Material strength and fatigue

Discussion in 'Boat Design' started by MikeJohns, Aug 9, 2006.

  1. Poida
    Joined: Apr 2006
    Posts: 1,188
    Likes: 51, Points: 48, Legacy Rep: 497
    Location: Australia

    Poida Senior Member

    Mike Johns would like us to sleep at nights.

    Except now I can't sleep worrying that when I take my boat out in the morning the next wave, due in 3 seconds, will be the one that fatigues my hull to bits.

    Thanks Mike

    Oh and geeks are called geeks because they are.

    And I am called an ignorant layman because I am.

    Always good to read your posts tho' very informative.
     
    1 person likes this.
  2. catsketcher
    Joined: Mar 2006
    Posts: 1,315
    Likes: 165, Points: 63, Legacy Rep: 790
    Location: Australia

    catsketcher Senior Member

    Cycles aint cycles

    Hello all,

    I remember being told as a 21 year old that my Twiggy's bows were going to fall off and it was all due to fatigue - I am now 39 and the boat's bows are still there. I think fatigue is still very poorly understood and I am going to put my two cents in.

    For a material to fatigue it has to be reverse loaded to a point where it undergoes sufficient stress to start being fatigued. In the Gougeon borther testing they don't seem to say how highly the loaded to materials when they tested them - low loading would increase fatigue resistance.

    Picture this - you have piece of wire and you wiggle it very slightly between your fingers. It is very lightly loaded and may not fail before you get athritis. However, work the wire back and forth hard and it may break in a minute or two. Obviously the numbers are stress and cycles not just cycles. A thin and highly stressed wooden structure may be more fatigue prone than a glass or aluminium structure more lightly loaded. (A sandwich one may be better than a ply hull)

    The DC3 plane (so I am told by a 747 pilot) has no fatigue limit - they are one of the few planes with no scrap date. This is because they are built heavy and so the alloy is cycled under a safe limit. Obviously if a designer can they should try to keep their laminates well away from yield. John Shuttleworth says 0.3 percent strain can lead to micro cracking and fatigue in glass.

    Obviously the Gougeon's waves and failure data can only apply in very particular circumstances.

    cheers

    Phil Thompson

    www.foldingcats.com
     
  3. SolomonGrundy
    Joined: Feb 2005
    Posts: 183
    Likes: 9, Points: 18, Legacy Rep: 51
    Location: lost

    SolomonGrundy I'm not crazy...

    fatigue @ failure

    An interesting coincidence occured last week while I was riding my mountain bike home from the boatyard. My '87 Gary Fisher (steel tube frame) started making noises, apearantly from the crank...distinct squeaking. It stoped as suddenly as it started while I was on the trail but nearly home. It was getting dark. The squeaking lasted less than 5 min. but afterwards the cranks seemed "squishy". That is to say it seemed that on each downward extention of the crank brought the pedal just a little more "inboard" than is normal.
    This is what I found the next morning...
    The pics are after I had preped the frame for welding, they clearly show a nearly 360 degree break in the tube. In actuality, there was less than 1/8" of metal unbroken. The break occured where the factory placed the welded (or brazed) on derailer tab. (This bike was made before clamp on derailers became popular.) The derailer tab, in fact, failed a few years ago, so I have been just manually switching those gears when I need to.
    The bike has been rewelded 100% but I now call it the Frankenbike.
    SG
     

    Attached Files:

  4. Figgy
    Joined: Feb 2006
    Posts: 315
    Likes: 12, Points: 18, Legacy Rep: 88
    Location: TN

    Figgy Senior Member

    You got lucky, that could've been bad...real bad.
     
  5. SolomonGrundy
    Joined: Feb 2005
    Posts: 183
    Likes: 9, Points: 18, Legacy Rep: 51
    Location: lost

    SolomonGrundy I'm not crazy...

    coulda been ugly...

    Roger that.
    I have used this bike continuously since '90. Gary Fisher bikes have a lifetime gurantee on their frames for the original owner that, unfortunately, is not my case. I am, however, a welder so maybe I can get another 20 years or so out of her before I turn her into art.
    What is of note is that the break occured where the tube had been weakened. Also of note is that the stresses induced during the break were not at all above normal, no jumps, bunny hops or anything (at the time).
    It just had enough weakening of the tube over time to eventually fracture.
    It just took 20 years of normal use.
     
  6. Raggi_Thor
    Joined: Jan 2004
    Posts: 2,457
    Likes: 64, Points: 0, Legacy Rep: 711
    Location: Trondheim, NORWAY

    Raggi_Thor Nav.arch/Designer/Builder

    Remember that fatigue in grp doesn't have to mean it breake. Solings are known to get soft after two years racing, then they can be used for recreational sailing or racing at a lower level for annother 30 years or so.
     
  7. MikeJohns
    Joined: Aug 2004
    Posts: 3,192
    Likes: 208, Points: 63, Legacy Rep: 2054
    Location: Australia

    MikeJohns Senior Member

    Phil
    You should read the thread!
    I posted this to try and get people to understand that You need to apply "SN" curves .....(S)tress on one axis (N)umber of cycles on the other.... ...failure on the curve.

    As you say it's all in the wiggle :)



    On Bikes..
    Mountain bikes often fatigue, the frames are highly stressed since weight and strength are opposites. The same problem as lightweight boats. Stress levels are high because the material is very thin. A lot of the carbon composite frames are failing too . I think the loads are just higher than the designer allows for particularly with large male riders, and fatigue is not properly addressed.
     
  8. PhilT
    Joined: Feb 2005
    Posts: 4
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Australia

    PhilT New Member

    Ignorance shown in public

    Hello Mike,

    I must admit I didn't really get the graph - I have looked at it for years and gone - yeah yeah. I thought it showed only cycles to failure but another look (and a gentle prod) got me to really see it. I thought it only showed how much stress you laminate could take when old.

    So obviously we want our boats to last a long time. Therefore we need to use the stress for our particular material at 10 to 7 cycles (a guess here)and use this to determine layups.

    Micro-cracking is a really interesting thing in glass laminates. As an old Laser sailor I am used to the idea of glass going soft. I have sailed a much older Tasar that is still stiff (it is foam sandwich and I guess the laminate is much lower loaded) Ragnar is right about old Solings and the like but be careful with old glass boats when still highly loaded.

    My kids sail a dinghy called a Flying 11. Out on the rescue boat one day I got to pick up an old 11 that had ripped in half. They had upped the rig tension to that used by the new boats and put two big kids in it. The boat just tore up in two. These boats have very light laminates and highly loaded gunwales. Obviously the glass was overloaded and was full of microcracks already. Cracks are interesting in their own right.

    As I understand it if you load up a structure to a high stress the critical crack length will be reduced. Try ripping a piece of paper a little and then increasing the tension you hold it with. At a certain tension the crack will release the stored energy by increasing the crack length which reinforces the situation. However if you keep the stress below the critical length then the crack can be easily dealt with.

    So in our laminates we have to keep the stress down so that the small cracks that are always in a structure do not increase in length.

    Ta Mike and its time to get back to the boat

    Phil Thompson
     
  9. Raggi_Thor
    Joined: Jan 2004
    Posts: 2,457
    Likes: 64, Points: 0, Legacy Rep: 711
    Location: Trondheim, NORWAY

    Raggi_Thor Nav.arch/Designer/Builder

    We also have several examples of grp speedboats that broke in two. The engines had been replaced and the power doubled, so they run in 35 knots instead of 25.
     
  10. MikeJohns
    Joined: Aug 2004
    Posts: 3,192
    Likes: 208, Points: 63, Legacy Rep: 2054
    Location: Australia

    MikeJohns Senior Member

    Phil
    When you get a crack you get a stress concentration, what happens next depends on the material and the design.
    In critical components we do fatigue analysis by presuming that a fracture has started and considering whether or not it will propagate, arrester strakes and other fracture propagation preventers can be added.

    Raggi
    The slamming loads produce very high stress levels and the drivers think their boats are unbreakable. Composite hulls don't give a clear indication pre-failure as metal hulls do. Alloy aluminium alloy hulls will often split along the welds from similar abuse and gives fair warning, GRP tends to "let go" in one catastrophic event.

    Metal fractures are generally very easy to spot with visual inspection.
    Steel allows close to perfect repair and has excellent fatigue resistance, Alloy is harder due to the weld strength loss issue but GRP hulls after such failure should become landfill.

    Full keel sail boats of yore tend to be stronger due to that massive girder (the keel) stiffening things up. I suspect modern short keels in GRP particularly will show some significant and nasty failures in the decades to come as well used boats flood the cheaper end of the market. Some of them flex so alarmingly at sea that failure is inevitable over time.
     
  11. rxcomposite
    Joined: Jan 2005
    Posts: 2,752
    Likes: 608, Points: 113, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Interesting topic

    I have seen this fatigue chart in several books and reference materials in my composite materials. I think it is very common except that we don't have any immidiate use for it. As Raggi has pointed out in reference to Gougeon and Dage Gerr books. 60% of the strength is retained after several million cycles.

    There are many theories that apply to composite fabrication. We have the ultimate stress theory, the stress strain theory, Tsai-Hill theory, and Tsai-Wu theory. As a boatbuilder, we are generally exposed to ultimate stress and stress strain theory.

    Class Societies would generrally accept stress to strain ratio of 33%, sometimes more. Meaning they would accept structures whose actual load is way below the allowable stress. Good enough for them, and the quality control tests are simple, such as hardness, thickness, tensile, flexural, and shear test.

    As the applied stress approaches the allowable stress, we enter the class of High Performance Composite or Aerospace Composites. In this class of fabrication, failure rate is integrated into the composite design. More test are carried out, both destructive and non-destructive. This is usually to ensure integrity of the process and repeatabily of the process and that the material will not fail as designed to the operating conditions.

    Fatigue test are normally concentrated on compression under what is normally called "hot-wet" conditions. It usually infers 90-100% humidity and 180-220 degree F operating conditions.
     
  12. MikeJohns
    Joined: Aug 2004
    Posts: 3,192
    Likes: 208, Points: 63, Legacy Rep: 2054
    Location: Australia

    MikeJohns Senior Member

    I am not entierly sure what you are trying to say here.

    But firstly if you are designing to scantling society rules then it is already applied. If you are designing from a load/section modulus approach then you need to apply a safety factor to allow for fatigue. The hard part is assessing the level of stress and the number of cycles in the design lifetime. As aboat repairer or surveyor you can also consider the fatigue damage to the hull given its use and age. So it is always useful.

    Yes SN curves are very common, every design engineer takes them into account. But many people are not aware of the fatigue strength of materials.

    As for an arbitrary 60% after several million cycles.......

    For example a comfortable yacht in a seaway will significantly load its keel and rigging attachments around every 4 seconds. If you have a calculator work out how many days at sea per million stress cycles.... you might be surprised.

    A smaller vessel with a high metacentric height can have a period closer to 2 seconds.

    Slamming loads on a power boat can occur every half second .


    As a builder you will be building to a designers plans ?

    I think you mean allowable stress to ultimate stress ratio
    what you are describing (stress/strain relationship) is the modulus of elasticity (E) which is not the same thing at all.

    Unfortunately there are no books for the layman that popularise this subject or I could recommend you read some.
     
  13. Vega
    Joined: Apr 2005
    Posts: 1,606
    Likes: 26, Points: 58, Legacy Rep: 132
    Location: Portugal

    Vega Senior Member

    I could not agree more.

    I have said something along those lines on another thread (How long lasts fiberglass?). I have pointed also that to the Stress weakening you have to join the weakening that comes from the fiberglass water absorption, that is practically inevitable in an old boat (if he stays in the water).

    http://www.boatdesign.net/forums/showthread.php?t=11455&page=2

     
  14. rxcomposite
    Joined: Jan 2005
    Posts: 2,752
    Likes: 608, Points: 113, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Composite fatigue

    MikeJ

    I agree with what you say. Sorry to confuse you.

    Class Societies furnishes a set of rules for every type and service of watercraft. Calculations are different for a high speed vessel, monohull, multihull, swath, yachts, ect.

    The Class Society is the “father” that stands behind our back, subject to “Periodic Survey” of course. Their set of rules has probably taken into account the cyclic stress that the vessel will be subjected to.

    For Class, the general Safety Factor is 3, for aircraft, it is 2. The safety factor can be expressed in several methods. In whole numbers such as 2 or 3. In percent, termed as Margin of Safety (MS) such as 50% MS or 66% MS, or expressed as limiting stress fraction (stress to strain) ratio such as 0.33 or 0.50.

    As most everyone here in this forum agrees, composites do not exhibit plastic behavior unlike metals. Composites stretch a little, and then fail catastrophically. For all practical purposes, we assume that the stress to strain graph is linear.

    In the Maximun Stress or Maximun Strain theory, Stress is applied to the laminate until it equals the “Allowable” stress, then the ply breaks.

    “Allowable” then depends on how much less the designer would design a design load. It could be anywhere from 0.1 to 99.9% of the ultimate stress.

    If say the ultimate stress of a ply is 10,000 lb/in2, and applying an SF of 3, then 10,000/3=3,333.33 the design load. A MS of 66.66%, or as stress to strain ratio of (3,333.33/10,000) of 0.33 or 1/3.

    When the stress applied reaches 3,333.33 lb/in2, the lamina is considered to have failed or when the actual stress of 3,333.33 divided by the ultimate stress of 10,000 equals 0.33, the ply has failed. In stacked laminate, the first laminate to fail is the first ply failure. In real world, the laminate did not “fail” but has equaled or about to exceed the “allowable/design” load.

    It is every engineers dream to build a racing boat that will break apart after the race or a rocket nozzle that will disintegrate after it has served its purpose. This is where the S/A curve is indispensable.

    If the S/A curve for a particular fiber/matrix combination will degrade by say 10% after experiencing an X number of cyclic stress in its working environment over a period of time the structure is serving its purpose, then he/she design the load (or allowable load) to a little under 90% of the ultimate strength. By how much under 90% depends on the experience of the designer or the available statistical data.
     

  15. alex fletcher
    Joined: Jun 2006
    Posts: 54
    Likes: 2, Points: 0, Legacy Rep: 18
    Location: Ettalong Beach NSW Aust.

    alex fletcher Junior Member

    Thank you Mike for posting such a relevant thread to boat design can you or anyone give a list of relevant texts on the subject wether or not for the layperson to aid forum members develop their knowledge of fatigue as it is of such a importantance in engineering Iam particularly interested in the consideration of the relationship of elasticity of materials and fatigue
    Without the Geeks and Boffins of the would we would all be paddling around in dug out canoes
     
Loading...
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.