Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Gelcoat Crazing Treatment

145 views
Skip to first unread message

joe volmar

unread,
Jan 17, 2002, 10:18:42 PM1/17/02
to
I am in the process of refinishing a 30 year old fiberglass boat which
has obvious gelcoat crazing. I have discovered a two-part polyurethane
primer specifically designed to fill gelcoat cracks and will be
applying this primer (without removing the damaged gelcoat) prior to
my two-part polyurethane finish coat. Having just been involved with
paying for a $16,000 refinish job on our club's ASK-21, I would like
to test my procedures on a glider gelcoat for the benefit of all those
owners out there who are observing these nasty little cracks. I am
looking for a specimen part of a glider with gelcoat crazing ....does
anyone have a discarded part that I can experiment with? This will be
a long range program, as I intend to place this refinished specimen
out of doors year-round for some time to test the longevity of the
finish. I will, of course, be observing the condition of my refinished
boat at the same time.
My test is being supported by Interlux Yacht Finishes and will be
followed with interest by several certified sailplane composite repair
stations.
I realize that the only 100% cure for gelcoat crazing is complete
removal and replacement, but if this system lasts for a number of
years, it may be a viable and less expensive solution.
Please let me know if you can help; any glider part will suffice:
aileron, empennage members or a part cut out of a damaged fuselage or
wing.

Joe Volmar, Adrian Soaring Club (Michigan, USA)

John Galloway

unread,
Jan 18, 2002, 6:12:10 AM1/18/02
to
Joe,

I think that to establish the validity of your procedure
for gliders you would have to include a flexing regime
for the test component and not rely simply on exposure
to the elements. Glider structures are subject to
a lot more flexing than boats are.

Also the extra weight and of a primer and then 2 pack
PU finish on a glider without first removing some gel
coat would be a more sensitive matter than in a boat.

John Galloway

joe volmar

unread,
Jan 18, 2002, 9:10:38 AM1/18/02
to
John Galloway <REMOVE_TO_R...@hotmail.com> wrote in message news:<a28vua$viucs$1...@ID-49798.news.dfncis.de>...

> Joe,
>
> I think that to establish the validity of your procedure
> for gliders you would have to include a flexing regime
> for the test component and not rely simply on exposure
> to the elements. Glider structures are subject to
> a lot more flexing than boats are.
>
> Also the extra weight and of a primer and then 2 pack
> PU finish on a glider without first removing some gel
> coat would be a more sensitive matter than in a boat.
>
> John Galloway
>
>
>
> John:
Your comments are very valid. I had planned to sand off enough old
gelcoat(as evenly as practicable) to compensate for the additional
weight. I don't know yet how to attack the flexing test but will work
out something depending upon the glider part available for the test. I
also plan to subject the item to some extreme temperature changes
during the summer to simulate high altitude conditions. I also should
duplicate low air pressure but do not have the facilities, as this is
more or less a "low tech" experiment.

Joe Volmar

Roger Druce

unread,
Jan 19, 2002, 7:51:30 AM1/19/02
to
A quick comment would be to suggest that you really need to think about what
scale of gelcoat cracking are you proposing to deal with. There is a huge
difference between the onset of the problem with fine initial cracking as
observed with a quality viewer, and full depth gel coat cracks which may
have existed for some time and are blindingly obvious to the naked eye.
Full depth gelcoat cracks can enter the underlying resin matrix and produce
structural cracks. If you just sand back part way in the severe case and
overpaint with some system albeit clever, how do you know what the story is
a the resin/fibre interface with the failed gelcoat, ie that you don't have
structrural cracking?

Nothing I have seen or heard about the gelcoat cracking issue with
sailplanes makes me at all comfortable with the idea of overpainting bad
gelcoat cracking. You could ruin a glider if you don't deal with bad
gelcoat failure by thorough & complete removal and refinishing carried out
competantly. On the other hand there maybe scope for a process which
addresses the other end of the scale of problem, ie. initial gelcoat failure
evident under magnifiacation but hardly visible to the eye except under
critical lighting.

Roger Druce

"joe volmar" <joe...@dundee.net> wrote in message
news:d5f5e963.02011...@posting.google.com...

airgonzo

unread,
Jan 19, 2002, 11:33:39 AM1/19/02
to
I'm curious, What causes you to believe that the gel coat crack can then
continues into the resin/glass,or(carbon fiber)?
I'm not flaming you, I wish to understand your belief.
thanks!

"Roger Druce" <rogdruceDEL...@melbpc.org.au> wrote in message
news:a2bp69$a59$1...@possum.melbpc.org.au...

snip


> Full depth gelcoat cracks can enter the underlying resin matrix and
produce
> structural cracks

snip


JohnPegase

unread,
Jan 19, 2002, 5:29:27 PM1/19/02
to
In article <Ddh28.7436$OS5.6...@newsread1.prod.itd.earthlink.net>, "airgonzo"
<airg...@earthlink.net> writes:

>I'm curious, What causes you to believe that the gel coat crack can then
>continues into the resin/glass,or(carbon fiber)?
>I'm not flaming you, I wish to understand your belief.
>thanks!
>

It happened to my glider, a Pegasus. We had a mid wing mounted wing dolly as
part of our one-man tow out kit. The wing flexed at twice its normal frequency
at this point. The gel was badly crazed, we thought due to trips to high
altitude early in its life. A friend re-gelled it for us at a very low price
and found that the crack went down into the resin. Charged us at cost price
for the glass and resin to repair it. Just painting over after a rough
rub-down could have made life interesting for us after a few more years.

But did the cracks from flexing the wing when cold and move down into the
resin, or up from the resin into the gel due to flexing around this point? Who
cares, have a good look, don't paint over it.


John Wright, 742

Al

unread,
Jan 20, 2002, 4:17:38 AM1/20/02
to
What usualy happens is sunlight can penetrate the cracks and UV decays the
resin matrix underneath
it is usualy not structural.

Al


"JohnPegase" <johnp...@aol.com> wrote in message
news:20020119172927...@mb-cl.aol.com...

Roger Druce

unread,
Jan 20, 2002, 9:18:22 PM1/20/02
to
It has been clearly established that full depth gel-coat cracks can lead to
problems in the resin-fibreglass matrix underneath. Gliding Federation of
Australia Airworthiness Notice 69 issue 3 of June 1987 records two of our
Australian professional workshops in 1985-86 finding clear evidence of crack
propagation from the bottom of the gel coat and through into the top skins
of a Hornet wing, Mosquito wing and Cirrus 75 tailplane.

Samples from the Hornet and Std Cirrus tailplane plus another sample from a
different Std Cirrus were analysed at the Dept of Civil Aviation Materials
Evaluation Laboratory by a materials specialist, and I have the report in
hand. (1986 Report X-5/86, "Deterioration of Glass Fibre Reinforced Plastic
Gliders" by A Romeyn.) Conclusions: Some deterioration of the glass fibre
reinforced epoxy laminates underlying the cracked gel coat was found. The
deterioration was limited to matrix microcracking and some localised
swelling and void formation, there was no evidence of fibre breakage.

Most of the information I have collected is from the 80's decade. The
Germans established a sub-group of their industry-academic Working Group for
New Fibre Reinforced Plastics ("Arbeitkreis Neue Fasserverstarktekunstoffe")
to look at the gel-coat cracking problem in 1987. It would be interesting
to get ones hand on the final outcomes of the work they did on the issue and
any more up to date info from the 1990's.

Roger Druce
Australia

"airgonzo" <airg...@earthlink.net> wrote in message
news:Ddh28.7436$OS5.6...@newsread1.prod.itd.earthlink.net...

0 new messages