Skinning foam cores (was Re: Hard shelling of cores ...)
G A Venkatesh
on homebuilding practices.
Anyway, here is all that I have assimilated so far on the subject. I would
appreciate *factual* corrections and or clarifications:
1. Pure epoxy (especially SafetyPoxy II with its lower viscosity) tends to
penetrate the foam cores. It doesn't really eat it (although there are some
claims that the presence of small amounts of styrene deforms the surface).
The penetration is to be avoided not because of the concern about bond
strength but because of the extra weight. Cutting with a hot wire tends to
seal the pores on the surface and helps somewhat but wherever you sand it,
there are open cells for epoxy to flow into.
2. To prevent the above from happening, you slurry the foam with a mixture of
micro-ballons and epoxy. The density of the slurry is much less than the
density of epoxy. The slurry either fills up some of the pores or blocks it
against penetration from pure epoxy. The method recommended by Burt Rutan had
the slurrying immediately followed by glassing.
3. People have been trying to reduce weight any way they can and they have
found that one way to achieve it is to let the slurry cure first. This method
is usually known as hard-shelling and some people have been using it for
aircraft construction for about 5 years and surfboard folks have been doing
that since early 80's. In this method, you sand the foam surface and align
neighboring pieces of foam exactly the way you want it. You use an epoxy such
as AlphaPoxy, West or Resin Research (which contains a sanding agent and cures
fast) epoxy with microballons. You use a runny mixture and squeege of all
excess. After cure, you sand the entire surface, apply a thin layer of slurry
as in item 2 above and glass it.
4. Worrying about the bond strength between slurry layers or between different
epoxies, etc., is unnecessary. The bond strength between those things (even if
unsanded) *is greater* than the bond strength with the foam, for the simple
reason that the foam tends to tear first (about 1/8" below the bond) before any
epoxy-epoxy epoxy-micro bond can come apart.
5. Some folks do a compromise between items 2 and 3 above and wait for the
slurry as applied in item 2 to set a little before glassing it. You avoid the
sanding and get some benefits from lesser penetration of the epoxy into the
slurry and into the foam.
6. The purpose of sanding a layup or a slurry application is not to "rough" up
the surface for a better hold. It is to remove the top layer of the cured
epoxy. Epoxies (especially laminating expoxies like SafetyPoxy) contain agents
that rise to the surface and allow it to cure. This is the thing that causes
the shine. Bonding with this layer is not as good as the bonding with the cured
epoxy below it. The bond strength with the unsanded surface is *still stronger*
than any bond with the foam surface. Finishing resins like AlphaPoxy either
don't have such a top layer or have very little of it.
7. Small dings and cracks in the foam can be repaired using a dry slurry. One
can do this as part of the slurrying in item 2 above and glassing immediately
over it. The problem is that some settling in the dings and cracks might
happen while curing and result in an uneven surface. The other option is to
fill the ding/crack with dry slurry and let it cure and sand it down level
before doing the glassing as in item 2 above. You can use a finishing resin
such as Alphapoxy for faster cure and easier sanding for this purpose. The
only problem is that sanding that part level with the foam is difficult since
the sand paper tends to dig into the foam around it. You make it as level as
possible when you fill it and sand very lightly over it. Remember that the bond
over that surface will still be better than the bond over foam.
venky
Bill Cox
<climbing up on his soapbox>
I'm concerned about a possible mis-interpretation of Venky's remarks:
"The bond strength between those things (even if unsanded) *is greater* .."
You could interpret this as a recommendation to slight the surface preparation.
DON'T DO IT!
I'm not exactly quoting Danny Maher, here, but very close.
"Your airplane is as strong as the care you take with your sanding".
I have experimented with scraps of layup which had almost invisible
layers of airplane-shop dust on them. Adherence is *very* poor. By
sanding before the next layup and/or carefully wiping the surface you
improve the layup strength greatly.
Burt Rutan, in his video, emphasizes this in another way. He recommends
that you keep your cloth in a sealed box until it goes on the plane, to
protect it from this dust, and that you wipe down the table you cut the
cloth on.
Of the bonds that hold your plane together, the glass-to-glass ones are
usually the critical ones. Examples: layups that hold the Velocity
bulkheads into the fuselage, the ones that hold the main spar in, or
(to a lesser degree) the trailing edges of the wings. Don't neglect
to sand these especially well.
<dusting off his hands, he steps off the soapbox>
Thanks for reading..
--
Bill Cox, happily bi...@cygnus.com Mr. President, CUT SPENDING, please.
Larry Johnson
>This thread is in support of my conclusion that there is very little consensus
>on homebuilding practices.
>
>Anyway, here is all that I have assimilated so far on the subject. I would
>appreciate *factual* corrections and or clarifications:
>
>
>6. The purpose of sanding a layup or a slurry application is not to "rough" up
>the surface for a better hold. It is to remove the top layer of the cured
>epoxy. Epoxies (especially laminating expoxies like SafetyPoxy) contain agents
>that rise to the surface and allow it to cure. This is the thing that causes
>the shine. Bonding with this layer is not as good as the bonding with the cured
>epoxy below it. The bond strength with the unsanded surface is *still stronger*
>than any bond with the foam surface. Finishing resins like AlphaPoxy either
>don't have such a top layer or have very little of it.
I accept your challenge :-)
Last night I epoxied 2 nail heads to a piece of glass, in one place
smooth and in the other sanded with 36 grit. This morning I pulled them
off and the roughed up glass let go much less easily. I wish I would have
used my fish scale (pun intended) but it didn't occur to me till now.
I'll do this again and report.
If the strength is a function of surface area (3 square miles is harder
to pull apart than 3 square microns), and roughing up a surface increases
surface area:
small area: ------------------------
large area: -- ----- ---- -
\ / | \ \__/
\/ \___|
then it seem reasonable that the bond will be stronger. Overhangs can
only help IMHO.
Do you have information on the nature of the agents? How many different
compounds are in epoxy? (more than two it would seem) Are these agents
not part of the reaction or left overs from imperfection in the mixing
ratios of epoxy? Perhaps a chemical wash would be better than sanding if
you are correct about the surface.
Let me ask a question:
Has anybody out there got first hand experience with delamination?
Is all this discussion a big waste of time?
Sid: In your posting (included), did you see delamination? Did you perform
an experiment?
-------------------------------------------------------------------------
From: si...@cpqhou.se.hou.compaq.com (Sid Lloyd)
I've been building a Cozy MkIV and have done layups over roughened slurry.
You definately DON'T get as good a bond as you do with doing a layup on
wet slurry. Maybe it's because you only have a physical bond when a
layup is done on roughened cured resin, but a chemical bond occurs (one
contiguous matrix) when you do the layup over wet slurry.
-------------------------------------------------------------------------
venky: Regarding your mail, did you ever resolve why you had problems
with hard shell? Are you currently using hard shell?
------------------------------------------------------------------------
(from venky)
> ....
> that after my experiment this weekend. I just talked to Duane about it. He
> seemed to support hard-shelling and was surprised that I did not get a good
> bond. He still suggests waiting two or three hours after slurrying to put
> the skin on. Apparently it prevents any kind of shrinkage due to the epoxy
> seeping into the wet slurry and provides for a smoother surface. I am
> still not sure if I am going to do that for all surfaces, I might try it
> for the winglets.
Bill: You have comments?
I'm about to start my velocity, so I greatly appreciate any comments from
those hardened epoxy jockies (pun intended again) out there.
Thanks
--
Larry D. Johnson
UUCP: la...@qedinc.com
G A Venkatesh
>
><climbing up on his soapbox>
>
>I'm concerned about a possible mis-interpretation of Venky's remarks:
>
>"The bond strength between those things (even if unsanded) *is greater* .."
>
>You could interpret this as a recommendation to slight the surface preparation.
>DON'T DO IT!
Thanks for the clarification. I didn't mean it to imply that one should not
sand. More bond strength you can get the better. However, there was one
post that talked about not doing a layup over cured slurry since it would be
difficult to sand every nook and cranny. When I talked to Scott Swing of
Velocity, Inc. about this, he made the assertion which I included. So I
wouldn't dismiss hard shelling for that reason.
>
>I'm not exactly quoting Danny Maher, here, but very close.
>
>"Your airplane is as strong as the care you take with your sanding".
I think that also refered to overdoing it and breaking strands?
>I have experimented with scraps of layup which had almost invisible
>layers of airplane-shop dust on them. Adherence is *very* poor. By
>sanding before the next layup and/or carefully wiping the surface you
>improve the layup strength greatly.
Dust on the surface is an orthogonal issue, of course. An air compressor
comes in handy here. But I don't have one. So I use a shop vac that can also
be converted to a blower. I alternate vacuuming and blowing a couple of times
to get the dust out before I do a layup.
venky
G A Venkatesh
>Last night I epoxied 2 nail heads to a piece of glass, in one place
>smooth and in the other sanded with 36 grit. This morning I pulled them
>off and the roughed up glass let go much less easily. I wish I would have
>used my fish scale (pun intended) but it didn't occur to me till now.
>I'll do this again and report.
Was the amount of epoxy exactly the same in both cases?
Actually, this issue is rather academic since you *must* sand the surface
anyway. However, the cure layer theory cautions strongly against roughing
up the surface without getting all the shine out.
>Do you have information on the nature of the agents? How many different
>compounds are in epoxy? (more than two it would seem) Are these agents
>not part of the reaction or left overs from imperfection in the mixing
>ratios of epoxy? Perhaps a chemical wash would be better than sanding if
>you are correct about the surface.
I have no information about the agents that create the surface layer except
that it is included precisely for that purpose, to seal the top layer so that
the promoted resin can cure. The local EAA tech counselor calls it "wax" but
I doubt that it is the technical term. :-)
> Has anybody out there got first hand experience with delamination?
> Is all this discussion a big waste of time?
Good question. I would be interested in first-hand information too.
>venky: Regarding your mail, did you ever resolve why you had problems
> with hard shell? Are you currently using hard shell?
I decided not to use the hard shelling process because of the extra effort
for minimal benefits and increased chances of screwing up. I *do* let the
slurry set for 3-4 hours before I put the skin on though. It definitely makes
the squeezeing much easier.
venky
(Velcity 173 under construction - One winglet skinned)
Sid Lloyd
>>>>>> stuff deleted <<<<<<<
: 4. Worrying about the bond strength between slurry layers or between different
: epoxies, etc., is unnecessary. The bond strength between those things (even if
: unsanded) *is greater* than the bond strength with the foam, for the simple
: reason that the foam tends to tear first (about 1/8" below the bond) before any
: epoxy-epoxy epoxy-micro bond can come apart.
Can't agree with you here. I've HAD delams between fiberglass and cured
slurry that was over foam. In this case, the foam bond held, and the slurry-
fiberglass layer let go. I will definately NOT do hard shelling...
Sid
Bill Cox
An embarassing follow-on to this: Danny Maher suggested, to keep the
ailerons weight down, that we lay up the lower skin on a piece of glass,
to get at least one surface absolutely smooth, and have no extra weight from
fillers. We auto-waxed the glass first. Our ailerons came off the glass
looking *gorgeous*! And we plan to sand them lightly but thoroughly to
remove any wax.
One of our friends was told by Nat Puffer (Cozy 3 plans) to lay his instrument
panel up on a sheet of glass, for cosmetic reasons, I think. He *didn't*
wax his glass. He nearly destroyed the panel separating it from the glass...
As he told us about this, someone said "well, fiberglass is made of glass,
I guess that's what would you expect, eh?". He didn't get hit, but that's
because our friend is the most even-tempered guy I know.
Bill Cox
> that after my experiment this weekend. I just talked to Duane about it. He
> seemed to support hard-shelling and was surprised that I did not get a good
> bond. He still suggests waiting two or three hours after slurrying to put
> the skin on. Apparently it prevents any kind of shrinkage due to the epoxy
> seeping into the wet slurry and provides for a smoother surface. I am
> still not sure if I am going to do that for all surfaces, I might try it
> for the winglets.
Larry Johnson says:
Bill: You have comments?
Sounds to me as if Duane suggests waiting until the slurry has 'gelled',
or begun to set up, which has the *potential* at least to provide that
one-giant-molecule chemical bond that was discussed earlier.
What we need is another experiment.. There now seem to be three cases.
Venky compared the hard-shell case with the cover-the-slurry-right-away case.
The wait 2-3 hours case should be tried also. With the same foam, slurry,
cloth, etc. Then do Venky's experiment of trying to pull the Triax off
the foam.
Is there an optimum time-to-wait before applying the wing skin? Does
waiting that much time produce a stronger bond than the hard-shell way?
Want to try the experiment, Larry?
Larry Johnson
>
>Sounds to me as if Duane suggests waiting until the slurry has 'gelled',
>or begun to set up, which has the *potential* at least to provide that
>one-giant-molecule chemical bond that was discussed earlier.
>
>What we need is another experiment.. There now seem to be three cases.
>Venky compared the hard-shell case with the cover-the-slurry-right-away case.
>The wait 2-3 hours case should be tried also. With the same foam, slurry,
>cloth, etc. Then do Venky's experiment of trying to pull the Triax off
>the foam.
>
>Is there an optimum time-to-wait before applying the wing skin? Does
>waiting that much time produce a stronger bond than the hard-shell way?
>Want to try the experiment, Larry?
>--
>Bill Cox, happily bi...@cygnus.com Mr. President, CUT SPENDING, please.
Yup, already in my plans, just as soon as I can tape this chip out.
The fat mutha is going out the door, but kicking and screaming the
whole way.
My kit arrives 7/12/93, I'll post the results of the above experiments
when completed.
Thanks to all the responders to my post for sharing their experiences.
G A Venkatesh
>ailerons weight down, that we lay up the lower skin on a piece of glass,
>to get at least one surface absolutely smooth, and have no extra weight from
>fillers. We auto-waxed the glass first. Our ailerons came off the glass
>looking *gorgeous*! And we plan to sand them lightly but thoroughly to
>remove any wax.
>
>One of our friends was told by Nat Puffer (Cozy 3 plans) to lay his instrument
>panel up on a sheet of glass, for cosmetic reasons, I think. He *didn't*
>wax his glass. He nearly destroyed the panel separating it from the glass...
Wouldn't placing a plastic sheet/cling wrap over the glass surface work?
Are you sure that the wax does not contaminate the epoxy (not just the top
layer), given the warnings about not using waxed cups for mixing?
venky
David Doshay
>Are you sure that the wax does not contaminate the epoxy
in order to get a nice surface.
In this case a small amount of wax contamination should not matter because the
panel is not structural. I do not think I would try wax for a surface on an
aileron, but mold release should work. I would think the part would weigh more
with solid epoxy filling the weave instead of a mix of microspheres. That could
be a problem on an aileron.
David
---------------------------------------------------------------------
David dos...@soma.arc.nasa.gov
=====================================================================
The thought police have a new policy. They insist I tell you:
My thoughts, not NASA's.
Rick Lafford
>Is there an optimum time-to-wait before applying the wing skin? Does
>waiting that much time produce a stronger bond than the hard-shell way?
>Want to try the experiment, Larry?
------------------------------
We also seem to have another confusion. This one conserns the use of a DRY
slurry to seal the foam and reduce the amount of resin used (weight). This is
a completely different issue and one you definitely do not want to do a wet
to wet layup on. If by 'hardshelling' we mean a fairly wet slurry, then I
would agree with waiting some fraction of the set-up time then laying the
first layer of glass down. Filler migration into the glass cloth should be
minor once the resin begins to cure.
If the slurry is of the DRY variety, I feel the only correct method is to
allow full cure with attendant shrinkage, sand smooth to contour, clean
thoroughly and only then go the the first glass layup. The reason for this is
that the filler content is high enough to create significant cloth
contamination problems. The sanded dry filled surface provides an excellent
bonding bed IF cleaned thoroughly.
One more quibble; this all can be changed if you use a 'waxed' resin. When
allowed to partial cure, the wax will migrate to the surface and will cause
bond problems. I"ll assume for this discussion that we are all talking about
'unwaxed' resins.
By all means do anything you can to reduce irregularities in the resin
content of the airframe. I have a set of glider wings made in 1967 which have
continued to cure since then. By 1981, every small resin concentration had
pulled in and created a significant dent in the surface. When these are on a
laminar flow wing.... well, I hope you like sanding, filling, getcoating and
sanding, sanding, sanding. This was particularly apparent along the spar to
skin junction. The fuselage of this particular glider used a nailboard to
mold the initial glass layers on. Every nail left a small resin pocket and
today the pattern is expressed on the fuselage as lines of tiny dimples.
Rick
"Gravitation cannot be held responsible
for people falling in love."
- Albert Einstein
Rick Lafford
> Wouldn't placing a plastic sheet/cling wrap over the glass surface work?
>Are you sure that the wax does not contaminate the epoxy (not just the top
>layer), given the warnings about not using waxed cups for mixing?
The answer here is no. When you make molded parts you routinely coat the
mold with several coats of mold release (carnuba) wax and then a coat of
PVA (Polyvinyl Alcohol) to enhance the release of the cured matrix.
You would also use a waxed resin for the top most layer to assure a
complete cureout of the air contact resin.
You can use any good all carnuba wax (NO SILICONES) as a mold release. I
use Johnsons Paste wax (you know, for wood floors).
Mike Best
You are correct, we are forewarned not to use such materials when mixing.
However, I have made engine cowling molds for a Defiant using mold release
wax (green looking stuff). I also used this wax for subsequent cowlings and
there was no apparent difference in strength from the originals. Of course,
these are not structural parts in the same way that the spar is.
Mike Best
Mark Hall
|> ------------------------------
|>
|> One more quibble; this all can be changed if you use a 'waxed' resin. When
|> allowed to partial cure, the wax will migrate to the surface and will cause
|> bond problems. I"ll assume for this discussion that we are all talking about
|> 'unwaxed' resins.
|>
|> By all means do anything you can to reduce irregularities in the resin
|> content of the airframe. I have a set of glider wings made in 1967 which have
|> continued to cure since then. By 1981, every small resin concentration had
|> pulled in and created a significant dent in the surface. When these are on a
|> laminar flow wing.... well, I hope you like sanding, filling, getcoating and
|> sanding, sanding, sanding. This was particularly apparent along the spar to
|> skin junction. The fuselage of this particular glider used a nailboard to
|> mold the initial glass layers on. Every nail left a small resin pocket and
|> today the pattern is expressed on the fuselage as lines of tiny dimples.
Although Rick does not specify the type of resin system being used,
the date he references and general description of the long term results
would indicate that the glider in question was probably built with a
polyester resin. The continual shrinking over a long period is not
atypical but is quit different from the epoxy based systems used in
most homebuilts today. Safety-Poxy for example tests at shrinkage of
less than 1% of its total volume and even that stabilizes within the
first several days.
In addition my conversations with my local Hexel tech rep.(mfr. of
Safety-Poxy) and a check of the literature on hand for DuPonts epoxy
systems reveal no reference at all to any sort of wax content in any
structural epoxy. More likely what is being observed is that surface
tension during cure will form an extremely smooth surface which
hardens to form a virtually impenetrable barrier. It is this glass-like
surface which must be broken down (by sanding) to allow a mechanical
bond with whatever goes on top of it whether that be another layup or
filler/paint. Once the epoxy quits flowing (this depends on temperature,
mix ratio, and etc.) the surface boundary is set and further chemical
bond (one big molecule) is no longer possible. (Please note that flowing
in this sense does not refer to the viscous state where epoxy may be
poured and wetted into the cloth).
Mark
Rick Lafford
> Although Rick does not specify the type of resin system being used,
>the date he references and general description of the long term results
>would indicate that the glider in question was probably built with a
>polyester resin. The continual shrinking over a long period is not
>atypical but is quit different from the epoxy based systems used in
> In addition my conversations with my local Hexel tech rep.(mfr. of
>Safety-Poxy) and a check of the literature on hand for DuPonts epoxy
>systems reveal no reference at all to any sort of wax content in any
>structural epoxy. More likely what is being observed is that surface
The system used was Shell Epon and yes, it was an older resin but not
a polyester. It's a true epoxy. You can rest assured that we don't have
any gliders flying around with polyester primary structures. If we're
putting carbon and kevlar cloth in to skins and sparcaps we really couldn't
put up with a lower strength resin system ;-).
As for the 1% shrinkage, what temperatures and times does the
Safety-Poxy literature say that number was arrived at. If you
store the aircraft outside and the surface gets to 120'F (my ship
was in a trailer but the temp still gets up there) what happens
after 20 years. I can tell you for a fact that all the 20 year old
gliders have post mold shrinkage marks anywhere there was
a higher concentration of resin. This especially occurres around
the spar to spar cap juncture.
The wax is used in polyester resin to aid curing of the surface layer.
Air inhibits curing of polyester resins. Withoug the wax you end up
with a forever tacky surface film.
Has anyone (mfg. or otherwise) done any type of serious age
testing with the current crop of room temp. cure epoxies?
Inquiring minds want to know.
Rick
-----------------------------------------------------------------
Rick Lafford | Kolb N93RD, Diamant HVB N11HS
Clinical Diagnostics | Discus B N154RS (HS) and real soon now
laf...@clpd.kodak.com | Berkut/Velocity N96HS
======================================================