Overweight takeoff / flight

(Koopas Ly) wrote
After reading NTSB reports that attribute the cause of the accident to
exceeding the airplane's maximum takeoff weight, I began wondering
about the effects of an overweight takeoff within C.G. limits.
Specifically, what would I have to do differently when flying an
airplane that's heavier than what the POH specifies. I am not
supporting the practice, of course, so let it be purely educational.

The first thing you should realize is that airplanes are LEGALLY
operated overgross all the time. In Alaska, Part 135 operators can
get the max gross raised by up to 15%, depending on the airplane. For
long overwater ferry flights, the FSDO will give you a ferry permit to
operate 20% overgross without so much as blinking, provided you sound
like you know what you're doing. Just for reference, on a plain
vanilla C-172 or Cherokee, that would be 400+ lbs overgross.

I would start by considering the increase in weight as comparable to
an increase in load factor. Hence, all your aoa-related speeds would
increase by the square root of the load factor. Vs, Vx, Vy, Vglide,
etc. would all increase.

So far, so good.

Va would also go up.

Not necessarily. Everything depends on where the weak spot defining
Va happens to be. The usual reason Va goes down on an airplane is
that the weak spot is the engine mount. If the weak spot is the
engine mount, the weight carried by the engine attach point (the
engine) is fixed, and thus maximum gee is fixed. Since at lower
weight you can exceed max gee at a lower speed without stalling, Va
goes down with weight.

If the weak spot is the wing attach point, then Va is constant. This
is because the weight carried by that point is NOT fixed. This is a
pretty common situation in gliders, but pretty rare in airplanes. The
real issue is this - once you exceed max gross, you don't really know
where the weak spot is anymore unless you do an engineering analysis.
Therefore, I would assume Va does not increase.

Now, by virtue of rotation speed being a function of stall speed, I
conjecture you'd have to liftoff at a faster airspeed which would
equate to a longer takeoff roll.

Correct, and this speed may be higher than what the square root of
weight correction would lead you to believe. Typically we rotate well
below best rate of climb speed, and count on being able to accelerate
in ground effect and climb out. However, once you load it up enough,
you may not have that luxury. You may have to wait until almost Vy
speed before you have excess power available to accelerate and climb
out. A too-early rotation may put you in the position of flyng in
ground effect without being able to accelerate enough to climb out.

Then, after pitching for your faster Vy airspeed, you'd notice a
decrease in climb rate at full power due to the increased power
requirement. During cruise, you'd notice a reduced cruise speed and
an increase in stall speed.

All correct.

At approach to landing, should you bump
up your approach speed, you'll find yourself sinking faster when
chopping off the power even though your glideslope will remain the
same.

Assuming you are still overgross.

Since your stall speed is invariably higher, you'll eat up more runway
when landing.

Maybe. Certainly if you want to minimize use of brakes. However,
your brakes will be more effective with more weight on wheels - you
will be able to use them at higher speed without locking them up.

You are also ignoring another important factor - the cg envelope
shrinks at higher gross weights. Because of this, just because you
are within cg limits for max gross does not mean you are still within
cg limits for the increased weight. Usually being forward is not too
bad - the plane will be nose heavy and will need to be landed at a
higher speed and/or with power to keep the nose up through the
touchdown. But if you are close to the aft limit for gross and are
overgross, beware. You are asking for stability problems in pitch,
and the plane may be uncontrollable.

So to sum up:

Takeoff: higher takeoff distance, higher rotation speed.

Climbout: lower climb rate at higher Vy speed, same angle of climb for
obstacle clearance at higher Vx speed. Should Vx not be flown faster,
a poorer angle of climb would result, making obstable clearance
doubtful. *I may be wrong here* I am not sure if the max. angle of
climb is constant regardless of weight...my calculations don't show
so...could someone clarify?

Max angle of climb will be reduced at higher weight, and Vx will have
to be increased.

Cruise/Maneuvering: lower cruise speed, higher maneuvering speed,
higher clean stall speed.

See above with respect to maneuvering speed.

Approach to maintain glideslope & descent profile: higher approach
speed, higher sink rate for a given power setting. Higher dirty stall
speed.

Landing: higher landing distance

Question (1 of 2): Seems to me that flying "overweight" is possible if
you're aware of the performance reductions.

Absolutely, it's done all the time - legally and illegally. It's a
rare piston freight hauler that doesn't routinely operate overgross.
What you have to realize is that all sorts of safety margins are
reduced. If you are aware of the reductions, it's not deadly. That's
why the FAA will give you a permit to operate overgross if you have a
need - as long as you demonstrate you understand what you're doing.

So why do you read so
many NTSB reports with probable causes listed as "overweight takeoff,
exceeded performance limitations"? As you slowly pull the yoke to
rotate, wouldn't a pilot *realize* through control forces, feel, gut
feeling that something is wrong?

Maybe - but if he's not expecting it, he may not realize it in time.

Question (2 of 2): When considering accidents due to exceeding maximum
takeoff weight, do the majority occur during takeoff? If so, is it
typically due to not reaching proper liftoff airspeed for that
increased weight, stalling, and spinning to the ground? Would this
scenario be consistent with failure to set the flaps/slats to their
takeoff value?

In general, the overgross accidents fall into two categories. First
there are the ones where climb speed is never reached and the plane
runs out of runway and hits something. Second is when an overgross
twin loses an engine and can't maintain flight. The first is usually
the result of the "It's always worked before" factor and the second is
betting the engines will both keep running.

Michael

The definition of Va in Part 23 is clear. It has nothing to do with
control surfaces and everything to do with stall speed and load
factor.

Then you haven't read Part 23.

Let me point out the sections to you:

-------------snip------------------
Horizontal Stabilizing and Balancing Surfaces

§ 23.423 Maneuvering loads.
Each horizontal surface...must be designed for the maneuvering loads
imposed by the following conditions:
(a) A sudden movement of the pitching control, at the speed VA...

(b) A sudden aft movement of the pitching control at speeds above
VA...

Vertical Surfaces
§ 23.441 Maneuvering loads.
(a) At speeds up to VA, the vertical surfaces must be designed to
withstand the following conditions....

Ailerons and Special Devices
§ 23.455 Ailerons.
(a) The ailerons must be designed for the loads to which they are
subjected -
....
(i) Sudden maximum displacement of the aileron control at VA. Suitable
allowance may be made for control system deflections.
-------------snip------------------

Now, the section that may be misleading you is this
---------snip-------------
§ 23.335 Design airspeeds.
(c) Design maneuvering speed VA. For VA, the following applies:
(1) VA may not be less than VS * sqrt(n) where -
---------snip-------------


Note that it says MAY NOT BE LESS THAN... In other words, it can be
more.


Oddly enough, many aircraft manuals bear this out, providing lower
Va speeds for lower weights.

Oddly, you didn't read what I wrote.

The point is that at Part 23 doesn't require this. And not all
aircraft publish such variations.

does not mean that the maximum speed at which you can fly
and be assured of not overstressing the airplane does not go down as
weight is reduced.

Again, you didn't read what I wrote. I said it doesn't scale UP.
Flying over max gross may increase maneuvering speed, but it doesn't
increase VA, because the increased weight won't protect control
surfaces from failure.


Even your control surface tangent isn't really relevant to this
particular thread

Tangent? It's the essence of what Va is.


I seriously doubt Todd has told you that Va remains the same
regardless of aircraft weight. You obviously misunderstood him.


Ok, you read what he wrote and tell me:

----------------snip----------------------
Note that this is a minimum Va ("no less than"). Thus the
designer can specify a higher Va and then protect the tail
surfaces by limiting stick throw, or by making the required
force at the stick to produce a damaging load on the
protected structure higher than a standard pilot could
exert. Note also that the regulatory definition of Va
*requires" that it be computed only at a stalling speed "at
the design weight" , i.e. max gross. Thus any lower speed,
or any lower weight cannot be Va as defined.
----------------snip----------------------

"Greg Esres" wrote in message
...

In my view, the most correct definition of Va will be it's the speed
above which you cannot make full or abrupt control movements, due to
control surface integrity.

This is way interesting & I've got the FAR's in front of me
now to get to the bottom of this.

First, I can't find a specific definition of "Design maneuvering speed" in
the FAR's, but my personal working definition is almost like yours.
I'd substitute "without risk of structural failure" for talk of control
surface integrity. Since control surface failure is indeed structural
failure, my definition would seem more restrictive than yours.

It looks like Va is mentioned twice in pt 23.

In 23.335 we get Va must be = Vs sqrt(n), with n the load
factor. We also get "Va need not exceed Vc" which
makes no sense to me, at least as far as a regulation goes.

Then, in 23.423 we see Va used in establishing the characteristics
of the (horizontal) control surfaces. Note that this doesn't
say this is how you calculate Va, it says you must use this speed
in the design of control surfaces to achieve certain rates of
response when they are used and/or to make sure you don't
break anything..............I suppose that manufacturers
do such a poor job of designing control surfaces that
they have to restrict Va just to meet this certification
requirement.....Well, bugger me Greg, looks like you're right!


New airplanes are supposed to come with a new Vo speed, which DOES
require that the airplane stall before exceeding the load factor.


Since control surfaces seem to be the limiting factor, I'd assume
that manufactures would design them for as low a Va as possible,
consistent with 23.335. So they'd choose Va = Vs.sqrt(n).

Vo does differs a little from pt 23 certification requirements, in that
Va isn't exactly Vo, because Va calculations assume that airfoil
lift does scale linearly with AOA and as the square of airspeed
when in fact these are only approximately true.

I'd bet that Vo and Va are pretty close. Allowing for the 1.5 safety
factor, I bet they're indistinguishable.

Here's a copy from a draft copy of an AC 23.something that I found.
The AC was intended to make this clear to test pilots, but I don't
think the draft was ever finished:

------------snip-----------------
VA should not be interpreted as a speed that would permit the pilot
unrestricted flight-control movement without exceeding airplane
structural limits nor should it be interpreted as a gust penetration
speed. Only if VA = Vs sqrt(n) , will the airplane stall in a nose-up
pitching maneuver at, or near, limit load factor. For maneuvers where
VAVS n , the pilot would have to check the maneuver; otherwise the
airplane would exceed the limit load factor.

Isn't this just a warning that Va "may not be less than Vs.sqrt(n)", and
so could be higher?

Va would be the same at any aircraft weight, which it certainly
isn't.

It is in some airplanes. My Piper arrow doesn't scale it with weight.

Moreover, Part 23 says that Va is *only* defined at max gross. Some
manufacturers do publish Va's at lower weight, but that appears to be
at their option. As written, it doesn't match Part 23 definition.

I don't see that in pt 23. I see it being defined as 'may not
be less than' some expression involving gross weight parameters,
but there is nothing to say that this applies only to gross
weight (to be pedantic). Nor does 23.423 - which we both
agree partially defines Va - say anything about the weight
of the plane during the certification maneuver.

I'd remind you how we got here. The suggestion was that
Va, should be scaled upward in an overloaded airplane. We
both claim that it should not. I'd also scale my maneuvering
speed downwards if underweight just to stay within load
factor limits, and I bet you would too. To my mind, the laws
of physics trump the FAR's. (and my Va is indeed pretty close to
Vs.sqrt(3.5)). After all, pt 23 just tells me how to certify a
plane, not how to fly it.

I'd claim that Va shouldn't be increased because it is really
the minimum of a number of different speeds where things
start to fall apart, and without further data we don't know
which one does the limiting.

Interesting discussion.


--
Dr. Tony Cox
Citrus Controls Inc.
e-mail:
http://CitrusControls.com/

"John Gaquin" wrote in message
...
An old pilot once told me, when I was a young pilot, "...sumbitch flies a
hell of a lot better overweight than it does outta gas..."

JG

I've got an old Flying Magazine (circa 1970 or so) where one of the editors
makes the comment that it is better to take off overloaded (with fuel) than
it is to try a launch with marginal fuel in order to stay under gross. The
comment was the same... It'll fly better over gross than outta gas.

I bet the magazine's lawyers wouldn't let them print that now...

KB

This is way interesting

I agree, and I appreciate and admire your open mind.

I'd substitute "without risk of structural failure" for talk of
control surface integrity. Since control surface failure is indeed
structural failure, my definition would seem more restrictive than
yours.

I can live with your defintion. I only used "control surface
integrity" in order to stress that it wasn't necessarily the main wing
we were talking about.

Vo does differs a little from pt 23 certification requirements, in
that Va isn't exactly Vo, because Va calculations assume that airfoil
lift does scale linearly with AOA and as the square of airspeed
when in fact these are only approximately true.

The only distinction I see between Va and Vo is that Va says "not less
than" and Vo is "not greater than". Where do you see the distinction
you are drawing?

All the lift slope curves I've seen for straight wings are pretty
linear, at least up until the stall. But that does lead us into the
concept of a dynamic stall. Airfoils rapidly rotated to a high angle
of attack can generate a much higher lift coefficient than when in
steady state. (References available upon request.) The whole concept
of Va, or even Vo, protecting the wing are a bit fraudulent.

I'd bet that Vo and Va are pretty close. Allowing for the 1.5 safety
factor, I bet they're indistinguishable.

I'd say you're right. A friend of mine, who spoke with the FAA's
Seattle Certification office, said that Va might be a maximum of 5
knots over what sqrt(n)*Vs would be.

Isn't this just a warning that Va "may not be less than Vs.sqrt(n)",
and so could be higher?

Yes, exactly. Some people need it spelled out. g

I don't see that in pt 23. I see it being defined as 'may not
be less than' some expression involving gross weight parameters,
but there is nothing to say that this applies only to gross
weight (to be pedantic).

If I understand what you're saying, I agree. I guess it depends on
what "defined" means. g

The suggestion was that Va, should be scaled upward in an overloaded
airplane. We both claim that it should not.

Agreed.

I'd also scale my maneuvering speed downwards if underweight just to
stay within load factor limits, and I bet you would too.

Yes. However, those knowledgeable about aircraft structures maintain
that load factors incurred in turbulence are less stressful on the
aircraft than what are incurred via flight control movements.
Turbulence penetration speeds are calculated allowing higher load
factors.

I'd claim that Va shouldn't be increased because it is really
the minimum of a number of different speeds where things
start to fall apart, and without further data we don't know
which one does the limiting.

Very well expressed.

"Peter Duniho" wrote in message
...

The definition of Va in Part 23 is clear. It has nothing to do with
control
surfaces and everything to do with stall speed and load factor.

Actually, it seems to depend on both. I'm all turned around
on this having scratched my head for a while. Greg is
essentially correct.

23.335 says that Va must be = Vs.sqrt(load-factor)
If we take the equality, then this is the load-factor
relationship we get assuming "Lift prop. to AOA"
and "Lift prop. airspeed**2".

23.423 (and others I'd missed) say how the control
surfaces must behave at Va and above. Designers
can set anything they want for Va as long as it
passes the control surface tests.

But since they are likely to want to minimize
complexity & weight of the control surface
mechanism, they are likely to choose Va to be
the minimum allowed by 23.335. But they don't
have to.

Greg is right. They really ought to have invented
another term for it. Va isn't the maneuvering
speed at all, and should be renamed to something
completely different.

--
Dr. Tony Cox
Citrus Controls Inc.
e-mail:
http://CitrusControls.com/

"Greg Esres" wrote in message
...

Vo does differs a little from pt 23 certification requirements, in
that Va isn't exactly Vo, because Va calculations assume that airfoil
lift does scale linearly with AOA and as the square of airspeed
when in fact these are only approximately true.

The only distinction I see between Va and Vo is that Va says "not less
than" and Vo is "not greater than". Where do you see the distinction
you are drawing?



I assumed that Vo was an actual speed determined by calculation or
flight test.

Va = Vs.sqrt(n) assumes (in the equality) lift linearity vs. AOA (which
we know drops off near the stall) and a quadratic relationship between
lift and AOA (which is only true for small AOA & probably off by
10% or more close to the stall).

That's why I called the distinction. Nothing profound.

--
Dr. Tony Cox
Citrus Controls Inc.
e-mail:
http://CitrusControls.com/

I assumed that Vo was an actual speed determined by calculation or
flight test.

Ah. I'm not sure how they determine Vo. They don't specify how it's
to be calculated, and the Part 23 Flight Test guide doesn't say how to
find it experimentally (unlike things like Vmc).

a quadratic relationship between lift and AOA (which is only true
for small AOA & probably off by 10% or more close to the stall).

I assume you meant "between airspeed and AOA" ?

"Greg Esres" wrote in message
...

Ah. I'm not sure how they determine Vo. They don't specify how it's
to be calculated, and the Part 23 Flight Test guide doesn't say how to
find it experimentally (unlike things like Vmc).

G-meter? Yank the yoke at different speeds on a calm day?


a quadratic relationship between lift and AOA (which is only true
for small AOA & probably off by 10% or more close to the stall).

I assume you meant "between airspeed and AOA" ?


Yes. slip of the keyboard.

Happy Thanksgiving!

John Gaquin wrote:

An old pilot once told me, when I was a young pilot, "...sumbitch flies a
hell of a lot better overweight than it does outta gas..."

That postulates a situation in which I those are my only two choices. I'm
betting that I can live my entire life without putting myself in that situation.

George Patterson
A man who carries a cat by the tail learns something that can be learned
no other way.