Tortured composite panels?

Discussion in 'Fiberglass and Composite Boat Building' started by Robert Biegler, May 7, 2021.

  1. Mr Efficiency
    Joined: Oct 2010
    Posts: 10,386
    Likes: 1,050, Points: 113, Legacy Rep: 702
    Location: Australia

    Mr Efficiency Senior Member

    I'm a bit surprised to hear that GRP is as elastic as claimed by some here, or as inferred by these supposed methods. If it is so elastic, why don't GRP hulls "dish" ?
     
  2. Dejay
    Joined: Mar 2018
    Posts: 721
    Likes: 138, Points: 43
    Location: Europe

    Dejay Senior Newbie

    Well 4.8% stretch is "relatively" elastic with enough force.
    And something like 400gsm without foam would indeed "dish", wouldn't it?
    I assume woven would also be more elastic than biaxial.

    PS: I'm mostly curious if I understand all this correctly. Please correct me if I'm wrong.
     
    Last edited: May 10, 2021
    fallguy likes this.
  3. rxcomposite
    Joined: Jan 2005
    Posts: 2,776
    Likes: 626, Points: 113, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    It is "elastic" to a point a little past 50% of the maximum (ultimate) force after which it becomes "plastic" where it will stretch but not return to its original state. Elastic means the behavior is like rubber. It will stretch when force is applied but will return back to its original state when the force is removed.

    Most laminate will be in its elastic state in 52 to 58% stress but will crack or deform after that.

    CSM laminate is about 90 N/mm2 ultimate strength. Apply 45 N/mm2 and it is still in the elastic stage. Note that it is force over area. If the laminate is 1 mm thick x 200 mm wide, you need to apply 9,000+ Newtons to get it to crack.

    Laminate will bend but refuses to stretch both ways. Our best effort is single laminate of CSM over foam in a single axis bend. Tighter bends, we kerf the foam and fill the void with fairing compound. This works even with the stiff honeycomb cores. Best method for compound curve is bend the foam by kerfing or heating over a form then laminate one side. Cure then laminate the other side.

    Bending is dominated by resin elongation and will be from 2% to 4.5%. Attached is typical resin curve.
     

    Attached Files:

    Last edited: May 11, 2021
    fallguy and Dejay like this.
  4. fallguy
    Joined: Dec 2016
    Posts: 7,975
    Likes: 1,802, Points: 123, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    My Skoota hulls were sitting outside upside down under a black vinyl tarp. I forgot to vent the forward compartments. Ambient temperatures got up close to say 95F and I had a moment of oh ****, forgot to hole my watertights in two places. I went and checked and the laminate had indeed been damaged (very sad builder).

    When I read your post, I wondered if there was a problem with my laminate. Then at the end, you comment that bending is 2-4.5%.

    Is there a way to quantify any of what happened for me? I ended up grinding out the delaminated areas of triaxial glass, bogging them over and laying brand new glass over the entire section and a lot more; the piece was like 2' square on each side. The hull was basically made to blow up like a balloon. Today, I have inspection ports made with pvc and when I open them, I still get a bit of pressure change. I have also considered pinholing them, but the boat is white. I just don't want to oil can her on a Texas day where it gets to 120F.

    Also, that laminate would have never failed off the core alone, so I was a little surprised it failed on the core, but I admit I was worried about it and was away on vacation when it happened. Is it because it ballooned and was being bent in two directions? The lines of delam were all on the lower first biax.

    I assumed I broked elongation limits of 8%, but reading your post makes me wonder if the limits were the 2-4.5% you mention at the end.

    thanks for any...it may be some thread drift, but it is rather relevant and I won't drone on if you remark
     
    Last edited: May 11, 2021
  5. rxcomposite
    Joined: Jan 2005
    Posts: 2,776
    Likes: 626, Points: 113, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Bending or stretching is when you apply force on a cured laminate. What you have experienced is thermoplastic deformation by application of heat after the laminate has cured. Black vynil tarp on a sunny day is like oven post curing,

    What you have experienced is this Large Depressions in Hull https://www.boatdesign.net/threads/large-depressions-in-hull.65009/page-3
     
    fallguy likes this.
  6. rxcomposite
    Joined: Jan 2005
    Posts: 2,776
    Likes: 626, Points: 113, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Some E glass fabric is listed as having 4.5% elongation. That must be the ultimate elongation to failure, after the laminate has entered the plastic state. There are resins that is listed as having 8% elongation such as Silver Tip epoxy but that is after the eglass fibers have cracked.

    2 to 2,5% strain/elongation of Laminate is a safe elastic range for eglass.
     
    Dejay and fallguy like this.
  7. fallguy
    Joined: Dec 2016
    Posts: 7,975
    Likes: 1,802, Points: 123, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    Okay...I went through a derivation and made a mistake.

    I think a panel held between two points stretches mostly in the middle. More like a triangle vs Dejays two circles.

    My bulkhead that failed was about one meter from the BH ends.

    A 5% change in the panel would happen more like two triangles

    If you have a 0.5m leg length, in order for a 5% stretch, you need the hypotenuse to be 2.5% longer than the leg. Or, hypotenuse is .5125m.

    This equates to 0.112m or 11.2 cm or ballooning of 4-5".

    I have a problem with the gutcheck. I don't believe the laminate would tolerate that. But also doubt it would stretch that far. And the more I think about it; the more I realize one meter was bigger than the dostance. And the shortest distance is the equation....and to make matters worse, tabbing shortens the distance even more because it bends less at the tabbing.

    So, let's say the BH distance is only 0.5m apart.. Then a 5% change is 2.5% longer than 0.25m or 56mm out. This is plausible with what I did. This also means panel failures can occur if I oilcan my hulls in superhot conditions, say on the hard...I am on the hard now and have pvc cleanouts in all the watertights...but some form of breather seems wiser...hmmm

    AND, it must not be forgotten, the panels are held fast up to the WL with extra fabric and so the actual area that moves is even less than that. And a taped deck above. The panels failed in their weakest locations.

    thanks guys, it is making good sense now, even when I disagree with dejay's derivation; it helped a ton
     
  8. Dejay
    Joined: Mar 2018
    Posts: 721
    Likes: 138, Points: 43
    Location: Europe

    Dejay Senior Newbie

    Yeah I think if you have an expanded compartment that is different than just a bend panel. I didn't know cured fiberglass laminate could thermoplastically deform. Also the PVC foam could have thermoplastic deformation.

    BTW I checked out delftship again but I remembered wrong... they just have this "unfold" for developable panels that show difference between bend / flattened interior edges, not some kind of "relaxation" to simulate how fiberglass or plywood would actually bend into a compound shape. So I still no clue how much fiberglass would have to stretch for this KSS method. They probably done a lot of trial and error.
     
    fallguy likes this.
  9. fallguy
    Joined: Dec 2016
    Posts: 7,975
    Likes: 1,802, Points: 123, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    well, this is a chemistry law

    comsider a closed box of fiberglassed core painted matte black and built at 70F and then set in abu dhabi. At some point, it is gonna heat up to 180? inside and stretch at all the box middle sections and a weaker bond/section will fail
     
    Last edited: May 12, 2021
    Dejay likes this.
  10. upchurchmr
    Joined: Feb 2011
    Posts: 3,287
    Likes: 259, Points: 83, Legacy Rep: 579
    Location: Ft. Worth, Tx, USA

    upchurchmr Senior Member

    I'm not sure about this "thermoplastic" discussion.
    Seems to me you don't understand what thermoplastic means.
    Thermoset materials change chemistry when they are formed - epoxy and polyester for example.
    They do not change stiffness with exposure to higher temperature. At some point (excess temperature for that material) they just start to degrade, loosing stiffness and strength.
    Thermoplastics do not change chemistry when exposed to temperature, they just gradually loose stiffness and therefore can permanently stretch.
    If the temperature is reduced, the material properties return to where they started. Again, if the temp is too high the material can be ruined.

    The definition of thermoplastic has nothing to do with "and a weaker bond/section will fail".

    Typically, either thermoset or thermoplastic composites strength are dominated by the reinforcement, especially if it has continuous strands like fiberglass cloth.
     
  11. fallguy
    Joined: Dec 2016
    Posts: 7,975
    Likes: 1,802, Points: 123, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    I assure you it is not me who is confused.

    First of all, I never mentioned thermoplastic unless in negating.

    What I am talking about is what happens to a sealed watertight compartment.

    What happens if you make a solid black box with epoxy and glass at ambient? Then you heat the box? The box expands under the laws of chemistry.

    pv/t=pv/t

    Pressure is unchanged; sort of, so we discount.

    v/t=v/t

    v/70=v/180

    volume of the hot box increases by 2.5 times

    In order for the volume to increase, the box expands. A lot. Any section that is weaker expands more. And this phenomena happened on my build with a sealed compartment that swelled on a 95 degree day under a black tarp.

    The hull experienced minor delamination which was repaired.

    All I attempted here was to understand the limits and how much it stretched and whether the failure made logical sense.

    After a few looks at it; it makes perfect sense.

    Don't believe me? Make an epoxy and foam composite box at ambient and seal it fully. Paint it black. Put it outside on a 95 degree day. Tell me what happens.
     
  12. upchurchmr
    Joined: Feb 2011
    Posts: 3,287
    Likes: 259, Points: 83, Legacy Rep: 579
    Location: Ft. Worth, Tx, USA

    upchurchmr Senior Member

    Solid glass and epoxy box? Do you mean the skins are "solid" or normal panels?
    I assume this box was empty except for normal atmosphere?

    If so, and you are heating the gas, then the pressure has to change. PV=nrt is not chemistry it is a description of steady state equilibrium for gases.
    Under that law, if V remains the same then pressure increases with temperature.
    So your analysis fails at the very first and simplest facts.

    Since your box has flat or relatively flat sides, any increase in pressure will cause the flat sides to bow out. The pressure rises immediately causing an increase in the volume of the box, which means the pressure does not rise as high as it would if the box was infinitely stiff.
    Since the sides are being bent there is an increase in mechanical stress due to the sides trying to bulge out like a sphere. This causes the corners to develop internal stress since they are attempting to straighten out.
    This corner stress very quickly becomes large, resulting in delamination of the material. Almost certainly the corners will fail first since the resin is under a great deal of shear. The resin being the weakest material in the box, it fails long before the glass will.

    If you are going to use a physical law you need to try to understand what it is about.

    This one is about gas, not solid epoxy/glass.

    Sorry, Dejay was the one who talked about thermoplastic, my mistake.
     
  13. fallguy
    Joined: Dec 2016
    Posts: 7,975
    Likes: 1,802, Points: 123, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    look Marc, I gave a caveat on pressure and I understand the physical law, I am an expert chemistry student and could have majored in it in college, a box, well bonded on the edges like a boat bulkhead does not split on the edges, but blows up like a balloon and the weaker areas blow up more which is precisely what happened for me...my only effort here was to quantify it somehow

    the volume of the box is forced to change when the temp rises...if the volume could not change, pressure would, as in pressure cooker, but what truly happens in somewhere in between

    I did not mean for polemics and drift to occur.

    I was merely trying to understand the limits of the stretch and then when I realized there is less stretch on tabbed areas and more stretch on areas wothout tabbing; the issue that occurred makes even more sense.

    Not sure if this picture shows the area where I had trouble, but first watertight. I need some time to find a picture of the event. The one I grabbed was partway repaired..

    All the best to everyone who contributed.
     
  14. fallguy
    Joined: Dec 2016
    Posts: 7,975
    Likes: 1,802, Points: 123, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    Here is a picture, for semantics, where I stressed the watertight compartment and blew it up like a balloon with temp and pressure and it tried to expand to eq...

    It was all ground out and filled with bog and the section reglassed.

    a horrible failure on my part and a commentary on the limits

    2C5D42A6-E441-4A47-8AAB-F037F4770B06.jpeg
     

  15. rob denney
    Joined: Feb 2005
    Posts: 959
    Likes: 353, Points: 63, Legacy Rep: 436
    Location: Australia

    rob denney Senior Member

    I did write that. KSS requires cutting the laminate, bending the triangular sections into place, glassing over them and fairing the result. Derek has some tricky but simple jigs for the bending, it does not take long. Unlike the fairing. The topsides are single curvature. It is an excellent system for building compound curved hulls, but far more work than building infused panels in single curvature panels with filletted (compound curved) edges which is what we do with Intelligent Infusion.

    To answer the original question, significant compound bending of cured fibreglass is effectively impossible. This is regardless of whether it is cored. There are several ways to get compound shapes but they require moulds and/or cutting and shutting and fairing.

    The bulges in the sides of a heated fibreglass box are compensated for by movement along the edges, not compounding the material. In sealed compartments, bulges could be caused by outgassing of the core, inadequate bonding to the core or movement around the edges.
     
Loading...
Similar Threads
  1. fallguy
    Replies:
    120
    Views:
    10,053
  2. mrdebian
    Replies:
    15
    Views:
    2,444
  3. Chotu
    Replies:
    8
    Views:
    1,607
  4. bajansailor
    Replies:
    13
    Views:
    1,735
  5. cando2
    Replies:
    10
    Views:
    1,718
  6. Markusik
    Replies:
    11
    Views:
    1,561
  7. fallguy
    Replies:
    12
    Views:
    1,971
  8. Mark C. Schreiter
    Replies:
    29
    Views:
    3,397
  9. Scott M..
    Replies:
    61
    Views:
    8,554
  10. Smokeyr67
    Replies:
    16
    Views:
    2,992
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.