Rolling a 172 - or not

Big John wrote:

Robert

You talk about doing a LOOP in association with a roll. A LOOP has no
roll in it.The ailerons are only used to keep the wings 90 degrees to
the plane of the loop. A IMMELMAN or CUBAN EIGHT has a roll
associated with part of a loop.

The description of making a corkscrew inside a tube is another way of
defining the flight of the airplane when it does a BARREL ROLL.

As a lurker here who has never and will never do a single loop or roll, it
seems to me you two are both describing exactly the same maneuver but using
slightly different words.

Both you and Bob agree that the barrel roll consists of having the airplane
follow a corkscrew path as if it were following the inside wall of an
imaginary barrel, or tube, in the sky.
If the plane were to roll while going straight down the long axis of the
tube it would be an ordinary "slow" roll. OTOH, if the plane were flying
along the inside circumference (i.e. at 90 degrees to the long axis) of the
tube it would be doing loops. The corkscrew path of a barrel roll is
halfway between these two situations (i.e. at about a 45 degree angle to
the long axis of the tube), so it doesn't seem unreasonable to describe it
as the combination of a loop and a roll.

Bob

Found and subscribed to alt.binaries.pictures.avation and down
loaded all the postings in that Group. About 500 or so. Sat down and
went through them untiL I found your posting. Drilled down and finally
got the picture you are so hot about.

My comments, No one could do a barrel roll as depicted in your
picture. You show a single plane (not airplane) picture and maneuver
is done along an extended length of airspace.

You don't enter at 90 degrees to the desired direction of flight. Your
picture looks more like a single leaf of a four leaf clover.

Maybe this is what you call a barrel roll today but sure isn't the
classic maneuver.

On your query on my credentials in this post. I have answered that in
another posting. However to follow along that line since you won't let
it go, who are you to question my credentials? You sound like some
'want a be' that's never spent time on his back maybe not even a
pilot?. I may well have more inverted time than your total time????.

I can do a double emmelman in a F-80/T-33 and a slow roll in aT-6/SNJ
without the engine cutting out when inverted. Also I can do a loop in
a T-6/SNJ starting from zero indicated airspeed. Want to try any of
those maneuvers? Lets see your buddy Kershner do those.He's got T-33
time so ask him if he ever did a double emmelman in it and if so how
he did it (airspeed, altitude, G's, etc)? I've got thousands of hours
of tail wheel time and never ground looped or let one of my students
ground loop on any kind of a field and in any type of weather/wind.

BOb and I may well have the highest tail wheel time of any on this
group and he's got a lot???

Among others I have instructed both Chinese and Norwegian Air Force
students and was awarded wings from both countries, which I am
authorized by our government to wear (you need permission you know).

I am a member of the caterpillar club (

When do you want me to quit??????

The more we talk the more I disagree with you so need to shut things
down before we get violent G

If you want, take the last shot (some people just have to) and that's
it.

And the best to thee from the wee ones.

Big John


On Mon, 10 Nov 2003 02:02:01 GMT, Robert Moore
wrote:

Big John wrote

Talk me through a loop doing a Barrel Roll. We must be using
different words or maneuvers?

John, have you looked at the barrel roll picture that I e-mailed
you and posted on alt.binaries.pictures.aviation?
If not, further discussion will be of no use.

Have you checked the "Willian Kershner" web site and read the
author's credentials?
If not, you just leave us wondering "who-the-hell" is Big John,
and what credence should we give to his unsupported assertions?
Mr. Kershner owns and operates an Aerobatic Flight School and
has authored the book, "The Basic Aerobatic Manual" which as he
points out, was an offshoot (with permission) of the manual he
helped write for Cessna Aircraft in 1969.

First let me try a different set of words for a GA barrel roll.

Here are William Kershner's words that accompanied the diagram of
the barrel roll in his The Flight Instructor's Manual in which he
devotes all of chapter 5 (66 pages) to aerobatic instruction.

Quote..............or should I say OCR'ed

THE BARREL ROLL
• Preparation. It's almost impossible to draw a barrel roll
on the chalkboard, but a model will give the desired results.
Have the trainee study the references.

•Explanation. The barrel roll is a precise maneuver in which
the airplane is rolled around an imaginary point 45° to
the original flight path. A positive-g level is maintained
throughout the maneuver, and the ball in the turn indicator
should stay in the middle.

You may wonder why the barrel roll is taught this late,
since it appears to be so simple. Well, it is a precise maneuver
requiring particular airplane attitudes at particular reference
points, which is difficult for the average trainee to do properly
at first.

This maneuver might be considered an exaggeration of the
wingover, but instead of starting to shallow the bank at the 90°
position, the pilot must steepen it continually until the airplane
has rolled 360° and is back on the original heading. The rate of
roll must be much greater than that used for the wingover
because the airplane must be in a vertical bank at 45° of turn,
and it must be inverted at 90° of turn. The roll and turn is
continued until the airplane is headed in the original direction with
the wings level. Compare the barrel roll in Fig. 23-12 with the same
view of the wingover in Fig. 20-7.

From behind the maneuver looks as though the airplane is being
flown around the outside of a barrel. This is a very good maneuver
for gaining confidence and keeping oriented while flying inverted in
balanced flight.
Good coordination is required to do the barrel roll properly and
the trainee will show an improvement in that area after a session of
barrel rolls.

The barrel roll is generally more difficult and precise than the
aileron roll, and he may have to work on this one awhile.

Why-
The barrel roll is one of the best maneuvers for improving
orientation.

Unlike the other acrobatic maneuvers covered thus far, the barrel
roll requires a constantly changing bank and pitch (with attendant
changing airspeed) and a radical change in heading (90°) while the
airplane is rolling. The average trainee probably will be looking at
the wing tip at a time when he should be checking the nose, or vice
versa.
When he is able to stay well oriented in the barrel roll, he is ready
to move on to the reverse Cuban eight or reverse cloverleaf.

How-
You might use the following explanation, or develop your own:
(1) Make sure the area is clear, then pick a reference on the
horizon off the wing tip as in the wingover and lazy eight.
(2) Set the throttle to low cruise rpm and ease the nose over to
pick-up about 10 K more than used for the wingover or set up
the airspeed used for a loop, whichever is higher. Power
adjustment should not be necessary during the maneuver. You
might have some of your sharper trainees apply full power as
the airplane approaches inverted and then remind them to
throttle back as the airspeed picks up in the last part of the
maneuver.
(3) Smoothly pull the nose up and start a coordinated climbing turn
(note that it will have to be at a much faster rate than was
used for the wingover) toward the reference point. (Assume that
at first the roll will be to the left.)
(4) When the nose is 45° from the original heading, it should be
at its highest pitch attitude and the left bank should be
vertical.
(5) When the nose is at 90° from the original heading, you should
be looking directly at the reference point that was originally
off the wing tipfrom a completely inverted position
(momentarily).
(6) When the airplane heading is again 45° from the original, the
bank is vertical but you will be in a right bank as far as the
ground is concerned; that is, the right wing is pointing
straight down at this instant of roll. The nose will be at its
lowest pitch attitude at this point.
(7) The roll is continued to wings-level flight as the nose is
raised back to the cruise attitude.

The maneuver must be symmetrical; the nose must go as far above the
horizon as below. The barrel roll requires definite checkpoints to
ensure that the airplane is at the correct attitude throughout. It is
interesting to note that if the barrel roll is to the left, all of
the airplane's path is to the left of the original line of flight and
the airplane's nose is always pointed to the left of the original
flight line (until it merges again at the completion of the
maneuver). The opposite occurs, naturally, for the barrel roll to the
right.

Another method of doing a barrel roll is to pick a reference on the
horizon, turn the airplane 45° to the reference point, and proceed to
make a wide roll around this real point. One disadvantage of this
method for the newcomer is that it depends on the pilot's own
judgment of how large the orbit around the point should be. For an
introduction to the maneuver, the first method is usually better, but
you may prefer the second and work out your own techniques of
instructing it.

Demonstration.
Try not to lose the reference point yourself while demonstrating
this one. You may find your explanation is not keeping up with the
airplane, which usually results in sputtering and stuttering while
the maneuver proceeds to its foregone conclusion -and then you have
to do a new demonstration. Don't worry, this will happen plenty of
times during your career of instructing aerobatics -when your mouth
can't keep up with your brain or the maneuver-and it can ease tension
if you react to it with humor.

Usually the trainee is surprised to see the same wing tip back on the
reference point and may confess that, like the first snap roll, the
earth and sky were blurred and he had no idea where the reference was
during the maneuver.

Practice.
You may rest assured the trainee will "lose" the reference point
during the first couple of barrel rolls. He'll usually stare over the
nose, seeingnothing but blue sky or ground and not really seeing the
point at all.

Common errors during barrel rolls include these:
1. Not pulling the nose high enough in the first 45° of the
maneuver, which means that the highest and lowest nose positions
are not symmetrical to the horizon.
2. Not maintaining a constant rate of roll. Usually things are fine
at the 45° position; the nose is at its highest pitch and the
bank is vertical.
As you approach the position of 90° of turn you will probably
find that he is not going to be completely inverted at that
point and will have to rush things a bit to make it. The usual
reason is that he did not maintain a constant rate of roll.
Remember that the nose is up and the airspeed is slower in this
segment of the maneuver, so the controls must be deflected more
to get the same rate. This is where coordination comes in. Watch
for it in particular.
3. Letting the nose drop after passing the 90° point; losing too
much altitude and gaining excess airspeed.
4. Failure to roll out on the original heading; having the wing tip
well ahead, or well behind, the reference when the maneuver is
completed.

Evaluation and Review.
Review each barrel roll briefly in the air, and have the trainee
use the model on the ground. This one can be hard to "see," so go
over it again as necessary after getting on the ground.

By the time a half-dozen barrel rolls have been practiced, the
average trainee should be oriented throughout the maneuver even
though he may still have minor problems of heading and symmetry.
After a dozen rolls he should be starting to work on a constant roll
rate and starting to ease his heading problems. After several
hundred, he may begin to be satisfied with his barrel rolls but will
realize that constant practice is required.

Unquote.................

But John...you really should look at that picture.

Bob

Dylan

From my good book (Owners Manual Model 172 and Skyhawk)

Normal category

Gross weight --- 2300#

Flaps up

+3.8 G
-1.52 G

Normal category is non acrobatic

You can do stalls (not whip stalls) and are limited to a max angle of
bank of 60 degrees.

Utility Category

Max gross weight --- 2000#

Flaps up

+4.4 G
-1.76 G

No aerobatic maneuvers are approved except those listed.

Chandelles
Lazy Eights
Steep Turns
Spins
Stall (Except Whip Stalls)

From these figures you can see that it would be touch and go if you
rolled the bird inverted. Of course the figures given are not ultimate
so might only bend things a little )

Big John

* Added note. "In the execution of all maneuvers, avoid abrupt use of
controls."



On Mon, 10 Nov 2003 08:13:29 -0000, Dylan Smith
wrote:

In article , Big John wrote:
On a 172,. I'd do a barrel roll in the bird but not a slow roll or
aileron roll .Either would over stress the bird 99 times out of a
hundred.

Really?

All the eileron rolls I've ever done have been distinctly low-G
manoevres. Dive a little to gain entry speed, pull the nose up to
about 30 degrees above the horizon, then full aileron until the world
comes the right way up again, at which point you'll be about 20 degrees
nose down. The G-meter has never shown more than 1.5G after
an aileron roll for me, and that was done in the initial pull-up.

I thought a C172 in the utility category was good for 4.2G, not 1.5G!

In article , Big John wrote:
On a 172,. I'd do a barrel roll in the bird but not a slow roll or
aileron roll .Either would over stress the bird 99 times out of a
hundred.

Really?

All the eileron rolls I've ever done have been distinctly low-G
manoevres. Dive a little to gain entry speed, pull the nose up to
about 30 degrees above the horizon, then full aileron until the world
comes the right way up again, at which point you'll be about 20 degrees
nose down. The G-meter has never shown more than 1.5G after
an aileron roll for me, and that was done in the initial pull-up.

I thought a C172 in the utility category was good for 4.2G, not 1.5G!

--
Dylan Smith, Castletown, Isle of Man
Flying: http://www.dylansmith.net
Frontier Elite Universe: http://www.alioth.net
"Maintain thine airspeed, lest the ground come up and smite thee"

Big John wrote

My comments, No one could do a barrel roll as depicted in your
picture. You show a single plane (not airplane) picture and
maneuver is done along an extended length of airspace.

Not "my" picture. Was scanned from Mr. Kershner's book.

Maybe this is what you call a barrel roll today but sure isn't
the classic maneuver.

Same maneuver that I was taught in the T-28 back in 1958.
Check-out this web site:

http://web.winco.net/~efildes/slowroll/barlroll.html

On your query on my credentials in this post. I have answered
that in another posting. However to follow along that line since
you won't let it go, who are you to question my credentials? You
sound like some 'want a be' that's never spent time on his back
maybe not even a pilot?. I may well have more inverted time than
your total time????.

20,000+ hours inverted??? Wow! That's a lot John.
My last aerobatic instructing was done as recently as 1998 in the
YAK-52. Great little airplane....good for +7g's. It combined the
performance of the Navy T-28B (1425 hp) in a T-34 size airframe.
BTW, in addition to a few thousand hours instructing in Boeing
jetliners, I have been an FAA certificated flight instructor since
1970. Spent the years 1993-99 teaching in C-172s.

BOb and I may well have the highest tail wheel time of any on
this group and he's got a lot???

Well, I built my own tailwheel airplane and flew it for 10 years.
It was exhibited at the Sun-n-Fun airshow in 1991 and 1992.
According to the factory that provided the kit,(TEAM) I was the first
and probably only person to spin the MiniMax.

Now....in summary, the following three web sites (documentation)
assert that the barrel roll is a combination of a loop and a roll.
But....Big John (no documentation) insists that there is no loop in
a barrel roll.

http://web.winco.net/~efildes/slowroll/barlroll.html
http://acro.harvard.edu
http://www.kershnerflightmanuals.com/

Bob Moore
ATP B-727 B-707 B-720 L-188
CFI A/I AGI IGI
USN S-2F P-2V P-3B

I get it and am smarter for it. Thanks.

My acro terminology is so rooted in RC flying from 30+ years ago, that I've
never realized how differently the terms are used in 'real' a/c.

In RC, the term aileron roll is commonly used for what is called a slow roll
here. It is a straight line roll and can happen very fast. Aileron and
elevator will produce a quite passable axial roll in a hot RC plane. Very
akin to what a full scale jet fighter might do. Pros like my brother will
use top rudder too.

A barrel roll in RC is typically done with rudder in a model without
ailerons. It looks like a full scale aileron roll. Less-than-hot
rudder/elevator only RC plance can do a passable barrel roll because of a
lot of dihedral. You don't even need the elevator if it is setup right.

And in a full scale glider, it's kind of hard to get all the way around a
pure aileron roll as described. You need to be pretty agressive with pitch
and rudder to do a passable 'aileron' roll. At least in my limited
experience.

"Robert Moore" wrote in message
...
"Maule Driver" wrote
I have some glider acro training and lot's of RC model acro. I
think of axial rolls and barrel rolls. Both benefit from
aileron, elevator and rudder input. I've thought that aileron
roles are synonymous with axial and slow rolls synonymous with
barrel rolls.

From the following web site about aerobatics:

http://acro.harvard.edu/IAC/faq_aerobatics.html

Aileron Rolls
Aileron rolls are flown with the rudder and elevator in the neutral
position during the roll. The aileron is fully deflected in the
direction of the roll. This is the easiest of the rolls to fly.
The aileron roll is started by pulling the nose up to 20 - 30 degrees
above the horizon. The elevator is then neutralized and the aileron
fully deflected in the direction of the roll. The controls are
maintained in that position till the roll is completed. After the
roll is completed the nose is usually 20 - 30 degrees below the
horizon.

The aileron roll is not a competition maneuver.


Slow Rolls
Slow rolls have to be flown normally on a straight line (exception is
the avalanche). The roll rate has to be constant and the longitudinal
axis of the plane has to go straight. This requires constantly
changing rudder and elevator control inputs throughout the roll.
Hesitation or point rolls include stops at certain roll angles. The
number on the base of the roll symbol describes the number of points
the roll would have if it were a 360 degree roll. Allowed are 2
point, 4 point and 8 point rolls. The fraction on the arrow of the
roll symbol describes what fraction of a full roll is to be executed.
If no points are specified, rolling is done without hesitations. If
no fraction is specified, a roll symbol that starts at the line
specifies a half roll (see description of the Immelman). A roll
symbol that crosses the line specifies a full roll (first figure).
The second figure shows the symbol for 2 points of a 4 point roll
(adding up to half a roll) from upright to inverted flight.

Snap Rolls
Snap or flick rolls also have to be flown normally on a straight
line. A snap roll is similar to a horizontal spin. It is an
autorotation with one wing stalled. In the regular snap, the plane
has to be stalled by applying positive g forces. In an outside snap,
the plane is stalled by applying negative g. In both cases rudder is
then used to start autorotation just like in a spin.

Barrel Roll
The Barrel Roll is a not competition maneuver. I The barrel roll is a
combination between a loop and a roll. You complete one loop while
completing one roll at the same time. The flight path during a barrel
roll has the shape of a horizontal cork screw. Imagine a big barrel,
with the airplanes wheels rolling along the inside of the barrel in a
cork screw path. During a barrel roll, the pilot experiences always
positive G's. The maximum is about 2.5 to 3 G, the minimum about 0.5
G.

Bob Moore

Or the corkscrew motion of an aileron roll must provide enough "outward"
(normal) force to counteract the downward force of gravity.

I think I understand what you're saying, Hamish. It may depend on the
airplane. An SNJ doesn't "corkscrew" very much. It would hardly be
enough to provide positive G in excess of 1.0.

The maneuver starts with a pullup...

It does start with a slight pullup, but then then back-prssure on the
stick is released, to produce what approximates a zero-G situation.
(Don't you do that in an Aerobat?)

I thnk the best description of an aileron roll is to call it a zero-G
maneuver. Not *exactly* zero, to the tenth decimal place, but close to
zero all the way around.

and that pullup is never really lost

I don't understand that. I need to use some right rudder after
passing the half-way point of a left aileron roll, or I'll end up
with the nose pointing down about 20 degrees.

(unlike the slow roll, which can result in negative G's in many aircraft).

I don't understand that, either. You're saying there are airplanes
that can perform a slow roll without experiencing negative Gs?

vince norris

In article ,
vincent p. norris wrote:

Or the corkscrew motion of an aileron roll must provide enough "outward"
(normal) force to counteract the downward force of gravity.

I think I understand what you're saying, Hamish. It may depend on the
airplane. An SNJ doesn't "corkscrew" very much. It would hardly be
enough to provide positive G in excess of 1.0.

Well, the *whole point* of a textbook aileron roll is to maintain
positive G's -- not 1+ G's, but *positive* -- all the way through the
maneuver, regardless of whether you do it in a Pitts or a puny Aerobat
(the Pitts is a lot more, erm, exciting in this respect :-)).

The maneuver starts with a pullup...

It does start with a slight pullup, but then then back-prssure on the
stick is released, to produce what approximates a zero-G situation.
(Don't you do that in an Aerobat?)

The textbook aileron roll in an Aerobat starts with a shallow dive to
120 KIAS, then a smart pullup to 30 degrees pitch, then a quick
simultaneous full-over on the ailerons and neutralization of the
elevator until pullout. Apply rudder as appropriate... The Pitts is
pretty much the same, but at least with the constant speed prop you
don't have to spend so much time obsessing about the throttle.

I thnk the best description of an aileron roll is to call it a zero-G
maneuver. Not *exactly* zero, to the tenth decimal place, but close to
zero all the way around.

In no plane I've ever done an aileron roll in is the G force "close to
zero all the way around". In fact, it's close to 2G's at two points in
the maneuver (pullup and pullout), and it's probably around .5 to 1 G
over the top.

and that pullup is never really lost

I don't understand that. I need to use some right rudder after
passing the half-way point of a left aileron roll, or I'll end up
with the nose pointing down about 20 degrees.

Remember, half way around your altitude is still quite a bit higher than
it was when you started the maneuver....

(unlike the slow roll, which can result in negative G's in many aircraft).

I don't understand that, either. You're saying there are airplanes
that can perform a slow roll without experiencing negative Gs?

No I'm saying that true textbook slow rolls result in negative G's (and
are nearly impossible in an Aerobat, but never mind...).

Hamish

An SNJ doesn't "corkscrew" very much. It would hardly be
enough to provide positive G in excess of 1.0.

Well, the *whole point* of a textbook aileron roll is to maintain
positive G's -- not 1+ G's, but *positive* -- all the way through the
maneuver....

Sorry, I didn't express myself well. I meant "It would hardly be
enough to provide positive G in excess of 1.0, enough to negate the
negative 1 G caused by gravity, plus a bit more to make it a positive
G maneuver."

BTW, I looked again at a tape of Tex Johnston's rolls, going and
coming. I cannot detect any "corkscrewing."

regardless of whether you do it in a Pitts or a puny Aerobat
(the Pitts is a lot more, erm, exciting in this respect :-)).

I'm sure it is. I've had a little time in a friend's Starduster 2
but never had the pleasure of flying a Pitts.

It does start with a slight pullup, but then then back-prssure on the
stick is released, to produce what approximates a zero-G situation.
(Don't you do that in an Aerobat?)

The textbook aileron roll in an Aerobat starts with a shallow dive to
120 KIAS, then a smart pullup to 30 degrees pitch, then a quick
simultaneous full-over on the ailerons and neutralization of the
elevator until pullout. Apply rudder as appropriate...

I would agree with that, except I neutralize the elevator THEN begin
the roll.

I don't think of the pullup as part of the maneurver, but as
"preparation" for it.

In fact, it's close to 2G's at two points in
the maneuver (pullup and pullout), and it's probably around .5 to 1 G
over the top.

I don't think of the pullup as part of the maneurver, but as
"preparation" for it. I don't pull up that sharply, and if I have to
"pull out" at the end, I figure I didn't do it right.

If by "over the top" you mean when 180 degrees inverted, I find that
very hard to imagine, based solely on my experience. (I've never
flown an airplane with a G-meter.) I don't hang on the seat belt, as
in a slow roll, but I feel a bit "weightless." I don't seem to be
pushing down (up?) on the seat very hard.

Remember, half way around your altitude is still quite a bit higher than
it was when you started the maneuver....

Yes, I think the airplane follows something like a "ballistic curve"
during the maeuver. Certainly, it gets dang little vertical lift as
it rolls past the 90 degree and 270 degree points.

Despite the fact that the nose is pointed up slightly, the airplane is
essentially "falling" and thus it, and whatever is in it, is
experiencing zero Gs. (One G from gravity, counteracted by one G from
the accelleration.)

vince norris

In article ,
vincent p. norris wrote:

An SNJ doesn't "corkscrew" very much. It would hardly be
enough to provide positive G in excess of 1.0.

Well, the *whole point* of a textbook aileron roll is to maintain
positive G's -- not 1+ G's, but *positive* -- all the way through the
maneuver....

Sorry, I didn't express myself well. I meant "It would hardly be
enough to provide positive G in excess of 1.0, enough to negate the
negative 1 G caused by gravity, plus a bit more to make it a positive
G maneuver."

This is a little frustrating -- the standard aileron roll is done to
cause continuous positive G's. That's an *aim* of aileron rolls. And the
net G forces don't have to be more than 1, just enough to cause *net*
positive G forces on the pilot 9and engine, etc.). That's the
*definition* of a positive G maneuver.

This started when you stated that aileron rolls don't cause positive G's
all the way around, -- and that's just wrong, unless by "aileron roll"
you mean something different from what the IAC, various texts, etc., all
mean by the term.

BTW, I looked again at a tape of Tex Johnston's rolls, going and
coming. I cannot detect any "corkscrewing."

I can't comment on that except to say the corkscrewing is not always
obvious, especially from the ground... (I've never seen the videos, if
you're talking about the Dash-80 "barrell roll").

regardless of whether you do it in a Pitts or a puny Aerobat
(the Pitts is a lot more, erm, exciting in this respect :-)).

I'm sure it is. I've had a little time in a friend's Starduster 2
but never had the pleasure of flying a Pitts.

It does start with a slight pullup, but then then back-prssure on the
stick is released, to produce what approximates a zero-G situation.
(Don't you do that in an Aerobat?)

The textbook aileron roll in an Aerobat starts with a shallow dive to
120 KIAS, then a smart pullup to 30 degrees pitch, then a quick
simultaneous full-over on the ailerons and neutralization of the
elevator until pullout. Apply rudder as appropriate...

I would agree with that, except I neutralize the elevator THEN begin
the roll.

Well, I'm lucky if I can get it all done at the same time :-).

I don't think of the pullup as part of the maneurver, but as
"preparation" for it.

It's a necessary part of achieving that ballistic corkscrew motion --
and (speaking from dumb experience) if you don't do the pullup properly,
you can end up in something more akin to a split-S or a nasty dive off
the end of the "roll", and / or a great deal of lost altitude.

In fact, it's close to 2G's at two points in
the maneuver (pullup and pullout), and it's probably around .5 to 1 G
over the top.

I don't think of the pullup as part of the maneurver, but as
"preparation" for it. I don't pull up that sharply, and if I have to
"pull out" at the end, I figure I didn't do it right.

Well, the standard aileron roll typically has you pointing down at the
ground at much the same pitch at the 360 degree point that you started
the roll from (as a consequence of the corkscrew motion), so if you
don't do the pullup, you find yourself gaining speed very rapidly...

If by "over the top" you mean when 180 degrees inverted, I find that
very hard to imagine, based solely on my experience. (I've never
flown an airplane with a G-meter.) I don't hang on the seat belt, as
in a slow roll, but I feel a bit "weightless." I don't seem to be
pushing down (up?) on the seat very hard.

For it to be a positive G maneuver, you don't have to be pushed down
very hard. Most people probably think it's negative G's when in fact
it's just reduced G's. Negative G's mean, yes, you're pressing up
against the straps....

Remember, half way around your altitude is still quite a bit higher than
it was when you started the maneuver....

Yes, I think the airplane follows something like a "ballistic curve"
during the maeuver.

Indeed -- the ballistic corkscrew curve is part of the definition of the
aileron roll I gave from Szurovy and Goulian earlier in this thread.
It's that motion that makes it a positive G maneuever. If, like me on a
typical day, you screw up and use the elevator or rudder wrongly, you
can certainly make it a negative G maneuver, but it stops being a real
aileron roll at that point (and starts becoming a cause for quick
roll-out-of-trouble action :-)).

Certainly, it gets dang little vertical lift as
it rolls past the 90 degree and 270 degree points.

It's not really supposed to.

Despite the fact that the nose is pointed up slightly, the airplane is
essentially "falling" and thus it, and whatever is in it, is
experiencing zero Gs. (One G from gravity, counteracted by one G from
the accelleration.)

No, the plane is experiencing *positive* G's in a decent aileron roll.
It's already starting to dive off the top at this point. Or should be,
if you're doing it right...

Hamish