Conclusions on Aluminum Adhesive Bonding Tests

[blakmax]

Hi Gary

Sorry about the lengthy PPRuNe thread. I really wish I did have the time to filter it. Essentially much of what I said is in DOT/FAA/AR – TN06/07, Apr 2007 BEST PRACTICE IN ADHESIVE BONDED STRUCTURES AND REPAIRS which is available through the FAA Tech Center Library. I was the primary author of that document. US residents should be able to access it on line, but we foreigners can not access .gov sites because of homeland security limitations.

The reasons I favoured the AC 130 system are that it comes pre-measured, it is commercially available, its packaging reduces the risk of contamination and it works with most epoxy systems on a number of metals and alloys. I don't have any specific data for 6061, but I do know it works on 2024, 7075 Titanium and some stainless steels. I urge people to grit blast rather than hand or machine abrasion.

I have a lot of experience with the Australian grit blast and silane coupling agent system which performs very well. The AC 130 system uses the same silane but has other additives which do make some difference. With the GBS system, we were able to reduce the repair in-service failure rate from 43% in 1992 to only three bond failures out of well over 4000 repairs since 1992.

By the way, lap shear tests and peel tests are not sufficiently discriminating to use for qualification of surface preparation processes. The best accelerated test is the wedge test ASTM D3762, BUT ignore the stated acceptance criteria. Use the ones in the reference above.

Regards

blakmax
​

 

[BBerson]

I think I read in the PPRuNe thread that Araldite was originally developed with phenolic resin. And it was mentioned that phenolic has a better bond than epoxy on aluminum.

If that is the case, why not use phenolic?

 

[blakmax]

Phenolics are good adhesives. The Fokker F27 used Redux bonding and never had problems and that was based ona phenolic adhesive system.

The reason phenolics are not popoular is that they are condensation polymerisiers, compared to addition polymerisers. In addition polymerisiers part A plus Part B make Part C. In condensation polymerisiers Part A plus Part B make Part C plus Part D which is a byproduct of the reaction and that is usually water. Because the adhesives require heat to react, the water turns to steam and results in voids in the bond. The way to minimise the effect is to apply very high pressures to crush the voids and that means it will be an autoclave process. Vacuum bagging will only increase the size of the voids.

Regards

blakmax

 

[Dana]

Couple of thoughts: People mention differences between 6061 and 2024, but in most cases 2024 skins are Alclad which is pure aluminum... not sure how that affects the cleaning and/or bonding process.

Second, when bonding to a low strength material like plastic foam, how critical is all this? Are any of the likely adhesive failure modes likely to be weaker than the foam material itself?

One other concern I have for my own application (aluminum capstrips over foam ribs) is the effect of the covering system (probably Stits) chemicals on the foam. One thought is to run a strip of some sort of tape over the corners of the foam shapes before bonding the capstrips on. This would provide some protection to the foam up against the surface, and also space the capstrips out by the thickness of the tape, slightly increasing the bond line thickness in a similar manner to scrim.

-Dana

Ask not what you can do for your country, but what your country is doing to you.

 

[GESchwarz]

Of course the foam is the weak link, therefore not as critical. It is critical if you're going after maximum strenght. Also, because the foam does such a good job of spreading the load, you have less risk of stress concentrations, which is where any structure begins to fail.

The difference between the two alloys is that 2024 is less resistant to corrosion and therefore may pose a greater risk of debond.

 

[Dana]

Yes, 2024 is less corrosion resistant, but my point was that 2024 is usually used in Alclad form with the coating of pure aluminum which is much more corrosion resistant than either 2024 or 6061.

I guess my real point is, what is the best adhesive system to use that's at least as strong as the foam and reliable, and is the special preparation necessary in this case? If the high strength isn't required, a less expensive and/or easier to apply adhesive would be the right choice.

-Dana

Give a man a fish and you feed him for a day. Teach him to use the Net and he won't bother you for weeks.

 

[GESchwarz]

Well that's an excellent point. I've actually been thinking the same thing. I was thinking about mixing the adhesive with micro balloon filler. I would imagine that the strength would go down proportionately, but that's adhesive strength is something we can afford to sacrifice if we're bonding to foam. What cannot be sacrificed is adhesion strength. I think that Proseal is a safer bet than any epoxy just because of epoxy's brittleness.

I don't know... I think I might go the micro balloon route to double my adhesive volume per dollar of epoxy.

I would not play around with the prep... I'll look into the AC130, but I wouldn't do anything less than what I've recommended already.

I too plan on using foam in my wings. 100h density PVC cap strips, of trapezoidal cross-section, over aluminum webs, forming an "I" channel. The rib spacing will be a bit tighter than conventional riveted construction as an added safety factor.

 

[blakmax]

Garry you wrote:

I would not play around with the prep... I'll look into the AC130, but I wouldn't do anything less than what I've recommended already.

I can assure you with absolute certainty that if you rely solely on scuff sand and solvent clean, you will eventually experience disbonding. You must use a process to develop resistance to hydration such as AC130, grit blast and silane or phosphoric acid anodising. Let me stress that if the structure is to carry ANY load, you must do this because the eventual strength of the bond without adequate preparation will be dead set zero. If your design can manage a zero shear strength, I'd like to see the calculations! There is no bond which is so lightly loaded that effective surface preparation can be ignored.

On another point, many epoxies are not brittle but are in fact quite ductile. There is reference material which shows that up to 80% of bond strength comes from ductile behaviour.

I am not familiar with "Proseal" but if it is as the name suggests a sealant rather than an adhesive, then you must recognise that while such materials have a high shear strain to failure and therefore are quite resistant to fracture, they also have a very low shear modulus which means that their load transfer capability is also quite low so they may result in quite flexible structures. I would urge caution in selecting such materials unless the design has been experimentally validated. It really is a question of the structure staying together but being a floppy system or using an actual structural adhesive

 

[arcticserv]

Hi, has anyone considered loctite h8600? I spoke to them today and that is what they recomended for aluminium to aluminium bonding.
Lokking at the speck sheet it is exceptionally strong and flexible, and improves a little in strength as it gets warmer from enviroment influence.
It does say that its for gal steel but the rep assured me it is fine for aluminium as well.

 

[GESchwarz]

blakmax,

I appreciate your words of caution. I'm not sure that your concern about hydration of the bond and delamination due to corrosion is as much of a concern to me. I live in sunny Southern California. Although it can pour down rain here, it's pretty bone dry on about 330 days of the year. And even if there was a source of moisture that could pose a threat, what is the method of ingress to the bondline? Is not the epoxy a fairly effective sealant against airborne moisture and an occasional puddle due to condensation? I will be taking measures to ensure adequate protection against water getting to the joints, and drainage away from the joints.

No hydration = no corrosion = durable bond joint. Isn't that right?

 

[Starman]

I think a little humidity is enough for hydration and from what I've heard of composite boat building, it will soak into epoxy.

 

[Tom Nalevanko]

Hydration will happen at the molecular level. Raincoats will not keep it out... And it may not take much.
Blue skies,
Tom

 

[GESchwarz]

The arguement that even slight amounts of moisture can cause corrosion is incongruent with the fact that paint on aluminum surfaces manage to prevent corrosion over the course of many many years, with no evidence of delamination or evidience of corrosion following paint strip. If paint doesn't delaminate, why must epoxy?

Although it may not take much moisture to cause corrosion, it doesn't take much to keep it out either.

 

[Tom Nalevanko]

There seems to be no logic in your argument. There are an awful lot of corroded painted aircraft surfaces out there. Just walk around the ramp at KCMA and you will see plenty right in dry SoCal. Look for raised and flaked paint. And there is essentially no stress on that paint.

Remember that there are water vapor molecules in the air during your painting/bonding process. Some people have tried to scuff their aluminum in a bath of epoxy and then go on to bonding without the piece "seeing" air/water vapor. Didn't help with the intro of the water molecules however with subsequent debonding. The water molecules are active and will get to the bond line as the epoxy absorbs water vapor over time.

It would be a good idea to read the reports and documents that blakmax cited.

 

[MadRocketScientist]

... Didn't help with the intro of the water molecules however with subsequent debonding. The water molecules are active and will get to the bond line as the epoxy absorbs water vapor over time.....

Doing the wet epoxy scuff won't help the bond at all if the epoxy itself doesn't bond to the aluminium by itself. I have always been under the impression that aluminium always needs a primer coat to have a good join. My suspicion is that even with the wet rub and without any moisture the epoxy can still break away from the aluminium due to insufficient bonding. Add to the the ability of epoxy to absorb moisture and the join would be certain to fail sooner or later.

The CriCri I am building uses aluminium/aluminium and aluminium/foam bonds throughout the structure. The 150+ CriCri that have been built use a manual scuff and primer for the join preparation and I have yet to hear of delamination issues with them. I am not entirely convinced that the process needs changing for my aircraft and will be using the plans process unless convinced otherwise.

Shannon.

 

[MadRocketScientist]

There seems to be no logic in your argument. There are an awful lot of corroded painted aircraft surfaces out there. Just walk around the ramp at KCMA and you will see plenty right in dry SoCal. Look for raised and flaked paint. And there is essentially no stress on that paint.

I wonder if the corrosion in SoCal has something to do with the overly dry air. Wash primers need moisture to convert their acids to the phosphate film when curing. If there isn't sufficient moisture in the air during the cure the acid remains under the paint to start corrosion. This may have caused the corrosion seen under the painted surfaces. If the corrosion is starting at the seams and around the rivet heads it would suggest the paint layer is fracturing and allowing the moisture to enter there.

Shannon.

 

[Topaz]

....The CriCri I am building uses aluminium/aluminium and aluminium/foam bonds throughout the structure. The 150+ CriCri that have been built use a manual scuff and primer for the join preparation and I have yet to hear of delamination issues with them. I am not entirely convinced that the process needs changing for my aircraft and will be using the plans process unless convinced otherwise.

Yep. I've yet to see this point adequately addressed in this thread. I can think of a half-dozen homebuilt aircraft types (Cri-Cri, several by John Monett, and the entire Schreder "HP" line of homebuilt sailplanes) that use adhesive bonding of aluminum to foams, composites, or to other aluminum parts - all with scuff-and-solvent-wipe (and/or primer) prep. The Cri-Cri's are often flown aerobatically.

With all due respect to the experts here, Moni's, Monarai's, Cri-Cri's, and HP-11, HP-14, HP-16, RS-15, and HP-18's are not falling from the sky with debond failures, and many, many examples have very significant service histories. A blanket dismissal of the technique flies in the face of reason - all evidence points to it being perfectly adequate for these relatively low-stress applications.

If scuff-and-solvent-wipe is "completely inadequate" and will "always" result in bond failure, why are we not seeing structural failures in these aircraft types after multi-decade service histories? If it turns out that you'll "always" see a bond failure but it takes sixty years for the breakdown to happen in these types of applications, how is this discussion any more than academic? There's a disconnect here somewhere, and I'd like to hear more explanation than vague generalities about "they are failing, but maybe just not bad enough yet" before we're encouraged to abandon a technique with good history of service.

 

[Topaz]

I wonder if the corrosion in SoCal has something to do with the overly dry air. Wash primers need moisture to convert their acids to the phosphate film when curing. If there isn't sufficient moisture in the air during the cure the acid remains under the paint to start corrosion. This may have caused the corrosion seen under the painted surfaces. If the corrosion is starting at the seams and around the rivet heads it would suggest the paint layer is fracturing and allowing the moisture to enter there.

Shannon.

I suspect corrosion here has more to do with ozone than moisture, at least initially. Surface ozone levels in CA (especially southern CA) are quite a bit higher than most other areas of the country. It's our much-vaunted smog. Polymer-based auto parts used to degrade very quickly here until the auto industry figured it out. You can still see ozone checking on auto tires if someone doesn't drive very often, and has more years than miles on the rubber. Smog has gotten much better here in the last twenty years, so it's not as bad as it used to be, but an airplane out in the sun since the '80's will have gotten the full brunt of both smog and sunlight. The stuff is quite a bit more reactive than regular oxygen, and will attack metals that much faster.

On a painted metal surface, the strong UV and ozone in this area attack the paint, allowing ozone and moisture can get through the paint barrier and start attacking the metal underneath. Worse, the degraded, cracked paint starts acting as a moisture trap (almost like a sponge), retaining moisture against the metal surface much longer than if it was unpainted. Once oxidation on the metal surface begins and pushes up the paint, the moisture and chemical ingress spreads and you get expanding areas of bubbled paint and corroded metal underneath.

 

[Norman]

The stuff is quite a bit more reactive than regular oxygen, and will attack metals that much faster.

There's alkali in the dust out there too, not to mention plain old sodium cloride

 

[Topaz]

There's alkali in the dust out there too, not to mention plain old sodium cloride

I'm not so sure that salt's as big a deal here as other parts of the country. We never see ice on the ground here, so we never salt the roads. And tourism marketing to the contrary, we're not all near the beach.