Conclusions on Aluminum Adhesive Bonding Tests

Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by GESchwarz, Mar 19, 2009.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Sep 2, 2011 #181

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    The other culprit could've been an adhesive the company used to tack the material down with. It's a spray adhesive (sort of like contact cement) that's used to hold prepreg in a particular place while other parts are pressed into the geometry of the mold. The adhesive keeps critical areas from shifting. In our case this was used to keep the material down in the trough of the corrugation mold while the webs and top surfaces were being shaped. In our case we think this too is a likely culprit because if the volatiles are not dissipated when the fabric is laid down, they could dilute or even dissolve the resin in the first couple of layers. Normally the procedures for this type of layup are set so that this doesn't happen but it is still a possibility.
     
  2. Feb 4, 2012 #182

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    I’ve done some experimenting with the mixing of polysulfide sealant and epoxy adhesive for the purpose of reducing the brittleness of epoxy and increasing the adhesive strength of epoxy.

    Each of these two compounds (the epoxy and the polysulfide) must be mixed separately (A mixed with B) before they are combined as a mix or blend.

    Here is what I discovered:

    1. Just a small percentage (~15%) of polysulfide added to epoxy greatly improves the bond strength of epoxy. Straight epoxy always fractures interfacially (brittle), never cohesively (ductile); when poly sulfide is added, the cured mix does not fail interfacially, or rather the adhesive always fractures cohesively. An interfacial fracture is where the break leaves no adhesive on the part being joined. A cohesive failure is where the fracture is through the center of the adhesive material, leaving material adhered to both mating parts.

    2. The shear strength of the Polysulfide/Epoxy mix is inversely proportional to the percentage of the polysulfide in the mix.

    3. 5140 epoxy is much tougher in peel strength than ES6228 epoxy, and is much stronger than ES6228/Polysulfide Mix in shear strength.

    4. There is some difference in strength between the various epoxies.

    5. Straight polysulfide is a very weak adhesive with low cohesive strength, but can stick well to the part if well prepared.
     
    Doggzilla likes this.
  3. Feb 4, 2012 #183

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    I should point out another benifit of this material blending. I've described how the polysulfide can improve certain properties of epoxy, that being adhesion and tuffness. But don't overlook the flip side of the coin...epoxy added to polysulfide increases the strength of the polysulfide.

    On to the next test...how well do these compounds withstand exposure to gasoline? All of these test coupons I made for doing the strength tests, I now have in a jar of gasoline. I suppose a month of exposure ought to have its effects.

    For you bond joint designers, I have another observation. Interfacial failure in peel tests always occur on the part that is curling away or otherwise deforming the most.
     
    Last edited: Feb 5, 2012
  4. Feb 4, 2012 #184

    addaon

    addaon

    addaon

    Well-Known Member

    Joined:
    Feb 24, 2008
    Messages:
    1,686
    Likes Received:
    100
    Location:
    San Jose, CA
    GE, if you're testing for gasoline exposure, can I request you test for ethanol (or ethanol/gas blend) exposure as well? Even just soaking in a pint of everclear would provide valuable information.
     
  5. Feb 5, 2012 #185

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    I suppose I could but I don't have a clue as to where to get ethanol, and I don't know what everclear is.

    I'd like to remind our readers that the highest performing adhesives I've had experience with is methacrylate, followed by the high-end toughened epoxies. Your run of the mill epoxies are quite brittle. By adding the polysulfide to the epoxy you are in effect creating a compound that has the good overall characteristics of methacrylate, but not quite as strong.
     
    Norman likes this.
  6. Feb 5, 2012 #186

    Voyeurger

    Voyeurger

    Voyeurger

    Well-Known Member

    Joined:
    Oct 9, 2010
    Messages:
    612
    Likes Received:
    3
    Location:
    Northern Phoenix, Arizona, U.S.A.
    Hello GE,
    Great info. Thanks! I'm using West Systems Epoxy. I'm very happy with it (man those "$30.00 a set" hand portion pumps are worth the hit). Please advise; Is polysulfide sold as a stand alone product? Is polysulfide a distinct chemical compound that can be ordered? The epoxy is so damned expensive, and at around a 15% mix for results, I want to get me some of that.
    Thanks,
    Gary
     
  7. Feb 5, 2012 #187

    dino

    dino

    dino

    Well-Known Member HBA Supporter

    Joined:
    Sep 18, 2007
    Messages:
    597
    Likes Received:
    82
    Location:
    florida
    You should be aware of the relatively, compared to aircraft structural epoxies, low glass transition temperature of West. Keep the aircraft surfaces cool by painting white.

    Dino
     
  8. Feb 5, 2012 #188

    Voyeurger

    Voyeurger

    Voyeurger

    Well-Known Member

    Joined:
    Oct 9, 2010
    Messages:
    612
    Likes Received:
    3
    Location:
    Northern Phoenix, Arizona, U.S.A.
    Thanks for the heads up Dino. It gets bloody hot here where I live so it's doubly important.
    Regards,
    Gary
     
  9. Feb 5, 2012 #189

    dino

    dino

    dino

    Well-Known Member HBA Supporter

    Joined:
    Sep 18, 2007
    Messages:
    597
    Likes Received:
    82
    Location:
    florida
  10. Feb 5, 2012 #190

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    Polysulfide is sealant sold by a number of different manufacturers under their own names. There are a variety of polysulfides that are formulatedd for different purposes, such as for sealing fuel tanks or sealing electronic components, etc.
     
  11. Feb 6, 2012 #191

    Dana

    Dana

    Dana

    Super Moderator Staff Member

    Joined:
    Apr 4, 2007
    Messages:
    8,621
    Likes Received:
    3,007
    Location:
    CT, USA
    In most states, just buy gasoline, it's got 10% ethanol. Just the right mix to test. Denatured alcohol (sold in hardware stores as paint thinner and stove fuel) is ethanol with a small amount of methanol added to make it poisonous so it can't be consumed. Everclear is grain alcohol, available in liquor stores, though I believe some states don't allow it.

    -Dana

    The difference between theory and practice is that, in theory, there is no difference between theory and practice, but in practice, there is.
     
    Last edited: Feb 22, 2012
  12. Feb 7, 2012 #192

    Voyeurger

    Voyeurger

    Voyeurger

    Well-Known Member

    Joined:
    Oct 9, 2010
    Messages:
    612
    Likes Received:
    3
    Location:
    Northern Phoenix, Arizona, U.S.A.
    Thanks Dino and Gary.
    Dino, that heat deflection occurs in testing as the part is stressed. I am left to wonder whether, should a laminate reach this temperature (very, very, easy to do here in Arizona), and if parts are unstressed (static), joint weakening will occur. Obviously I wouldn't allow my baby to stay exposed to such heat, nor stress it through use approaching these temps. Since resin parts are autoclaved routinely at 200 degrees, the deflection hazard much coincide solely with the duration of the 123F and above temps.
    I'm answering my own question. Don't be a fool, wait till it's cool. (Or something like that).
    Thanks again,
    Gary
     
  13. Feb 21, 2012 #193

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    By way of microscopic observation of a bonding failure in progress, it was learned that the adhesive separates from the metal as the metal stretches beyond the ability of the ability of the adhesive to stretch. As the metal bends, deforms, or otherwise peels away from the adhesive it stretches relative to the adhesive. The strength of an adhesive joint is maintained only so long as the entire joint shares the load. But when the metal deforms, there is a concentration of load where the deformation is taking place, thus depriving the joint of its key characteristic, that being that the load is shared across a large area.

    The key is to prevent this deformation of the parts through careful part design so that there is no peel or deformation under load and/or rivets placed at the ends of the bond line.
     
  14. Feb 21, 2012 #194

    Tom Nalevanko

    Tom Nalevanko

    Tom Nalevanko

    Well-Known Member

    Joined:
    Oct 10, 2007
    Messages:
    1,308
    Likes Received:
    127
    Location:
    Alpine, WY
    That's crazy; the adhesive will always stretch (be more elastic) than metal, say aluminum. It is not the metal stretching that is the problem. How do you make those microscopic observations? Resins (even the one's you say are brittle), and adhesives, in general, are quite elastic -- much more so than metals. Make up some samples and put them in a Universal Test machine and you will see.
     
  15. Feb 21, 2012 #195

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    How about a microscope? My most recent shear strength tests were on a universal. Maybe if you saw what I saw during some peel and bend tests using thick sheet aluminum you wouldn't be such a Doubting Thomas, Tom. These tests were with epoxy. The thick sheet aluminum stretches very noticably on the outside radius, more so than the epoxy that was on it.
     
    Doggzilla likes this.
  16. Feb 21, 2012 #196

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,370
    Likes Received:
    2,300
    Location:
    Rocky Mountains
    That's crazy;

    Maybe poorly worded, but not so crazy. If this link works it explains what I think he is observing.
     
  17. Feb 21, 2012 #197

    Dana

    Dana

    Dana

    Super Moderator Staff Member

    Joined:
    Apr 4, 2007
    Messages:
    8,621
    Likes Received:
    3,007
    Location:
    CT, USA
    No, what he says makes sense. It's not a matter of the elasticity, but in a peel test, as the thin metal bends and pulls away, it can bend without failing far more than the adhesive is capable of.

    -Dana

    A child of five could understand this! Fetch me a child of five.
     
  18. Feb 21, 2012 #198

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    I'm not sure if I mentioned it earlier but when you have two pieces (substrates) joined by adhesive, the adhesive maintains a much better bond to the stiffer of the two; the one that peels away is the one that lets go of the adhesive, because that's the one whose surface has stretched, being on the outside of the bend radius. The rigid piece of course does not stretch at all. The stretching of the substrate is a sure way to induce cavitation as described in Hot Wing's link.

    The lesson learned is to prevent part deformation at bond lines by design. Don't have the joint in peel or tensile. Pure shear is the ideal application.
     
  19. Feb 21, 2012 #199

    Tom Nalevanko

    Tom Nalevanko

    Tom Nalevanko

    Well-Known Member

    Joined:
    Oct 10, 2007
    Messages:
    1,308
    Likes Received:
    127
    Location:
    Alpine, WY
    "the metal stretches beyond the ability of the ability of the adhesive to stretch"

    Does this make sense?
     
  20. Feb 21, 2012 #200

    bmcj

    bmcj

    bmcj

    Well-Known Member HBA Supporter

    Joined:
    Apr 10, 2007
    Messages:
    12,912
    Likes Received:
    4,855
    Location:
    Fresno, California
    I follow GE's logic. He is saying that deformation of the metal shape under load allows the failure to start in peel mode from the point of deformation and progress out from there (or remain localized as a local bond failure). It's much like peeling a glued layer of fabric from a wood surface... you could't easily pull the entire fabric surface apart all at once, but you can start at a point of separation (analogous to the local failure at the point of deformation) and easily peel the fabric back from the wood's surface.
     

Share This Page

arrow_white