Article 8299 of alt.surfing:
Path: sol.ccs.deakin.edu.au!munnari.oz.au!news.uwa.edu.au!rhino!rd005.251.54.192.in-addr.arpa!conrad
From: conrad@jtec.com.au (Conrad P. Drake)
Newsgroups: alt.surfing
Subject: FibreGlass FAQ Part 2 of 2
Date: Mon, 17 Oct 1994 05:23:32 GMT
Organization: QPSX Communications Ltd
Lines: 301
Sender: news@qpsx.oz.au (Usenet Admin)
Message-ID:
NNTP-Posting-Host: uniwa.uwa.edu.au
X-Newsreader: Trumpet for Windows [Version 1.0 Rev B final beta #4]
===== Fibreglass repair FAQ 0.2 17 Oct 94 ===== Part 2 of 2
2. Other Stuff
--------------
Epoxy and Other Materials
--------------------------
Not suprisingly, materials have come a long way since 1950. Almost everyone knows
about Carbon fiber and Kevlar. Not so well known are the changes in the foam and resin.
Most high-tech fibreglass jobs (ie commercial and homebuilt aircraft, and some sailing boats)
are constructed using a two-part epoxy, with a stronger glass fibre.
Some surfboards have been made, using a special closed cell (to stop waterlogging) polystyrene
or polyvinylxloride (PVC) foam, both in the West Coast of the USoA and on the east coast of Oz.
The few reviews (Aust.Surfing life had one recently) seem impressed by the weight and strength
of the boards - down to 2.5kg for a 6' board you can jump on. One other advantage is that they
can be repaired with 2-part expoy glues (such as Araldyte).
There are rumoured to be people making kevlar (reef-proof) boards in South West Oz.
Many of the newer resins come preimpregnated in carbon/glass/kevlar cloth (to order) and require
autoclaving (cooking) to set.
I'm not going to go into detail on my own, but rather I'll include a few articles from
regular rec.aviation.homebuilt contributors. Tune in there for more info (unfortunatley
they don't seem to have an all singing/all dancing composites FAQ). Also try rec.autos.racing.*
rec.boats,.building and rec.models.rc
Strengths of materials
----------------------
Tm Tensile Modulus (Giga Pascals 1gpa = 10^9 Newtons/m^2 ~= 140,000 psi, i think)
Ts Tensile Strength (Giga Pascals)
D Density (grammes per cubic centimeters)
Material Tm Ts D
E-glass 77 2.5 2.54
S-Glass 85 3.5 2.48
Graphite AS4 190 2.8 1.80
Kevlar 49 130 2.8 1.45 - note kevlar is shitty in compression.
Aluminum 70 .14-.6 2.7
Note these are unidirectoinal free fibre, with no expoxy. Reality is only 1/4 as good.
There are lots of different grades of each material.
Some books - various people
----------
(I haven't been able to find these at my local library - I haven't tried the Uni yet.)
"Graphite Fibers and Filaments" by M. S. Dresselhaus et al, Springer
Series in Materials Science 5, (Springer-Verlag, Berlin, 1988), chapters 1, 2,
and 12.
ASM Engineered Materials Handbook
Volume 1: Composites
ISBN 0-87170-279-7 (v. 1)
(c) 1987 ASM International
'Composite Construction for Homebuilt Aircraft' Jack Lambie's
"Composite Basics". Andrew "Bud" Marshall
"Composites Design", by Steve Tsai
"The Behavior of Structures Composed of Composite Materials"
J.R. Vinson and R.L. Sierakowski
Composite workshop review (by Daid Parrish)
-------------------------
This weekend, I attended the composite basics workshop put on by
Alexander Aeroplane in Griffin, Georgia and I thought that others on
the net would like to hear my impressions on the workshop and what I
think are some very useful techniques presented.
First off, a little about the workshops and in particular, the
instructor for the composite workshop, Stan Montgomery. Currently,
Alexander is running three - two day workshops on basic composite
construction, welding and fabric covering that are held in Griffin,
Syracuse, Oshkosh, Trenton, Bloomington, Houston, and Lakeland and
will soon be adding workshops on other topics like wood and
metalworking. Stan also mentioned the possibility of three and five
day advanced composite workshops and another workshop on finishing.
Stan Montgomery is a very good speaker, has a masters in chemistry, so
he knows his resins, and was a military pilot and has built numerous
homebuilt aircraft, so he knows airplanes and composite construction.
He's also VERY passionate about composites. This has caused some - er
- disagreement with illuminaries such as Dick Rutan, but anyone that
can make a spar shear web out of heavy, finely woven bid glass and
achieve 40% resin, BY HAND, and still use peel ply should be listened
to seriously.
For those that don't have the foggiest what I just said, some extra
information. The spar carries the weight of the plane through the
wings, and the spar web carries the load between the upper and lower
spar caps, forming a strong and very stiff I-beam inside the wing. In
fiber/resin composites, the fibers carry the load while the resin
keeps the fibers together. If there is too little resin, the fibers
don't stick together as well, and the part is weakened. If there is
too much resin, which is much weaker than fiberglass, then the resin
starts taking some of the load, and produces a weaker product.
The problem is, most homebuilders think that if resin is good, more is
gooder. But anything above 60% resin, 40% glass, is actually weaker
than 50-50 or 40-60 and is both heavier than it should be and more
expensive, since any resin, from $20 a gallon polyester to $100 plus a
gallon epoxies are expensive to buy.
Bid glass is a fiberglass cloth that has nearly equal number of fibers
both along and across the bolt of cloth. Peel ply is a light weight
and finely woven nylon or polyester cloth that is used as the last
layer in a layup. When peeled off, it fractures the resin surface for
a stronger bond with subsequent layups without using sandpaper, which
damages the glass fibers on the surface.
Back to the workshop.
The syllabus had an ambitious schedule that we were unfortunately
unable to keep up with because of time constraints. Even saying that,
the most important points and procedures were hit upon. The first part
of Saturday was a lecture that covered what composites are, safety
with composite materials, and various techniques used in composite
construction. Most of the rest of the two days was hands on building
of a short section of a canard with interspersed discussions on other
composite techniques, including a hands on vacuum bagging.
The construction of the canard started with a block of blue
polystyrene foam and a pair of templates, so we had to hot wire out
our own canard cores. Hot wire cutting the foam is a very intense few
of minutes where you have to be aware of wire temperature, cutting
speed, wire drag, tension against the template and exactly where your
partner on the other side of the wire cutting frame is at on the
templates. This is done by numbers around each template that one calls
off and the other gives faster or slower cues as the pair cuts around
the templates.
Once cut out, the leading edge just in front of the spar cap recess is
hot wired off and the front edges of the trailing edge part were
rounded so there are no sharp edges for the spar web to go over. Once
rounded, the spar web area is filled with epoxy filled with
microscopic glass balloons or Alexander's superlite filler. Both are
very light in weight and prevent raw epoxy from soaking into the foam.
A quick, unsolicited ad for the superlite epoxy filler. This stuff is
a very light weight epoxy filler that can be used in place of either
Bondo or microballooned epoxy and was developed by Stan. It has a
density at least half that of Bondo, doesn't shrink, is very sandable,
and contains a built in chromate conversion for use on aluminum. It
also contains an agent that helps tiny entrapped bubbles rise to the
surface and burst so pinholes are greatly reduced. The only thing I've
tried it on so far is to fill an aluminum nose gear fork, and it has
very nice working properties. I'm definitely going to try it when I
start finishing the wings on my Velocity.
Anyway, next Stan showed us how to a layup with 40% resin, yet still
be fully wetted out. First he cut out two ply of glass cloth to
approximately the right size, weighed them, and then placed them on a
piece of plastic and poured on a weighed amount of epoxy to an exact
40% resin, 60% glass by weight ratio. He then put another piece of
plastic over the layups and worked the epoxy into the glass. To evenly
distribute the epoxy, he would occasionally fold the glass and work it
some more, making sure not to crimp any folded glass and carefully
applied heat from a hair drier to thin the epoxy out. When he was
done, the layup was totally saturated with no white streaks indicating
dry spots. Once this was done, it was applied to the canard core and a
sheet of peel ply was squeegeed onto the surface, further reducing the
epoxy content of the layup.
According to Boeing, this is impossible. They can achieve a 37-63
ratio, but only by using multi-million dollar autoclaves. While a
40-60 ratio takes quite a bit of experience, I was able to do a 50-50
layup with no problems on my first try. Wake up guys. Homebuilders are
at least ten years ahead of anything that comes out of the big iron
plants. The only thing that may come close would be the formerly Beech
Starship, and it was designed by the homebuilder's homebuilder: Burt
Rutan.
I do have one nit to pick with Stan on epoxy though. Being a chemist,
he wants exact molecular ratios of resin to hardener, therefore the
only way to do this is by weighing both the resin and hardener before
mixing instead of using an epoxy pump, which does do ratios by volume.
For me, working alone, all that extra weighing of resin and hardener
just takes to much effort and time. Assuming the pump is working
correctly, the volume ratio is based on the weight ratio of the two
components and the only weight ratio change would be from the
DIFFERENCE in the expansion rates of the components with temperature.
Since most epoxy systems has a 5% margin of error, I'm not overly
concerned about this difference, but I am going to retest my pumper at
various stroke lengths. Also, if you have a scale that can only
register to 2 grams and you're doing a batch of ten or twenty grams
for a small layup, you may end up with an error greater than 5%
anyway. Knowing the weight of the glass and the epoxy used in most
layups I totally agree with.
Another neat technique he showed us was with unidirectional spar cap
tapes. The tapes are only a few inches in width and produce a thick
layup. The rovings are held together with a sparse cross thread, but
any weaving in a glass cloth reduces the strength. What he did was
find the single fine thread on the edge that held the cross thread in
place and removed it after the tape was placed on the canard. Once it
was removed, the cross thread was carefully removed, leaving straight,
flat fibers in the spar cap. That one even surprised the epoxy
manufacturer that sat in on our workshop on Sunday.
Recommended tools were scales, layup rollers and a hair dryer. The
scales are used for weighing the glass and epoxy to calculate their
weight ratios. The rollers are for working out air bubbles and to
distribute the epoxy. He was against using a paint brush to remove air
(a process known stippling) because it tended to break up bubbles
instead of removing them. Layup rollers are shaped something like a
small paint roller, except the roller is plastic or aluminum and has a
grooved surface that allows entrapped air to escape. The hair dryer is
probably his favorite tool. With it he can drastically thin the epoxy
to improve wetting of the glass and speed up the setting time.
Another point he made was that all epoxy layups should be post cured
at an elevated temperature. All epoxies have what's called the glass
transition temperature, where it looses it's strength. They all have a
maximum transition temperature, such as 190 degrees Fahrenheit, but
the actual temperature that it weakens is only thirty or forty degrees
above the temperature the resin was cured at. If it's 60 degrees when
you make a wing, the wing will sag when the skin reaches only 90 or
100 degrees. Not good if you fly down to Sun 'N Fun!
To fix that, after the initial cure is done, Stan post cures the part
by heating it to 130-150 degrees for a few hours with the part
supported so it doesn't bear any weight. This can be done by painting
the part with black tempera and leaving it out in the sun or by
putting it in an 'oven' made out of cardboard boxes and a small
forced air heater.
All in all, I really enjoyed the workshop. There were some problems
that I think were due to the newness of the workshop and limited time,
but over all, it was well worth the $150.
I finally got to try the techniques from the workshop, and the short
answer is: It's harder than it looks.
That's not really a fair statement, as I didn't really allow for the
differences in my application. My first trial was the installation of
my canard bulkhead into the fuselage. To do that, the bulkhead is
aligned plumb and square with the proper station in the fuselage and
then taped in place with 2 ply of bid cloth cut at a 45 degree angle.
The tape goes about an inch and a half up on the bulkhead and down on
the fuselage, both fore and aft, along the full joining line.
So I used a string to find approximately what length of bid tape to
use for the four sections. (Fore and aft and left and right of the
nose gear door cutout.) I cut the tapes to length, weighed them, and
poured an equal weight of epoxy over them on plastic film. With the
layup roller, I spread the epoxy out. With narrow bid tapes, this can
be a bit difficult without the tapes distorting, but the roller did a
good job as long as I didn't move the epoxy ahead of the roller too
quickly. They're rather expensive, but I think they do a better job
than stippling with a paint brush.
Once wetted out, I cut the film to rough size and carried the whole
thing over to the plane and put it in place. One thing though. The bid
tape conforms to the surface much better than the plastic film that
keeps it from stretching, so you have to carefully peel the cloth from
the film as you put it in place. Once in position, I used the corner
and layup rollers to press the cloth in place and covered that with
peel ply squeegeed in place.
When I peeled the peel ply off the next day, there were streaks were
there was no epoxy between the bulkhead and the bid tape, mostly on
the vertical surfaces of the bulkhead. The problem has to do with the
surface of cured triax cloth that was used on the bulkhead. In triax
cloth, there are three layers of fiber bundles, stitched together like
a quilt instead of woven, giving it greater strength. But it also
makes the surface more uneven, with valleys between the bundles of
glass fibers. The streaks I saw were the valleys that had not filled
with resin.
My mistake was not heating the layup with a hair dryer to thin the
resin out. I had even prewetted the triax with resin on one side as a
test before taping and it didn't seem to make much difference. As a
second test, I cut out a 3 by 3 inch piece of bid and laid it over the
original streaked tape and out onto the bulkhead, this time using
heat. When I peel the peel ply off this time, the dry streaks were
OVER the bundles of glass, indicating I'd used too much heat and
pressure and had worked too much resin out of the cloth. The valleys
were filled nicely though. Ah well. Live and learn.
One thing I feel fairly certain about is the actual best resin to
glass ratio will depend heavily on the weight and weave of the cloth
and how much work you want to put into thoroughly wetting it out.
The next time I try layups like this, I'll try the other technique he
showed us. In it, he marked off the size and shape of the layup on the
plastic film with a Sharpie pen and cut the cloth to approximately the
correct size and shape. After weighing the cloth, he poured an equal
weight of epoxy directly on the film and spread it evenly with a
plastic bondo squeegee. He then laid the cloth on the spread epoxy and
worked the epoxy into the cloth. When the glass was completely wetted
out, he used a razor blade to cut both the cloth and plastic film to
the marked line, leaving a layup ready to be used.
Within the next couple of weeks, I'll be ready to skin my left wing,
so I'll be reporting again on how well the layup rollers do when
working on (very!) large areas of glass.
David Parrish
