6061 vs 2024

Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by GESchwarz, Dec 31, 2008.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Dec 31, 2008 #1

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    Until very recently it was my understanding that all aluminum aircraft structures are of 2024. I have sence learned that 6061 is used here and there, and not just in areas where there are welds.

    Given that 6061 is 2/3 the strength of 2024, where is 6061 most often used?

    Could I get away with using 6061 for all my skins without increasing the thickness by 1/2 to match the strength of the 2024?

    FYI, I will be bonding my skins to the airframe using a two-step process of applying Proseal to the substrate using a very fine tooth (80 tpi) trowel (hack saw blade). Let that set up for a day. Then prior to mating, apply a coat of Hysol epoxy over both Proseal surfaces, then mate. This hybrid joint has proven to to be superior to a Hysol joint in peel strength, and superior to Proseal in lap shear strength, but not superior to Hysol in lap shear; but that's okay, because if the Hysol joint fails in peel (unzip), all the lap shear in the world won't help you.

    The only places where I will be riveting the skins to the airframe will be at certain edges where the bond line would most likely start to peel. A typical location would be where the bondline is plainar, as opposed to around a curved rib or former where a bondline is virtually unable to "unzip". Some would call these "cheater rivets", I just call it good design.

    So given that in the bonded areas where there are no stress concentrations associated with rivet joints, can I afford to go with 6061 in place of 2024?
     
  2. Dec 31, 2008 #2

    addaon

    addaon

    addaon

    Well-Known Member

    Joined:
    Feb 24, 2008
    Messages:
    1,686
    Likes Received:
    100
    Location:
    San Jose, CA
    6061 is easier to work, more corrosion-resistant (compared to unclad or damaged-clad 2024), cheaper (significantly!), and polishable. There are small designs (Zenith's, for sure; probably others) that are basically pure 6061.

    You should always do an analysis when replacing materials. Specifically, just going to 1.5x thicker material is not always okay, because the extra mass may change loads. However, in places where 2024 is being used because "it's always used", even though you're at minimum gage, it's likely that a drop-in replacement of 6061 will turn out okay.

    For airplane design, the most important characteristic difference between the aluminum alloys is strength (although corrosion, brittleness, etc do matter). You can use whatever alloy you want, as long as it has predictable properties (basically anything but pot metal). But you have to do the structural analysis with those properties in mind.
     
  3. Dec 31, 2008 #3

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    Thanks Addaon.

    I jave done a little more research and have come up with the following bits of information...

    Chris Heintz knows that for thin sheet metal structure the failure mode of many parts is by buckling. The load that a part will take before buckling depends on the thickness of the metal and its elastic modulus, not the yield strength. All aluminum alloys have the same elastic modulus, so a thin sheet metal structure made from 6061 will have the same strength as one made from 2024, but the one made from 6061 will be cheaper and more resistant to corrosion.

    The Modulus of Elasticity of 2024T6 is 10500 ksi, 6061 is 10000 ksi. The two are quite comparable.

    An elastic modulus, or modulus of elasticity, is the mathematical description of an object or substance's tendency to be deformed elastically (i.e., non-permanently) when a force is applied to it.

    Would it be safe to say that modulus of elasticity is a measure of a material's stiffness under compression?

    For almost all other physical properties, and just about every other failure mode, 2024-T3 is quite measurably better than 6061-T6

    6061 is widely used for construction of aircraft structures, such as wings and fuselages, more commonly in homebuilt aircraft than commercial or military aircraft. [2]

    6061 is used for yacht construction, including small utility boats. [3]

    6061 is commonly used in the construction of Bicycle Frames and components. The Cycling industry also uses 7005 and 7075 aluminium alloys.

    The Sonex is all built of 6061 T6.

    2024 is widely used for construction of aircraft structures, such as wings and fuselages. [2] The superior metal fatigue resistance of 2024 makes it popular for aircraft structures which are under tension, such as the bottom of wings. [3]

    6061 alloy is seldom manufactured as alclad, because it is naturally corrosion resistant.
    We construct ocean dock parts from 6061 with no need for corrosion protection in a salt spray environment, with great success and longevity. Commercial fishing boats use it for net spools and guides.

    So it makes sence that if the part is not a high tension member, you’re ok going with 6061. Is that true?
     
  4. Dec 31, 2008 #4

    Topaz

    Topaz

    Topaz

    Super Moderator Staff Member

    Joined:
    Jul 30, 2005
    Messages:
    13,826
    Likes Received:
    5,461
    Location:
    Orange County, California
    That highlighted line is a gotcha, too. Not only is 6061 less strong overall than 2024, the lower fatigue resistance will lower the allowable stress level even more.

    This thread is timely - I was looking at a particular tubing that is only available (from Aircraft Spruce) in 6061.

    The kicker, GESchwarz, is that while you may indeed be able to use 6061 for an airframe structure, the entire and complete structural analysis has to be done with 6061 in mind - the "can I use it" question is not simply "yes" or "no", or even "yes, under these circumstances."
     
  5. Dec 31, 2008 #5

    addaon

    addaon

    addaon

    Well-Known Member

    Joined:
    Feb 24, 2008
    Messages:
    1,686
    Likes Received:
    100
    Location:
    San Jose, CA
    Well, to be more optimistic, the answer is "yes, definitely"; but sizing has to be done with that in mind.

    That said, having worked with 6061 a bit now, and 2024 slightly less... I'm definitely basing my designs around 6061 (or 5086!) wherever reasonable, going to 2024 only with good reason. Workability matters, in my opinion, since I'm the one who has to work it.
     
  6. Dec 31, 2008 #6

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    Thanks Topaz.

    Is the fatigue resistance only a major issue for members primarily under tension, for instance longerons and spar caps? What I'm getting at is, is this less an issue for skins where modulus of elasticity is the critical issue? Is fatigue resistance a major issue on skins, particularly skins that are bonded to the airframe, rather than riveted?
     
  7. Dec 31, 2008 #7

    addaon

    addaon

    addaon

    Well-Known Member

    Joined:
    Feb 24, 2008
    Messages:
    1,686
    Likes Received:
    100
    Location:
    San Jose, CA
    Fatigue resistance is always an issue, because cumulative failure sucks. Rivets definitely provide failure points, but they're also easy to inspect. I don't think going to bonded is a huge difference, as long as you're really really sure that the bottom (hidden) surface has no crack propagation points. As Sonex and Zenith has shown, this is not a show-stopping issue; it's just yet another thing to be considered.
     
  8. Dec 31, 2008 #8

    Topaz

    Topaz

    Topaz

    Super Moderator Staff Member

    Joined:
    Jul 30, 2005
    Messages:
    13,826
    Likes Received:
    5,461
    Location:
    Orange County, California
    Structures is getting outside of my area of knowledge, but by my rather limited understanding, any structural member that is subjected to cyclic loads in tension or compression - or even shear - is subject to fatigue.

    IIRC, the fatigue curves or forumulas that I've seen only specifiy the level of working stress without defining the type of stress. That implies that the type of stress is not a factor, only the amount. I'm not absolutely sure about this and don't have the time at the moment to go crack open my Bruhn, so perhaps someone more knowledgable can confirm what I remember.
     
  9. Dec 31, 2008 #9

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    The choice of a particular alloy, especially in the homebuilt industry, is one of tradeoff where the variables are often strength and durability versus cost. For small, lightly loaded airframes the 6061-T6 often turns out to be a fairly good compromise, even though the structure may end up somewhat heavier than if the more optimal 2024-T3 was used.

    However, as a bit of my recent work has shown, when looking to increase the structural integrity to aerobatic loading, the weight penalty becomes evident (and significant) rather quickly, especially once you start accounting for the fatigue properties as well as the stress levels encountered due to the presence of bolted or riveted connections.

    But regardless of which material you use, the structure must be analyzed for those specific conditions and alloys. There is no simple formula that can indicate a simple substitution so, if the plans call for 2024, substituting anything else without running a full analysis would be foolhardy at best.

    And one side note: For those interested in a more precise definition, the Modulus of Elasticity of any material is a measure of how much strain (stretch) is encountered for a given amount of stress. Illustrating the case, if you plot the stress-strain behavior of a material where the stress is listed on the vertical axis and the strain is plotted on the horizontal, then the Modulus of Elasticity is the slope of the plotted line. The stress is plotted per unit area (pounds per square inch) and the strain is plotted as distance per unit length (usually micro inches per inch). The resulting units for the Modulus are therefore "pounds per square inch", which must never be confused with material stress.

    And finally, how much data and/or experience do you have with the proposed bond scheme? This sounds downright scary, especially considering that we've conducted several tests for bonding with ProSeal, all coupons resulting in what I would definitely rule out as a structural bond. As a friend of mine pointed out to one of his associates, "What about the term "seal" makes you think this is a structural bonding agent?"

    About three years ago I detected a small leak in one of my wingtip tanks. We weren't sure where the leak was but were able to identify the specific area it came from. The leak was not too serious so I decided to delay the repair 'till after a short trip I had to go on. As such, we temporarily cleaned the area off and prepped the parts (fiberglass and metal) as required, and then applied a fairly healthy coating in the region.

    This did stop the leak but when I got back we took the tip off and set about doing a permanent fix (it turned out to be a fuel line fitting so it was easy). The thing that struck me as significant though was that the removal of the ProSeal was quite easy. All it took was a bit of fingernail under a thin edge and the whole coating peeled off in three or four pieces. The material was relatively weak in shear to the point where peeling by hand was sufficient to tear the coating.

    This poor structural behavior sort of confirmed a series of bonding tests i did about five years ago, aided at the time by the gentleman who headed up the structural branch of Seattle FAA office. In those tests we worked with about five different bonding materials and three different surface prep techniques. The result were interesting and somewhat varied but pertinent to this discussion, the ProSeal coupons were very consistent in that they delivered the worst performance for all conditions. As a matter of fact, two of the coupons literally fell apart the moment I picked them up.

    In the three test series I ran where I was looking for suitable aluminum bonding compounds, the only material that delivered any appreciable potential for airplane application was the Methylcrylate. I still hold that this has potential however the test showed an uncomfortable level of inconsistency so I have not committed to it yet, nor will I recommend it just yet, at least until one of these days I can identify what created the difference from one set of coupons to the next.
     
  10. Dec 31, 2008 #10

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    Thanks for the words of caution about Proseal adhesion.

    My experience has been very different. I prepped my coupons by creating a coarse finish which completely removed any of the original mill finish. So I was bonding to freshly exposed, coarsly-textured aluminum. I never had a delamination between the Proseal and the substrate, I did sometimes with the epoxy.

    I think of the Proseal more as an elastomeric primer for the Hysol. The Proseal has a greater peel strength than the Hysol. The Hysol makes the joint stronger in shear than if it were all Proseal. My test coupons of 2 square inches each were tested head-to-head against a pair of AN -4 rivets and the rivet joints failed. In some tests the rivets sheared cleanly at the shear line accompanyed by a loud bang, and the other joints failed when the rivet heads tore through the 2024 T3 coupon.

    What your findings tell me is that I need to do more testing to find out all the performance limits of these materials. At this point I am guessing that your surfaces were comparatively smooth. Even an 80 gt sanded finish leaves aluminum fairly smooth.

    How coarse or abraided were your finishes?
     
  11. Dec 31, 2008 #11

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    We did three surface preps - the first was a simple degrease of the coupons. The pieces were about 6" long and three inches wide. One edge of one coupon was bent 90 deg. so it would be easy to grab both edges and pull apart. The material tested was 2024 Clad 6061 bare and 7075 bare.

    The second coupon set used a slight abrasion that was applied with a random orbit sander and a 120 grit medium. After sanding the coupon was cleaned with Acetone.

    The third set of coupons used an 80 grit abrasion, also applied with a random orbit sander, then cleaned in Acetone.

    The results were interesting in that they followed the theory of bond strength as a function of surface condition quite closely: The coarse sanded coupons resulted in the worst peel strengths, while the smooth and 120 grit surfaces were fairly close in value for the different materials. Epoxy bonds did best with the smooth material while the Methylcrylate bonds did best with finer grit.
     
  12. Dec 31, 2008 #12

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    That's interesting that the coarser finish did not perform as well. Pulling at 90 degrees to the bond line is a sure test of peel strength. No adhesive or "sealant" performs well in that mode, especially epoxy. What I believe happens there is that the epoxy cannot tolerate the expansion of the substrate as the outside radius is formed during deformation. It's this deformation and expansion of the subsrate that fractures the bond, initiating a zipper-like debond. Deformation of the substrate is the Achellies Heel of epoxies. So a good epoxy bond design avoids putting the bond in peel. Most of my tests were all in shear only. Proseal in shear always separated down through the proseal itself, always leaving sealant material on the substrate. When I pulled the epoxy joints appart in peel, only the very best joints separated down through the epoxy; many of the joints delaminated at the bondline, which is a total failure of adhesion.

    One thing I noticed about the Proseal is that because it is so thick, it traps air readily, both during mixing and mating. No matter how good your adhesion is, or how much squeeze-out you have, you may still have great big voids. This is why fuel tanks leak at the seams. I recommend applying a coat to each mating surface with the above described trowel and letting it setup. then before mating applying a single tall bead down the center before mating. And whatever you do, don't let them separate once they've been mated, or you'll have bubbles galore. It will be the trapped air that is producing all the squeeze-out of sealant.

    I don't know for sure but I've read that acetone leaves a residue. I've used the orbital sander. It works well for applying a very fine texture. The texture I applied on about half of my coupons was made using a high speed radiac abrasive cutting disk, laid down at a about 20 degrees to the surface. I would apply only two or three light passes, just enough to obliterate the original mill finish. This created a texture similar to a phonograph record. To the touch it felt like lava stone. I used acetone before the abraiding operation.

    In peel, these joints can be destroyed by hand and a couple pair of pliers. In shear, failure requires several hundred pounds of force, which I applied with the use of a hydraulic jack setup.
     
  13. Jan 1, 2009 #13

    BBerson

    BBerson

    BBerson

    Well-Known Member HBA Supporter

    Joined:
    Dec 16, 2007
    Messages:
    12,158
    Likes Received:
    2,382
    Location:
    Port Townsend WA
    I agree with Gary, wiping with acetone could cause problems. Any solvent wipe tends to smear the remaining oil contaminates on a molecular level.

    I use phosphoric acid scrubbed with scotchbrite and flushed with clean water to remove oil before painting. Seems to work well. But I would not bond aluminum... just my opinion.
    BB
     
  14. Jan 2, 2009 #14

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

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

    please believe me when I recommend using chemical cleaning for aluminum bonding substrates. Abrasive cleaning, acetone rinses are not consistent or reliable.

    Degrease, followed by an alodine etch of sufficient duration to produce a golden appearance, followed by a distilled (or rainwater) rinse. Air dry and use within twenty-four hours.

    The strength of the adhesive bond will be superior to any other surface preparation (other than using a chromic acid etch - but that's a nasty operation and not recommended outside of an industrial facility).

    The surface does NOT have to be roughened for a good adhesive bond. Acetone (if not reagent grade) is likely to contain trace amounts of oily substances which will reduce bond strength.

    Phosphoric acid etch is actually a recommended surface preparation method for bonding and painting aluminum.

    Commercial and military aircraft have been assembled using adhesives (mainly epoxies) for almost fifty years, and these aircraft stay together under conditions that GA aircraft never see. It is more than just 'accepted,' it is a preferred assembly method.

    Methacrylate adhesives have an excellent track record for aluminum truck body assembly, and hold up well under brutal environmental conditions. Again, the key is proper surface preparation for optimal performance, and the same surface prep system used for epoxies works well for methacrylate adhesives.

    I have previously posted much information on adhesive bonding for those who are interested.

    Fly safe.

    MalcolmW
     
  15. Jan 2, 2009 #15

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    Thank you all. And thank you MalcolmW, I get feeling lonely sometimes pioneering a 'new' process. If I couldn't build a plane the way I want it, it wouldn't be worth doing at all. I can buck a rivet with the best of 'um, but I just think it's so dang labor intensive and monotonous. I also think I can succeed at applying skins to an airframe and have it as strong or stronger and look absolutely cherry, without a single ripple. I expect to run into unforseen problems, and I'll deal with them when I get there. The tail feathers are a good place to get my feet wet. I'm looking forward to being able to skin an entire wing panel with just two or three sheets

    I'm going to have to look into these Methacrylate adhesives, and pick up an alodine kit too. It would be nice to not have to apply a coarse finish to all my bond surfaces. If it is true that there need not be any rough texture to achieve an optimal bond, then my joints performed as well as they did because the abraiding process exposed so much "fresh" aluminum. That's what the etch process does, it produces a "fresh" aluminum surface. Two ways, one mechanical, the other chemical, to skin the cat.

    Before I actually begin actually assembling with these adhesives, I will have proven and documented my process. I will produce test coupons that represent worst-case assembly fit-ups. I will have accept/reject criteria based on the testing I've done, and I will share it with all of you.

    Do you have any recommendations for the right methacrylate adhesive?

    I am just a few days away from completing the assembly and part drawings for my tail surfaces. From there I'll begin drawing the empenage. You always want to design a couple of assemblys ahead to discover all the constraints and gotchyas that have a way trickling back, dictating changes an assembly or more away.
     
  16. Jan 2, 2009 #16

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    The Methacrylates I've tried were from Plexus and Extreme Adhesives, the latter delivering better working qualities and results. But keep in mind that these adhesives are not compatible with fuel so wet wings will not work without some substantial protection.
     
  17. Jan 2, 2009 #17

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    Thanks for the recommendation Orion. So what is the primary advantage of Methacrylates over Epoxies. I know that the big down side to epoxies are their relative brittleness. Are methacrylates tougher and more elastic?
     
  18. Jan 2, 2009 #18

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    From what we saw in our tests, the Methacrylates seemed to exhibit a form of self-etching in that the aluminum at point of contact was somehow modified by the adhesive and thus a better bond was formed. In most of the peel tests the epoxy (Hysol EA9430), once separated, peeled in a rather brittle and catastrophic fashion.

    Not so with the Methacrylates. Most of those coupons were rather difficult to separate, and on one occasion to the point of actually tearing the aluminum.

    But as I indicated, we also had a couple of the Methacrylate coupons that failed much in the same manner as the epoxy so until I can determine for sure why that happened, I really don't want to commit to a bonded aluminum structure.

    It is quite possible that maybe the surface prep was the issue - I did not realize that Acetone could leave a film. Well, I'll look into this more as time allows.
     
  19. Jan 2, 2009 #19

    GESchwarz

    GESchwarz

    GESchwarz

    Well-Known Member

    Joined:
    Oct 23, 2007
    Messages:
    1,179
    Likes Received:
    108
    Location:
    Ventura County, California, USofA.
    I'm in the process of getting samples sent to me. Parson Adhesives has some interesting methacrylites and rubberized (shore 85) epoxies. Plexus and Extreme Adhesives seem to be out for the holidays.

    I would have never thought that acetone leaves a residue either. The sales rep at Parson says that their methacrylics resist gasoline, others manufacturers don't. I'll be looking for second and third opinions on the gasoline resistance claim.
     
  20. Jan 3, 2009 #20

    MalcolmW

    MalcolmW

    MalcolmW

    Well-Known Member

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

    Thank you for your kind words. Yes, I agree with Orion about the use of methacrylate (acrylic) adhesives – they should not be used to assemble gas tanks. Acrylic adhesives are attacked by aromatic solvents (slowly, but surely) and 100LL contains benzene, which is a strong aromatic solvent. If you want to make an aluminum gas tank, consider using epoxies or polysulfides (ProSeal), which hold up well against gasoline.

    The main difference between acrylics and epoxies is ultimate tensile strength and application. Epoxies are a little stronger and have more heat resistance (consider using epoxies to bond sheet metal around the engine compartment).

    Yes, Orion, acrylics are less demanding as far as surface preparation are concerned, however, for maximum bond strength and longevity, do prepare aluminum surfaces with a chemical etch followed by a distilled water (or filtered rainwater) rinse. Air dry and use within twenty-four hours. Cleanliness is next to Godliness in the world of adhesive bonding.

    Acrylics have a good track record in truck body (aluminum) assembly, and hold up well in delivery service in all kinds of weather without any special care. Tougher environment than your typical aircraft sees.

    The largest manufacturer (I believe) of acrylics is the Lord Chemical Corp., and their largest distributor is Ellsworth, see their website: Lord Corp. Adhesive and Surface Preparation Products
    They also carry epoxy and urethane adhesives, which reminds me, if you have fiberglass parts which need bonding together, urethane adhesives are clearly superior to any other adhesive (this I know for I was involved with their development).

    Oh, yes, if you do contact them, don’t say the application is for aircraft… er, say you’re building a race car or an aluminum boat. There are other manufacturers of acrylic adhesives.

    Good luck with your project, it sounds as though you are taking a rigorous and careful approach, which should produce sound results. I could add more, however, I’ve posted a lot of material already on adhesives, and I don’t want to clutter up this site.

    Fly safe,
    MalcolmW
     

Share This Page



arrow_white