epoxy weakens at moderately high temps?

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

messydeer

Member
Joined
Mar 17, 2010
Messages
20
Location
Bellingham, WA
I've been looking for the best glue and moisture barrier for wood. Planning on building an Osprey. Below is from the Falco construction manual dated 2002.

'The principal limitation is that epoxies soften with heat, and the room temperature cure epoxies used on boats and homebuilt airplanes have poor performance at elevated temperature. Tests by Bellanca Aircraft showed that the two most popular epoxies used by homebuilders, Chem-Tech T-88 and FPL-16A, begin to soften at 125°F, have about 25% of their original strength at 150°F, and have a shear strength of only 40 psi at 175°F—about the same strength as library paste. (The epoxy industry is notorious for its lack of ethics in sales literature: Chem-Tech's literature for T-88 claims a shear strength of 1,000 psi at 180°F. Don't believe any claims unless you have run tests.) Tests by others have indicated that the West System epoxies have slightly better performance...And what does Gougeon Brothers say about our position on epoxy? The head of the testing department says that our position is exactly the same as theirs: that concerns about the temperature performance of room temperature cure epoxies are valid even though they are not aware of anyone having problems with white-painted epoxy-built wood aircraft. They know from tests that the temperature performance of West System is better than high viscosity flexible epoxies (such as T-88). They have not run enough tests at elevated temperatures to chart the performance of their epoxy system. They are hard at work developing an epoxy system with better temperature performance, but it is a difficult task. The reactive diluents which are needed to thin the epoxy for good penetration into the wood fibers lower the temperature performance, and the additives that raise the temperature performance have undesirable characteristics.'

This surprised me since I'd thought the most popular glue was T-88. Comments?

Dan
 

DangerZone

Well-Known Member
Joined
Sep 5, 2011
Messages
2,183
Location
Zagreb HR
Been using West Sytem epoxies cause these showed to be the best in harsh boat environments of the Adriatic for more than 50 years. They are somewhat more expensive than others but when spending hour and hours to build something it would be a shame to see it go to waste just because of trying to spare some money on good quality adhesives.

However, you have to know your work place when building. In time you moght start to chose days when working on your project. If the weather is moist, I skip epoxying and go to a bar with friends instead or prepare/cut/arrange stuff around the workplace. If it is warm and dry, then it's 'glue time', the penetrations showed to be deeper and the bonds are stiff solid.

However, that's just me, I've never worked with this T-88 epoxy and maybe some have different experiences. I was offered discounts by some to use other cheaper epoxies with good test specifications but I'd rather stick to what I know and understand well. On a hot summer day, you could fry an egg at an epoxied boat deck made of dark wood, so this might answer this thread's topic.
 

dino

Well-Known Member
HBA Supporter
Joined
Sep 18, 2007
Messages
698
Location
florida
Thousands of sailplanes have been built with MGS epoxies. They publish data sheets that to my knowledge have not been refuted.

Dino
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
15,868
Location
Memphis, TN
Contact Kevin Kimball that does the Pitts M12 and he will tell you why he recommends the T-88 for his kits. His shop has also built and restored about 200 airplanes. He uses resorcinol on the certified ones but wishes to use the T-88 if it did not require hoops. If you have a heat worry, you can use the resorcinol in those areas.
 

dcstrng

Well-Known Member
Joined
Oct 17, 2010
Messages
913
Location
VA or NoDak
T-88 seems to just about own the homebuilt market unless the builder has some marine experience (where T-88 is usually know by its System Three family name). In the marine world my personal observation is the WEST System is more prevalent with the few commercial wood boat builders I’m aware of (along the Chesapeake).

I use WEST for almost everything and never have had a failure (from plywood water tanks to aviation…), but on one of these forums someone posted that although WEST is designed as a laminating resin, T-88 is superior as an adhesive – not being an engineer (or even knowing one) I have no idea whether that is accurate, but both T-88 and WEST clearly have a pretty good track record assuming the joints/mating-surfaces are anywhere close to appropriate.

I’ve never tried it with the System Three folks, but with WEST I could email and get answers to almost anything – they have bunches of pdfs (most on their website, but some unpublicized) and will talk on the phone, so I image System Three would as well…
 

djschwartz

Well-Known Member
Joined
Jun 21, 2008
Messages
982
Location
Portland, Oregon
First of all, any of these epoxies are, if properly mixed and applied, dramatically stronger than the wood they're gluing together. The glue can lose a substantial fraction of its strength and still result in a joint in which the wood will fail before the epoxy. This is a very different situation from a composite structure where the resin is being used to hold together very high strength fibers of glass or graphite. In those cases any loss of strength in the resin can translate directly to a loss in strength of the structure.

My Stevens Akro wing was built with FPL-16 over forty years ago. I am building a new wing because the main spar is cracked from excessive load. It is the wood that has failed. Even after all this time any attempt to break a glue joint results in failure of the wood on either side of the joint.

Dave
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,767
Location
Rotterdam, Netherlands
Thousands of sailplanes have been built with MGS epoxies. They publish data sheets that to my knowledge have not been refuted.

Dino
Indeed. Most of these kind of statements are little more than blanket statements.

Without information about post-cure and the exact resin used, even testing is worthless.

Sailplanes have to do all structural tests at 54C, which is equivalent to 125F air temp and soaking heat up for hours on a hot day (white skin obviously). Dark red or blue can heat up to 200-220F in the same circumstances.

Without a post-cure any epoxy will be worthless at elevated temperatures. Most structural ones are post-cured at 60C.

I'm pretty sure most homebuilts survive because they get their heat-treatment in the sun, sitting on the tarmac. Not the way to go (sag, uncontrolled), but usually works fine in the real world.

Any color other than white or very light colors are a no-go unless you do a post-cure though...
 

BJC

Well-Known Member
HBA Supporter
Joined
Oct 7, 2013
Messages
14,007
Location
97FL, Florida, USA
I'm pretty sure most homebuilts survive because they get their heat-treatment in the sun, sitting on the tarmac. Not the way to go (sag, uncontrolled), but usually works fine in the real world.Any color other than white or very light colors are a no-go unless you do a post-cure though...
I've heard of people putting wings and fuselags of composite homebuilts in a black plastic tent under the sun and monitored the temperature trying to get a good soak. 160F could easily be reached that way.

Anyone here done that?


BJC
 

messydeer

Member
Joined
Mar 17, 2010
Messages
20
Location
Bellingham, WA
Ah, yes.... Post curing. I really work with fire her class so vergara this item. Checkout the link and search for post cure on that page. Below the link is an excerpt.
http://www.systemthree.com/sys3news/category/technicalresources/

Resin Properties
In foam or balsa cored epoxy/fiberglass laminate boat the skins must take the load – the core has little strength or stiffness. A load on the hull (falling off a wave, for example) will generally put the loaded side laminate in compression and the opposite side in tension. Like rope, reinforcing fabrics are great in tension but poor in compression. It is the epoxy resin that must take the compression load. So, the first requirement of the epoxy here is that it be very stiff so that it does not buckle in compression. Most wood boat building epoxies have sufficient compressive strength for this when at room temperature. However, they rapidly loose stiffness and compressive strength as the temperature rises and, finally, about 145ºF they are totally inadequate and a sufficient compressive load will result in a catastrophic failure – first on the compression and then on the tension side. The laminate literally blows apart. So, the second requirement for this type of construction is that the epoxy maintain adequate compressive strength throughout the expected operating temperature range – generally considered to be up to about 175ºF (higher if the boat will be painted a dark color.)
As the temperature of a cured resin is increased the modulus (a measure of stiffness/resistance to bending) decreases. At first, the decrease is very gradual. As the temperature increases the rate of modulus decrease accelerates. Finally, at some point the rate of decrease hits a maximum and then begins to diminish with further increases in temperature. If one were to graph this with modulus on the vertical axis and temperature on the horizontal axis, both increasing as one moves away from the origin, a reverse “S” shape curve would result. At the point of maximum modulus rate change the curve would change from concave downward to concave upward. The lower temperature end of the curve is where the resin acts as a glass-like material while the resin acts as a rubber-like material. The point of maximum modulus change is called the Glass Transition Temperature (Tg pronounced as “tee-sub-gee”). (This is an important concept so take a break now, reread this paragraph and draw things out.)
 

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
5,192
Location
capital district NY
Just wondering how much strength the wood looses at elevated temperatures as according to 43.13 it looses 25% of it's strength at 125 degrees F so at temperatures concerning the glue's strength perhaps the wood is substantially weakened as well? Effects of temperature on wood are discussed starting on page 4-36 of the link. https://www.fpl.fs.fed.us/documnts/fplgtr/fplgtr113/ch04.pdf

image_2021-08-30_190532.png
 

AeroER

Well-Known Member
Joined
Oct 6, 2021
Messages
53
Elevated temperature and water absorption degrade the strength of epoxy, post cure or not. However, post cure improves the situation.

This is sufficiently important that it is Incorporated into the material allowables for high performance structures in fighters and airliners. It is not negligible.

Take a peek at the temperature placard in the spar carry through in an Extra aerobatic airplane if you get a chance. I was surprised the first time I saw the placard and temperature limit.

The problem with epoxies is the humongous number of products on the market, ranging from crap to remarkably good. As a professional, I would never apply blanket statements about their strength compared to wood. For every example that confirms the claim, another can be created to refute the claim.

Orville, Wilbur, and Glenn probably would have used epoxy given a choice.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,491
Location
Saline Michigan
Who's zooming who here? This is sort of like folks pointing out that Yellowstone is a volcano, like it could go BOOM tomorrow...We have all this huge history of using the stuff and airplanes are not falling from the sky with failed epoxy structures.

The epoxies we typically use in joints bonding wood parts is way stronger than the wood. This is true at room temp, low temps, and high temps. Does epoxy weaken with time and temperature? Sure. So does the wood, and all of the approved adhesives for wood. Properly designed, the parts are still stronger than they need be. Don't believe me? Make some sample glue joints with typical airplane materials - Spruce, birch plywood, mahagony plywood, etc. Store them at 0F and 125F and then break them. They will pull base wood from the joints with boring monotony.

Get into composites and the needs of the resin change a little depending upon what it is used for:

Laminates are stiff and strong because of the fibers in them, with the cured resin holding them in alignment with each other. Do a mesh analysis of stiffness or strength without even considering the resin and you will be pretty close in most directions. The one exception is unidrectional fiber laminate in tension across the fibers - at 90 degrees to the fiber direction. Then we have strength that reflects the resin strength. But no one comes even close to putting large tension across unidirectional laminates.

As for adhesives bonding together pieces into larger assemblies, well, let's just think about that. Most of us use seriously overbuilt joints. Min factor of safety is around 8 and most of us get way beyond that. For instance, my outer wing skin panels carry about 5400 pounds of lift, and the faying surfaces for that skin are about 1200 in^2, for an average loading of 4.5 psi on the adhesive. Yeah, the airloads pulling up on the skins forward of the spar are more like twice that, but even conservative estimates of mechanical bond strengths are 1000 psi, and if you sand the faying surfaces with 220 grit just prior to bonding, you can get 4000-5000 psi through some chemical bonding. Are you really concerned about the epoxy strength going down to even 500 psi when the live loads are 10 psi?

When you find birds disassembling in flight, call us. I predict the presence of incredibly bad design/construction... The ones I have heard of includes one fiberglass airplane bonded together with all of the peel ply still in the airplane and a wooden airplane with continuous ribs and the spar shear web discontinuous at each rib. Darwinists can easily predict the outcome when this level of behavior was applied to airplane building.

Billski
 

mcrae0104

Well-Known Member
HBA Supporter
Log Member
Joined
Oct 27, 2009
Messages
4,152
Location
KBJC
As for adhesives bonding together pieces into larger assemblies, well, let's just think about that
Since we're talking about wood construction here, I'm glad you drew the distinction between epoxy used as a composite matrix vs. as a bond between assemblies. Let's take your wing skin example, and translate it to wood. This suggests that we should not be concerned with the exterior color of the wing skin, so long as the epoxy strength is sufficient at the projected elevated temperature. (Of course the faying area to wood ribs might tend to be smaller than foam ribs.) Am I right so far?

On the other hand, with a composite wing such as yours (bonded skins, not Rutan-style) we want to keep it white, not because we're concerned about the skin-to-rib bond (which will tend to be sufficient even at elevated temps) but because we don't want to heat the matrix of the upper spar flange to the point of insufficient strength? Or are you saying that we're not even concerned about that, since the strength of the spar comes from the UNI fibers, loaded along their axis, and not the matrix, in which case, we don't need to concern ourselves with paint color nearly as much as we do?

I hope these questions are not too elementary--and thanks for your patience, Billski--but I'm trying to draw out some clarity here (and maybe save myself from needlessly being a resorcinol Chicken Little on my own project). Even after all the discussions we've had on this topic, your post above seems to have turned on a light bulb in my head. Sometimes I am a slow learner.
 
Last edited:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,491
Location
Saline Michigan
Since we're talking about wood construction here, I'm glad you drew the distinction between epoxy used as a composite matrix vs. as a bond between assemblies. Let's take your wing skin example, and translate it to wood. This suggests that we should not be concerned with the exterior color of the wing skin, so long as the epoxy strength is sufficient at the projected elevated temperature. (Of course the faying area to wood ribs might tend to be smaller than foam ribs.) Am I right so far?
Sort of. My friend, perhaps you should make a post with your theories, instead reading into my comments so strongly. LOL.

I am talking in terms of keeping within what already we know works... not stretching the envelope. This thread was started by someone trying to tell us that wood and epoxy is at best really limited because the epoxy becomes less strong at the upper end of the temperature range we work in with little airplanes. Harrumph! We have over 50 years of successful wood and epoxy airplanes, and another 30 or so years with resorcinol adhesives before that. And yes, they all get less strong as temps go up, but so does the lignen that holds cellulose fibers together as wood. And they all do just fine when designed and built as we know how to build them.

On top of that, most of our wooden airplane structures have skins that are overbuilt for carrying torsion and bending loads (the job of skins) because they must be sturdy to handling and airshow morons and because they are designed to be stiff enough to prevent whole wing flutter. The temp rise is not a terribly big deal with wooden wings, partly because there is quite a bit of heatsink in the plywood, and partly because the working stresses are pretty low. Then we taxi about, and takeoff flying at 1 g with fast moving air cooling the sun warmed side from whatever it was back towards ambient. I sure would not worry over the epoxy much. All the same, I also would not paint the top surfaces of any wooden wing flat black or go after taking every little bit of weight out of those wings I could.

On the other hand, with a composite wing such as yours (bonded skins, not Rutan-style) we want to keep it white, not because we're concerned about the skin-to-rib bond (which will tend to be sufficient even at elevated temps) but because we don't want to heat the matrix of the upper spar flange to the point of insufficient strength? Or are you saying that we're not even concerned about that, since the strength of the spar comes from the UNI fibers, loaded along their axis, and not the matrix, in which case, we don't need to concern ourselves with paint color nearly as much as we do?
I suppose it is possible to have a resin with a Tg of 140F that will move a little when you park the airplane in full sun on one of those 100F windless days, but I doubt even that will move much. The internal stuff is lower temp, usually close to ambient temps, while mostly the sun warms the outer skin laminate and the bond lines at the spars and leading edge, while the rest of it is still quite cool and sturdy under just gravity pulling on stuff. So it won't move much even if an IR thermometer reads the paint above the published Tg of the resin.

In composite wings, we again have all sorts of history - more than 40 years of successful airplanes made of synthetic fibers and resins - but we stay white or at least pastel hues for most homebuilts. We keep the colors white (for polystyrene foam cores) or very light (for PVC cores) to keep from collapsing the foam cores. There are epoxies that have Tg high enough to stand more absorbent paints, but unless you are building with Nomex honeycomb cores, our usual cores won't stand dark colors, 100F days, and full mid-day sun.

We hear talk of post curing. If your resin post cures, and it spends three years sitting around while you do the rest of the build, most of the molecular loose ends will have found other loose ends and effectively post-cured anyway, but a nice warm day in a black visqueen tent out in the sun ensures that the parts are good to whatever temp you took them too. And post curing in this manner cooks the whole assembly, so it it is stronger at temperature than it would have been otherwise. Then we taxi and take-off at 1g and the skins come down to ambient temperature pretty darned quick. No Eagle Team snap rolls on takeoff, OK?

The next point goes back to composites (and wood too) being bundles of fibers held in formation by glue. To put some scale on all of this, we design our composite structures to have min FOS of 2.0. Let's say it is a 4.2g design, so at gross weight, the weakest part of the structure is designed to carry 8.4 times the level flight loads on the parts. If we designed our wing to right up there at the edge, the weakest spot on the wing has the fibers at 12% of first fiber failure load in 1 g. Sitting on the ground, we are not even that high... Next thing to know is that first fiber failure in most composites is somewhere around 1% strain (some more, others less, but closer to 1% than to 2%) while failure in the cores and epoxy happens up around 5-8% strains. These structures are all designed to move together under load, so long before we can fail cores and resins, the fibers have given up. Now back to our structure taking a hot soak. If the fibers in the worst spot on the airplane are around 12% of first fiber failure stresses, and the cores and resins will stand say six times as much strain before they fail, that means the cores are sitting there loaded to about 2% of their first failure. That is pretty darned lightly loaded to be freaking out about.

Is it real? Yes. Can it matter much? Yes. Have we been doing airplanes successfully this way? Yes. How? Stick to our known best practices and be happy. If you really want to build a TIE Fighter replica, paint it black, and wear a Darth Vadar helmet when you taxi to the ramp, well, you might need Nomex cores, high temp resins, and cooled underwear.

I hope these questions are not too elementary--and thanks for your patience, Billski--but I'm trying to draw out some clarity here (and maybe save myself from needlessly being a resorcinol Chicken Little on my own project). Even after all the discussions we've had on this topic, your post above seems to have turned on a light bulb in my head. Sometimes I am a slow learner.
Build your wooden winged airplane with System Three or West System, use the covering systems per the manuals, and don't go all Batman on the paint colors, and we know you will have a sturdy and exciting airplane. Likewise build and paint the way we know works depending upon your cores and resin system, and again, be happy with a sturdy and exciting airplane.

Billski Out.
 

mcrae0104

Well-Known Member
HBA Supporter
Log Member
Joined
Oct 27, 2009
Messages
4,152
Location
KBJC
Sort of. My friend, perhaps you should make a post with your theories, instead reading into my comments so strongly. LOL.
Fair enough, although I hesitate to post theories until I'm fully convinced of them myself. 😂 Theories are the white lies we tell ourselves until we really grasp reality...

Often, without a complete graduate-level education on a topic, folks treat new information by testing it against what they (think they) know, and then look for the areas of incompatibility. If the inconsistencies can't be reconciled, sometimes they choose to throw out the new (good) information because they already "know" another set of (poor) facts. Thanks for indulging my testing/"known" fact reconciliation process--you've broadened my understanding and now I need to throw out a few things I thought I knew. Excellent post.
 
Last edited:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,491
Location
Saline Michigan
It is easy to see that we have fibers and resins and cores and then imagine them to be links in a chain with the weakest one limiting us. False picture.

The reality of any structure of cores with fibers and resin through it is more like three ropes holding up a weight, with one rope a soft material, and a rope on either side of stout wire bundles, and then paint the whole thing. Add or subtract weight, and all three move together, but they almost move the same as if the core and paint were not there. The core and paint mostly go along for the ride while the stout wire bundles elastically carry the loads.

It is not that this stuff is easy, it is kind of tough to get your brain around at the beginning. But once you get that the pieces all strain together under load, you will see that the load is distributed per the stiffness of the components. Resin, whether it be epoxy or phenolic or lignen, is soft compared to the fibers. Get to plastic foam, and that stuff only matters at all because we have so much of it.

We can do tensile numbers. Let's imagine a 1/2" foam core with 0.010" facings of unidirectional glass and epoxy of equal volume to the glass.
Glass - A = 0.020 in^2, E = 10e6 psi, AE = 200,000;
Resin - A = 0.020 in^2, E = 0.50e6 psi, AE = 10,000;
Foam - A = 0.5 In^2, E = 20e3 psi, AE = 10,000

Total A = .54 in^2, AE = 220,000. All that foam and resin added only 10% more stiffness than the glass alone. The glue keeps things in formation, while the core is there to spread the fibers out for bending stiffness.

Sitting on the ground, a well designed airplane has stresses at a tiny fraction of strength and we avoid colors that get us up near Tg, so the gadgets should be OK. Over 40 years of airplanes made this way and experience fits in with the theory... This stuff really works!

Billski
 

speedracer

Well-Known Member
Joined
Feb 4, 2020
Messages
215
I remember when Danny Meyer owned Velocity they painted one of the fleet orange and called it The Totally Orange Velocity. They said that after sitting in the hot sun they would try not to exceed 1 G till it cooled down. Also, Dick Rutan painted his Long EZ (the second customer built one) light blue. Rumor had it that when Burt saw it he fired Dick - LOL. I painted both the Long EZ's that I built yellow, the second best color for non heat absorption. White's just kind of boring.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
15,868
Location
Memphis, TN
It would be pretty easy to make test coupons, thermometer, and a heat gun. On models, I have softened epoxy to pull wood joints apart and used heat to get them to cure. You end up putting a lot of heat in the surrounding wood to get the epoxy to let go. A lot more that I would feel comfortable for a real airplane.
 
Top