Tandem wing pitch stability.
Don Stauffer
aircraft. I cannot find much on longitudinal stability of
these beasts, and do not understand it. The only thing even
close I have is an analysis of 'lifting stab' free-flight
models, and that section of the book I have has been soundly
criticized by many. Any references folks can suggest (ones
available/obtainable)?
I have done a pretty thorough web search on the 'flying
flea,' but have not come up with anything sufficiently
technical to be useful in understanding and design.
Can a tandem wing be stable in pitch with the same airfoil
on front and rear?
--
Don Stauffer in Minneapolis
stau...@gte.net
http://home1.gte.net/stauffer/
Tom Speer
>
> ...
> Can a tandem wing be stable in pitch with the same airfoil
> on front and rear?
Sure. You need to make sure that the front wing is more heavily
loaded than the aft wing. It must trim with the front wing at a
higher angle of attack than the rear wing.
Find the longitudinal location where the pitching moment does not
change with angle of attack, taking into account the effect of
the front wing's downwash on the aft wing. This point is the
neutral point. Your center of gravity must be ahead of that
point.
The stability relationships should really be no different than
those for an aft tail or canard. Just plug in the respective
areas. The trickiest bit might be deciding on an appropriate
reference area. But if you work out the stability in dimensional
terms instead of nondimensional equations, this shouldn't matter.
Good luck,
--
Tom Speer
tsp...@gte.net
http://home1.gte.net/tspeer
fax: +1 206 878 5269
Ib Therkelsen
>
> I am interested in trying to design a model tandem wing
> aircraft. I cannot find much on longitudinal stability of
> these beasts, and do not understand it. The only thing even
> close I have is an analysis of 'lifting stab' free-flight
> models, and that section of the book I have has been soundly
> criticized by many. Any references folks can suggest (ones
> available/obtainable)?
>
> I have done a pretty thorough web search on the 'flying
> flea,' but have not come up with anything sufficiently
> technical to be useful in understanding and design.
>
> on front and rear?
I believe that your topic is covered in the book "The design of the
aeroplane" by Darrol Stinton. At least the Flying Flea is discussed at
some length.
Ib Therkelsen
Dennis M. Straussfogel
> I am interested in trying to design a model tandem wing
> aircraft. I cannot find much on longitudinal stability of
> these beasts, and do not understand it. The only thing even
> close I have is an analysis of 'lifting stab' free-flight
> models, and that section of the book I have has been soundly
> criticized by many. Any references folks can suggest (ones
> available/obtainable)?
>
> I have done a pretty thorough web search on the 'flying
> flea,' but have not come up with anything sufficiently
> technical to be useful in understanding and design.
>
> Can a tandem wing be stable in pitch with the same airfoil
> on front and rear?
Isn't a tandem wing configuration the same as a conventional wing-tail
combination with a very large tail, i.e., the same size as the wing? I
don't think any assumptions are made in the derivation of the
longitudinal stability equations for a conventional configuration that
would invalidate them for a situation where the tail area approaches
that of the wing. Even if some such assumption were made, it should be
easy enough to rederive the equations without making that
assumption. Given this view point, a tandem wing configuration would
be extremely stabile, the same as a conventional configuration with a
huge tail. Controllability might be another matter.
---------------------------------------------------------------------
Dennis M. Straussfogel, Ph.D.
Aerospace Engineering Consultant
dm...@nh.ultranet.com
Don Stauffer
> In article <3A0427E5...@gte.net>, Don Stauffer <stau...@gte.net> wrote:
> > I am interested in trying to design a model tandem wing
> > aircraft. I cannot find much on longitudinal stability of
> > these beasts, and do not understand it. The only thing even
> > close I have is an analysis of 'lifting stab' free-flight
> > models, and that section of the book I have has been soundly
> > criticized by many. Any references folks can suggest (ones
> > available/obtainable)?
> > Can a tandem wing be stable in pitch with the same airfoil
> > on front and rear?
> Isn't a tandem wing configuration the same as a conventional
> wing-tail combination with a very large tail, i.e., the same size as
> the wing? I don't think any assumptions are made in the derivation
> of the longitudinal stability equations for a conventional
> configuration that would invalidate them for a situation where the
> tail area approaches that of the wing. Even if some such assumption
> were made, it should be easy enough to rederive the equations
> without making that assumption. Given this view point, a tandem wing
> configuration would be extremely stabile, the same as a conventional
> configuration with a huge tail. Controllability might be another
> matter.
The tandem wing designs had both wings at positive angle of
attack at trim. This is indeed like a 'lifting stab'
free-flight model airplane, but that type of freeflight does
not use the normal method of achieving stability. In fact,
some books on aerodynamics insist that the lifting stab ff
is not stable.
I know in flying them many years ago they were sure tricky.
You either got a really good flight or you got disaster :-(
I used to love hand launch glider flying. These were
typically set up zero-zero (zero decalage). If your launch
was perfect, and it rolled out of the launch thrust level,
then you got a good flight. Heaven help you if it rolled
out nose high.
brian whatcott
>Don Stauffer wrote:
>>
>> ...
>> Can a tandem wing be stable in pitch with the same airfoil
>> on front and rear?
>....
>
>Sure. You need to make sure that the front wing is more heavily
>loaded than the aft wing. It must trim with the front wing at a
>higher angle of attack than the rear wing.
>
>Find the longitudinal location where the pitching moment does not
>change with angle of attack, taking into account the effect of
>the front wing's downwash on the aft wing. This point is the
>neutral point. Your center of gravity must be ahead of that
>point.
>
>The stability relationships should really be no different than
>those for an aft tail or canard. Just plug in the respective
>areas. The trickiest bit might be deciding on an appropriate
>reference area. But if you work out the stability in dimensional
>terms instead of nondimensional equations, this shouldn't matter.
I seem to recall that for a canard, the stick fixed margin is more
critical than the stick-free, the reverse of the aft tail arrangement.
Brian
--
Brian Whatcott Altus OK
<in...@intellisys.net>
Eureka!
peter....@baesystems.com
> I am interested in trying to design a model tandem wing
> aircraft. I cannot find much on longitudinal stability of
> these beasts, and do not understand it. The only thing even
> close I have is an analysis of 'lifting stab' free-flight
> models, and that section of the book I have has been soundly
> criticized by many. Any references folks can suggest (ones
> available/obtainable)?
> Can a tandem wing be stable in pitch with the same airfoil
> on front and rear?
Having designed & built several successful "canard" models (photos
available on request, but it might take me a day or two to get them
scanned) I would make some comments. The following is very much a
practical, as opposed to theorectical treatment:
In general the important issue is to arange for the foreplane to stall
before the mainplane. If the mainplane stalls first the aircraft
"falls" backwards and was to wait for the directional stability (ie
the fin) to point it in the right direction before it can start to
recover into normal flight and this takes a lot of height. You can
also get some very dangerous asymetric stall modes which lead to spin
configurations which are essentially irrecoverable. You can arange for
the foreplane to stall first in a number of ways:
1. Give it a higher loading than the mainplane, which essentially
means trim the CG such that the foreplane ends up at a slightly higher
angle of attack than the mainplane in straight & level flight.
2. Give it a higher aspect ratio than the mainplane (only really works
if the mainplane is a very low-aspect ratio surface like a "delta"
which doesn't really exhibit a stall per se).
3. Give the foreplane a radically different section to the
mainplane. Choose something with a very low Cl(max) - usually a thin,
low-camber section and/or a very sharp leading edge.
There are other ways, but you get the general idea. While we're on the
subject, it is much safer to use an "all flying foreplane" for pitch
control than to use a fixed foreplane with "elevators". This is
because the elevators are actually "flaps" which may actually increase
the Cl(max) of the foreplane section and delay its stall.
The most practical technique is the first. To achieve this, consider
the foreplane and mainplane as seperate wings. Take a "reference CG"
for each wing; say 25% mean aerodynamic chord (where MAC is adjusted
for sweep, taper and washout in the usual way) and then measure the
distance between them. The "equal stall" CG point will be the point
on this line where each wing has equal moment - if the foreplane is
25% of the mainplane area (ie 20% of the total surface area) and the
distance is 500mm then this point will be 100mm in front of the point
which represents 25% of the mainplane MAC (gosh that sounds
complicated - it isn't really and I hope my meaning is clear!). Put
your CG 5-10% further along this line (ie 25-50mm) as a starting
point. As a general rule, if the model trims out with the foreplane
rigged at between +1.5 and +3 degrees with respect to the mainplane
you're probably safe.
Don't be tempted to overdo it! I have seen canard models with the
foreplane mounted at more then 7 or 8 degrees with respect to the
mainplane, and this produces a model which has excessive
speed-stability (ie any change in speed requires a change in pitch
trim - the model seems very "zoomy") and also severely limits the
looping-plane manoeuverability. This latter effect feels like a does
of "elevator understeer"; whenever you pull on the stick the foreplane
stalls and the model carries on in a straight line rather than going
around the corner!
Note also that it is in general possible to achieve a stable canard
model with the CG positioned such that the foreplane trims out
anywhere from about -3 degrees to +10 degrees, but few of these
configurations have "safe" stalling characteristics!
Some people have difficulty getting to grips with the directional
stability requirements (ie fin area and moment arms). These aren't
really any different for a canard than they are for a "conventional"
configuration except that:
1. Any dihederal on the mainplane effectively ADDS to the fin area
(because it's behind the CG) instead of subtracting from it as it does
with a conventional configuration, and dihederal on the foreplane
SUBTRACTS from it for the same reason.
2. Sweep on the mainplane generally reduces the fin area requirement,
although this isn't as cut and dried as it is for dihederal.
In general the best way to test a particular configuration is to build
a simple chuck-glider version of your chosen configuration of around
450mm in wingspan, using simple slabs of 1.5mm thick balsa for the
mainplane, foreplane and fin, and a profile of 3mm balsa for the
fuselage. Move the CG around with lumps of plasticene ballast and
throw the model into deep stalls and other unusual attitudes to see
how it recovers. Remember that (for a radio-controlled model) you
aren't looking for absolute "revover from any attitude" positive
stability, merely the absence of nasty "departures" from controllable
situations.
One final thought. Canard gliders are easy to get right. Powered
canards can be more difficult if you want to put a single engine on
the front. This is because the propwash can end up dominating the
airflow over the foreplane which can prevent it from stalling properly
- this way lies disaster! The solutions are either:
1. Put the engine on the back (which can make it difficult to achieve
the required propeller/ground clearence and still have a reasonably
short undercarriage length).
2. Put the foreplane a long way behind the engine (which can be
difficult to achieve without needing several tons of ballast in the
tail)
3. Use two engines mounted in nacelles on the mainplane. This makes
for a particularly elegant design, but you usually have to put the
fins and rudders on booms on the back of the nacelles (as per Rutan's
"Voyager") if you want halfway decent rudder effectiveness.
In my experience other factors (like minimising the effect of the
foreplane wake on the mainplane) which challenge "full size" designers
can be ignored for models because you simply ain't interested in
wringing the last 5% of efficiency out of the typical model.
Hope this is of interest.
--
Peter D Rieden
(aka PDR)
Peter....@baesystems.com
Some further thoughts after some reflection...
I forgot to talk about the foreplane area. It is possible to design a
canard with a foreplane area anything from 0% to 100% of the mainplane
area. The former is called a tail-less, and beyond the upper limit
convention dictates that we call the foreplane the mainplane(!). In
practice if you go much below 10% two things happen:
1. The aeroplane becomes very sensitive to CG location.
2. The aeroplane can become short of pitch (elevator) authority using
the foreplane alone, and it has to be augmented with some other form
of pitch control (usually elevons, but this only works if the
mainplane is sweapt or deltoid).
A good-handling canard with a broad CG tollerance can be achieved with
a foreplane of 20-25% of the maiplane area. I once built a true
"tandem wing" aerobatic glider which had identical main &
foreplanes. The model was controlled using full-span elevons on each
wing and had the foremost one rigged at about half a degree more
positive than the rearmost one (actually 2mm at the leading edge on a
225mm root chord). The model was equipped with mechanical control
mixers such that the righthand transmitter stick controlled the
"aileron mode" of the foreplane and the "elevator mode" of the
rearplane whilst the left hand transmitter stick (with throttle
ratchet replaced with centring springs) controlled the "aileron mode"
of the rearplane and the "elevator mode" of the foreplane. This
produced a model which could be flown by anyone, whether left or right
handed or mode 1 (split stick) or mode 2 (combined stick) preference.
The model was built to experiment with control configurations and if
it hadn't been for the difficulty in providing rudder control at the
transmitter I might have built a powered version, but I couldn't ever
see a way of doing this! Rolling the model using BOTH sets of ailerons
produced a rather fast and satisfyingly axial roll, whilst using full
aileron in opposite direction on each stick yeilded extremely powerful
airbrakes. Pulling back on the RH stick whilst pushing forwards on
the LH stick effectively dropped "flaps" on all four corners and
resulted in a nice steep, draggy landing approach. Doing the opposite
resulted in 50 degrees of NEGATIVE flap all round, which yeilded a
brick simulation second only to a landing approach in a Jaguar...
One day (when time permits) I want to return to this project because I
feel there is more to gain from the control configuration. Incidently,
did you know that the current crop of canard fighters (EFA/Typhoon,
Rafale, Grippen etc) don't use the canard surface for elevator
control? They use mainplane elevons as the primary pitch control and
the canards are just "trim controls".
"Not a lot of people know that..."<G>
I have built lots of canard gliders, but only two powered ones. The
first was a twin (for reasons given in my previous posting) with
slight forward sweep on the mainplane. This had a 25% foreplane and
flew extremely well. The second was based on the BAe P1233/1 SABA
(Small Agile Battlefield Aircraft) which had a single pusher
engine. The necessarily high thrust line of the SABA (to achieve
adequate propeller/ground clearance) made take-off runs (especially
from grass) rather long.