Mysterious Gyroplane Accidents.

Vance

Gyroplane CFI
Staff member
Joined
Oct 30, 2003
Messages
18,363
Location
Santa Maria, California
Aircraft
Givens Predator
Total Flight Time
2600+ in rotorcraft
Ferran posted on a thread with what I find to be a particularly annoying title “Engine out causing fatality?”

I am not familiar with many of the European gyroplanes accidents.

I feel it is important to understand these accidents.

I have observations from a distance.

Gentlemen, I´ve been thinking a lot in the last months about gyrocopter accidents. I think that Europe´s gyrocopter accident record is really bad. And this is common to all countries with modern stable gyros. Anyway this is only my subjective opinion and I may be wrong...

From my point of view, if gyros are safer than planes, our fatal accident record should be much better... and this is not the case. I think (my subjective opinion again) that UK has the best safety record. And the main difference with other countries is the training requirements. In spite of this, their record is not good enough, it should be much better than fixed wing planes...

Have you seen the last accident reported in the forum? Have you seen that the blades are bended UPWARDS? The same that the last MT accident. The same that the Mallorca accident.

The main question is Why?

If the blades are bended upwards is because the lift force acting on the blade has done it. I think that the only possibility is a strong decay in rotor rpm… or maybe not so strong.

One of my friends who is an engineer has calculated that a drop of 10% in rotor rpm reduces 21% the centrifugal force acting on the blade. So a decay of only 30% in rotor rpm would produce a reduction of 63% of the acting centrifugal force…

Our teetering rotors cannot afford this lack of centrifugal force, and the result is a flexion force that bends the rotor blades upwards. I believe that this is the substitute of PPO on modern gyros…

This can explain a lot of badly explained accidents that has happened lastly. My main concern is that we don´t know nothing about how long can be unloaded a rotor before that rotor rpm decay will be lethal. I think that this is a very important issue that has not been analyzed in our certification rules (BCAR T).

I guess that time elapsed before that the rotor rpm has decayed too much is closed to 1 second, and less than two seconds… Of course I can be wrong, but this is something we should study in a wind tunnel (real flight is too dangerous)

The teetering rotor cannot work against flexion. Its individual blade flapping freedom is 0.
An average C-30 had a flapping freedom from -10 to +40º. This is a big difference. Most probably this is the difference between a safe gyro and our wonderful modern stable ones…

Did you know that the average rotor speed of a C-30 was about 200 rpm? I don´t know of a single C-30 accident because of unloaded rotors.

I know that we can fly a gyro with a teetering rotor for ages without any problem. But if you unload significantly the rotor you will be dead. And I believe that the gyro community is not aware of that.

What do you think?

Ferràn


Hello Ferran,

I feel you are too focused on low Gs.

In my opinion the low G event is usually very short in duration and it would be very hard to find a way to get to negative Gs.

On the other hand it is not that hard to stall a blade and they will slow down very fast if that happens.

I feel an unintended control input from an out of control fuselage is also a possible way to stall a rotor blade.

In the absence of lift it is not hard to imagine a descent rate that is enough to fold up the blades if they are missing the centripetal force of rotor RPM.

I feel that a stalled rotor blade can easily become a non recoverable event.

I don’t have enough data points to know which gyroplanes are falling out of the sky.

It is my impression that the Magni Gyroplanes are not falling out of the sky as much.

I hope if this is an incorrect impression someone will correct me. There are not enough Magni gyroplanes flying in the USA to get a statistical read on it.

It this is a correct impression it appears to me that the people who copied the Magni made a mistake that we should try to identify.

These copied gyroplanes seem to have a similar rotor system and divergent from Magni’s rotor system.

When I flew two different models from AutoGyro they both had a much more responsive rotor than any of the Magnis I have flown.

I found the Calidus needing rudder input with power changes and to a lesser degree the MTO Sport.

I did not notice this as much with the Magni.

The AutoGyro instructor I flew with seemed quite taken with slips as is another AutoGyro CFI that I have watched fly at several events. I am not a gyroplane slip enthusiast.

It is reasonable to expect that with the rapidly expanding market that putting the training infrastructure in place will take some time and mistakes will be made.

In my opinion any time there a lot of low time pilots flying around the accident rate will go up.

I feel that as more data points become available it will be easier to pinpoint the gyroplane models that have the most unexplained accidents. Perhaps the cause will immerge.

I cannot imagine what to do without sufficient data to even validate a hypothesis.

I expect as the financial stakes escalate so will the efforts to find and fix gyroplane challenges.

Thank you, Vance
 
Vance, thank you to put this question in the right place. It is true; I’m very focused on low g events. But this is because I’m seeing several accidents that show the rotor blades bended upwards.

In your opinion this can be the result of stalling the rotor blades… I don’t know about this possibility. What I know positively is that unloading the rotor at or closed at 0g will imply a very significant rotor rpm drop. Furthermore, in all these accidents the gyro has ended pointing deeply down…

Besides, I think that stalling a rotor blade is really very hard. Autorotating rotors are working with AOA’s much lower than helicopter ones. And I don’t know about helicopter stalled rotors… The closest situation is ring vortex state, and the cure is to enter an autorotation.

The very low G event can be very short… or not. In my experience, very low G events due to turbulence are very short. But, Can we say the same thing about pilot commanded low G situations?

You are right about Magni gyrocopters: we don’t know nearly anything about their accidents. We only have information about accidents occurred in air shows. Anyway their rotors are heavier, so probably they will have more inertia than others. Magni’s have heavier controls than others gyros, too, and this can prevent them to be unloaded because of overcontrol.

However, we have still the same problem; we don’t know how long they can be in an unloaded situation before it becomes irreversible. And this is something we should know about all our rotors.

You are right, Vance, we have lots of low experience pilots now. These are the ones that are suffering most of the accidents. However in UK they have been able to reverse the fatal accidents rate in 6 years’ time. The recipe has been putting the gyros as PPL and monitoring closely training, licenses, manufacturers, incidents and accidents. Probably we should learn something from them.

Anyway, the main question here is that the teetering rotors have a weakness that has not been addressed in airworthiness regulations.

Ferran
 
What fundamental weakness in the teetering rotor system did they solve in the UK via monitoring of training, licensing, incidents and accidents? I can see training deficiencies solved by that. I can't see fundamental weaknesses in the teetering rotor system solved by that.
When I look at these accident reports http://gyroaccidents.blogspot.com/2012/10/gyrocopter-accidents-2012.html

The vast majority are wire strikes and flying over terrain that is dangerous to land i.e. water/trees. There are a number of landing accidents and several over rotating leading to torque roll crashes. There are still bunts even though the reasons and the solution for that is widely known. There appears to be several aggressive side slip crashes. The number of unexplained accidents doesn't seem over represented.

What exactly are the specific accidents that have led you to your conclusion about teetering rotors?
 
0g does not stop the rotor blades that fast, does it?

When you land a gyro, the "g-load" on the rotor goes from 1g to 0g. You can easilly see what rate the RRPM slows down.

For an MTO, normal RRPM is 320 or more.
Lets say that 260 RPM is safe to carry 1G without stalling.

I't my impression that I have approx 10 seconds to speak with my students after full stop before RPM has dropped below 200 RPM, so I believe I have much more than 2 seconds before RPM has dropped below 260.

One quite interesting test we did was "what does slow down the rotor". Quickly changing the stick from one side to the other (back and forth) slows down the rotor rapidly. (We tested on ground with a stop-clock and a rotor prerotated to 200 RPM).

Maybe doing this in the air could slow down the rotor to a stall?
 
Just a naive question.

Could a rapid pull-up out of a steep dive cause blades to fold up?

The pull-up will increase the blade loading above 1G.
In addition, the resulting increase in the blades angle of attack will increase the induced drag, which will start reducing the RRPM.


NoIdea.gif


Dave
 
Just a naive question.

Could a rapid pull-up out of a steep dive cause blades to fold up?

The pull-up will increase the blade loading above 1G.
In addition, the resulting increase in the blades angle of attack will increase the induced drag, which will start reducing the RRPM.


NoIdea.gif


Dave

You would have to know what the load of the Rotor was. Meaning check with the manufacturer of the blade to know what the max load is.
In the case of a well designed kit they should have a VA or design maneuvering speed. If you exceed this speed you can have failure with full deflection of controle.
So If you are below this speed an abrupt deflection should be no problem as long as you do not do any of the normal no no's.

Negative G, Power push over, And all the rest.

Typically I try to teach not to dive in a Gyro at high speed. It is better to slow down to loose altitude then dive. Now in a engine out vertical decent to landing you need to dive before flair to pick up some speed to land safe.
 
Even the more aerobatic helicopters are limited to about 3.5 g's by blade stall, most will be a lot less.
 
The only bent hub bar accident I've heard about in the U.S was with an RAF gyro that had a 2.5L Subaru flown with a lightweight pilot by himself hanging on the prop. Any powerful 2 seat gyro flown at a very light gross weight has the potential to have the blades so lightly loaded they could slow down to a dangerously low speed.

I've read about cracked hub bars down under, but I think those incidents need to have a close examination of the gyros, the disc loads, the type of flying, etc. etc. to get a better read on the material failures.
 
The only bent hub bar accident I've heard about in the U.S was with an RAF gyro that had a 2.5L Subaru flown with a lightweight pilot by himself hanging on the prop. Any powerful 2 seat gyro flown at a very light gross weight has the potential to have the blades so lightly loaded they could slow down to a dangerously low speed.

I've read about cracked hub bars down under, but I think those incidents need to have a close examination of the gyros, the disc loads, the type of flying, etc. etc. to get a better read on the material failures.

I have actually seen 2 cracked hub bars. One was on the forum. The other was the same brand being used as a temp training set at Larry's with Butterfly
 
0g does not stop the rotor blades that fast, does it?

When you land a gyro, the "g-load" on the rotor goes from 1g to 0g. You can easilly see what rate the RRPM slows down.

For an MTO, normal RRPM is 320 or more.
Lets say that 260 RPM is safe to carry 1G without stalling.

I't my impression that I have approx 10 seconds to speak with my students after full stop before RPM has dropped below 200 RPM, so I believe I have much more than 2 seconds before RPM has dropped below 260.

One quite interesting test we did was "what does slow down the rotor". Quickly changing the stick from one side to the other (back and forth) slows down the rotor rapidly. (We tested on ground with a stop-clock and a rotor prerotated to 200 RPM).

Maybe doing this in the air could slow down the rotor to a stall?

I think you gave you the answer yourself. Rotating disc in calm air mass shouldn't stop so easy but when travelling at "normal" airspeed, the dissymmetry of lift causes blades to move in another manner. I see it as energy consuming, causing RRPM to decrease more rapidly, at the similar way as you could provoke in your experiment.
Talking about Zero G risks - here's how I feel: Imagine another experiment: Pull 1,3 G and make some stick movements sideways. Can you feel how quickly your gyro answers? Try to make a manouver in 0,7 G and do the same stick movement. Mushy in comparison? Think about it people. Less G - less stick authority. The question is if it's going be enough to pull back once you need it when you for some reason reach uncomfortably low Gees. So just don't go there. And don't fly in heavy weather.
 
I'm not ready to use bent blades or nose-down returns to earth as proof of some low-G event.

Blades bend in surprising ways in mere ground strikes. Often they will bend up. If they do, sometimes they fold sharply at mid-span, and other times they adopt a gradual curve like a scimitar. Still other times, they bend back or down -- or one of each.

Any gyro with a H-stab will travel nose-down in a freefall.

Ferran is correct, though, that we don't know where the "edge" is for any given blade design. On the one hand, we have pilots falling out of the sky for no apparent reason. On the other, we have Jim Vanek's intentional zero G demonstrations, and a few others who report flying through intentional zero G. Then there is the famous 1930's incident in which a Pitcairn gyro was pulled over onto its back by drag from floats, but landed normally. We have all flown into zero and maybe neg G caused by turbulence -- you can't avoid it.

I would suggest that a few manufacturers get together and build a "poor man's wing tunnel."

This testing device can be a truck with a heavy overhead boom and tower arrangement. Rotors could be mounted at the upper/forward end of the boom on a controllable rotor head and flown in the relatively undisturbed air above and ahead of the truck. The truck cab and windshield would need to be reinforced against bent and broken blades' penetrating the crew area.
 
Besides, I think that stalling a rotor blade is really very hard.
Not true.

Maybe doing this in the air could slow down the rotor to a stall?
No chance.

Could a rapid pull-up out of a steep dive cause blades to fold up?
If the rrpm was alowed to drop too much, yes.

Any powerful 2 seat gyro flown at a very light gross weight has the potential to have the blades so lightly loaded they could slow down to a dangerously low speed.
Low disc loading will only slow the rate of responce.
Within reason, you cant have a 'too low' disc loading in regard to rpm and centriphical force. Centriphical tension will increase/decrease proportinatly with rpm, so as long as you dont yank the stick like your yankn your d1c, but give the blades time to keep up, theres no hazard.

I've read about cracked hub bars down under,............
Im pretty confident that most, if not all these cracks were caused by miss handleing on the ground, misshandleing in the air or aggressive prespin start up pulse, or all three.
Most people wouldnt hava clue that you gota give heavy blades more time.
 
The Magic of Rotors.

The Magic of Rotors.

Vance, thank you to put this question in the right place. It is true; I’m very focused on low g events. But this is because I’m seeing several accidents that show the rotor blades bended upwards.

In your opinion this can be the result of stalling the rotor blades… I don’t know about this possibility. What I know positively is that unloading the rotor at or closed at 0g will imply a very significant rotor rpm drop. Furthermore, in all these accidents the gyro has ended pointing deeply down…

Besides, I think that stalling a rotor blade is really very hard. Autorotating rotors are working with AOA’s much lower than helicopter ones. And I don’t know about helicopter stalled rotors… The closest situation is ring vortex state, and the cure is to enter an autorotation.

The very low G event can be very short… or not. In my experience, very low G events due to turbulence are very short. But, Can we say the same thing about pilot commanded low G situations?

You are right about Magni gyrocopters: we don’t know nearly anything about their accidents. We only have information about accidents occurred in air shows. Anyway their rotors are heavier, so probably they will have more inertia than others. Magni’s have heavier controls than others gyros, too, and this can prevent them to be unloaded because of overcontrol.

However, we have still the same problem; we don’t know how long they can be in an unloaded situation before it becomes irreversible. And this is something we should know about all our rotors.

You are right, Vance, we have lots of low experience pilots now. These are the ones that are suffering most of the accidents. However in UK they have been able to reverse the fatal accidents rate in 6 years’ time. The recipe has been putting the gyros as PPL and monitoring closely training, licenses, manufacturers, incidents and accidents. Probably we should learn something from them.

Anyway, the main question here is that the teetering rotors have a weakness that has not been addressed in airworthiness regulations.

Ferran

The point I was trying to make Ferran was that if the two blade teeter rotor is the problem than some two blade teeter rotors are less problematic than others.

This is of course assuming I am correct in my assumption that Magni gyroplanes are falling out of the sky less than AutoGyro gyroplanes or the other Magni copies.

I feel I perform some pretty extreme maneuvers in The Predator and the rotor rpm is not inclined to drop below 270 rotor rpm.

I feel if I am too aggressive with the cyclic it is possible to stall the blades even at 270 rotor rpm. I feel this would cause a rapid reduction in rotor rpm and quickly become a non recoverable event.

I feel that a tail strike is preceded by a blade stall.

Because a gyroplane depends on the airframe to command the rotor in my opinion it is possible to stall the blades with an uncomanded roll or pitch.

It seems sort of magic to me that the rotor responds in so many positive ways. The rotor makes rpm before it makes lift. The rotor rpm gives blades that can barely support themselves the strength to support 1,400 pounds. I can ask this huge spinning disk to change its orientation and with very little effort it will change quickly. This of course requires energy but there seems to me to be a lot of energy stored in the rotor that we can use for all kinds of things.

Thank you, Vance
 
I am Confused by Your Statements Desmon.

I am Confused by Your Statements Desmon.

You would have to know what the load of the Rotor was. Meaning check with the manufacturer of the blade to know what the max load is.
In the case of a well designed kit they should have a VA or design maneuvering speed. If you exceed this speed you can have failure with full deflection of controle.
So If you are below this speed an abrupt deflection should be no problem as long as you do not do any of the normal no no's.

Negative G, Power push over, And all the rest.

Typically I try to teach not to dive in a Gyro at high speed. It is better to slow down to loose altitude then dive. Now in a engine out vertical decent to landing you need to dive before flair to pick up some speed to land safe.

You have touched on a subject that I am confused about and you have added to my confusion Desmon.

I am starting with the premise that you are correct and then trying to understand my experiences that seem divergent.

It appears to me that with The Predator when I add load the blades spin faster and carry more load.

In the pictures I have seen of The Predator in aggressive maneuvers the conning angle seems fairly consistent as the load and rotor rpm increases.

It appears to me that as the speed increases the rotor’s ability to manage the loads increases. I understand maneuvering speed in a fixed wing where the load carrying capability is fixed but I don’t see how this applies to a gyroplane where the rotor gets its strength from rotor rpms and the rotor rpms increase with speed.

What is the maneuvering speed of the MTO and at what load?

Why do you typically teach not to dive at high speed in a gyroplane?

It feels to me that the Predator in more stable and controllable at 90 kts indicated air speed than she feels at zero kts indicated air speed. I find in gusting conditions particularly with wind shear that The Predator is easier to keep aligned with the runway with more rather than less indicated air speed. Her response to power changes seems more benign with more indicated air speed rather than less.

What am I missing?

Thank you, Vance
 
Last edited:
Divergance of my Experience.

Divergance of my Experience.

I think you gave you the answer yourself. Rotating disc in calm air mass shouldn't stop so easy but when travelling at "normal" airspeed, the dissymmetry of lift causes blades to move in another manner. I see it as energy consuming, causing RRPM to decrease more rapidly, at the similar way as you could provoke in your experiment.
Talking about Zero G risks - here's how I feel: Imagine another experiment: Pull 1,3 G and make some stick movements sideways. Can you feel how quickly your gyro answers? Try to make a manouver in 0,7 G and do the same stick movement. Mushy in comparison? Think about it people. Less G - less stick authority. The question is if it's going be enough to pull back once you need it when you for some reason reach uncomfortably low Gees. So just don't go there. And don't fly in heavy weather.

I am puzzled by the divergence of my experience with some of your thoughts.

What sort of gyroplane are you flying?

How many hours to you have in her?

How do you typicaly fly her?

In a landing flair it is not unusual for me to see the rotor speed in The Predator go from 330 rotor rpm to 270 Rotor RPM in less than two seconds so I do not understand where your ten seconds comes from.

She has 30 foot eight and a half inch chord Sport Copter aluminum blades and solo typically weighs around 1,100 pounds. She typically flies at 310 rotor rpm.

I agree that it takes energy to tilt the rotor but I have not witnessed a significant reduction in rotor rpm during aggressive maneuvers in The Predator.

When The Predator is at low indicated air speed I don’t do aggressive maneuvers because her response is vague.

I have not found a way to sustain low Gs in The Predator so I am not able to imagine your experiment.

Thank you, Vance
 
The rotor makes rpm before it makes lift.
Not quite Vance.
Wen you pull back, the disc tilts back first, and as soon as this happens, the AOA increases, so lift increases, THEN the rrpm increases.
Thats where the delay in responce in an autorotating disc comes from.
And its this sequence that means you MUST allow the blades time to keep up with cyclic commands, or your rrpm will get left behind, which means your centriphical tension is low.
Cone stress.
The higher the blades inertia, the longer the time lag.
 
Thank you for the clarification David.

Thank you for the clarification David.

I agree that it is an important distinction.

What is your opinion on load vs. coning?

It appears to me that more load creates more rotor rpm and the disk cones about the same.

Thank you, Vance
 
Last edited:
I am puzzled by the divergence of my experience with some of your thoughts.

What sort of gyroplane are you flying?

How many hours to you have in her?

How do you typicaly fly her?

In a landing flair it is not unusual for me to see the rotor speed in The Predator go from 330 rotor rpm to 270 Rotor RPM in less than two seconds so I do not understand where your ten seconds comes from.

She has 30 foot eight and a half inch chord Sport Copter aluminum blades and solo typically weighs around 1,100 pounds. She typically flies at 310 rotor rpm.

I agree that it takes energy to tilt the rotor but I have not witnessed a significant reduction in rotor rpm during aggressive maneuvers in The Predator.

When The Predator is at low indicated air speed I don’t do aggressive maneuvers because her response is vague.

I have not found a way to sustain low Gs in The Predator so I am not able to imagine your experiment.

Thank you, Vance

Vance,
I'm not analytic the same way you are. 90% of my airtime is without instruments. My experience is limited to only a few different types of rotors but I can easily feel differences and what I feel in most cases confirms the way I see the whole picture of rotor behaviour. I'm aware about differences in construction and behaviour and I have enough experience not to push the limits.

I believe that your figures are correct in your case. I didn't mention any figures in my comment above but I feel that Andreas is on the right course.

About stick in low G: If you lower the nose carefully on the top of the climb following a kind of ballistic curve but not allowing the gyro to go under 0,7 G, there you can have a few seconds to test rotor's response on your stick movements.
You can make your own conclusions. I made mine. Disclaimer: I don't recommend anyone to do it. When I say imagine the situation, i don't mean it necessarily has to be demonstrated. I demonstrate it sometimes when I feel that students have difficulties to understand basic principles in theory. Even if students sometimes have difficulties to feel the difference in manouvers as described in my answer to Andreas, I feel that comments and conclusions we make during such lessons make good points.

There are people on this forum that could put this in words much better than I do.

Roman
 
These are the three accidents I’m referring:

http://www.rotaryforum.com/forum/showthread.php?t=36839

The only information is the image, but it seems that the two blades are bended upwards. The blade in first position is bended from the end of the rotor hub. Because of that I think that it has been bended in flight.

German MT
The image below is self explaining. No doubt in this case, the bending has happened in flight: both blades from the end of the hub. So the crash cause was unloading the rotor, decay of rotor rpm and the reload bended the rotor. I think this bending happened in the very last seconds. If this has happened before one blade would be broken.

http://www.rotaryforum.com/forum/showthread.php?t=31503

fhMX8n7.png



Andy’s accident in Mallorca
The manufacturer, who saw the remains, explained that the broken blade was bended upwards. He stated that this happened when it hitt the ground, and explains that because the teeter stop in the other side was broken. I guess that the other blade was bended upwards too…
http://www.rotaryforum.com/forum/showthread.php?t=32113&page=2

Probably, if we coul review more unexplained accidents we coul find this upwards beded blades...

About the rotor rpm decay regime in unloaded situations, the 10 seconds reported are until the total rotor stoppage. The rotor rpm decay much quicker at the beginning that at the end… And we don’t know what the safe boundary is in rotor rpm… 260 rpm seems to me a dangerous low regime for a teetering rotor.

Ferran
 
I agree that it is an important distinction.

What is your opinion on load vs. coning?

It appears to me that more load creates more rotor rpm and the disk cones about the same.

Thank you, Vance

Hi Vance, although your question is directed at David, allow me to chime in. I have done a numerical study on rotor blade bending followed by experiments on a real rotor instrumented with numerous strain gauges.

Your suspicion that coning angle is independent of load is correct in the case where the rotor has had enough time to find its equilibrium rpm. More load results in higher rpm which will generate more centrifugal force so that in the end the ratio of lift and centrifugal force, which determines the coning angle, will be the same.

In the transient time until the rotor has achieved its equilibrium rpm, however, coning will be different. The worst case scenario is one in which you reduce rrpm (i.e., by partial unloading of the rotor) followed by a sharp turn, which increases the load in a very short time. Until the rrpm catch up, the coning angle will be markedly larger.

Greetings, -- Chris.
 
Top