Uncontrolled Loops Elevator failure

Just thought I would shift this thread. Not being an aerodynamicist, wouldn't a glider with a jammed elevator run out of energy in short order and end up in a spin within a tight loop and a half or two?

I've been mulling this over for years now. We have had 3 similar accidents, Ivens/Egon, Naddler/cadet, Sergio/Jim......all high speed dive, wings bent up into a U shape followed by one or more wing separation!............all in the same basic fuselage (SH 2 place), all with motors. The motors were not in use, so we can eliminate the actual motor, but all fuselages had a large opening in the aft fuselage to allow for the engine. The fuselages were strengthened around the hole, but maybe not enough for a high speed pullout? Let's say these ships got into a high speed dive for whatever reason, maybe as simple as both pilots thought the other guy was flying! Anyway, they realized the problem and pulled hard to get the nose up. Let's say they pulled about 5 G's. What is the aft boom going to do under a 5 G load? It's going to try and bow up a little, but the designed strength prevents any noticeable movement except around the engine hole. What might be going on there? The lower skin is holding in tension, but the upper skin is trying to fail under compression load. Where is it likely to fail? Around the engine hole that buckles slightly inward. This buckling allows the boom to bow upwards, slightly. Now let's look at what this buckling does to the horizontal stabilizer, it digs is slightly and therefore increasing the G load, which increases the bowing in of the engine hole............more G's, more buckling.......you get the picture! I could see this resulting in a 2 second loop
Something to think about,
JJ

On Friday, September 28, 2018 at 10:42:00 AM UTC-5, wrote:
I've been mulling this over for years now. We have had 3 similar accidents, Ivens/Egon, Naddler/cadet, Sergio/Jim......all high speed dive, wings bent up into a U shape followed by one or more wing separation!............all in the same basic fuselage (SH 2 place), all with motors. The motors were not in use, so we can eliminate the actual motor, but all fuselages had a large opening in the aft fuselage to allow for the engine. The fuselages were strengthened around the hole, but maybe not enough for a high speed pullout? Let's say these ships got into a high speed dive for whatever reason, maybe as simple as both pilots thought the other guy was flying! Anyway, they realized the problem and pulled hard to get the nose up. Let's say they pulled about 5 G's. What is the aft boom going to do under a 5 G load? It's going to try and bow up a little, but the designed strength prevents any noticeable movement except around the engine hole. What might be going on there? The lower skin is holding in tension, but the upper skin is trying to fail under compression load. Where is it likely to fail? Around the engine hole that buckles slightly inward. This buckling allows the boom to bow upwards, slightly. Now let's look at what this buckling does to the horizontal stabilizer, it digs is slightly and therefore increasing the G load, which increases the bowing in of the engine hole............more G's, more buckling.......you get the picture! I could see this resulting in a 2 second loop
Something to think about,
JJ

Fuselage will bend down both in front of and behind the wing during high g loading. Tail pushes down to make the nose go up. Top of fuselage should be in tension during high positive g maneuvers. So, I don't think fuselage flexure is a contributor. Could be wrong, but I think that is the wrong tree to bark up.

Steve Leonard

On Friday, September 28, 2018 at 8:42:00 AM UTC-7, wrote:
I've been mulling this over for years now...

Very interesting, good catch, JJ! I think you're off base about the direction of fuselage bending, but you might be on the right track in general.

* Up elevator tends to bend the tailboom downward.

* Bending the tailboom downward applies compression to the bottom half and tension to the top half.

* In many gliders, the elevator push-pull tube presses aft for up elevator, and pulls forward for down elevator. (Known exceptions are ASW24-29, and the later single-seat Schleicher cockpits based on them)

* In some gliders, the elevator push-pull tube runs along the bottom half of the tailboom

Given a glider with a limber tailboom, the elevator push-pull tube running along the bottom of the tailboom, and the push-pull tube pressing aft for up elevator, the system described will have some measure of positive feedback.

As the pilot applies up elevator, the increased downward force at the tail bends the tailboom downward. The bending compresses and shortens the lower half of the tailboom structure. The shortening effect reduces the distance between where the elevator push-pull tube starts and where it meets the bellcrank at the bottom of the fin. Since the push-pull tube does not shorten as well, it tends to apply additional up elevator force at the aft bellcrank and consequently the elevator.

The big question is whether the effect is pronounced enough to become divergent at any point within the glider's operational flight envelope. My suspicion is that it's not. In most gliders, the tailboom is so stiff, and the elevator push-pull tube so close to the tailboom neutral axis, that the effect will be so small as to be barely noticeable. However, if perhaps the tailboom is more limber than normal (maybe because of an engine cutout), and the elevator push-pull tube is lower than normal (maybe relocated downward to accommodate an engine installation), the effect might be significant.

However, as B-47 aficionados will recognize, there is one important additional factor: As the tailboom bends downward, the relative incidence between the wing and horizontal stabilizer decreases, reducing the downward force applied by the horizontal stabilizer. This effect will tend to negate the elevator input effect described above, and might in fact completely overpower it.

Overall, this would be a great topic for one of our friends at Akaflieg Cal Poly San Luis Obispo. Some simple FEA on typical glider shapes and structures should be enough to indicate whether further study is warranted.

--Bob K.

On Friday, September 28, 2018 at 2:26:57 PM UTC-4, Bob Kuykendall wrote:
On Friday, September 28, 2018 at 8:42:00 AM UTC-7, wrote:
I've been mulling this over for years now...

Very interesting, good catch, JJ! I think you're off base about the direction of fuselage bending, but you might be on the right track in general.

* Up elevator tends to bend the tailboom downward.

* Bending the tailboom downward applies compression to the bottom half and tension to the top half.

* In many gliders, the elevator push-pull tube presses aft for up elevator, and pulls forward for down elevator. (Known exceptions are ASW24-29, and the later single-seat Schleicher cockpits based on them)

* In some gliders, the elevator push-pull tube runs along the bottom half of the tailboom

Given a glider with a limber tailboom, the elevator push-pull tube running along the bottom of the tailboom, and the push-pull tube pressing aft for up elevator, the system described will have some measure of positive feedback.

As the pilot applies up elevator, the increased downward force at the tail bends the tailboom downward. The bending compresses and shortens the lower half of the tailboom structure. The shortening effect reduces the distance between where the elevator push-pull tube starts and where it meets the bellcrank at the bottom of the fin. Since the push-pull tube does not shorten as well, it tends to apply additional up elevator force at the aft bellcrank and consequently the elevator.

The big question is whether the effect is pronounced enough to become divergent at any point within the glider's operational flight envelope. My suspicion is that it's not. In most gliders, the tailboom is so stiff, and the elevator push-pull tube so close to the tailboom neutral axis, that the effect will be so small as to be barely noticeable. However, if perhaps the tailboom is more limber than normal (maybe because of an engine cutout), and the elevator push-pull tube is lower than normal (maybe relocated downward to accommodate an engine installation), the effect might be significant.

However, as B-47 aficionados will recognize, there is one important additional factor: As the tailboom bends downward, the relative incidence between the wing and horizontal stabilizer decreases, reducing the downward force applied by the horizontal stabilizer. This effect will tend to negate the elevator input effect described above, and might in fact completely overpower it.

Overall, this would be a great topic for one of our friends at Akaflieg Cal Poly San Luis Obispo. Some simple FEA on typical glider shapes and structures should be enough to indicate whether further study is warranted.

--Bob K.

Interesting thoughts, Bob! One point in JJ's post that caught my eye was the fact that all these accidents happened in SH gliders with a motor. I have to claim ignorance about the way the motor is held in the stowed position but is there a possibility that the engine assembly pushes down onto the elevator control rod when subjected to high g-loads?
Uli
'AS'

On Friday, September 28, 2018 at 12:04:10 PM UTC-7, AS wrote:

...is there a possibility that the engine assembly pushes down onto the elevator control rod when subjected to high g-loads?

I am curious about that as well. My worry is that there might be some subtle and unanticipated failure mode that brings the engine support system into conflict with the elevator system. But I am not familiar enough with the gliders of concern to know. Clearly the factory considers them safe.

--Bob K.

Very unlikely. The elevator pushrod is mounted to the left side of the engine box, not underneath.

Le vendredi 28 septembre 2018 21:04:10 UTC+2, AS a écritÂ*:
On Friday, September 28, 2018 at 2:26:57 PM UTC-4, Bob Kuykendall wrote:
On Friday, September 28, 2018 at 8:42:00 AM UTC-7, wrote:
I've been mulling this over for years now...

Very interesting, good catch, JJ! I think you're off base about the direction of fuselage bending, but you might be on the right track in general.

* Up elevator tends to bend the tailboom downward.

* Bending the tailboom downward applies compression to the bottom half and tension to the top half.

* In many gliders, the elevator push-pull tube presses aft for up elevator, and pulls forward for down elevator. (Known exceptions are ASW24-29, and the later single-seat Schleicher cockpits based on them)

* In some gliders, the elevator push-pull tube runs along the bottom half of the tailboom

Given a glider with a limber tailboom, the elevator push-pull tube running along the bottom of the tailboom, and the push-pull tube pressing aft for up elevator, the system described will have some measure of positive feedback.

As the pilot applies up elevator, the increased downward force at the tail bends the tailboom downward. The bending compresses and shortens the lower half of the tailboom structure. The shortening effect reduces the distance between where the elevator push-pull tube starts and where it meets the bellcrank at the bottom of the fin. Since the push-pull tube does not shorten as well, it tends to apply additional up elevator force at the aft bellcrank and consequently the elevator.

The big question is whether the effect is pronounced enough to become divergent at any point within the glider's operational flight envelope. My suspicion is that it's not. In most gliders, the tailboom is so stiff, and the elevator push-pull tube so close to the tailboom neutral axis, that the effect will be so small as to be barely noticeable. However, if perhaps the tailboom is more limber than normal (maybe because of an engine cutout), and the elevator push-pull tube is lower than normal (maybe relocated downward to accommodate an engine installation), the effect might be significant.

However, as B-47 aficionados will recognize, there is one important additional factor: As the tailboom bends downward, the relative incidence between the wing and horizontal stabilizer decreases, reducing the downward force applied by the horizontal stabilizer. This effect will tend to negate the elevator input effect described above, and might in fact completely overpower it.

Overall, this would be a great topic for one of our friends at Akaflieg Cal Poly San Luis Obispo. Some simple FEA on typical glider shapes and structures should be enough to indicate whether further study is warranted.

--Bob K.

Interesting thoughts, Bob! One point in JJ's post that caught my eye was the fact that all these accidents happened in SH gliders with a motor. I have to claim ignorance about the way the motor is held in the stowed position but is there a possibility that the engine assembly pushes down onto the elevator control rod when subjected to high g-loads?
Uli
'AS'

No. The engine bay's lower enclosure is the shell of the fuselage.

Interesting thoughts, Bob! One point in JJ's post that caught my eye was the fact that all these accidents happened in SH gliders with a motor. I have to claim ignorance about the way the motor is held in the stowed position but is there a possibility that the engine assembly pushes down onto the elevator control rod when subjected to high g-loads?
Uli
'AS'

No. The engine bay's lower enclosure is the shell of the fuselage.

So how are the control rods/cables routed past the engine bay? On either side of the engine bay? Between the outside shell and the engine compartment?
As I said, I am not familiar with these designs.

Uli
'AS'

On Friday, September 28, 2018 at 8:42:00 AM UTC-7, wrote:
I've been mulling this over for years now. We have had 3 similar accidents, Ivens/Egon, Naddler/cadet, Sergio/Jim......all high speed dive, wings bent up into a U shape followed by one or more wing separation!............all in the same basic fuselage (SH 2 place), all with motors.

It's a bad thing anytime the tail feathers aren't pointing the way you like.. These accidents all ended the same way with the wings coming off from very high G-loads, but the evidence (such as it is) would seem to indicate mostly different starting points.

The Engen/Ivans Nimbus 4DM accident did not seem to involve any control circuit issue but, in the NTSB's judgement and according to witnesses, resulted from excessive Gs associated with recovery from a departure (spin or spiral dive) that started while thermalling.

The Nadler/cadet Arcus M incident apparently started with an uncommanded yaw (I believe flying straight) followed by an unrecoverable spiral dive. Maybe they will find the fuselage and get more information, but that sounds like a different root cause.

We may never know what happened with the Colacevich/Alto DuoDiscus accident, but a series of very high-G loops sounds more like a pitch control problem than a yaw control problem or a departure, though the chain of events up until these final maneuvers is unknown.

Sadly, there are a lot of different ways to end up with the wings coming off. It's important to learn from these tragedies, but for the moment I'd be hard-pressed to go down the path of some common root cause or a single design or construction defect with S-H two-seat gliders. These feel more like isolated incidents to me.

Andy Blackburn
9B