Why have fabric covered control surfaces…

Many thanks for all the responses…..there is a reason but it certainly isn’t a simple one! 🙂

Personally I think the purpose of having fabric covered controls is to keep me busy…:D

Now of course the question is what is being sewn…? 😉

[QUOTE=old shape;1312073][QUOTE=chumpy;1312033]

I’d be interested to know a present design with a dimpled fastener, and I would also like to see the spec. docs for reference. I’m not doubting you, I have never worked on or seen such.

Hi there,
Hmm dimples and current designs, I can but quote the DHC Dash-8. I spent many ‘happy’ years at GKN Westland working on the D-8 engine nacelle programme, horrified to look back and realise it was 1990!! Also at the time similar dimpling of Shorts 360 wing sections. Today I work elsewhere however the Dash-8 contract continues at GKN, as far as I know still done pretty much the same.

As previously stated still much dimpling to be found in the vintage / light aircraft sector, the only way to acheive flush rivets on skins that are too thin for a ‘cut’ countersink.

As for specs etc the various companys had they own techniques / processes. In the case of GKN a large floor standing, pneumatic machine was used. This fitted with electrically heated male / female dies, of various sizes. Pressure, temp, dwell-time varied according to the material being dimpled. ( The D-8 being a combination of light-alloy and Titanium parts).

The skins came with the holes drilled and carefully deburred, usally no probs with cracks. Seem to recall a test piece had to be done for each batch, these closely looked at by the inspection dept.

The machine was operated by foot switch, the part to be dimpled located on the spigot of the male-die. Every now and again I missed the hole, hit the switch..this resulted in the machine punching a hole through the part in question…my foreman was NOT ammused!!

Dimpling can also be acheived by the method mentioned by Pongo..slower but safer!!

Chumpy.

[QUOTE=old shape;1312073][QUOTE=chumpy;1312033]

I’d be interested to know a present design with a dimpled fastener, and I would also like to see the spec. docs for reference. I’m not doubting you, I have never worked on or seen such.

I didnt suggest dimpling was used on a current design and to be honest I cannot remember exactly what type I did use it on, but It was possibly VC10 although it could be Comet, Piston Provost, Hurricane, Hart, Chipmunk, or Viscount. We used to make a tool from a bolt, just drill a hole in the head counter sink it. Polish it up a bit and its dome.

Richard

Something that has not been picked up on the subject is the practicality of manufacture.
Given that with most types of aircraft rivets you have to be able to get a bucking bar to the inside of the stucture to set the rivet, think of the problems that creates with actually building the component and think how will it be prepared.
There are sequences of assembly which will allow this but they can be involved especially on something with an eliptical wing.
Fabric on the other hand is cheapish and does not require the same amount of training to use.
My bit to the thread…..

That problem exists all over the shop, not just in narrow T/e wedges. It is overcome by using Blind Rivets. Basically an airworthy version of a Pop Rivet. Larger versions are, for example, Huck bolts. They can run to dozens of dollars PER HOLE.
I was looking at some bolts the other day which are $1,750 each. The nut was another $250, washers $15. 36 of them required, per side. 72 per A/c.
$150 grand to assemble 2 parts. Important parts though 🙂

Something that has not been picked up on the subject is the practicality of manufacture.
Given that with most types of aircraft rivets you have to be able to get a bucking bar to the inside of the stucture to set the rivet, think of the problems that creates with actually building the component and think how will it be prepared.
There are sequences of assembly which will allow this but they can be involved especially on something with an eliptical wing.
Fabric on the other hand is cheapish and does not require the same amount of training to use.
My bit to the thread…..

Just out of interest…if one scrolls down this thread there is a nice photo of an early type external mass balance weight on an old a/c .

http://forum.keypublishing.co.uk/showthread.php?t=54858

cheers baz

[QUOTE=chumpy;1312033][QUOTE=old shape;1311992]

In 35 years messing with A/c and military equipment I’ve never seen dimpled skins on an A/c. I’ve seen it on vehicles and trains. The dimpling is a metal deforming process which is bound to induce cracks, so I doubt if that practice is used today….but we are talking olde worlde days of Historic A/c. I stand to be corrected.

Dimples not used today!!!….you should get out more.:)

Still a lot of it about, both in old and new aircraft, no does not normally induce cracks.

Cheers Chumpy.

I’d be interested to know a present design with a dimpled fastener, and I would also like to see the spec. docs for reference. I’m not doubting you, I have never worked on or seen such.

Flutter of a badly balanced moveable is actually more like a shake.

I’ll say !!:rolleyes:
Control surface flutter is actually a really serious potential problem,but guess what !! We know how to avoid it !!
The original question was about control surfaces not airbus wingtips.
Flutter of a badly balanced movable (control surface in english ??) potentially has serious consequences,but as i said ,the cure is historically known about ….ergo not a problem nowadays (hopefully !!)

cheers baz

[QUOTE=old shape;1311992][QUOTE=pogno;1311823]

In 35 years messing with A/c and military equipment I’ve never seen dimpled skins on an A/c. I’ve seen it on vehicles and trains. The dimpling is a metal deforming process which is bound to induce cracks, so I doubt if that practice is used today….but we are talking olde worlde days of Historic A/c. I stand to be corrected.

Dimples not used today!!!….you should get out more.:)

Still a lot of it about, both in old and new aircraft, no does not normally induce cracks.

Cheers Chumpy.

[QUOTE=pogno;1311823]

Also, if made of ally, the rivets would have had to be mush-heads, which is not nice to the air on a wing. The ally wouldn’t be thick enough to take the countersink of a flush-head. You could of course chemi-etch the ally down from say 16’s SWG (.063″) to paper thin between the ribs (Leaving the 16’s thickness on the ribs to take the c/sink)

When rivetting very thin skins to very thin ribs the usual process was to ‘Dimple’ the two items, this in effect pressed a rivet head shaped depression into the skin and rib without removing any metal, often it was done cold but hot dimpling was available and was used, especially on magnesium skins.

In my experience flutter nearly always effected control surfaces but it can be a problem on a very poorly designed wing,tailplane or fin.

I remember a Vanguard that the crew were complaining about vibration through the controls, that was badly worn rudder tab bearings allowing the tab to flutter . Even tabs are balanced in their own right.

Richard

In 35 years messing with A/c and military equipment I’ve never seen dimpled skins on an A/c. I’ve seen it on vehicles and trains. The dimpling is a metal deforming process which is bound to induce cracks, so I doubt if that practice is used today….but we are talking olde worlde days of Historic A/c. I stand to be corrected.

I’m mixing me flutters up……or we are divided by a common language :-).
The flutter I speak of (Because I dealt with the ramifications of it on the A320 early flights) was a resonating vibration at cruise. Cause by minute errors in the wing tip design. It was corrected, then corrected further with the “Fence”.
Flutter of a badly balanced moveable is actually more like a shake.
The Erik Hokuf answer is the best here.

Every has hit several points that all apply, but it hasn’t been hit on the head completely.
The reason the control surfaces are balanced is to save overall weight on the aircraft, but it is not by saving a few ounces in the difference between painted fabric and aluminum sheetmetal.
Yes the reason the flight controls were balanced was to prevent flutter. This is done by adding extra mass to the forward side of the hinge point of the surface and making it balanced on the hinge. This prevents the trailing edge of the surface from sagging when the surface is subjected to g-loads(a turn, a bump in the air, etc,) If the trailing edge sags, the surface will now create more lift (becuase of increase in angle of attack) thus creating lift and rising the trailing edge of the surface, as it rises it looses lift, then the G-load takes over and then of course the cyle of flutter starts.
So what does this have to do with saving weight? (these numbers are not accurate, just off the top of the head) If you can save 10 ounces on the trailing edge of a surface, that may equal a reductions of balance mass of 15 lbs on the leading edge of the surface do to the short arm fwd of hinge line. If say this is an elevator and you have 2 of them, that equals 30 lbs, than add the rudder which has twice the length of the elevator meaning you have to add 25 lbs to leading edge to make balance, you now have added 55lbs to the tail of the airplane. Now that you have added 55lbs to the tail of the airplane you may have to compensate at the front of the airplane with more weight to compensate for the added wieght in the tail.
Point is that a few ounces on the trailing edge of the control surface creates many pounds of extra overall weight on the entire aircraft. Later designs either sacrificed the weight or reduced internal weight of the surface so as to not require the high amount of mass to balance. Other ways they reduced flutter was by creating friction in the control system(like holding on to the stick when the airplane oscilates) to prevent the flutter from starting.

From my direct experience that’s why the control’s were made of fabric.
Erik Hokuf

Eric
I accept that any reduction in structural weight behind a control surfaces point of balance will help reduce the total mass required, but I disagree with your discription of how flutter is instigated. If you read the explanation shown here http://www.cs.wright.edu/~jslater/SDTCOutreachWebsite/aerodynamic_flutter_banner.pdf you will see that it is an aerodynamic function that can effect an aircraft regardless of gravity. Speed is a prime factor. As for adding friction to the control circuit, that I think is a complete no no, it might have been tried but is certainly not the answer, ask a pilot.

Richard

Every has hit several points that all apply, but it hasn’t been hit on the head completely.
The reason the control surfaces are balanced is to save overall weight on the aircraft, but it is not by saving a few ounces in the difference between painted fabric and aluminum sheetmetal.
Yes the reason the flight controls were balanced was to prevent flutter. This is done by adding extra mass to the forward side of the hinge point of the surface and making it balanced on the hinge. This prevents the trailing edge of the surface from sagging when the surface is subjected to g-loads(a turn, a bump in the air, etc,) If the trailing edge sags, the surface will now create more lift (becuase of increase in angle of attack) thus creating lift and rising the trailing edge of the surface, as it rises it looses lift, then the G-load takes over and then of course the cyle of flutter starts.
So what does this have to do with saving weight? (these numbers are not accurate, just off the top of the head) If you can save 10 ounces on the trailing edge of a surface, that may equal a reductions of balance mass of 15 lbs on the leading edge of the surface do to the short arm fwd of hinge line. If say this is an elevator and you have 2 of them, that equals 30 lbs, than add the rudder which has twice the length of the elevator meaning you have to add 25 lbs to leading edge to make balance, you now have added 55lbs to the tail of the airplane. Now that you have added 55lbs to the tail of the airplane you may have to compensate at the front of the airplane with more weight to compensate for the added wieght in the tail.
Point is that a few ounces on the trailing edge of the control surface creates many pounds of extra overall weight on the entire aircraft. Later designs either sacrificed the weight or reduced internal weight of the surface so as to not require the high amount of mass to balance. Other ways they reduced flutter was by creating friction in the control system(like holding on to the stick when the airplane oscilates) to prevent the flutter from starting.

From my direct experience that’s why the control’s were made of fabric.
Erik Hokuf

[QUOTE=old shape;1311750] Also, if made of ally, the rivets would have had to be mush-heads, which is not nice to the air on a wing. The ally wouldn’t be thick enough to take the countersink of a flush-head. You could of course chemi-etch the ally down from say 16’s SWG (.063″) to paper thin between the ribs (Leaving the 16’s thickness on the ribs to take the c/sink)

When rivetting very thin skins to very thin ribs the usual process was to ‘Dimple’ the two items, this in effect pressed a rivet head shaped depression into the skin and rib without removing any metal, often it was done cold but hot dimpling was available and was used, especially on magnesium skins.

In my experience flutter nearly always effected control surfaces but it can be a problem on a very poorly designed wing,tailplane or fin.

I remember a Vanguard that the crew were complaining about vibration through the controls, that was badly worn rudder tab bearings allowing the tab to flutter . Even tabs are balanced in their own right.

Richard

I once saw a great example of why control surfaces are fabric covered when I was asked to fly a kit built A/C. All went well until I took my hand of the stick to pick up a map. The aircraft, by it self suddenly started an oscillatory side to side roll, quite violent as well, together with the stick banging from one side to the other. I grabbed the stick and the whole motion stopped, released and it started again. After a precautionary landing I found the ailerons were not to drawing, in that they had been plywood skinned on a foam core. The aileron trialing edge weight was just under twice what it should have been. The owner (and not the person who had constructed the ailerons, these were bought in from another source) removed plywood skinning, foam core and replaced it with fabric over ribs as per the original drawing. The next time I flew it, absolutely no sign of this earlier problem.

Hence the reason for the fabric covering is to get the control surface cg as far froward as possible so as to mass stabilise the surface against aerodynamic induced flutter. Helicopter rotor blades have used a similar principle the past.

I agree with all of that, apart from the bit about it being for flutter. Flutter is actually more likely to happen with a lightweight moveable.
As stated, it’s all about weight-off.
Flutter is not really a problem on a flap / aileron because these structures are hinged near each end and in the middle if there is room. Each hinge will be fixed to a rib inside the moveable, and as you say there will be a counterbalance of some description, forward of the hinge line but not necessarily behind the hinge. The C/balance could be in the fuse, on the end of the hinge rod for example.
Doped fabric was strong enough for this purpose and is considerably lighter than the thinnest SWG of Ally. Also, if made of ally, the rivets would have had to be mush-heads, which is not nice to the air on a wing. The ally wouldn’t be thick enough to take the countersink of a flush-head. You could of course chemi-etch the ally down from say 16’s SWG (.063″) to paper thin between the ribs (Leaving the 16’s thickness on the ribs to take the c/sink) but that is a process which is a pain in the A, is expensive and time consuming, the weight saved is actually not worth the cost if there is a simple alternative. Further, I’m not sure the chemi-etch process was in full use during the War years. I’ll sit corrected on that one. So the ally would have to be machined down, this would be an even bigger pain in the A.
Flutter is usually apparent at wing tips or Fin tips.

Sorry.. got to go with pogno on this one… Mass balance is to prevent Flutter on a control surface,control surface flutter was a huge problem in the 20’s/30’s when a/c design airspeeds were increasing,it could destroy the a/c structure.
The centre of gravity of the control surface has to be forward of the hinge line,and if you repair/paint the control surface then you have to do a balance check before refitting to a/c.

cheers baz

Many flying control surfaces are balanced, ailerons, rudder and elevator which means adding weights foreward of the hinge point. So to reduce the amount of weight foreward of the hinge the trailing structure has to be as light as possible. The whole purpose was to avoid flutter.
Even on fully powered controls the surfaces are balanced, including modern airliners. The number of times paint is added is controlled, after a certain number you have to strip it all off and start again, a point the marketing people wouldnt understand.
Fabric was the lightest covering known for many years, and was very suitable on the light structure it was attached to.
Aerodynamic balance is a portion of the control surface ahead of the balance point, this was used to reduce pilot stick forces.

Richard

I agree with all of that, apart from the bit about it being for flutter. Flutter is actually more likely to happen with a lightweight moveable.
As stated, it’s all about weight-off.
Flutter is not really a problem on a flap / aileron because these structures are hinged near each end and in the middle if there is room. Each hinge will be fixed to a rib inside the moveable, and as you say there will be a counterbalance of some description, forward of the hinge line but not necessarily behind the hinge. The C/balance could be in the fuse, on the end of the hinge rod for example.
Doped fabric was strong enough for this purpose and is considerably lighter than the thinnest SWG of Ally. Also, if made of ally, the rivets would have had to be mush-heads, which is not nice to the air on a wing. The ally wouldn’t be thick enough to take the countersink of a flush-head. You could of course chemi-etch the ally down from say 16’s SWG (.063″) to paper thin between the ribs (Leaving the 16’s thickness on the ribs to take the c/sink) but that is a process which is a pain in the A, is expensive and time consuming, the weight saved is actually not worth the cost if there is a simple alternative. Further, I’m not sure the chemi-etch process was in full use during the War years. I’ll sit corrected on that one. So the ally would have to be machined down, this would be an even bigger pain in the A.
Flutter is usually apparent at wing tips or Fin tips.

Many flying control surfaces are balanced, ailerons, rudder and elevator which means adding weights foreward of the hinge point. So to reduce the amount of weight foreward of the hinge the trailing structure has to be as light as possible. The whole purpose was to avoid flutter.
Even on fully powered controls the surfaces are balanced, including modern airliners. The number of times paint is added is controlled, after a certain number you have to strip it all off and start again, a point the marketing people wouldnt understand.
Fabric was the lightest covering known for many years, and was very suitable on the light structure it was attached to.
Aerodynamic balance is a portion of the control surface ahead of the balance point, this was used to reduce pilot stick forces.

Richard

galdri,

Next, again using the Spit as a measurement, all the way up to the end of the production, elevators and rudders were fabric covered. The Spit, and certainly the early ones were very marginally stable (some might say very unstable) in the longitudinal (how do you spell that) axis due to aft center of gravity. A number of Mk.V broke up in midair. The fix was to add bobweights in the contol run to fool the control forces felt by the pilot, and assist in bringing the problem under control. It worked a treat! No more inflight brake ups. So the reason an aircraft like the Spit has fabric covered elevator and rudder is due to weight saving measures at the tail, to have the center of gravity as far forward as possible. To fit metal surfaces with (possible) mass ballance would have made the instability even worse! I´m open to suggestions

It has been years since I read “Spitfire – The History” and other books on the subject. I remember reading about the bob-weights in the context of the trainer. The second seat moved the CG backwards and the aircraft had a tendency to “tighten” the turns. It became so light in pitch that the pilot(s) pulled more than intended. To counter act this tendency, an adjustable bob-weight was fitted to the control stick in the aft cockpit. It was attached forward of the hinge point. The result was that when G-load builds up, the bob-weight is also affected by the G-load and counter acts the “tightening” by pulling on the stick forwards. As you point out, it also “adds” to the control feel of the pilot(s).

Christer

Ive not mentioned aerodynamic loading, and I’ve never heard of fabric covered control surfaces on higher performance aircraft either.
One misconception seems to be coming up and that as a strong skinning fabric is inferior. Fabric and dope covered surfaces still have a great deal of surface rigidity, and along with the internal structure integrity, forms a strong enough structure.