Study of flight dynamics of a gyroplane in gliding flight

Jean Claude

Junior Member
Joined
Jan 2, 2009
Messages
1,882
Location
Centre FRANCE
Aircraft
I piloted gliders C800, Bijave, C 310, airplanes Piper J3 , PA 28, Jodel D117, DR 220, Cessna 150, C
Total Flight Time
About 500 h (FW + ultra light)
In German Unfortunately not translated https://elib.dlr.de/88528/1/IB-Seitengleitflug.pdf
Autor: Falk Sachs from DLR
Comment: It will take to me some time to understand how the "yaw angle" of a rotating rotor is not simply like the angle of attack of a cylindrical profile

Sans titre.png
 

Brian Jackson

Platinum Member
Joined
Jul 17, 2004
Messages
2,853
Location
Hamburg, New Jersey USA
Aircraft
GyroBee Variant - Under Construction
I tried translating the PDF through a couple of services but the results were poor. Will attempt a different method when I get time later today. Thank you for posting this. It looks like a wealth of information to study.
 

XXavier

Member
Joined
Nov 13, 2006
Messages
912
Location
Madrid, Spain
Aircraft
FW Savannah XL + ELA R-100 and Magni M24 autogyros
Total Flight Time
1394 FW + 526 gyro (Oct. 2019)
In German Unfortunately not translated https://elib.dlr.de/88528/1/IB-Seitengleitflug.pdf
Autor: Falk Sachs from DLR
Comment: It will take to me some time to understand how the "yaw angle" of a rotating rotor is not simply like the angle of attack of a cylindrical profile

View attachment 1145343
The title is (...) in gliding sideways flight...


The key issue is mentioned in page 18 of that paper:

Unter den heutigen Tragschrauberpiloten ist man sich einig, dass gerade im Seitenflug einige beunruhigende Effekte beobachtet wurden, die mit einigen schweren Unfällen in Verbindung gebracht werden können.

'There is, among present-day pilots of autogyros, a general consensus pointing to some worrying effects observed in sideways flight, which are probably connected with some serious accidents.'
 
Last edited:

Jean Claude

Junior Member
Joined
Jan 2, 2009
Messages
1,882
Location
Centre FRANCE
Aircraft
I piloted gliders C800, Bijave, C 310, airplanes Piper J3 , PA 28, Jodel D117, DR 220, Cessna 150, C
Total Flight Time
About 500 h (FW + ultra light)
At first glance, the stability of the airframe at the big yaw angle (sideways flight) seems to me the main cause to analyze.
The inversion of the lift slope of the stalled vertical stabiliser seems to me capable of producing a complete loss of control of the yaw, while the pitch and roll remains easily controllable whatever the angle of "yaw" of the rotor
 

XXavier

Member
Joined
Nov 13, 2006
Messages
912
Location
Madrid, Spain
Aircraft
FW Savannah XL + ELA R-100 and Magni M24 autogyros
Total Flight Time
1394 FW + 526 gyro (Oct. 2019)
At first glance, the stability of the airframe at the big yaw angle seems to me the main cause to analyze.
The inversion of the lift slope of the stalled vertical stabiliser seems to me capable of producing a complete loss of control of the yaw, while the pitch and roll remains easily controllable whatever the angle of "yaw" of the rotor
The problem of a sideways glide is –at least that's what this paper mentions– that in such a glide, the relative wind may cease blowing upwards through the rotor disk, the blades will slow down, and an often fatal crash follows...
 

Tyger

Active Member
Joined
Nov 25, 2017
Messages
281
Location
Germantown, NY
Aircraft
Magni M16
Total Flight Time
225
Are any accidents given as possible examples of the mooted phenomenon, or is this all just theoretical?
 

XXavier

Member
Joined
Nov 13, 2006
Messages
912
Location
Madrid, Spain
Aircraft
FW Savannah XL + ELA R-100 and Magni M24 autogyros
Total Flight Time
1394 FW + 526 gyro (Oct. 2019)
Are any accidents given as possible examples of the mooted phenomenon, or is this all just theoretical?
Five accidents are mentioned. Two in Germany, one in France, one in Spain and another in the US.

Captura de pantalla 2019-09-20 a las 7.16.12.png

I remember quite well the accident at Mallorca, that got much attention here and in Germany, since the pilot, Andy Tille, was a German who resided in Mallorca, where he managed a gyro flight school. He was a highly qualified instructor (if a little too daring, some said...). I remember reading an interesting comment in the German gyro forum, where a pilot with long experience wrote that he was seriously thinking in stopping flying gyros, because he thought that, within the flight envelope, there might be points of instability that were conducive to crashes, and the knowledge of where those points were was impossible, as the pilots usually died in those crashes...
 
Last edited:

Jean Claude

Junior Member
Joined
Jan 2, 2009
Messages
1,882
Location
Centre FRANCE
Aircraft
I piloted gliders C800, Bijave, C 310, airplanes Piper J3 , PA 28, Jodel D117, DR 220, Cessna 150, C
Total Flight Time
About 500 h (FW + ultra light)
After trying to understand, I think that the author has greatly complicated his analysis, because of the poor choice of spatial reference.
Relative to the direction of the relative wind, the longitudinal flapping a1 (about 2 °) and the lateral flapping b1 (about 1 °) simply have a different orientation during a slided flight, which just requires the pilot a weak and instinctive correction to balance at again the flight.
This can not result in any difficulty of control.
Also, transient throttle action from the idle does not cause difference in the behavior of the disc seen the pilot.
Half of the report is about this non-event.
Sader is that the stall of the vertical empennage during strong slides (60 degrees) is only evoked in a few lines. Il my opinion, this can produce a dangerous yaw unstability, and the lateral displacement of HS relatively to the trajectory can produces roll effects.
The main is just ommitted.
Can Juergen confirm this?
 
Last edited:

kolibri282

Active Member
Joined
Aug 2, 2009
Messages
2,780
Location
Duesseldorf
Dear all,
the discussion of this very interesting report (thanks goes out to the author and Holger Duda from DLR!) seems to be somewhat hampered by the fact that the text is in German. I had the impression that some important parts of the text have not been clearly understood, which does not come as a surprise since the author is using the German language in a pretty complex way.

To give a better basis for discussing the text I translated, what I consider, some key parts of it to the best of my abilities. Of course any proposal for improvement is highly appreciated. Both google Translate and deepL produced text that was unitelligeable for major portions, so the whole translation took about three hours.

Notes reagarding the translation.

Words in round brackets ending with trans. are additional remarks by the translator in an effort to further clarify the text. Up- and downwind have two very different meanings in this text and it is important to realize which is the one intended in the context. They may refer a) to the windward (upwind) and lee (downwind) side of the fuselage with respect to the oncoming air or they may b) refer to an additional upward or downward flow component through the rotor. Fortunately an upward flow component exists in the upwind sector of the fuselage while downwind we have a downwind...;-). The author consistently uses the German word "Seitengleitflug" which would translate to yawed gliding flight, indicating that the aircraft descends. I have omitted the term gliding for brevity, please keep this in mind while reading the text.

I have split the text in several sections giving the German and English text for each one to make it easier for the reader to identify the words and phrases that have been translated.

The main point the author makes is, that yawed flight for an autogyro is highly asymmetric due to the flow field over the fuselage and the rotor and therefore the autogyro will behave markedly different depending on whether the aircraft is flying nose left or nose right with respect to the resultant air speed. I have translated two key passages of the text, first the one where he describes the different behaviour of the rotor in the two cases and than the conclusions from the findings in the test flights. The whole text assumes a rotor turning counter clock wise and the main finding is, that with a CCW rotor an inexperienced pilot may very quickly enter an unrecoverable flight state if he is flying with the nose to the right w.r.t straight flight and entering abrupt control commands with too large an amplitude.
 

Attachments

Last edited:

Jean Claude

Junior Member
Joined
Jan 2, 2009
Messages
1,882
Location
Centre FRANCE
Aircraft
I piloted gliders C800, Bijave, C 310, airplanes Piper J3 , PA 28, Jodel D117, DR 220, Cessna 150, C
Total Flight Time
About 500 h (FW + ultra light)
Thank you very much, Juergen, for the time you spent translating this DLR text.
I had roughly understood what it was and, in my opinion, this study is based on a few misunderstanding of the rotor's operation by the author. he writed:
"The first important insight from the experiments is that the flapping angle in the ? =180?- as well as in the ? =90?-position in yawed flight with nose left w.r.t straight flight increases and in yawed flight with nose right w.r.t straight flight decreases. This is caused by a wind field induced by the fuselage.
How could the asymmetry of skid behavior be related to the presence of the fuselage since it is also symmetrical?
In fact, in a symmetrical flight, the maximum angle of beat is not in the axis of the gyro, which the author seems to ignore:

Sans titre.png
Angle b1 is due to the non-uniformity of the induced speed, and to the coning.
The flow on the fuselage is for nothing in this story .

he also writed:
"The wind field induces an upwind on the upwind (windward trans.) side (of the fuselage trans.) and on the opposite side a downwind of equal strength which results from stall and a dead water zone with considerable turbulence phenomena forming on the downwind side of the fuselage. Upwind and downwind both meet the rotor."
How can the wake of the fuselage meet the rotor?
It's really not convincing
Sans titre.png
 
Last edited:

kolibri282

Active Member
Joined
Aug 2, 2009
Messages
2,780
Location
Duesseldorf
Quote: How can the wake of the fuselage meet the rotor? /Quote
If one adds a few more streamlines and a rotor to Abbildung 6.9 page 64 it becomes immediately clear why there is a rotor fuselage interaction.
rotor_fuselage_interaction.png
 

Attachments

Last edited:

C. Beaty

Gold Supporter
Joined
Apr 16, 2004
Messages
9,656
Location
Florida
I’m aware of three fatal accidents resulting from yawed flight.

Gyros with substantial fuselage side area below the CG produce a rolling moment in yawed flight that at the limit can roll the fuselage beyond the ability of cyclic control to correct.

The Wind Ryder was a very attractive, streamlined design by the designer of Skywheels rotor blades.

Ron Herron, designer of the LittleWing gyro test flew the Wind Ryder for its owner and cautioned him not to let it get sideways, otherwise he might exceed control limits which is exactly what happened during the new owner’s first flight.


Factual
07/03/2003
Final
09/30/2003

Final Report PDF | HTML

Data Summary (PDF)

04/13/2000

LAKELAND, FL

Air & Space. 18A

N905AS

MIA00LA133

Fatal(1)





Final
08/25/1993

Final Report PDF | HTML

Data Summary (PDF)

03/08/1992

LITTLE ROCK, AR

WIND RYDER

N123ZT

FTW92DPG01

Fatal(1)
 

Jean Claude

Junior Member
Joined
Jan 2, 2009
Messages
1,882
Location
Centre FRANCE
Aircraft
I piloted gliders C800, Bijave, C 310, airplanes Piper J3 , PA 28, Jodel D117, DR 220, Cessna 150, C
Total Flight Time
About 500 h (FW + ultra light)
Yes, Juergen. However not efficient on the active radius (ie > 0.7R). At this distance, the shell is too far.
 

kolibri282

Active Member
Joined
Aug 2, 2009
Messages
2,780
Location
Duesseldorf
Quote: the shell is too far /Quote
The problem is similar to that of calculating ground effect. Now an accepted formula for this case is
vi/viN = (1 - exp(h/(2*R)/c))
where h is the hight of the aircraft above the ground R is the rotor radius and c is an empirical constant of about 0.2
if you enter h=0.7R you get 0.82 which means that even at this distance you already have a difference of 18% in induced velocity, but the rotor is much closer to the fuselage than 0.7. For a Calidus it is rather 0.3 and for this value the difference is 48%. The author also states that the simulation model they have was modified to incorporate the effect and gave very good results. I am afraid your gut feeling in this case is wrong. Actually applying mathematical methods like potential theory to engineering problems had been invented because gut feelings turned out to be wrong too often..;-)
 

Jean Claude

Junior Member
Joined
Jan 2, 2009
Messages
1,882
Location
Centre FRANCE
Aircraft
I piloted gliders C800, Bijave, C 310, airplanes Piper J3 , PA 28, Jodel D117, DR 220, Cessna 150, C
Total Flight Time
About 500 h (FW + ultra light)
Juergen,
My feeling tells me that the inter-action of the very localized fairing far from the effective areas of the blades is certainly not comparable to the ground effect. The solid angle, seen from the efficient blade portions, is very low, and the oriention from this vision is almost perpendicular to the lift
Sans titre.png
Moreover, how does it explain that the effect of a contrary sliping on the fairing does not give an equal and opposite flow on the disc?
 
Last edited:

kolibri282

Active Member
Joined
Aug 2, 2009
Messages
2,780
Location
Duesseldorf
It turned out that it was not really difficult to show that the assumptions in the report are perfectly valid. In the little octave/Matlab program below the formulae for the velocities in potential flow around a cylinder are implemented. The components in radial (uR = (1 - (R/r)^2)*Uo*cos(thta); and circumferential (uthta=-(1 + (R/r)^2)*Uo*sin(thta)) direction are calculated and then transformed to a body fixed cartesian coordinate system uX (flow parallel to the rotor disk) and uY (flow perpedicular to the rotor disk). Since we are interested in the velocities far away from the cylinder the flow can be assumed inviscid and the solution of the potential flow equations gives a very good approximation to the real flow speeds. The height of the rotor head for a Cavalon is about four times the fuselage radius (r=4*R) and a point on the windward side 60° from the negative x axis was calculated. Here the distance to the fuselage is 4.0/sin(60°) = 4.662*R. For a lateral gust of 25mph = 40km/h = 11.11 m/s the flow velocity at this point through the rotor disk is 0.45 m/s. At a flight speed of 60mph the constant induced velocity through the rotor is 1.18m/s thus the upward flow speed ist almost 40% of the induced velocity, which, of course, *is* significant. This actually is a lower bound since, due to the boundary layer, the distance of the flow cylinder to the rotor head is smaller. Assuming a thickness of the boundary layer of 5% the distance to the point of interest now is 4.4*R and the flow perpendicular to the rotor is 0.5 m/s

potential_flow.png


The important result line looks like this
Uo= 11.11 m/s -> uX= 11.40 uY= 0.45


The octave/Matlab program is this one:
clc
clear all


R = 0.50;
rN= 4.0*R;
psX = 60;

fprintf('\nrN = %3.2f*R\n',rN/R);

thta2= psX*pi/(180);
r = rN/sin(thta2);
fprintf('r = %3.2f*R\n',r/R);

thta = pi-thta2;

Uo=1.0;


uR = (1 - (R/r)^2)*Uo*cos(thta);
uthta=-(1 + (R/r)^2)*Uo*sin(thta);

uX = uR*cos(thta) - uthta*sin(thta);
uY = uR*sin(thta) + uthta*cos(thta);

fprintf('uthta %10.3f\n' , (uthta));
fprintf('uR %10.3f\n' , (uR));
fprintf('uRes %10.3f\n',sqrt(uR^2 + uthta^2))
fprintf('uX %10.3f\n' , (uX));
fprintf('uY %10.3f\n' , (uY));
uRes=sqrt(uX^2 + uY^2);
fprintf('uRes %10.3f\n',uRes)
Uo= 40/(3.6);
fprintf('Uo= %6.2f m/s -> uX= %8.2f uY= %8.2f\n',Uo,uX*Uo,uY*Uo);
fprintf('uRes %10.3f\n',uRes*Uo)
 
Last edited:

Jazzenjohn

Gold Supporter
Joined
Oct 9, 2004
Messages
2,708
Location
Milan Mich.
Aircraft
Gyrobee, My design
Total Flight Time
350
It seems to me to refer to deep side slip flight as opposed to more ordinary yawed flight with relatively stable altitude. There was a rash of accidents in Europe where some fixed wing instructors were teaching deep side slips to reduce altitude just before landing with several notable crashes as a result.
 

kolibri282

Active Member
Joined
Aug 2, 2009
Messages
2,780
Location
Duesseldorf
Quote: yawed flight with relatively stable altitude /Quote
You are right, John, as I had pointed out in post #9 the author consistently uses the term "yawed gliding flight" indicating that the aircraft looses altitude.
 
Last edited:

Jean Claude

Junior Member
Joined
Jan 2, 2009
Messages
1,882
Location
Centre FRANCE
Aircraft
I piloted gliders C800, Bijave, C 310, airplanes Piper J3 , PA 28, Jodel D117, DR 220, Cessna 150, C
Total Flight Time
About 500 h (FW + ultra light)
Juergen,
Here, the flow around a cylinder of infinite length according to JavaFoil.

Sans titre.png

It is obvious that a correction of about 2 degrees on the control plate would be enough to cancel this asymmetry.
In addition, it must be considered that the model cylinder is only 5 ft in length, and the deflection on the disc parts in front and in rear is null. Then it is not difficult to understand that the correction required is still much weaker .
 
Last edited:

kolibri282

Active Member
Joined
Aug 2, 2009
Messages
2,780
Location
Duesseldorf
Quote: ....2 degrees .../Quote
This does not come as a surprise at all, Jean-Claude. Arctan(0.45,11.11) is 2.26° so your Java Foil plot confirms the result I had obtained, thank you for posting it! Please keep in mind that the effect on the rotor is doubled by the fact that we also have the same angle as a down flow on the lee side of the fuselage. The effect is even multiplied since the rotor sweeps across this perturbed flow field all along its entire revolution and so we do agree now, that the findings of the report are correct, don't we? I ask that question because from your first posts I briefly had the impression that you were seriously questioning those findings. Now professionals somtimes do make mistakes. I found glitches in naca reports, Nikolsky's book (I found those ones thanks to the symbolic math capabilties of octave's syms module) and Bramwell is escpecially sloppy in his examples, but before I would assume that someone like Holger Duda, who has spent his whole professional life investigating rotary wing aircraft and especially autogyros, had signed a report that was seriously mistaken I would definitely calculate some really hard and fast figures that prove my point. Fortunately, now that we found that the results of report are correct we can discuss the implications of those results and I am really looking forward to reading the thoughts of the experienced pilots in the forum on what recommendations should be derived from the insight that has been gained.
 
Last edited:
Top