturbations away from trim, not what it is in the trimmed condition.

[Ben Lanterman]

he forgot to draw in the tail lift vectors. That might be a source of trouble, when I looked at it I guess I mentally filled in the correctly located arrows! Got to quit doing that some day:-)

[BMatthews]

Hence my earlier description of the rotational force due to flight trim between the wings and tail that counters the CG rotational force.

[Salto]

Stability deals with how the moment changes with perturbations away from trim, not what it is in the trimmed condition.

So Mark, can you describe for us how, when after a disturbance, the moments change to produce stable/neutral/unstable behaviour.

For me, understanding this point is the last piece of the jigsaw puzzle. I can see how it all works for very forward CG locations and a down lifting stabiliser, but for the life of me I can't see what forces and moments produce stable or unstable behaviour with a lifting stabiliser.

Graham.

[Ben Lanterman]

.............. "but for the life of me I can't see what forces and moments produce stable or unstable behaviour with a lifting stabiliser."

and I just hit the wrong key, hence the edit thing.

Anyway think of a model with a really long tail (as in a glider). The wing lift is at wing25%chord, the tail lift at tail24%chord and the CG is at wing40%chord.

The Neutral point for this stable airplane is about wing50%chord.

The tail is lifting upward to balance the moments about the CG and is set for level flight trim. The exact numbers would depend on fine tuning of tail lift and direction to keep it stable and working efficiently. It also needs to be able to come back to the trimmed lift/angle of attack condition. The incidence angle between the lifting tail and the wing will determine that effect.

A lot of us old timers have seen a freeflight power ship dive all the way to the ground when the stab trim timer didn't change from the climb setup to the glide setup. Your airplane diving for the dirt can be very stable in the dive but you don't want that in a freeflight airplane.

The increments in tail lift restoring the disturbed airplane work just as they do in the more conventional case of our "sport" airplanes.

All that is required to be stable is that the tail lift response to an up gust or a stick rap nose up is a nose down moment. It doesn't say anything about the direction that the trim lift is set.

We looked at a lot of airplanes in full scale work that are both unstable and have lifting tails. They depend on a finely tuned autopilot system though.

You can make any combination of the effects - stable, unstable, uplifting tails, and downlift tails. What you chose to do depends on the relative aero efficiencies of each surface in that particular design.

For instance you wouldn't want to load a tail with a bigger up load than a wing. It isn't an efficient design. Make the tail bigger and you have a tandem and some more and it is a canard which can be very efficient.

Something like that.

[markdrela]

For instance you wouldn't want to load a tail with a bigger up load than a wing. It isn't an efficient design. Make the tail bigger and you have a tandem and some more and it is a canard which can be very efficient. Something like that.

Ummm... "canard which can be very efficient" ??? :rolleyes:

Here's a pretty accurate description of a canard:
* conventional layout
* small, highly-loaded wing
* enormous lifting tail, with 300% to 500% of the front wing's area
* CG at 400-900% chord location. Such a far-aft CG is made possible by the huge tail.

That doesn't sound very efficient to me. :D

[Ben Lanterman]

Such an "interesting" way of looking at things :-)

[vintage1]

BUt reductio and infinitum is a valid way to look at things.

I am not gong to get involve in what a nuetral pint ism because I have seen posts where teh tai is lumped in, and where its cindiered sepoarately, and teh terms becme different.

The easiets way to LOOK AT IT in my headm is as follows.

For the machine to be in stable flight, all the aerodynamic foirces keeping iot up muts exactly balance the weight. So lift = weight.

For the machine to not deviate by pitching up or down, the vector sum of the aerodynamic forces must act thriugh the exact center of gravity. That means that what I TINK is called tehnutral point must lie excatly over the center of gravity.

For teh machine to RETURN to that stable state after a minor perturbation - i.e. have some stability - the change in neutral pont with airspeed must be negative. That is incresing airspeed must pull the neutral point forwards. So that in a dive, the nose gets pulled up, and in a climb the nose gets pulled down.

Where teh CG is with respect to the two wings can vary enormously according to teh size and layoiut of the wings. All yoiu can say is it will never be in front of the fromnt wing or behind the back wing. It can CERTAINLTY be BETWEEN the wings. Canards show that. Also contest gliders and FF models with big tails and long fuselages. CG can be aft of the TE on that sort of layout.

iTS NIOT where the NP is that determines stability its HOW IT MOVES.

Those that are considering the average lift vector of te wimng alone, and then saying the 'CG has to be in front' are not necessarily correct. A lifting tail is stable PROVIDED THAT the lift on the tail does not go up as FAST as the lift on the main wing does, with increasing speed. That's a function of decalage. Or wing section I suppose.

[Spinner]

Hey guys, Mark's got it right. We're getting confused by terminology big time. Sounds like the concepts of stability and trim are causing some problems. :(

The neutral point of an airplane is the c.g. location for which the airplane becomes neutrally stable in pitch--i.e., no restoring moment when the angle of attack is perturbed from trim, stick fixed. It does not move around in flight for unpowered gliders and rigid airframes. The location of the n.p. is usually predicted through geometry. However, the location is subject to effects such as power (slipstream on tail, direct prop "fin" effects, etc.). The classic engineering textbooks step through the power-off location first, then discuss effects of power. The n.p. is not the slope of the pitching moment curve vs. angle of attack, it is a discrete c.g. location and is related to the slope of the pitching moment curve. The slope of the pitching moment curves vs. alpha become zero as the c.g. reference point is moved aft to the location of the n.p. The resultant forces acting on an airplane in flight will not act through the n.p. unless you design it to be neutrally stable.

The center of pressure is the location of all the integrated aerodynamic forces acting on the airplane, and it must act through the c.g. for trim during unaccelerated flight. As angle of attack is perturbed from a trimmed condition, the individual forces acting on various parts of the airplane change, and the location of the c.p. moves, creating a restoring moment (for a stable airplane--c.g. in front of neutral point) or a destabilizing moment ( for an unstable airplane-c.g. further aft than the neutral point). Motion will occur until the c.p. again acts through the c.g. and the airplane retains trim.

It's awfully important to grasp the principles that Mark describes. One way to understand the concepts better is to remember that the airplane rotational motions occur around the c.g. location. Think of the airplane as being "pivoted" there for your analysis. You can interpret the results of aerodynamic forces a lot easier if you think of the c.g. as the point to sum the effects of individual forces, rather than the c.p. You should not think of the c.g. as "causing a moment", as this can cause confusion.

Hope this perspective helps a bit--and great job, Mark! :)

Joe

[Ben Lanterman]

Hi Joe, "The neutral point of ...............effects of power. "

You're OK up to here.

"The n.p. is not the slope of the pitching moment curve vs. angle of attack, it is a discrete c.g. location and is related to the slope of the pitching moment curve. "

I said the NP = xcg - dCm/dCL. (If I got the signs right) you'll find it in your textbooks. It is a location on the airplane where xcg (which is the static margin) goes to zero. If the wind tunnel reference point was chosed to be the final CG, then at that point the slope goes to zero. I can chose any point to be the wind tunnel reference point that still get the right NP.

To measure it in the wind tunnel we read the slope of the dCm/dCL curve and add the xcg appropriately. That gives the exact location of the NP. When you know the actual CG to be used in flight you can change the way the dCm/dCL curve slope is plotted and have a visual indication what happens when the airplane gets into the non linear portion of the curve at high or low angle of attack. The final CG corrected plot is used to determine the stabilator trim needed and you can work performance from there. I really do know what I am taking about (except for the appropriate signs). I have been a working aero eng for 30+ years. I always hesitate to put in an equation because off the cuff I generally mess up a sign or notation somewhere :-)

"The slope of the pitching moment curves vs. alpha become zero as the c.g. reference point is moved aft to the location of the n.p. The resultant forces acting on an airplane in flight will not act through the n.p. unless you design it to be neutrally stable."

Right.

"The center of pressure is the ............ nt). Motion will occur until the c.p. again acts through the c.g. and the airplane retains trim."

Keep in mind that the center of pressure is a mathematical construct and it is conveinent to think of it that way but as has been noted in the literature through the years it falls apart at several conditions. That is the reason that the Neutral Point concept came into favor.

"It's awfully.........should not think of the c.g. as "causing a moment", as this can cause confusion."

Pretty well agree but again remember you can sum forces and moments about any point on the airplane (or off) and get the same answers.

[markdrela]

For the machine to be in stable flight, all the aerodynamic foirces keeping iot up muts exactly balance the weight. So lift = weight.
Yep.


For the machine to not deviate by pitching up or down, the vector sum of the aerodynamic forces must act thriugh the exact center of gravity.
Yep.



That means that what I THINK is called the neutral point must lie exactly over the center of gravity.
NO. Wrong terminology. The Center of Pressure lies over the Center of Gravity in trimmed flight. The Neutral Point is at some other location, behind the CG if stable, and in front of the CG if unstable.

The CP moves as the airplane deviates from trimmed flight, but the NP stays very nearly fixed.

[Spinner]

Ben,

I think we agree on most everything, except your discussion of :

"I said the NP = xcg - dCm/dCL. (If I got the signs right) you'll find it in your textbooks. It is a location on the airplane where xcg (which is the static margin) goes to zero. If the wind tunnel reference point was chosed to be the final CG, then at that point the slope goes to zero. I can chose any point to be the wind tunnel reference point that still get the right NP."

The neutral point is where dCm/dCL goes to zero. Further, xcg is not the static margin. Static margin is the difference between xcg and xNP in percent mac. I also don't understand what you are saying in the next sentence.

Hope we can clarify our differences for the benefit of others.

My background: Retired 65-year old NASA aero engineer who spent 36 years in stability and control at Langley.

Thanks,
Joe

[Gonzalo]

There IS a point on the model where if you balance it there the model will show neutral pitch stability. I've always seen that described as the neutral point, a usage that matches with the other pundits I'm refferring to above. "My" neutral point is fixed by the wing and tail areas and the moment arms and the neutral point is determined by the tail volume coefficient and other fixed charactaristics of the aircraft design. Flight loadings or elevator trim does not enter into it so it cannot move around to follow the CG or produce pitching forces when you use the elevator.

Do Simons and/or Lennon describe such a point in their books?

It's still an error in sematics but perhaps it's my error.... :o


I searched both books and didn’t find a description for a point as your neutral point but on Annex 1 of martin Simons’s book I found a formula to calculate the “Neutral Point” position and it uses de position of the wing MAC, the stabilizer efficiency and volume coefficient and the slopes of the lift curve for both wing and stabilizer. So I guess he is talking about the same neutral point as you but describing it on a different way.

[Gonzalo]

The other two figures show stable and unstable cases and apparently have been ignored by the guys reading the book. What is missing is that the detail that the Lift vector (which is really the aerodynamic center including wing and tail needed to trim effects) does move around a little with angle of attack and tail deflections. The Neutral Point doesn't move though.

Actually it was picture C that made me initiate this thread. After reading the text definition for “Neutral point” I got the – now I know wrong – conclusion that the “Neutral Point” was where the lift vector was located and it should move with changes on speed and elevator incidence.

Remember NP (in percent of chord) = dCm/DCL - static margin. (if I have the signs right). It is a fallout of being able to accurately measure the slope of the pitching moment curve, knowing the wind tunnel reference center and knowing the CG location.

In flight you can work backwards and find it also but it is easier to see and work with by using wind tunnel data (my first aero job was reading NP slopes from wind tunnel data, hundreds of them, arrrrg - bad memories)

I'll attach a graph I made of the way the aero stable case would plot up for a 0-0-0 angle airplane with a randomly selected reference point (again it is chosen at random but is close to the final guess of CG) for ease of reading the slopes of the curves.

The blue line is the stable case at zero tail deflection. As the airplane is displaced to the high angle of attack it finds itself with increased lift and a negative pitching moment about the chosen Reference center to restore it to the original trim condition. The wing increased lift and the tail increased lift in the up direction. Again note that the figure is a pure aero case. When we apply the data to a real airplane the CG gets into the figure and you can end up with an initial tail setting that is either down or up depending on the airplane layout just as Mark Drela said (by the way, when Mark mentions something do pay attention, he knows wherein he speaks, he is not like one of those professors I never liked when in college - a prof that flies models would have been great to have - I am always at the wrong place at the wrong time).

For the real airplane case with a CG at the Neutral Point then the ac is shifted by the required elevator deflection to trim.

Remember

1. The ac definition is where the sum of aero forces end up to have zero moment when not considering the real CG effects. It will change with tail deflection and airloads.

2. The Neutral Point does not change as a function of stabilizer or angle of attack. It's the slope of the dCm/dCL curve. As the elevator changes the slope of the curve doesn't change. The position of the curve on the plot does change however. But Neutral Point is about slopes.

Note that the changes in the Lift vector in Simons's figures - that to be perfectly accurate they should include a moment due to the tail. I don't know if he intends that to be the case or not, lets assume so. I would prefer the figures to show wing and tail vectors but the figures are close enough to get the general idea.

The NP is a function of the geometric areas and their relative aero efficiencies. The aero center is the geometric location of the final aero configuration and takes into account all the lift and pitching moments due to deflections, aero efficiences and angle of attack and varies everywhere depending on those items.

I'm not too sure what I have left out - its early for us retired guys.



The information on this thread is going beyond my current aerodynamic understanding capabilities (I’m sure they will improve with the time), but I’ll try to summarize what I thing I have learned from it:

Before I thought that a flying object would react to the moments pitching around it’s CG, now I’m starting to think that it only apply to a body moving on free space and an object moving on a fluid like air will pitch around a different point that depends on it’s geometry and is called the Neutral Point. The point where the vector resultant of the combination of all lift and drag forces is located is called Aerodynamic Center (or maybe Center of Pressure or maybe both) and it will move with changes on speed and/or elevator incidence.

I believe that when the authors of those book says that the Neutral point is where the resultant of all lift and drag forces act it should be interpreted in the same sense as the combinations of forces act over the CG of an object moving on free space. Agree that they are very good books and I’m learning a lot from them.

[Ben Lanterman]

Hi Joe, I was OK until I called the Xcg the static margin, I shouldn't do these things watching TV. That was entirely wrong (I hate screwing up). The fingers should have typed sm not Xcg in the actual text :(

Xcg is the distance from the leading edge to the CG in percent chord.
dCm/dCL is actually the static margin converted to percent chord.
The sum of the two gives the NP in percent chord.
I'll do it again and see if I get it right this time............

I said the NP = xcg - dCm/dCL. (If I got the signs right) you'll find the correct equation in your textbooks if I got them wrong :D

The NP is a location on the airplane where sm (which is the static margin) goes to zero. (ie, the NP and Xcg are both the same location in percent of wing chord).

If the wind tunnel reference point was chosed to be the final CG, and if in testing you find the dCm/dCL slope goes to zero you will have neutral stability on the real airplane. I can chose any point to be the wind tunnel reference point as the slope of dCm/dCL will reflect the difference between the reference point and the airplane's real NP.

............ is that better - well hopefully - correct wording ?

[Spinner]

Hi Ben,

Excellent! We're in agreement and in the groove, man!

As retired aero guys, we can appreciate how some of this technical stuff is hard for others to interpret, especially since many of the non-technical books and articles have misleading, if not wrong explanations. I hope our discussion helps someone understand this concept a little better. Now, if only someone could teach me to fly my models without leaving parts at the field! :)

Keep 'em flying,
Joe

[Ben Lanterman]

Joe don't you hate it when you give a big explainaton and say how great you are and miss something in the typing!! Grooose (or however it is spelled!).

I do try to get things right though - it is so much easier for guys to understand what is actually happening if it can be made clear. The actual concepts are easy, sometimes explaining them with words is so hard. Two minutes with an airplane in the hand could explain most things on lift and stability.

It is a lot of fun. I was grousing earlier about not having Mark Drela for a professor. It probably would have made going to college a heck of a lot more fun.

Bruce and Gonzalo - another look -

Bruce said, "My" neutral point is fixed by the wing and tail areas and the moment arms and the neutral point is determined by the tail volume coefficient and other fixed charactaristics of the aircraft design. Flight loadings or elevator trim does not enter into it so it cannot move around to follow the CG or produce pitching forces when you use the elevator."

Bruce is right in saying the elevator pitch setting doesn't affect NP. The degree of the precision of NP can be increased or decreased by the loss of tail efficiency due to position and distance aft of the wing.

"...... Flight loadings or elevator trim does not enter into it ...... "
Of couse the elevator produces pitching moments for maneuvering but the NP during that maneuvering (due to elevator pitch setting) basically doesn't change.

Then Gonzalo said....

"I searched both books and didn’t find a description for a point as your neutral point but on Annex 1 of martin Simons’s book I found a formula to calculate the “Neutral Point” position and it uses de position of the wing MAC, the stabilizer efficiency and volume coefficient and the slopes of the lift curve for both wing and stabilizer. So I guess he is talking about the same neutral point as you but describing it on a different way."

Note that what it is saying about the NP doesn't use the actual settings of the elevator - just the lift curve slopes and the stabilizer efficiencies (due to elevator positions and downwash at the tail). Lift curve slopes and pitching moment curve shopes don't change with stabilizer settings. It really is describing the same point in the same way. Just getting into the details and accuracies a little bit more .