Why a Rear Engine For the P-39?

[rinkol]

One issue with the mid engine configuration is the weight and complexity of the extension shaft. One has to route it below the pilot and employ careful engineering to avoid torsional resonances. There is a flexible coupling at the engine, ball bearings to support the shaft near its middle and separate arrangements for providing oil to the gearbox, which is located behind the propeller. All this made for a very noisy cockpit.

Other aircraft designs depending on extension shafts often had problems with vibrations - the XB-35 comes to mind.

 

[wuzak]

One issue with the mid engine configuration is the weight and complexity of the extension shaft. One has to route it below the pilot and employ careful engineering to avoid torsional resonances. There is a flexible coupling at the engine, ball bearings to support the shaft near its middle and separate arrangements for providing oil to the gearbox, which is located behind the propeller. All this made for a very noisy cockpit.

The solution to that is to make a pusher...

[imghttps://upload.wikimedia.org/wikipedia/commons/4/46/G%C3%B6ppingen_G%C3%B6-9.png[/img]

Other aircraft designs depending on extension shafts often had problems with vibrations - the XB-35 comes to mind.

This was not necessarily to do with the extension shafts, but could also have been caused by interference from the wake of the wing.

 

[MIflyer]

One unanticipated consequence of the rear engine configuration was illustrated during a P-39 delivery flight over Alaska. The pilot brought along a ham sandwich and, flying along, unwrapped it and took a bite. He diced it was spoiled, rolled down the window and threw it out. The sandwich entered the carb air intake just behind the canopy and the engine quit.

And when you think about it, that intake probably was susceptable to icing.

If you look at the P-63 pilot's manual you'll see that the airplane has a number of alternate air sources.

 

[swampyankee]

Hello Ivan,

Because of the location of the engine in the P-39, the wing is located relatively farther aft than on an aircraft with a conventional engine location. This has three effects which tend to require an increase in the size of the tail surfaces:

1. The distance from the center of gravity to the tail surfaces is less than in a conventional aircraft; this means that the tail surfaces need more area to get the required tail volume.
   
2. The distance from the center of gravity to the propeller is greater, so the propeller will be more destabilizing; this will drive up the size of the tail surfaces.
3. There is more fuselage ahead of the center of gravity; this is also destabilizing, so this will be another increase in the size of the tail surfaces.

I wasn't aware of any particular problems with control harmony; usually this was something that US designers were pretty good at, possibly because the USAAC and USN were pretty fussy about it. On the other hand, overly sensitive elevators could come about because the center of gravity was too far aft.

 

[MIflyer]

According to the RAF report on the Airacobra I:

1. There was excessive friction in the aileron controls on the ground. (Note that pilots also reported that the P-39 ailerons required very small movements as compared to other aircraft; they were geared high.)
2. In the air the rudder and elevators were found to be light, quick, and responsive.
3. The ailerons (fabric covered in this model) were light but tended to become much heavier at speeds over 300 mph. Lateral control was considered to be comparable to the early Spitfire I with fabric covered ailerons.
4. Aerobatics could be flown with ease and there were no undesirable handling qualities.

Anyone know if they went to metal ailerons with the later P-39's as they did with the Spitfire?

I find it interesting that fighter planes of 1941 usually had fabric covered ailerons (the F4U was even wood) while the Ercoupe had metal flush riveted ailerons. I guess that the Ercoupe ailerons, although much larger, did not have to be nearly as strong and the metal ailerons were much easier to build, consisting mainly of two pieces of 0.020 aluminum. .

 

[XBe02Drvr]

• The distance from the center of gravity to the tail surfaces is less than in a conventional aircraft; this means that the tail surfaces need more area to get the required tail volume.

• The distance from the center of gravity to the propeller is greater, so the propeller will be more destabilizing; this will drive up the size of the tail surfaces.

• There is more fuselage ahead of the center of gravity; this is also destabilizing, so this will be another increase in the size of the tail surfaces.

Nice analysis. One question. You tiptoe around but never use the operative term here, namely: moment. Why? Moment is not just a weight and balance term; it applies to any force acting at a distance from the CofG pivot point, whether it's stabilizing or displacing.
You shorten the tail and you're reducing the moment and thus the effectiveness of both stabilizing and displacing functions of it, as you said.
Not quarreling with your analysis, just curious about the language.
Cheers,
Wes

 

[Peter Gunn]

The P-35, P-36 and P-43 were radial engined fighters, so they are always going to look fat.

The P-40 was developed from a radial engined fighter (the P-36), so it retains some of the flab.

But I don't think the P-39 could be considered fat compared to the Mustang.

"HEY... YOU UP FRONT... Yeah the little guy in olive drab, does this polished aluminum make our butts look big?"

 

[Ivan1GFP]

Hello Swampyankee,

Regarding your assertions:

Because of the location of the engine in the P-39, the wing is located relatively farther aft than on an aircraft with a conventional engine location.

I do not believe this is a natural result of the engine location.
I believe part of the problem here is that we do not have a common frame of reference.
I suggest we use the CoG as the reference point. That avoids all the weird datum points one finds with different aeroplanes.
Basically the CoG will be located somewhere around 25% to 30% MAC unless there is something weird going on and with the P-39 there was definitely something weird.

The distance from the center of gravity to the tail surfaces is less than in a conventional aircraft; this means that the tail surfaces need more area to get the required tail volume.

How is this determined by the location of the engine?
With the Tricycle landing gear, there really was no particular limitation on what the distance from the CoG to the Tail Surfaces needed to be. Choice of length of the moment arm is much less restricted.

The distance from the center of gravity to the propeller is greater, so the propeller will be more destabilizing; this will drive up the size of the tail surfaces.

With the engine behind the CoG, there is actually less requirement for the nose to be a particular length. Normally the engine and accessory area determines the minimum length ahead of the cockpit, but there was no such problem with the P-39. The length was determined by what equipment the designers wanted to install in the nose, so this is not an inherent issue of a rear engine design.
Why would having the propeller further forward be less stable? It would seem that the thrust vector would tend to correct any directional or pitch deviations with greater effect because of the longer moment arm.

There is more fuselage ahead of the center of gravity; this is also destabilizing, so this will be another increase in the size of the tail surfaces.

There is no inherent connection between needing more fuselage ahead of the CoG and a rear engine setup. As I see it, the P-39 didn't have a particularly large nose. Please see the attached image for a comparison between a P-40E and P-39. The drawings are to the same scale and I believe they are aligned on each aircraft's CoG.

I wasn't aware of any particular problems with control harmony; usually this was something that US designers were pretty good at, possibly because the USAAC and USN were pretty fussy about it. On the other hand, overly sensitive elevators could come about because the center of gravity was too far aft.

In reading flight evaluations, it appears that some aircraft got control harmony more right than others and a lot of it depends on air speed at which the comparison is made. A CoG that is too far back should result in a lot of instability but not necessarily overly sensitive controls.
An example of this lack of connection would be with the Spitfire. In early models, it was slightly unstable longitudinally. That issue was corrected by increasing the size of the aerodynamic balance area of the elevator. This made the control more sensitive but apparently fixed the instability.

Note in the attached drawing how large the nose of the P-40 was in comparison to the P-39. I combined a couple drawings to see for myself the relative differences between the two aircraft that seem to be of such great interest at the moment.
Length of the P-39Q is 30 feet 2 inches as it is for most variants.
Length of the P-40E is 31 feet 8.75 inches.

Update: Disregard the comment that the two drawings are lined up on the CoG.
I cannot confirm the CoG location for the P-40 at the moment.

- Ivan.

 

[XBe02Drvr]

Hello Swampyankee,

Regarding your assertions:

Hello Ivan.
Regarding your assertions:

I do not believe this is a natural result of the engine location.
I believe part of the problem here is that we do not have a common frame of reference.
I suggest we use the CoG as the reference point. That avoids all the weird datum points one finds with different aeroplanes.
Basically the CoG will be located somewhere around 25% to 30% MAC unless there is something weird going on and with the P-39 there was definitely something weird.

Anything unorthodox, successful or not, seems weird. Don't let that cloud your vision.
If you move the engine back in the fuselage, you relocate the CofG for the fuselage assembly aft and your 25-30% MAC CofLift has to migrate rearward to compensate. Result: longer nose, shorter tail, unless you're willing to accept the increased wetted surface and heavier tail moment that lengthening it entails. Eventually you wind up with a BF-109 tail, but that was a front engine plane. Remember, the P-39 was supposed to be a " big engine, small airfame" machine.

How is this determined by the location of the engine?
With the Tricycle landing gear, there really was no particular limitation on what the distance from the CoG to the Tail Surfaces needed to be. Choice of length of the moment arm is much less restricted.

BZZZZT, WRONG! For the reasons stated above, the innate tendency to tailheaviness limits how much tail moment you can add, unless you move fixed weights forward in the fuselage, like batteries, radios, etc. Unfortunately there was little unused space up there and technical limitations of the equipment tended to preclude that.

With the engine behind the CoG, there is actually less requirement for the nose to be a particular length. Normally the engine and accessory area determines the minimum length ahead of the cockpit, but there was no such problem with the P-39. The length was determined by what equipment the designers wanted to install in the nose, so this is not an inherent issue of a rear engine design.
Why would having the propeller further forward be less stable? It would seem that the thrust vector would tend to correct any directional or pitch deviations with greater effect because of the longer moment arm.

BZZZZT AGAIN! (Going to wear out that dang buzzer!)
The equipment they wanted to install (a big honking cannon and two HMGs) was a major reason for the aft engine configuration, as the Allison wasn't well suited to a through-the-hub cannon installation.
You're on the horns of a dilemma here, as a longer nose would give you more room and more moment to counter the tailheaviness tendency, but at the cost of more destabilizing keel area forward and more leverage arm for propeller effects such as P-factor and gyroscopic precession.

There is no inherent connection between needing more fuselage ahead of the CoG and a rear engine setup. As I see it, the P-39 didn't have a particularly large nose. Please see the attached image for a comparison between a P-40E and P-39. The drawings are to the same scale and I believe they are aligned on each aircraft's CoG.

Don't get too fixated on CofG. Sure, it's the pivot point around which an aircraft maneuvers, but it's not a fixed location, it's a moving target according to aircraft loading. It does have to stay within a certain range forward of the Center of Lift.

In reading flight evaluations, it appears that some aircraft got control harmony more right than others and a lot of it depends on air speed at which the comparison is made. A CoG that is too far back should result in a lot of instability but not necessarily overly sensitive controls.
An example of this lack of connection would be with the Spitfire. In early models, it was slightly unstable longitudinally. That issue was corrected by increasing the size of the aerodynamic balance area of the elevator. This made the control more sensitive but apparently fixed the instability.

Note in the attached drawing how large the nose of the P-40 was in comparison to the P-39. I combined a couple drawings to see for myself the relative differences between the two aircraft that seem to be of such great interest at the moment.
Length of the P-39Q is 30 feet 2 inches as it is for most variants.
Length of the P-40E is 31 feet 8.75 inches.

Update: Disregard the comment that the two drawings are lined up on the CoG.
I cannot confirm the CoG location for the P-40 at the moment.

View attachment 492186

- Ivan.

BZZZZT! YES, OVERLY SENSITIVE CONTROLS! If the CofG gets too close to the CofL the stabilizing moment of the delicate balance between inherent noseheaviness and stabilizer down load deteriorates and the controls get really sensitive, especially pitch.
I once inadvertantly flew a Beech 1900 from Boston to Burlington VT with the CofG 4 inches out of limits aft. Can you spell SQUIRRELLY? The plane was borderline divergent stable, and reacted abruptly to any pitch inputs. Fortunately the electric trim had a slow speed setting which minimized overshoots. We had to replace all the barf bags in the back. Turns out the rampies in Boston had loaded 400 pounds of company shipments in aft baggage without listing it on the manifest. We got in hot water with the Feds over that one.
Cheers,
Wes

 

[Ivan1GFP]

Anything unorthodox, successful or not, seems weird. Don't let that cloud your vision.
If you move the engine back in the fuselage, you relocate the CofG for the fuselage assembly aft and your 25-30% MAC CofLift has to migrate rearward to compensate. Result: longer nose, shorter tail, unless you're willing to accept the increased wetted surface and heavier tail moment that lengthening it entails. Eventually you wind up with a BF-109 tail, but that was a front engine plane. Remember, the P-39 was supposed to be a " big engine, small airfame" machine.

By weird, I was referring to a CoG that went back to around 35% MAC.
Nothing really wrong with a Long Tail such as on a Messerschmitt 109 if you need to increase stability.
If it is a choice between a longer tail with more wetted area and larger stabilizers / fin, which would you choose?
I am not saying one is better than the other, but in the case of the Airacobra, there was no extra shock load from a tail gear that had to be taken into account.
Lengthening the tail would be much like lengthening the tail for the late model P-40.
As for moving the masses around in the fuselage, you figure out where you need to put heavy items and then put the wing where it needs to be to maintain a proper CoG and Center of Lift relationship. That kind of thing should have been much better thought out at the design stage along with consequences of disposable loads in the wrong place.

BZZZZT, WRONG! For the reasons stated above, the innate tendency to tailheaviness limits how much tail moment you can add, unless you move fixed weights forward in the fuselage, like batteries, radios, etc. Unfortunately there was little unused space up there and technical limitations of the equipment tended to preclude that.

As mentioned before, the tail surfaces and supporting structure are really rather light in weight and would not move the CoG much.. Increasing the tail moment should not greatly affect anything else A lot of choices of where things go would be made at the design stage before there really was a "fuselage" form to limit what could go where.
As for heavy items that could have moved forward, there is also the cooling system which didn't really have enough space where it ended up.

BZZZZT AGAIN! (Going to wear out that dang buzzer!)
The equipment they wanted to install (a big honking cannon and two HMGs) was a major reason for the aft engine configuration, as the Allison wasn't well suited to a through-the-hub cannon installation.
You're on the horns of a dilemma here, as a longer nose would give you more room and more moment to counter the tailheaviness tendency, but at the cost of more destabilizing keel area forward and more leverage arm for propeller effects such as P-factor and gyroscopic precession.

That buzzer is a bit enthusiastic, isn't it?
That cannon and pair of HMG's is an armament consideration and not an engine location consideration. You pick a suitable armament package for the kind of aeroplane you intend to build If you need a shorter cannon or more HMG's, then it should have been designed that way. There needs to be some kind of flexibility in what armament can be installed because weapons tend to evolve.
Assuming the cannon and armament were not negotiable, then perhaps the wing itself needed to move forward to match the anticipated CoG. I understand that the basic concept of the Airacobra was to keep weight down by keeping the engine mounting structure close to the wing, but other changes in the wing such as reducing or increasing sweep can be used to move the Center of Lift. An example of this in real life would be the IL-2 Sturmovik.

Don't get too fixated on CofG. Sure, it's the pivot point around which an aircraft maneuvers, but it's not a fixed location, it's a moving target according to aircraft loading. It does have to stay within a certain range forward of the Center of Lift.

It is a very good thing to get fixated on when designing because everything literally revolves around it.
In service things will migrate around a bit but if the proper locations are picked for items that may change in weight, then the problems don't become as severe later on as they did with Airacobra. This can become more of an interesting issue with multi engine aircraft there is the space to move a lot more things around.

BZZZZT! YES, OVERLY SENSITIVE CONTROLS! If the CofG gets too close to the CofL the stabilizing moment of the delicate balance between inherent noseheaviness and stabilizer down load deteriorates and the controls get really sensitive, especially pitch.
I once inadvertantly flew a Beech 1900 from Boston to Burlington VT with the CofG 4 inches out of limits aft. Can you spell SQUIRRELLY? The plane was borderline divergent stable, and reacted abruptly to any pitch inputs. Fortunately the electric trim had a slow speed setting which minimized overshoots. We had to replace all the barf bags in the back. Turns out the rampies in Boston had loaded 400 pounds of company shipments in aft baggage without listing it on the manifest. We got in hot water with the Feds over that one.

It sounds like this kind of thing would depend quite a lot more on the general aircraft design, size of tail surfaces, etc. I am not necessarily thinking of a horizontal stabilizer that needs to provide down force. The point I was trying to make was that there may be other reasons for overly sensitive controls besides a CoG that was a bit far aft.
Just out of curiosity: What is the CoG range on the Beech 1900 in relation to MAC?

- Ivan.

 

[XBe02Drvr]

By weird, I was referring to a CoG that went back to around 35% MAC.
Nothing really wrong with a Long Tail such as on a Messerschmitt 109 if you need to increase stability.
If it is a choice between a longer tail with more wetted area and larger stabilizers / fin, which would you choose?
I am not saying one is better than the other, but in the case of the Airacobra, there was no extra shock load from a tail gear that had to be taken into account.
Lengthening the tail would be much like lengthening the tail for the late model P-40.
As for moving the masses around in the fuselage, you figure out where you need to put heavy items and then put the wing where it needs to be to maintain a proper CoG and Center of Lift relationship. That kind of thing should have been much better thought out at the design stage along with consequences of disposable loads in the wrong place.

You're right. A major drawback of the "big engine, small airfame" approach is lack of room for growth and evolution. Couple that to an unorthodox layout, and you've doubled your problems. I can't see much fuctional difference between a 109 tail and the approach Bell used except the 109 version might have added more usable keel area aft for stability.

As mentioned before, the tail surfaces and supporting structure are really rather light in weight and would not move the CoG much.. Increasing the tail moment should not greatly affect anything else A lot of choices of where things go would be made at the design stage before there really was a "fuselage" form to limit what could go where.
As for heavy items that could have moved forward, there is also the cooling system which didn't really have enough space where it ended up.

The problem was all the "growth" that came along after metal was cut. Cockpit armor, self sealing tanks, more and heavier radio equipment, a battery capable of repeated engine starts without a GPU, wing guns and ammo, and (the biggie!) removal of the turbocharger system. The net effect of all these tended towards increased aft moment. Not a good thing. In a borderline tailheavy aircraft, increasing aft moment DOES affect everything else.
So what about the cooling system? You could put the coolant and oil radiators up front like the P-40, but this entails a lot of drag and unwanted additional keel area forward, plus with all the additional plumbing, turning the cockpit into an oven, and guaranteeing a scalded pilot if a single bullet finds its way under the cockpit floor.

That buzzer is a bit enthusiastic, isn't it?
That cannon and pair of HMG's is an armament consideration and not an engine location consideration. You pick a suitable armament package for the kind of aeroplane you intend to build If you need a shorter cannon or more HMG's, then it should have been designed that way. There needs to be some kind of flexibility in what armament can be installed because weapons tend to evolve.
Assuming the cannon and armament were not negotiable, then perhaps the wing itself needed to move forward to match the anticipated CoG. I understand that the basic concept of the Airacobra was to keep weight down by keeping the engine mounting structure close to the wing, but other changes in the wing such as reducing or increasing sweep can be used to move the Center of Lift. An example of this in real life would be the IL-2 Sturmovik.

The buzzer burned out. Have to do without.
Your assumption about cannon armament non-negotiable is right on, but I don't know if it was internal or external. USAAC WAS very fond of their 37MM. An upright V-12 is an awkward customer when it comes to a hub-mounted cannon (the DB was inverted), so if you're stuck with an Allison, you've got to figure out somewhere else to put it.
Move the wing FORWARD to match CofG??? Tell me that's a typo! If anything it needs to move aft to get the lift behind and near where the CofG is.

It is a very good thing to get fixated on when designing because everything literally revolves around it.
In service things will migrate around a bit but if the proper locations are picked for items that may change in weight, then the problems don't become as severe later on as they did with Airacobra. This can become more of an interesting issue with multi engine aircraft there is the space to move a lot more things around.

What I was trying to point out is CofG isn't a fixed point but a range of points and that range has to be positioned forward of the CofL an appropriate distance to give the aircraft a slightly nose heavy condition throughout it's CofG range.

It sounds like this kind of thing would depend quite a lot more on the general aircraft design, size of tail surfaces, etc. I am not necessarily thinking of a horizontal stabilizer that needs to provide down force. The point I was trying to make was that there may be other reasons for overly sensitive controls besides a CoG that was a bit far aft.
Just out of curiosity: What is the CoG range on the Beech 1900 in relation to MAC?

- Ivan.

Stabilizer with no down load? What are you smoking, man?? We're talking circa 1940 here, and fly-by-wire computer enhanced active stability (F-16, F-18, A120) is still a half century away. For any conventional planform aircraft (ignoring canards, flying wings, and tail-less curiosities) stability requires an inherent nose-down balance matched by an equivalent down load generated by the tail to achieve stable steady-state flight.
As for the 1900, it's been 27 years since I last flew one, and 23 since I last flew anything, and my 1900 manual went away years ago, so I don't remember picayune details like that. In any case figuring CG vs MAC was a ground school exercise we did twice a year in recurrent. In practice we used a pre-computed "cheat sheet" where we entered the weights in various parts of the plane which gave us a moment number which we plotted on a graph opposite the GTOW and if the dot fell in the cross-hatched area, we were good to go. You'd be surprised how quickly you can pull that off when you're being yelled at by the Ramp Supervisor, the Captain's got the right engine turning already, the last little old lady is hobbling across the ramp to your door, there's a DC-9 idling at the edge of the ramp waiting for your spot, and Ground Control is advising you have two minutes remaining on your clearance void time!
Cheers,
Wes

 

[Ivan1GFP]

Hello XBe02Drvr.

You're right. A major drawback of the "big engine, small airfame" approach is lack of room for growth and evolution. Couple that to an unorthodox layout, and you've doubled your problems. I can't see much fuctional difference between a 109 tail and the approach Bell used except the 109 version might have added more usable keel area aft for stability.

We are getting quite a bit off the original topic here.
I never stated that the Airacobra was a good design. Interesting? Yes. Good? Not so much.
It had a lot of designed in flaws that were hard to work around.
I like it because I think it looks cool and not because I think it was a particularly good aeroplane.

My original response to Swampyankee was that the long nose and short tail was not necessarily a consequence of the engine location.
I wasn't advocating for lengthening the tail so much as I was proposing how it could easily be done if necessary AND how it could be done with less consequence than with a tail dragger design.
I was also commenting that the relatively long nose was also not necessarily a consequence of the engine location but because of other factors.

The problem was all the "growth" that came along after metal was cut. Cockpit armor, self sealing tanks, more and heavier radio equipment, a battery capable of repeated engine starts without a GPU, wing guns and ammo, and (the biggie!) removal of the turbocharger system. The net effect of all these tended towards increased aft moment. Not a good thing. In a borderline tailheavy aircraft, increasing aft moment DOES affect everything else.
So what about the cooling system? You could put the coolant and oil radiators up front like the P-40, but this entails a lot of drag and unwanted additional keel area forward, plus with all the additional plumbing, turning the cockpit into an oven, and guaranteeing a scalded pilot if a single bullet finds its way under the cockpit floor.

I am in general agreement with you here about the no room for stretch. I do not see how removal of the Turbocharger would make the aeroplane more tail heavy or adversely affect anything other than performance at altitude.
Regarding the cooling system:
I wasn't suggesting something along the lines of the P-40 which would destroy aerodynamics. I was thinking their buried setup might still work move a bit further forward by where the nose wheel was.
If you think about it, the P-51 and just about every other fighter that had the coolant radiators or oil coolers somewhere other than right next tot he engine had to have some substantial plumbing between the engine and Radiators. I am thinking of Hurricanes, Ki 61, Yak fighters et al. Others managed to deal with the issue of remote radiators without great difficulty.

Your assumption about cannon armament non-negotiable is right on, but I don't know if it was internal or external. USAAC WAS very fond of their 37MM. An upright V-12 is an awkward customer when it comes to a hub-mounted cannon (the DB was inverted), so if you're stuck with an Allison, you've got to figure out somewhere else to put it.
Move the wing FORWARD to match CofG??? Tell me that's a typo! If anything it needs to move aft to get the lift behind and near where the CofG is.

An upright V-12 isn't really the problem with mounting a motor cannon. The Hispano Suiza and Klimov engines both had motor cannon.
As I understand it, the issue is more because the space between the cylinder banks has to be have enough room and there needs to be room behind the engine block itself for the cannon. Of course the engine reduction gear has to have enough offset for the cannon bore to clear the engine block as well.
With the Allison, there was a really convoluted intake manifold in the way. With the Merlin, there was the supercharger. The Daimler Benz and JuMo engines got around the supercharger issue by offsetting their superchargers to the side.

Forward? Oops! The problem here is that the CoG of the Airacobra moves around quite a lot depending on the amount of disposable loads remaining. Perhaps we really needed a variable sweep setup like F-111 or MiG 23?

What I was trying to point out is CofG isn't a fixed point but a range of points and that range has to be positioned forward of the CofL an appropriate distance to give the aircraft a slightly nose heavy condition throughout it's CofG range.

I believe what we are really getting at here is to properly locate the heavy equipment and variable loads in a manner that allows only for minimal changes with lighter pieces of equipment. In other words, it helps to plan for the inevitable growth very early in the design. Nose heavy will be addressed shortly.

Stabilizer with no down load? What are you smoking, man?? We're talking circa 1940 here, and fly-by-wire computer enhanced active stability (F-16, F-18, A120) is still a half century away. For any conventional planform aircraft (ignoring canards, flying wings, and tail-less curiosities) stability requires an inherent nose-down balance matched by an equivalent down load generated by the tail to achieve stable steady-state flight.

With a conventional layout, the forward CoG limit is limited by the amount of down force that the tail surfaces can apply to offset pitch down moment.
As the CoG moves aft, the aircraft stability gradual decreases.
When the CoG moves behind the CoL, the tail surfaces will need to provide lift instead of downforce.
There is nothing inherently wrong with that except for the reduction in stability.
As the CoG moves further aft, at some point stability will become nil and this is the aircraft "Neutral Point".
As the CoG moves even further aft, there gets to be a situation in which the tail surfaces which are providing lift will stall before the wing will.
When this happens, the whole world blows up because the aircraft will continue to pitch up and increase the Angle of Attack and load on the wing and likely cause a structural failure.

Regarding the conventional versus canard versus something like Langley's Aerodrome. They all follow the same basic rules.
The aeroplane will fly most efficiently if forward and aft horizontal surfaces are providing lift. Stability may be poor in certain load conditions but life gets really bad if the aft surface stalls and results in the aircraft pitching up.

As for the 1900, it's been 27 years since I last flew one, and 23 since I last flew anything, and my 1900 manual went away years ago, so I don't remember picayune details like that. In any case figuring CG vs MAC was a ground school exercise we did twice a year in recurrent. In practice we used a pre-computed "cheat sheet" where we entered the weights in various parts of the plane which gave us a moment number which we plotted on a graph opposite the GTOW and if the dot fell in the cross-hatched area, we were good to go. You'd be surprised how quickly you can pull that off when you're being yelled at by the Ramp Supervisor, the Captain's got the right engine turning already, the last little old lady is hobbling across the ramp to your door, there's a DC-9 idling at the edge of the ramp waiting for your spot, and Ground Control is advising you have two minutes remaining on your clearance void time!

My apologies. I didn't know your history. I know nothing about the Beech 1900 and was trying to get a data point to understand better.

- Ivan.

 

[swampyankee]

Regarding the conventional versus canard versus something like Langley's Aerodrome. They all follow the same basic rules.
The aeroplane will fly most efficiently if forward and aft horizontal surfaces are providing lift. Stability may be poor in certain load conditions but life gets really bad if the aft surface stalls and results in the aircraft pitching up.



- Ivan.

Actually, canard aircraft tend not to be more efficient than conventionally configured aircraft; they tend to be less so. Indeed, Rutan proved this with a canard sailplane having performance that was far from competitive.

 

[KiwiBiggles]

The aeroplane will fly most efficiently if forward and aft horizontal surfaces are providing lift.

Actually, because the tail plane is almost certainly of lower aspect ratio than the main plane, the aeroplane will fly most efficiently if the tail plane has zero lift. The main plane is a more efficient lifting surface, and so should be left to do all the work. The tailplane does its stabilizing job best if it provides no lift either way.

 

[XBe02Drvr]

The aeroplane will fly most efficiently if forward and aft horizontal surfaces are providing lift. Stability may be poor in certain load conditions but life gets really bad if the aft surface stalls and results in the aircraft pitching up.

You just don't get it, do you?
When CG gets back to CoL, you've arrived at neutral, and your controls have become seriously twitchy. There's NO stability. If CG gets behind CofL you now have divergent stability. If any displacement occurs, it will accelerate away from equilibrium rather than seeking to return to it, resulting in overstress and catastrophic failure if allowed to continue.
Tail surfaces providing positive lift is attractive for efficiency reasons, but practical only in a canard. In a conventional tractor airplane it's a recipe for disaster as the tail is gauranteed to accentuate any pitch down or pitch up.
That's what happened to pioneer aviatrix Harriet Quimby. She was flying a lifting-tail plane (Bleriot, I think) and entered a shallow fast descent. The increased lift on the tail due to the increasing speed steepened the dive until the plane "tucked under" and pitched her right out of the cockpit. The rules of stability are written in blood.
Cheers,
Wes

 

[XBe02Drvr]

Actually, because the tail plane is almost certainly of lower aspect ratio than the main plane, the aeroplane will fly most efficiently if the tail plane has zero lift. The main plane is a more efficient lifting surface, and so should be left to do all the work. The tailplane does its stabilizing job best if it provides no lift either way.

Actually some fly-by-wire planes with active stability do derive useful positive lift from the tail, but this only works safely with a computer making constant tiny corrections with the control surfaces. That's why the A320 is more fuel efficient than the very similar 737-800.
Cheers,
Wes

 

[Ivan1GFP]

Hello Gentlemen,

Actually, canard aircraft tend not to be more efficient than conventionally configured aircraft; they tend to be less so. Indeed, Rutan proved this with a canard sailplane having performance that was far from competitive.

I don't think I made any statement about relative efficiency of different planforms because I simply don't know.
Thanks for the note about a Canard Sailplane. I will have to look for it.

Actually, because the tail plane is almost certainly of lower aspect ratio than the main plane, the aeroplane will fly most efficiently if the tail plane has zero lift. The main plane is a more efficient lifting surface, and so should be left to do all the work. The tailplane does its stabilizing job best if it provides no lift either way.

There are other reasons why the horizontal tail isn't as efficient a lifting surface: They tend to use symmetrical airfoils.
Even without providing any lift, the tail plane is providing drag, so your argument as I see it is whether zero lift from the tail is better than a little positive lift from the tail which would offset some of the aircraft weight.

You just don't get it, do you?
When CG gets back to CoL, you've arrived at neutral, and your controls have become seriously twitchy. There's NO stability. If CG gets behind CofL you now have divergent stability. If any displacement occurs, it will accelerate away from equilibrium rather than seeking to return to it, resulting in overstress and catastrophic failure if allowed to continue.

Tail surfaces providing positive lift is attractive for efficiency reasons, but practical only in a canard. In a conventional tractor airplane it's a recipe for disaster as the tail is gauranteed to accentuate any pitch down or pitch up.

It sounds to me like you just stated that the Neutral Point of an aircraft always coincides with the Center of Lift....
This is certainly not the case.
If you look for the definitions of Neutral Point and Static Margin along with diagrams, you will see what I mean.
I have also attached a couple images which show the relationships as I understand them.

That's what happened to pioneer aviatrix Harriet Quimby. She was flying a lifting-tail plane (Bleriot, I think) and entered a shallow fast descent. The increased lift on the tail due to the increasing speed steepened the dive until the plane "tucked under" and pitched her right out of the cockpit. The rules of stability are written in blood.

Ms. Quimby should have worn a seat belt!
Seriously though, it sounds to me that trim or angle of incidence was set too high on the tail surfaces and while at low speed, everything was in equilibrium, as speed and dynamic force increased, the trim / incidence effect was enough to cause uncontrollable nose down pitch.
This is a different problem as I see it.
A CoG too far forward can get you killed too. That was what happened to Wiley Post and Will Rogers.

- Ivan.

 

[KiwiBiggles]

whether zero lift from the tail is better than a little positive lift from the tail which would offset some of the aircraft weight.

The tails zero-lift drag is what it is. But as soon as the tail starts to provide lift, its induced drag is going to be higher than the extra induced drag on the main, if the main were to supply the same lift.

 

[XBe02Drvr]

It sounds to me like you just stated that the Neutral Point of an aircraft always coincides with the Center of Lift....
This is certainly not the case.
If you look for the definitions of Neutral Point and Static Margin along with diagrams, you will see what I mean.
I have also attached a couple images which show the relationships as I understand them.

Well I looked up the definitions and read the explanations, and I've got to admit, I'm embarrassed! It appears I was taught and have been preaching all these years a highly over-simplified understanding of the whole stability problem. And largely downright wrong! Mea Culpa! And I never updated myself and discovered the errors.
Well, I guess a little humble pie never hurt anyone. Thanks for showing the way.
Cheers,
Wes

 

[Ivan1GFP]

The tails zero-lift drag is what it is. But as soon as the tail starts to provide lift, its induced drag is going to be higher than the extra induced drag on the main, if the main were to supply the same lift.

Hello KiwiBiggles,
I actually can't find any fault in your logic, but that doesn't agree with the descriptions I have seen in the past.
Then again, I am not claiming I know the subject all that well beyond basic principles.

Hello XBe02Drvr,
I actually found this link last night after posting, but didn't read it until this morning.
6 Angle of Attack Stability, Trim, and Spiral Dives
Section 6.1.4 gives a really good summary of what we have been discussing.
The interesting thing is that last night when I was looking for reference diagrams, the ones I was finding for actual aircraft tended to have the CoG range entirely ahead of the CoL.

- Ivan.