Pull off testing

Discussion in 'Fiberglass and Composite Boat Building' started by fallguy, May 11, 2018.

  1. rxcomposite
    Joined: Jan 2005
    Posts: 1,790
    Likes: 115, Points: 63, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    No that is Correct. How can it be wrong when it is just statement of the facts. You are just being creative in pulling me into an endless debate. Write something, maybe a procedure or analysis or the proper way to do it so that other members will learn from your "expertise" on the subject matter.
     
  2. gonzo
    Joined: Aug 2002
    Posts: 13,066
    Likes: 290, Points: 93, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    The delamination happens because the resin cracks.
     
  3. Ad Hoc
    Joined: Oct 2008
    Posts: 5,692
    Likes: 248, Points: 63, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Good god....you still don't get it.

    Delamination is simply a physical separation or loss of bond between two layers. That's it!
     
  4. fallguy
    Joined: Dec 2016
    Posts: 1,112
    Likes: 40, Points: 48, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    I would like to know what the peel pull test ought to be in 18oz biax. Pretty likely it will fail there.

    The epoxy manufacturer is telling me Volan coated fabric would alleviate this issue.

    A couple other guys say normal.

    A few others get into polemics or almost.

    Is there an objective number here?

    Again, thanks.
     
  5. gonzo
    Joined: Aug 2002
    Posts: 13,066
    Likes: 290, Points: 93, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    The separation occurs because the resin breaks. Otherwise, there would be no physical separation. The fiberglass layers are the reinforcement of a matrix (the resin) which fractures.
     
  6. Ad Hoc
    Joined: Oct 2008
    Posts: 5,692
    Likes: 248, Points: 63, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Gess...round and round in circles.

    Ok...what is a crack??
     
  7. ondarvr
    Joined: Dec 2005
    Posts: 1,765
    Likes: 83, Points: 48, Legacy Rep: 506
    Location: Monroe WA

    ondarvr Senior Member


    There are too many variables to give an answer, mostly due to the flexibility of the laminate being peeled off.

    You could have the exact same adhesion with a flexible resin as a rigid resin, but the rigid resin will have higher numbers in a peel test, being resin rich could do the same thing.

    The pull test Gurit recommended would give a better comparison, but still, the number will only be what the core can handle, once the core fails you won't get a higher number.

    If you look closely at the surface of the laminate after its been peeled off you will typically see tiny broken pieces of the core, just the edges of the cells that were in contact with the epoxy. The core is still the weak link, it just fails in a slightly different way when you do a peel test. What it shows is that the bond didn't fail, the fragile core did due to the point loading during the peel.
     
  8. fallguy
    Joined: Dec 2016
    Posts: 1,112
    Likes: 40, Points: 48, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    So, ondarvr, if you bump the edge of a laminated panel; is shear or peeling expected?

    I about fell over when it happened.

    I doubt very much I will test at the core psi.

    This is the same resin that shears concrete off the shop floor. The difference, of course, is lots more resin per sq in.
     
  9. rxcomposite
    Joined: Jan 2005
    Posts: 1,790
    Likes: 115, Points: 63, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Call it shear or peeling or whatever. There are three basic modes of loading as shown in the attachment.

    1. If you pull the skin perpendicular or near perpendicular to the core laminate (Mode 1), the core will fail first as Ondarvr is describing. The laminate will never experience this. This is just a poor man's test to see if bits of pieces of the core attaches to the skin. Bondline failure, but you are not interested in this.
    2. If you pull the top skin and the bottom skin away from each other, that is Mode 2. The greatest shear will occur at the middle of the core and the core will fail first. There is some shear in the bondline, between the core and the laminate but it is much less. And there is a much lesser shear in between the laminate if more than 1 ply of skin. This is the interlaminar shear. I believe this is what you are looking for and you want to know how much force. The lap shear test is best to know the shear strength of the adhesive between the laminates. Same test they use for glued, riveted, or bolted substrate.
    3. Mode 3 will happen if you try to rotate opposite skins. A scissor like effect. but again this is unlikely loading. I need to mention this although not much of importance. Mode 3 happens in a laminate with a 90, +45-45, degree layup. The +45,-45 moves and closes in shearing the resin in between. This force is very small and is used only in very high end laminate analysis. This is macromechanics theory, don't worry about it.

    Give me the brand of epoxy resin you are using and the density of Gurit 12 mm core and I will come up with some numbers as soon as I am free from my paid work.

    And by the way, Volan is a treatment (chemical sizing) given to the fiber to ensure wet out and reduce voids. It is a treatment to increase the STRENGTH/MODULUS of the laminate and is no way related to interlaminar or bondline strength.
     

    Attached Files:

  10. Ad Hoc
    Joined: Oct 2008
    Posts: 5,692
    Likes: 248, Points: 63, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    A general method for calculating the interlaminar shear strength in E-glass laminates is given by

    Interlaminar Shear Strength = 22.5 - 17.5Gc

    Where Gc = (N/mm^2)

    upload_2018-5-19_7-59-26.png

    and T = normal laminate thickness in mm and W = weight of reinforcement in g/m^2
     

    Attached Files:

  11. SamSam
    Joined: Feb 2005
    Posts: 3,779
    Likes: 139, Points: 63, Legacy Rep: 971
    Location: Coastal Georgia

    SamSam Senior Member

    [​IMG] This jpg is labeled 'delamination' but then the phrasing of the illustrations call them 'cracks', so the term for the problem does seem flexible.

    In this photo, the edges of the delamination are so straight it looks to be some sort of mechanical/fabrication problem as opposed to resin failure. Is this piece the same that you didn't vacuum until way past gel time? It certainly is nothing I would not expect or want in polyester laminates, much less epoxy.
    [​IMG]

    In your other thread Mould parting delam https://www.boatdesign.net/threads/mould-parting-delam.60241/ you said you were using super slow hardener and releasing in 16 hours. I'm assuming when you say release that you mean de-molding or taking it off the table. In the Ampreg literature, it says
    Maybe you tried pulling the part too soon.

    I wouldn't use anything like that in a boat, but as long as you have it, you might cut a strip 3" wide and a foot long from a good section, support the ends and press down in the center until it breaks. It will be a gradual process in that it starts to be compromised and damaged before it actually breaks, but it will show you things.

    If you cut a few strips of similar size and do it in a somewhat consistent way, such as supports the same distance every time, weights applied to the same area in the same way every time, you can have you own rudimentary testing and comparison system. If you can correlate it to a known material, such as the breaking strength of a 1" square stick of oak in the same test, you might even be able to put some rough numbers to the strength of whatever you are testing.
     
  12. Ad Hoc
    Joined: Oct 2008
    Posts: 5,692
    Likes: 248, Points: 63, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    No, it is the understanding: of what how and why. Otherwise we woudl not have terms like: shear strength, tensile strength, delamination, exfoliation etc etc...because at the molecular level these are all ostensibly "cracks" - but the mechanism for them is very different, hence these terms to aid the engineer to distinguish what may or may not have caused the failure. It would be rather immature to call everything a crack, as that is a child's interpretation. Since a child would not understand slip bands, or strain energy or covalent bonds etc etc.

    All materials have a theoretical value of strength. This is stress = E/(2.Pi)

    But this assumes a homogeneous arrangement of atoms and that the whole crystal matrix is in fine balance. In other words the distance between each atoms is equal thus the whole system is in equilibrium, no net out of balance forces.

    However, real materials are not like this, they are not "perfect" or "theoretical", they are real. The crystalline matrix has 1 or 2 (or more) atoms that are just ever so slightly further away from its neighbor than the others. This is therefore a 'weakness', because it is a localised stress concentration. Because several atoms have different bond strength to the surroundings. Thus under an applied load this localised region does not behave as the rest does, the atoms then get pulled further apart than the equi-spaced surroundings. This then pulls the atoms further away as the bond strength gets weaker the further apart these atoms are and on an atomic level the atoms are then simply "pulled' apart...leaving a void. This void becomes what is called a 'crack'.

    Thus either we distinguish failures from their mode or mechanism of such, or we just ignore engineering and be like a child and default everything to "it's a crack"...and not understand the hows and whys.
     
  13. fallguy
    Joined: Dec 2016
    Posts: 1,112
    Likes: 40, Points: 48, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    Not using Ampreg. I was incorrect about a mould parting issue. I tested a 40 hour parting and it behaved the same. In fact, the piece shown was 40 hours. I think that is the top side in the picture. Both sides seem weaker interlaminar than I would expect.

    I am using an extra slow hardener with 120 minute working time.

    I don't wish to name the epoxy at this time. I'd rather not get hauled into court.

    The parting that seems too easy is interlaminar.

    The likelihood is that there is a bit too much resin exiting the part is my primary concern.

    I left parts on the table 40 hours now with no difference.

    However, it was pointed out post curing might help and post curing is needed for the extra slow epoxy.

    I plan to run some tests next week.

    The reason I want to work this out is the designer spec'd biax for the cabin exterior, so want to make sure the bonds are good. That and I realized this interlaminar shear/peel by accident.

    Rx - I will message you the epoxy. Please avoid mentioning it. I think the epoxy is great, but the process might need tweaking.
     
  14. rxcomposite
    Joined: Jan 2005
    Posts: 1,790
    Likes: 115, Points: 63, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Most biax or triax has light CSM or SM on one side. This increases the bondline as it reinforces the resin with fibers. The area is no longer "pure resin" and is reinforced by fibers. I am sure Ondarvr made some test before. Even Class rule would require CSM in between WR over WR only if the laminating process has stopped and the resin has cured. The small fibers not only increases the resin strength but "digs in" into the undulations of the woven or knitted fabric.
     

  15. rxcomposite
    Joined: Jan 2005
    Posts: 1,790
    Likes: 115, Points: 63, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Ok will not. Need also core density. Nothing to worry as it probably has the shear strength it specified.
     
Loading...
Similar Threads
  1. brokensheer
    Replies:
    5
    Views:
    1,508
  2. hyboats
    Replies:
    8
    Views:
    3,180
  3. brokensheer
    Replies:
    17
    Views:
    4,319
  4. ahender
    Replies:
    6
    Views:
    2,345
  5. Phil Locker
    Replies:
    2
    Views:
    1,293
  6. ericlarnold
    Replies:
    1
    Views:
    3,105
  7. IronPrice
    Replies:
    55
    Views:
    1,557
  8. youngfella87
    Replies:
    0
    Views:
    756
  9. Fabuglas74T
    Replies:
    3
    Views:
    919
  10. Nicko1985
    Replies:
    7
    Views:
    1,225
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.