DuoDiscus Wing

First of all, it is not the epoxy which bears the load but the cotton
threads. Epoxy is just the matrix, and the rated load of cotton/epoxy is
around 7N/mm.

Microballons are much heavier and the composite formed of balloons/epoxy
doesn't hold the same load as cotton threads/balloons (balloons are spheres
so in this case it's the epoxy which finally bears the load). Microballoons
are rather used for surface cosmetics.

Seperated glass rovings would be ideal - but the beasts don't bend easily
into small radii if threaded and a paste made up of them would leave many
mm-sized voids. so it just doesn't work.

I think that people proposing utrasonic quality checks don't have an exact
idea how a wing is constructed. Utrasonic QC basically detects interfaces,
and a composite glider wing is made up from interfaces all over the place.
I'd say that the error rate in an utrasonic QC would be completely through
the roof.

I think it would be more easy (and straightforward) to do just a structural
load test up to 1.5 times max rated load :-)

--
Bert Willing

ASW20 "TW"


"Martin Gregorie" a écrit dans le message de
...

Does anybody on here know what advantage cotton threads would have
over, say, microballoons or separated glass rovings?
--
martin@ : Martin Gregorie
gregorie : Harlow, UK
demon :
co : Zappa fan & glider pilot
uk :

One simple fix would be to have the entire assembly on a rotating jig and
have the leading edge
45degrees nose down while curing
"John Galloway" wrote in
message ...
My understanding is that any part of the rear flange
that is not actually bonded will show as a 'white'
area even if there is no visible or probe-able gap.
The thin flange is being bonded to a black spar cap
and resin bonded to both surfaces gives the see through
effect.

As the resin gets squeezed out from the front to the
back apparently if no voids reach near the shear web
there aren't going to be any farther forward.

The factory have looked at ultrasonic inspection and
have been that it won't work for the spar problem because
of all the diferent layers it would have to look through
- apparently.

If there are continuous stalactites or curtains of
resin hanging from the rear spar flange and no visible
voids or white areas the spar is safe.

All this is stuff from various reliable sources at
various times but there is a need for a full statement
IMHO - I hope it will happen once all the inspection
data have been analysed

John Galloway


At 06:00 15 October 2003, Slingsby wrote:
'John Morgan' wrote in message news:...
'Slingsby' wrote in message

It would be interesting to know if there is a commonality
between all
of these discrepancies. Was the bonding paste always
too thin or did
they mix it according to the clearances of each wing?
If the mixture
really was too thin causing a void of 14-15 inches
long then there are
probably many other voids which can't be seen on
video. There needs
to be an ultrasonic inspection procedure developed
which can map the
bonding interface.


I have no first hand knowledge. Talked to the owner
of a well known,
respected composite repair shop who said that cotton
threads are added to
the epoxy to thicken it and keep it from running.
And that apparently they
had failed to add enough cotton fiber and this resulted
in epoxy running
out, creating the voids.

According to the DG website the bonding paste should
be about as thick
as cake icing. I suppose too much cotton would also
weaken the bond
as cotton fibres aren't as strong as epoxy. I still
believe that if
there are voids which can be seen and can have wires
poked into them
then there must also be voids which can't be seen and
are far to thin
to allow wire into them. A .0005 to .008mm thick gap
between the spar
cap and shear web could not be seen on a video but
it would still be
an area where there is no bond. Ultrasound would still
reflect off of
the interface and show a gap. The ultrasonic technique
would need to
be proven and calibrated on actual wings where voids
have been found.
Filling the voids immediately makes the chances of
developing another
inspection method unlikely.

Simple fix was to make sure the mix of chopped cotton and resin was thick
enough!
Maybe the problem is when they mixed the resin and cotton together, the
temperature of the resin was relativity low which made it more viscous and
while the whole thing was heat cured, the viscosity of the resin decreased
and ran before gelling. The rotating jig is a good idea but the size of the
reinforcing to keep the wing true would be problem. ( and expensive :-)
Paul
"goneill" wrote in message
...
One simple fix would be to have the entire assembly on a rotating jig and
have the leading edge
45degrees nose down while curing
"John Galloway" wrote in
message ...
My understanding is that any part of the rear flange
that is not actually bonded will show as a 'white'
area even if there is no visible or probe-able gap.
The thin flange is being bonded to a black spar cap
and resin bonded to both surfaces gives the see through
effect.

As the resin gets squeezed out from the front to the
back apparently if no voids reach near the shear web
there aren't going to be any farther forward.

The factory have looked at ultrasonic inspection and
have been that it won't work for the spar problem because
of all the diferent layers it would have to look through
- apparently.

If there are continuous stalactites or curtains of
resin hanging from the rear spar flange and no visible
voids or white areas the spar is safe.

All this is stuff from various reliable sources at
various times but there is a need for a full statement
IMHO - I hope it will happen once all the inspection
data have been analysed

John Galloway


At 06:00 15 October 2003, Slingsby wrote:
'John Morgan' wrote in message news:...
'Slingsby' wrote in message

It would be interesting to know if there is a commonality
between all
of these discrepancies. Was the bonding paste always
too thin or did
they mix it according to the clearances of each wing?
If the mixture
really was too thin causing a void of 14-15 inches
long then there are
probably many other voids which can't be seen on
video. There needs
to be an ultrasonic inspection procedure developed
which can map the
bonding interface.


I have no first hand knowledge. Talked to the owner
of a well known,
respected composite repair shop who said that cotton
threads are added to
the epoxy to thicken it and keep it from running.
And that apparently they
had failed to add enough cotton fiber and this resulted
in epoxy running
out, creating the voids.

According to the DG website the bonding paste should
be about as thick
as cake icing. I suppose too much cotton would also
weaken the bond
as cotton fibres aren't as strong as epoxy. I still
believe that if
there are voids which can be seen and can have wires
poked into them
then there must also be voids which can't be seen and
are far to thin
to allow wire into them. A .0005 to .008mm thick gap
between the spar
cap and shear web could not be seen on a video but
it would still be
an area where there is no bond. Ultrasound would still
reflect off of
the interface and show a gap. The ultrasonic technique
would need to
be proven and calibrated on actual wings where voids
have been found.
Filling the voids immediately makes the chances of
developing another
inspection method unlikely.

"
I think that people proposing utrasonic quality checks don't have an exact
idea how a wing is constructed. Utrasonic QC basically detects interfaces,
and a composite glider wing is made up from interfaces all over the place.
I'd say that the error rate in an utrasonic QC would be completely through
the roof.

I think it would be more easy (and straightforward) to do just a structural
load test up to 1.5 times max rated load :-)
Bert Willing
************************************************** ********************************
I understand your point about ultrasonic testing. The sound wave
would have to go through the outer skin, foam, inner skin, carbon
fibre rovings spar cap, bonding paste, shear web of differing
compositions. I guess there could be too many interfaces although I
would think that one could find an ultrasonic frequency which didn't
reflect off of the interfaces but did reflect off of voids. Or
reflects different frequencies and giving a picture.
Any idea how composite structures are inspected on large airplanes? I
assume x-ray wouldn't be very sensitive on nonmetallic structures.
How about neutron radiation? Acoustic emission transducers? Ballistic
Recovery Chutes?

Also, if the cotton fibres carry the load, is the repair done with an
epoxy/cotton mixture? How do they get it into a blind air bubble?

John Galloway wrote in message ...
My understanding is that any part of the rear flange
that is not actually bonded will show as a 'white'
area even if there is no visible or probe-able gap.
The thin flange is being bonded to a black spar cap
and resin bonded to both surfaces gives the see through
effect.
I wondered, as the fibreglass on my glider is a translucent green but
I am not sure I am looking at bonding paste which has cotton fibres in
it.

As the resin gets squeezed out from the front to the
back apparently if no voids reach near the shear web
there aren't going to be any farther forward.

I would guess that if the spar caps didn't get bonded correctly on
several wings, then the ribs in the D-tube, the tops of other ribs,
the air brake box and other areas of the wing didn't get bonded
correctly on some wings also. Shemp-Hirth should be looking into this
and possibly cutting some wings open. Especially wings with known
voids in them.

The factory have looked at ultrasonic inspection and
have been that it won't work for the spar problem because
of all the diferent layers it would have to look through
- apparently.

It could also be that the only entities with ultrasonic equipment
capable of doing the inspections are major airlines and the military.

If there are continuous stalactites or curtains of
resin hanging from the rear spar flange and no visible
voids or white areas the spar is safe.

Its at least as safe as the G103 fuselage which is now speed and acro
limited because the "margin of safety" is not as great as they shought
it was. Until a couple of weeks ago I would have said that the G103
was one of the strongest, most durable, over engineered, tolerant of
abuse soaring trucks ever produced.

All this is stuff from various reliable sources at
various times but there is a need for a full statement
IMHO - I hope it will happen once all the inspection
data have been analysed

John Galloway


At 06:00 15 October 2003, Slingsby wrote:
'John Morgan' wrote in message news:...
'Slingsby' wrote in message

It would be interesting to know if there is a commonality
between all
of these discrepancies. Was the bonding paste always
too thin or did
they mix it according to the clearances of each wing?
If the mixture
really was too thin causing a void of 14-15 inches
long then there are
probably many other voids which can't be seen on
video. There needs
to be an ultrasonic inspection procedure developed
which can map the
bonding interface.


I have no first hand knowledge. Talked to the owner
of a well known,
respected composite repair shop who said that cotton
threads are added to
the epoxy to thicken it and keep it from running.
And that apparently they
had failed to add enough cotton fiber and this resulted
in epoxy running
out, creating the voids.

According to the DG website the bonding paste should
be about as thick
as cake icing. I suppose too much cotton would also
weaken the bond
as cotton fibres aren't as strong as epoxy. I still
believe that if
there are voids which can be seen and can have wires
poked into them
then there must also be voids which can't be seen and
are far to thin
to allow wire into them. A .0005 to .008mm thick gap
between the spar
cap and shear web could not be seen on a video but
it would still be
an area where there is no bond. Ultrasound would still
reflect off of
the interface and show a gap. The ultrasonic technique
would need to
be proven and calibrated on actual wings where voids
have been found.
Filling the voids immediately makes the chances of
developing another
inspection method unlikely.

In article ,
says...
I understand your point about ultrasonic testing. The sound wave
would have to go through the outer skin, foam, inner skin, carbon
fibre rovings spar cap, bonding paste, shear web of differing
compositions.

Is the spar cap on the Duo Discus bonded to the inner skin or the
outer skin? If it is bonded to the outer skin, it would not have the
foam to go through, which should ease the problem considerably.
--
!Replace DECIMAL.POINT in my e-mail address with just a . to reply
directly

Eric Greenwell
Richland, WA (USA)

On Thu, 16 Oct 2003 08:00:49 -0700, Eric Greenwell
wrote:

In article ,
says...
I understand your point about ultrasonic testing. The sound wave
would have to go through the outer skin, foam, inner skin, carbon
fibre rovings spar cap, bonding paste, shear web of differing
compositions.

Is the spar cap on the Duo Discus bonded to the inner skin or the
outer skin? If it is bonded to the outer skin, it would not have the
foam to go through, which should ease the problem considerably.

The problem is that the bond in question attaches the top spar to the
web. Any problems with this bond will drastically affect the spar
strength by allowing the top spar to peel off the web. In the DG-style
of assembly with carbon cloth between the spar and the web the bond
between the spar and the cloth is equally as critical as that between
the web and the cloth.

In the model world we prevent spar peel failures by assembling both
spars to the web and then wrapping to lot with Kevlar thread or an
epozy-wetted woven carbon tube before building the wing round the
complete spar. This gives a 30G wing. I accept that this is overkill
for full size, but my teeth still itch a bit at the lack of any
binding round both the spars except on the protruding stubs at the
root.

I flat out would not fly a glider with foam between the web and the
top spar and would hope that such a structure has never been made or
flown.

--
martin@ : Martin Gregorie
gregorie : Harlow, UK
demon :
co : Zappa fan & glider pilot
uk :

I wasn't clear enough. The problem I thought would be eased was the
ultrasonic inspection problem. So, is the spar cap attached to the
outer wing skin, or is there actually foam between the outer skin and
the spar cap?

In article ,
ess says...
On Thu, 16 Oct 2003 08:00:49 -0700, Eric Greenwell
wrote:

In article ,
says...
I understand your point about ultrasonic testing. The sound wave
would have to go through the outer skin, foam, inner skin, carbon
fibre rovings spar cap, bonding paste, shear web of differing
compositions.

Is the spar cap on the Duo Discus bonded to the inner skin or the
outer skin? If it is bonded to the outer skin, it would not have the
foam to go through, which should ease the problem considerably.

The problem is that the bond in question attaches the top spar to the
web. Any problems with this bond will drastically affect the spar
strength by allowing the top spar to peel off the web. In the DG-style
of assembly with carbon cloth between the spar and the web the bond
between the spar and the cloth is equally as critical as that between
the web and the cloth.

In the model world we prevent spar peel failures by assembling both
spars to the web and then wrapping to lot with Kevlar thread or an
epozy-wetted woven carbon tube before building the wing round the
complete spar. This gives a 30G wing. I accept that this is overkill
for full size, but my teeth still itch a bit at the lack of any
binding round both the spars except on the protruding stubs at the
root.

I flat out would not fly a glider with foam between the web and the
top spar and would hope that such a structure has never been made or
flown.

--
martin@ : Martin Gregorie
gregorie : Harlow, UK
demon :
co : Zappa fan & glider pilot
uk :



--
!Replace DECIMAL.POINT in my e-mail address with just a . to reply
directly

Eric Greenwell
Richland, WA (USA)

Is the spar cap on the Duo Discus bonded to the inner skin or the
outer skin? If it is bonded to the outer skin, it would not have the
foam to go through, which should ease the problem considerably.

Modern ultrasonic non-destructive evaluation (NDE) should have no
problem inspecting the joints in question. Rolls-Royce has developed
standoff NDE equipment that must contend with a 20 mm air gap. They
got excellent results detecting defects in carbon fiber samples:

Advances in air coupled NDE for rapid scanning applications
Farlow, R.; Kelly, S.P.; Hayward, G.;
Ultrasonics Symposium, 1994. Proceedings., 1994 IEEE , Volume: 2 , 1-4
Nov. 1994
Page(s): 1099 -1102 vol.2

Applications of through-air ultrasound for rapid NDE scanning in the
aerospace industry
Kelly, S.P.; Farlow, R.; Hayward, G.;
Ultrasonics, Ferroelectrics and Frequency Control, IEEE Transactions
on , Volume: 43 Issue: 4 , July 1996

S-H is just talking to the wrong people.

Tom

Eric Greenwell wrote in message ...
I wasn't clear enough. The problem I thought would be eased was the
ultrasonic inspection problem. So, is the spar cap attached to the
outer wing skin, or is there actually foam between the outer skin and
the spar cap?



Is the spar cap on the Duo Discus bonded to the inner skin or the
outer skin? If it is bonded to the outer skin, it would not have the
foam to go through, which should ease the problem considerably.

The problem is that the bond in question attaches the top spar to the
web. Any problems with this bond will drastically affect the spar
strength by allowing the top spar to peel off the web. In the DG-style
of assembly with carbon cloth between the spar and the web the bond
between the spar and the cloth is equally as critical as that between
the web and the cloth.

In the model world we prevent spar peel failures by assembling both
spars to the web and then wrapping to lot with Kevlar thread or an
epozy-wetted woven carbon tube before building the wing round the
complete spar. This gives a 30G wing. I accept that this is overkill
for full size, but my teeth still itch a bit at the lack of any
binding round both the spars except on the protruding stubs at the
root.

I flat out would not fly a glider with foam between the web and the
top spar and would hope that such a structure has never been made or
flown.

************************************************** ****************************
I don't think anyone said that there is foam between the spar cap and
shear web. Foam within the shear web adds a great deal of stiffness,
however.

The spar cap of German designs is built into one wing skin while the
spar web
and opposite spar cap is bonded into the other skin. When the two wing
halves are mated the one spar cap is glued to the spar web and cap of
the
other wing half. It's pretty much a blind operation and everything had
better well match or there will be a poor or no bond.

The Genesis 2 and the LAK-17a both have the spar completely built
outside the
wing. The spar is totally wrapped in glass fiber insuring it's
integrity.
It will not debond. Following is a site which has pictures of a
Genesis spar.

http://www.aviation-salvage.com/airc...enesis%202.htm

On the Genesis wing it appears that the spar is bonded to an inner
skin which is a foam sandwich with an outer skin. The Genesis spar in
the picture had a metal fence post cut through it and it looks like
Graphlite rods have been bashed into more of a round shaped bundle. I
believe they are spread out into more of an I-shape than the picture
would suggest. At least with this design, if the spar is not bonded
to the skin there is still a very strong spar.