bonded aluminum construction

Discussion in 'Sheet Metal' started by Georden, Dec 5, 2006.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Jan 23, 2007 #21

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    Wow! Great stuff. You must've been really immersed in this subject.

    I remember some of this from way back too. Before I got my engineering education I worked on the 747 production line in sheet metal fabrication and machining. The surface prep chem plant was right next to us so some of this is very familiar.

    But having had that experience, and having seen the handling requirements of the materials after they exited the prep site (a finger print in the wrong place was enough to scrap a part), I didn't think that this was very conducive to the light airplane industry, especially to homebuilding. As such, the primary thrust of most of my work in examining bonding agents and surface prep techniques was to eliminate as much of this as possible. I'd like to see this evolve in such a way that only basic surface cleaning would be required of the builder before the bond takes place.

    The most promising bonding agents I've seen thus far are the Methylcrylates (Plexus, Extreme, etc.). From what I've been told and what I've seen in our tests, as it cures the material actually does a bit of its own etching of the surface, negating the need for any significant prep work to be done on the part of the builder. So far, peel tests have demonstrated strengths that exceed some of my epoxy tests by at least a factor of two to three. The Methylcrylates are already used in the automotive industry for everything from bonding body panels to mounting door hinges onto the vertical king-posts. The aluminum structure of the new new Ford GT40 is reportedly all bonded with this material.

    But before I commit to any process, I'd like to see more test data so thus far I've been a bit slow in jumping on the bandwagon. But it does look promising.
     
  2. Jan 23, 2007 #22

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

    Joined:
    Jan 22, 2007
    Messages:
    118
    Likes Received:
    6
    Yes, a fingerprint in the wrong place can really screw up an adhesive bond. Many years ago, I built (but never raced) a car made with aluminum honeycomb panel construction. After a few false starts, the adhesive bonding procedure became routine - cleanliness was essential - and the epoxies cured at room temperature. Very strong, and exceptionally light.

    Methacrylates? I have no knowlegde or experience with them. Self-etching? How interesting! Yet I suspect that their ultimate performance comes on an etched / chromate converted surface. I've read that the Jaguar car is also assembled with adhesives (unspecified) on a production line. If adhesives can be made to work in that environment, then the production engineers have established a rigorous methodology (today, I believe the word is 'robust') for adhesive bonding.

    As far as test data on adhesive bonding, at one time I was privy to it, however, no longer. I do remember seeing very impressive numbers on the Instron for both peel and lap joint testing for a number of epoxy adhesives.

    Yet, I suspect the automotive requirements would also include impact resistance, weathering, oil and fuel tolerance, plus high peel strength - very much the qualities sought for aircraft construction.

    Do you have any links or references to the methacrylate adhesives which discuss its chemical qualities?

    I also have some thoughts on composite panel construction, and ways to aleviate the 'boxy' appearance, while still exploiting the high strength to weight qualities. I'm limited in putting sketches together, having no cad or other skills.
     
  3. Jan 23, 2007 #23

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    A couple of individuals at Extreme Adhesives (www.extremeadhesives.com) seem quite knowledgeable and have been willing to talk in the past, although it's been a while since I last had any dealings with them. But regrading papers and technical documents, no I have not found any useful sources, which is why we've been doing all these tests ourselves.

    Back when I was a member of the SAE (Society of Automotive Engineers) I do remember a number of articles in their magazine that referred to the bonded assembly line, but those discussion tended to concentrate on discussing the advantages rather than any real physical data. But at least that might be a place to start if you're interested in tracking some of this down in more detail.

    Of the few very sketchy technical sources I have run across, the problem I've been running into is that despite assurances of specific bonding properties for most materials, the vast majority of the presented numbers represent bonding in a laboratory environment - something that your average homebuilder does not have. As far as kit airplanes are concerned then, the solution therefore would be to do much of the bonding prior to sending the parts and subassemblies out to the customer. The question though is how much can you get away with. Too much and the FAA will take a dim view of the practice since it can run counter to the 51% rule.

    This is why I like the potential of the Methylcrylates - if they'll deliver the performance with little or no surface prep, maybe in conjunction with a few "chicken rivets", then we'd have a potentially good solution that combines the inexpensive aluminum materials with the redundancy of bonded components.
     
  4. Jan 23, 2007 #24

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

    Joined:
    Jan 22, 2007
    Messages:
    118
    Likes Received:
    6
    Okay, I didn't get anything from Extreme's website, then clicked to a couple that I know. One thing jumped out from www.ipscorp.com info - aluminum does need surface prepping:

    'Prepare metal for bonding by cleaning and removing all loose material and other surface residues including oil and grease. Metal primer should be used to prepare aluminum and stainless steel (for bonding with methacrylate adhesives)'

    Also, they provided strength data for methacrylate adhesives:
    Tensile 2 - 2.5 kpsi
    Modulus 60 - 75 Kpsi
    Lap shear 2.4 - 2.8 kpsh (ASTM D1002)

    This tells me that methacrylate adhesives have about 50% of the strength of epoxies. They do have much higher elongation (up to 200%), which means their impact strength (and glass transition temperature) is probably greater than epoxies. see also www.gluguru.com

    Hmm, why do we find it hard to prepare a surface for adhesive bonding when we are willing to sweat cobs to prepare for a paint job? I believe the surface preparation for adhesive bonding is no more difficult than for a paint job... mebbe.

    Well, if methacrylates proved strong enough in test, then epoxies probably would also. I have no data on the strength of polysulfide adhesives. I know that their forte is as a sealant and binder agent for rocket fuel, but their is a lot of anecdotal comments about how well the stuff sticks.
     
  5. Jan 23, 2007 #25

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    I also did a bit of searching this morning and found links similar to yours. One that really caught my eye was http://www.gluguru.com/Hysol Data Sheets/HPrimer2000.htm. Imagine, just wipe on the surface prep and bond - I gotta get some and try this stuff out. Might work with epoxies and Methacrylates. We'll see.

    The best results I've gotten previously out of epoxy was with the Hysol EA9430, with a peel strength that exceeded the published numbers by about 20% (about 45 to 60 pli - scuffed and degreased surface). Now, my test may not have been exactly per ASTM but it was enough to really present the Hysol products as superior to just about anything else I tried at the time. But the results with the Extreme Methacrylate was in the neighborhood of 80 to 110 pli (bare and degreased standard sheet, and bare and degreased Clad aluminum - 6061-T6 and 2024-T3).
     
  6. Jan 24, 2007 #26

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

    Joined:
    Jan 22, 2007
    Messages:
    118
    Likes Received:
    6
    Yes, that primer is fascinating, and I hope you have a chance to check it out. I'd sure like to hear the results. If you could also test for tensile and lap shear, that would also be very informative.

    I can understand why methacrylate adhesives have greater peel strength than epoxies, primarily due to their greater elongation (stretchiness). However, what is required for aircraft construction - peel strength or lap tensile strength? I guess it depends upon the design of the joint.

    From the little I know about adhesives, I still believe that cleanliness of the bonding surface is of utmost importance for a strong joint. I still remember the basics that the shear strength of oil and grease is far lower than that of any adhesive, which is why all bonding surface must be cleaned.
     
  7. Jan 24, 2007 #27

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    No question - no substitute for a clean bond line.

    Regarding which property is more important, I think the answer is "it depends". About ten years ago a US company imported several European gliders (ten I think) that were constructed using a form of aluminum bonding (I was never able to discover which materials were used or find the bonding spec). The gliders were fully Part 23 legal, including for aerobatic flight. One of the gliders, during a demo flight, was seen to exhibit a small amount of skin wrinkling during moderate G turns. Upon landing, the subsequent inspection discovered that the aluminum bonds in the wing had deteriorated and failed. An inspection of the other aircraft had revealed a virtually identical failure where in all cases bond line failure was evidenced over nearly 70% of the joints (only 30% of the bond was still functional).

    Being curious, the company decided to test the wings in order to see the effect of the deterioration - the wing was loaded to limit, and then to ultimate load. It was actually able to achieve about 10% above ultimate load before total failure - that's with 70% of the bond gone!!!!

    So, given the characteristics of a well designed bonded structure, one can see that there is a substantial amount of redundancy in the load paths due to the high area of contact, and that the bonds are not under high stress, even near the ultimate condition of the service envelope. For this reason I think I'd tend to hedge on the side of toughness rather than pure strength.

    I don't recall why the bond line failed but it is my gut feel that it was as a result of gradual failure in peel, onset by either contamination or service factors. For this reason I'm more interested in edge stability (peel resistance) than pure strength. But this is only my opinion for now - more testing may prove me wrong.
     
  8. Jan 24, 2007 #28

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

    Joined:
    Jan 22, 2007
    Messages:
    118
    Likes Received:
    6
    I'm surprised at the extent of bond failure in the gliders. I'm of the same opinion as Orion that adhesive bonding does spread the loads and reduces stress points or stress risers. Still, a 70% failure is almost criminal.

    I'd also love to learn the cause of the bond line deterioration. Was it due to creeping corrosion? Or, perhaps UV caused degradation? or, was it due to peel from substrate (which might suggest poor surface preparation)? And lastly, if it were due to cohesive failure, then that would suggest improper adhesive selection (cross-linked or polymerized adhesive not having sufficient inherent strength for required role).

    'Peel resistance' may or may not be directly linked to tested peel strength, for the form of joint failure may point to some other deficiency in the adhesive bond.

    Testing candidate adhesives requires a combination of consistently prepared test specimens, a rigorous testing of the range of adhesive properties (using an Instron), plus testing after accelerated environmental aging. Having done this in the past, I recognize that this is easier said than done.

    As for prevention for edge failure in peel, the common suggestion is to install some rivets to hold the edges together. Yep, probably would work. However, as one familiar with adhesives, I would also want to know that the adhesives were up to the job selected for them.

    Consequently, going back to previous experience, I do know that epoxies (amine cured), applied over a properly prepared surface, will hold tenaciously, with tensile strength in excess of 3.5 kpsi, and lap shear in excess of 4.0 kpsi.

    Yes, if I were flying in an aluminum aircraft dependent upon adhesive bonding, I would want to be very sure that it was done properly. And I agree with you about the concept of using a 'foolproof' adhesive that didn't depend upon a tricky surface preparation procedure (unless I did it myself).
     
  9. Jan 26, 2007 #29

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

    Joined:
    Jan 22, 2007
    Messages:
    118
    Likes Received:
    6
    I guess we got onto adhesives in 'Boxy Aircraft...' This is an answer that could also go there.

    The term 'Acrylics' is again another one of those either contractions or same usage term. The correct name for these adhesives is (I pretty sure of this) is polymethyl methacrylate, however, if you say 'acrylic' or 'methacrylate,' I'm sure the adhesive manufacturers will understand.

    Here’s an example of replacing rivets and other fasteners in a demanding truck body application:

    http://www.thefabricator.com/Assembly/Assembly_Article.cfm?ID=1476

    If I read the Lord Corporation data sheets correctly, the above adhesive used in truck body assembly is a two-component polyurethane which can be adjusted for different speeds of cure (open time). The only drawback to polyurethane bonding agents is that the metal substrate definitely needs a primer for ultimate, long-term adhesion.

    As for methacrylates, I hear you (Orion) about the problem of using moisture cure adhesives with metal substrates, and I agree that's not a good idea. Perhaps two-component adhesives - even though more of a complication - are a safer approach. I can speak with experience about bonding aluminum with epoxies, and do know that their ultimate performance is hard to beat.

    Perhaps polyurethanes, methacrylates (two component) and epoxies might fulfill roles in safely assembling aluminum aircraft. Polyurethanes certainly make sense as far as polyester resin fiberglass composites are concerned.
     
  10. Jan 26, 2007 #30

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    Out of curiosity Malcolm, what do you do? Your chemical/bonding process background seems to be quite a bit more extensive than that of a casual user. And BTW, thank you for your in-depth input here - great stuff.
     
  11. Jan 26, 2007 #31

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

    Joined:
    Jan 22, 2007
    Messages:
    118
    Likes Received:
    6
    Orion, as I mention in 'New Members,' I've had three careers and am now retired. With the crappy weather in N.E. Ohio, I'm not getting much flying in at the moment.

    I worked with adhesives, primarily in the research area and focused mainly on polyurethanes and epoxies for a number of years. Then career path changed - fickle finger of fate beckoned... y'know the story, I'm sure.

    BTW - I must compliment you on your willingness to share your obviously extensive experience in aeronautics (if that's the correct term), and your breadth of knowledge about aircraft. Makes this a lively site. I do enjoy it, and I'm learning from it, too. Thanks!
     
  12. Jan 27, 2007 #32

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

    Joined:
    Jan 22, 2007
    Messages:
    118
    Likes Received:
    6
    Okay, I did a little clicking around and have found some items which may be of use to those who are interested in using adhesives in the assembly of aluminum aircraft. First, there is a glossary of adhesive terms (if you end up talking to an adhesive vendor, you’ll use the same words as him and avoid mis-communicating). Also, some adhesive vendors shy away from letting their products get used in experimental aircraft. If you run into that problems, tell them that it’s for an aluminum boat, or you’re building an aluminum race car:

    http://www.conprotec.com/pdf/glossary.pdf

    Next, this is a good description of the three classes of structural adhesives:

    http://www.manufacturingcenter.com/dfx/archives/1003/1003adhesives.asp

    And here’s a link to more than you’d ever want to know about surface preparation of adhesive bonding (comes from England, so, the terminology is a little different that what you are used to hearing). It has methods that even experts haven’t seen before! Blimey!

    http://www.on-hand.com/Manuals/SurfacePreparation.pdf

    Here is a listing of adhesives made by Lord Corporation, some of which have been used for assembling aluminum truck bodies (a demanding environment):

    http://www.lordfulfillment.com/upload/PB3036.pdf

    And another…

    http://www.sikaindustry.com/tds-ipd-sikafast3141-us_rev_4-07-05.pdf

    In the above link, there is reference to the LACK of chemical resistance by methacrylate adhesives to gasoline. Hmm, that could be a real concern in aluminum aircraft assembly. Also checked chemical resistance of acrylics, and found that aromatic solvents do cause damage (softening and swelling). Recognize that some gasolines contain aromatic hydrocarbons which disqualifies methacrylate adhesives in any location that comes in contact with gasoline.

    So, do NOT use a methacrylate adhesive where gasoline can contact it.

    I hope that this proves useful.

    My personal preference for bonding aluminum remains epoxy due to its long track record and preference by the commercial and military aircraft industry. Nevertheless, methacrylates do seem to possess the necessary properties to assemble light aluminum aircraft.

    Do some testing before using.
     
  13. Jan 29, 2007 #33

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    I've also noticed the gas problem with the Methacrylates - in all my work nowadays though I tend to use bladders so hopefully that shouldn't be an issue.
     
  14. Jan 29, 2007 #34

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

    Joined:
    Jan 22, 2007
    Messages:
    118
    Likes Received:
    6
    Hello, Orion;

    the chemical degradation problem in adhesives has been overlooked by many users, and in aircraft that is a dangerous practice. The problem is that most of the vendors tend to skim over such weaknesses, which could cause later problems if the user 'assumes' that it is safe to use.

    Methacrylate adhesives appear to be doing a fine job with structural aluminum bonding for truck bodies, which suggests they should be fine for many aircraft bonding applications. However, if there is any chance of gasoline dripping, overflowing or fumes coming in regular contact with the methacrylate adhesive joint, then other adhesives or joining methods should be employed. Other adhesives or sealing materials with long track records of satisfactory application include epoxies and polysulfides.

    You mention bladders, which I assume you mean for fuel. Do you fabricate these yourself? If so, what polymer do you employ? I have some familiarity with fuel containers, particularly from racing applications, know nothing about aircraft bladders.

    Do some testing before using.
     
  15. Jan 29, 2007 #35

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    The bladders we've used recently come from an Oregon company, however as I am not in my office now I don't have the info on them at hand.

    Having seen all sorts of problems with wet wings over the years, even composite ones, and given the amount of labor it takes to track down and seal every pinhole, I've come to the conclusion that bladders could potentially eliminate much of the headaches, even though the cost can be a detriment. And since virtually all my work is on new structures, incorporating volume and access is relatively simple since it is done during the design process.
     
  16. Feb 1, 2007 #36

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

    Joined:
    Jan 22, 2007
    Messages:
    118
    Likes Received:
    6
    I have mixed feelings about bladders, having seen frequent flecks of 'black rubber' floating in the gasoline from aircraft tanks (Cessnas primarily). I've also been warned that bladders can fold or buckle, leading to misleading readings while 'sticking' a tank. I really don't know if this is a serious problem or not.

    I suspect the problem might have something to do with the nature of 100LL, which contains aromatic hydrocarbons (to get octane rating, and high vapor pressure). Unfortunately, aromatics tend to be tougher on polymers than straight-cut gasolines (aliphatics), causing many types to swell and soften.

    The data I have seen on adhesives suggest that polysulfides (especially the higher molecular weight varieties) are generally quite resistant to chemical attack. In addition, rotationally molded containers of high density polyethylene have a long track record of service as gasoline tanks and containers.

    I do know that bladders are used in race cars (made from polyurethanes), but I don't know if they have the same service requirements as aircraft. Certainly they don't have as long a service life as airplanes.
     
  17. Feb 1, 2007 #37

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    I've also heard about the contamination issue with bladders but so far have not run across anything conclusive. One gentleman however did suggest using two transparent in-line filters of fairly generous size. That way preflight can include an inspection of flow and filter quality.

    The system we recently finished was for Kerosene so incompatibility was not an issue.

    For a larger tank, the issue of collapse is a real one and can be significant with a tank that uses a float type level indicator. We're looking into using some of the racing technology where the tank is filled with either a stainless steel wool or a foam product. The filler takes up two to three percent volume, but it goes a long way toward preventing issues of sloshing and mass fuel release in case of rupture. Furthermore, the filler's own structure would fill up the bladder sufficiently to prevent any collapse. However this combination only works with an induction or optical level sensor.
     
  18. Feb 2, 2007 #38

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

    Joined:
    Jan 22, 2007
    Messages:
    118
    Likes Received:
    6
    The anti-slosh material used in fuel tanks in race cars used to be reticulated polyurethane foam (made with polyester polyols and HMDI, blown with freon). Under a microscope, this looks like a mass of bubble that have lost their wall except where they touche each other. In other words, a complex geodesic mass of highly resilient polymer - open cell foam. I don't know what is used currently.

    I don't remember the life of the reticulated foam, however, in aircraft that would be of far greater importance than in race cars. As for stainless steel wool (the stuff like stainless Brillo pads?), I have no knowledge of how well that works.

    I think there was an article in the AOPA magazine this month about a fuel tank liner collapsing and causing a crash.

    Have you considered lining the fuel tanks with polyurethane? There are many polyurethane polymers that come as two-component liquids which have open times of upto several hours. They cure into very tough membranes with very low permeability. The aluminum interior of the tank would need to be cleaned, and perhaps even primed to be absolutely sure of a tenacious bond.

    Do some testing before using.
     
  19. Feb 2, 2007 #39

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    A polyurethane lined wing sounds intriguing but I'm assuming that if this was to work you'd need to do sort of the same motion as a roto-mold. Difficult to do with an assembled wing panel. Or I guess you could put a lot in, slosh it manually, then pour the excess off.
     

Share This Page

Group Builder
arrow_white