Designing a 9 meter sailing catamaran

Discussion in 'Boat Design' started by Pammie, Jul 25, 2018.

  1. Ad Hoc
    Joined: Oct 2008
    Posts: 5,752
    Likes: 265, Points: 83, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Because the questions you ask, imply to me that such information is either missing or just not understood.
    Structural design is very straight forward, but, methodical....so perhaps your methodology is slightly different (because this is perhaps your first time doing it) and as such, draws me to conclude differently from what is actually being done. That's all.

    When doing structural design you must know where such "breaks" or "discontinuities" are, as these either aid you or hinder you, when you begin the overall arrangement. And thus, knowing this information from an holistic overview tends to drive the whole procedure in how you achieve your objective.

    In layman's language, it is like putting the cart before the horse! :D

    And that's why....QED
     
  2. Pammie
    Joined: Dec 2015
    Posts: 87
    Likes: 5, Points: 8, Legacy Rep: 10
    Location: Eindhoven, Netherlands

    Pammie Junior Member

    @RX: great explanation!
    concerning point 2: ISO part5 defines bottom as: below the waterline. ISO part7 defines bottom the way you do. That way would be more logical to me, it also reduces calculated loads at the plates just above 50 degree angle. With bottom defined as waterline the calculated shear stress in this area is bigger than that of the bottom (because curve is smaller).
    point 3b. OK, that means I'm not allowed to mix longitudinal and transversal sections? Also not if a transversal section with an aspect ratio of around 1 is shown to be able to handle the pressure? Ofcourse this has to be seen, but what I mean is this: I've been looking at pictures of catamaran hull for quite some time and the only ones where you see longitudinal stiffeners is in plywood boats. In foam core (and massive glass) boats they build bunk boxes (longitudinal stiffeners) mostly connected with an internal structure that connects front and aft bunkbox. Of course I don't know the laminating schedule of these boats.
    Point 4. I think it will get more clear when I start to calculate.
    Point 4b. I didn't fully understand this rule but do now. I can use that!
     
  3. rxcomposite
    Joined: Jan 2005
    Posts: 1,812
    Likes: 122, Points: 63, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Pammie-
    I will have to answer this in two part because the 2nd part is lengthy and rather confusing.
    2. Anything below the waterline is calculated as bottom pressure. In the simplified ISO rule, there is no distinction where in the bottom, depth that is. In the basic Jasper and Heller rule and any class rule, the pressure in the bottom is defined from the distance above the base. The base distance has the highest pressure and diminishes as it reaches the waterline. Above waterline, the pressure greatly diminishes as there is no hydrostatic pressure.
    2a. Any panel occurring after "A" that is below waterline is treated as bottom pressure. It does not care whether that "B" panel is partially below waterline ie above waterline. The higher pressure is used.
    2b. If the panel starts from the bottom and ends up in the deck to hull knuckle "A", mainly because it is almost flat or the knuckle bend is greater than 150 degree, it is treated as the bottom panel pressure. This occurs mostly in the forward section where the bilge radius diminishes or completely vanishes.
    3. You are allowed to mix transversely framed or longitudinally framed boats. A mixture of both. Since a panel stiffness is defined by its boundaries, you can add more transverse or longitudinal to make the panel smaller and stiffer. For example, if the panel span is too wide, the addition of a floor, furniture, box,or cabinet whose edge is fixed to the panel becomes an additional longitudinal or transverse.
    3a. With a smaller panel then, you can calculate the load the panel will support. Just be wary of the AR rule which follows next.
     
  4. rxcomposite
    Joined: Jan 2005
    Posts: 1,812
    Likes: 122, Points: 63, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Part 2
    The ISO standard bending moment computation for the non isotropic panel is 83.33 x kc^2 x 2k2 x P x b^2 x 10-6 (Eq 34). Look closely at the coefficient 83.33. That is the result of k2 value of a 2:1 panel ratio and is the result of of the long form computation. For >2:1 aspect ratio it will not matter much but if you go <2:1, then it matters a lot. That is the Alpha Beta rule. That 83.33 coefficient changes. Try the long form computation and you will see that eAR (effective Aspect Ratio factor) is dependent on that. The 2 way rule can be programmed in Excel.

    There are other bending moment rule I see. Eq 52 for dissimilar materials or wood. There maybe others as well for sandwich but I see only the calculations for section modulus. I see on other Class rule, the bending moment formula is different for sandwich laminate vs a single skin. I have to read again ISO rule.

    Eq H4 and H5 is for the 2 way rule, the Alpha Beta rule, or the long form for an isotropic panel. An isotropic is a laminate whose strength is the same in an x,y direction, that is 0/90 degree laminate. A +45-45 laminate is also isotropic (but with degraded properties) in the sense that strength is the same for x, y direction.

    When the AR of the panel becomes square, or almost square, the primary and stiffeners supporting that panel becomes nearly or of the same height. This becomes a grillage structure whose joints of the frame/stiffener becomes critical weak points (not discussed in ISO). A stiffener height should not exceed the primary stiffener height by more than 50%, (Not an ISO rule)
     
    Last edited: Aug 9, 2018
  5. Pammie
    Joined: Dec 2015
    Posts: 87
    Likes: 5, Points: 8, Legacy Rep: 10
    Location: Eindhoven, Netherlands

    Pammie Junior Member

    @Ad Hoc: well : in layman's language: at first I had a nice Gerr cart with the horse in front of it. Then you paraded all different kinds of other cart for the nose of the horse from which I choosed the ISO cart. As it has much more ropes and beams on it than the first one I better make sure that all is OK before I start riding :D
     
  6. Ad Hoc
    Joined: Oct 2008
    Posts: 5,752
    Likes: 265, Points: 83, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    A trot or a gallop.??? :rolleyes:
     
    rxcomposite likes this.
  7. Pammie
    Joined: Dec 2015
    Posts: 87
    Likes: 5, Points: 8, Legacy Rep: 10
    Location: Eindhoven, Netherlands

    Pammie Junior Member

    @Ad Hoc: both!! I learned: first do it good, then do it fast.
     
  8. Pammie
    Joined: Dec 2015
    Posts: 87
    Likes: 5, Points: 8, Legacy Rep: 10
    Location: Eindhoven, Netherlands

    Pammie Junior Member

    I made a spreadsheet of Annex with alpha beta rule which shows that my laminate might not be sufficient. Main problem is in the sides because of panel size. As you all said. Still working on stiffeners.

    @RX: 2/ 2a/ 2b: I have stiffeners slightly above the waterline in almost all sections. So no problem with that. But as I understood from 6.2.5 the load is taken in the middle, or as an average of below/ above waterline area's.
    part 2: as I understood it Eq. 33 and 34 apply for all plates. Calculation with section modulus is just a special situation.

    I have some questions about Annex H and interpretation of its data:
    1. Why is +-45 degrees laminate excluded from Annex H? Because it behaves different in combined loading situations? Pressure, torsion, temperature, moist? I suppose that if Annex H shows that such a combined laminate is critical that CLT analysis would not make it better?
    2. I have learned a lot of my escapade the last weeks, but 'm still not sure that I'll able to perform a CLT analysis that does make sense. So I was wondering or it might be easier to make a test section and measure failure load? Ofcourse this is not explicitly specified in ISO. An argument for this could be that there are many assumptions in the proces: the used laminate properties are for polyester laminates where I use a high quality epoxy. Discussion about Part5/ Part7 loads (part 7 could be official in a few months), etc.
    3. If a plate section has a compliance factor for a ply of 0,8 in one direction and 1,8 in the perpendiculair direction does it fail? Is there a simpified failure criterium for laminates?

    Section G3. my round bilges and hard chines don't fit in the table so I have to calculate them myself. My questions is: why does Part5 calculate on a 2x 150 mm basis? I would say that effective plating width (11.6) would make more sense but is much smaller in my case (38 mm).
     
  9. rxcomposite
    Joined: Jan 2005
    Posts: 1,812
    Likes: 122, Points: 63, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Pammie,

    Aah, the Alpha Beta rule. Quite complex topic. Bending moment rule depends on isotropy of materials and AR of panel. You will have to jump from one rule to another to come up with a meaningful analysis.

    In the example cited in 6.25-2, it was said 30% bottom, 70% side, pressure should be averaged. I was thinking along the line based on example as 70% bottom, 30% side. I would choose the one with the higher pressure. Better err on the safe side rather than assign an average for bottom panel analysis.

    Jump then to rule 8.4 where there is a limiting pressure load for very large panels where kAR is mentioned.

    1. "Why is +-45 degrees laminate excluded from Annex H?" Because it says "The method is strictly limited to laminate schedules which are largely composed of 0/90". and "This annex is intended to be used to analyze laminate stacks for cases were the laminate schedule is complex and cannot be regarded as quasi-isotropic or to analyze a stiffener either made of dissimilar material or not". Well, biax +-45 is a balanced laminate, isotropic in the sense that 0 or 90 degree, the strength/modulus is the same except that it is degraded compared to 0/90 WR. Where it excel is shear, where it is most useful as web of a stiffener or torsion in a plate laminate.

    2. The mechanical property of the biax +-45 is given in table C.7 though it would be rather confusing as it says 0/90 degree direction and the strength is higher than a WR. Other test data shows the strength/modulus goes down as the ply is rotated at an angle. ISO rule H.2.1.11 eq H.2 and H.3 should prove this. It may be daunting to calculate but if you plot the strength/modulus vs angle of rotation, it will show that it reaches its lowest at 45 degree. I trust this formula as it is common in BV, Filament Winding, and CLT theory. I use this a lot and it can even be accurate for different materials like Eglass, S2 glass, Aramid, or Carbon Fiber. The slope varies for different material owing to the different mechanical properties. I will try to post a graph of the results later as it is embedded in my spreadsheet(s). Once you prove this, you can substitute the type of material (biax) and its mechanical property in Table H.2.

    There are two forms of CLT analysis. One uses the standard ply angle rotation (eq H.2 and H.3) and strain method analysis as outlined in most classical lamination analysis (including ISO) and the other is the macro mechanics theory. Macro mechanics uses the matrix algebra and the ADB/Inverse ADB presentation table. I hope you are not punishing yourself yet with the matrix algebra method. This is the highest form of composite analysis using formulas. I remember it uses 6 material property inputs.

    ISO does not have properties for Eglass epoxy laminate. It should be 5 to 10% stronger. I guess you should rely on test data or other published data. Another conflict is they have properties for Aramid/Carbon Fiber but polyester is not compatible with Aramid/CF. Was the result used epoxy laminate?

    3. I am not aware of the 1.8 compliance factor for the perpendicular direction or I maybe misreading you. Where or what part rule is this?

    Where is "Part5 calculate on a 2x 150 mm basis?"? Is it 2 x the 150mm per side illustration on V type hull? Total 300 mm keel width? There is a different rule on effective width of plating. ISO is looking for an equivalent section modulus of the keel so if it does not fit in G.3, you have to calculate it. Rule of thumb width is 10% of breadth and 1.5 x thickness (minimum) of bottom laminate (single skin for keel).
     
    Last edited: Aug 26, 2018
  10. Pammie
    Joined: Dec 2015
    Posts: 87
    Likes: 5, Points: 8, Legacy Rep: 10
    Location: Eindhoven, Netherlands

    Pammie Junior Member

    Bottompressure for lower side: I see I already took full bottompressure

    I have no Kar below 0,25. Would it make sense to remove ringframes to reach a better l/b? That is: at the side and deck as that is where most square panels are. Reduces Kar and load, but I suppose it will cost me more somewhere else.

    2. Strength biax (table 7: 95 MPa) more than WR (table 4: 277 MPa)?
    a. My tests on biax : strength: 111-2*4,8= 101 MPa (tensile, tensile perp., compression). I have no sufficiënt data for modulus.
    b. H2/H3 is easy for 45 degrees: (Table 4a) Vf=0,6, E1=E2=38000*0,6-5000=17800, G=700*0,6+2240=1344,Eb=El=1/(0,707^4*((2-2*0,3)/17800+1/1344))=6015 MPa
    c. Table C7: Et = 0,45Et (WR) = 0,45*17800=8010. Which is not exactly 6015.
    Because of a) I interpreted Table C7 for biax as +-45. For calculation used 95 tensile strength and 8010 modulus. As tests were not complete (some because of strain measurement problems, some because of test appendage setup) I used the table according to EL-b.

    Had a look at some CLT software but they took NA in the middle so stopped with that. Won't be doing that for myself.

    3. Sorry I wasn't more clear about hard chines. It was about Figure G.4 indeed. The V is not for keel but for deck chines. In the front side the distance between the 2 chines is less than 300 mm so both chines will influece each other. I have some idea's about that but will report later.

    I added my hull load spreadsheet. Has some loose ends still. The section with the light and dark blue background is a modification of Annex H. The lightblue is the longitudinal calculation which uses Eq H4, the darkblue is the transversal using Eq H5. Both give a bending moment compliance factor in Annex H column 22. Eg for ply 2, long cF = 1,74 (cell B258) transv cF = 0,506. The question is: will ply 2 fail? Or does it imply that ply 2 is mostly supported in longitudinal direction, so transversal load will not occur? And what is the extent of that?
     

    Attached Files:

  11. rxcomposite
    Joined: Jan 2005
    Posts: 1,812
    Likes: 122, Points: 63, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    I am showing the load angle rotation. The first graph is the CLT method with formula as used by ISO/BV/Filament winding VS the formula used by LR. Note that the slope varies but arrives at the same modulus at 45 deg angle.

    The slope do varies depending on the material properties of the material and the type of weave. Woven roving usually have low strength/modulus to start with but does not lose much as the load is rotated. About 35 to 45% remain because the fibers tug on each other. Unidirectional, on the other hand has a very strong strength/modulus to start with but quickly loses its properties as it reaches 45 deg. 9 to 10% remains. That is because the resin (weakest element) is the only one holding the fibers together. For Uni, it is usually called the 10% rule. Only 10% of the original property remains.

    That is why I am still scratching my head how the DB in ISO has 95 N/mm2 VS the WR which has about 91 N/mm2.

    Table H3 and H4 of ISO can be considered the long form of CLT. I am attaching the CLT in what I called the short form or matrix algebra presentation with ADB and inverse ADB results. I wrote it in Excel.

    Any software that would designate the mid portion of the laminate as the Neutral Axis, I throw away. NA is very sensitive to modulus and strength. It shifts from the midplane quickly.
     

    Attached Files:

    Last edited: Aug 27, 2018
  12. rxcomposite
    Joined: Jan 2005
    Posts: 1,812
    Likes: 122, Points: 63, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    I revisited my ISO spreadsheet and tried to extend it using the proportiona method where the panel extends over two general areas. Drawing an imaginary hull sections, it seems that in station 5, I would be using slightly higher than calculated pressure for the side. Station 6 is a no brainer. I use bottom pressure for the bottom and side pressure for the sides. Sta 7 where the panel extends all the way from the sides to the bottom, ISO is not explicit about that. Do I use bottom, side, or average?

    The manual does not seem to have allowances for slamming factors. Based on Class Rules (you also started calculations based on Class Rules) The vertical acceleration increases as you move (forward) of the CoG thus increasing pressure. The bow buries itself in the oncoming waves.

    Standard reinforcements requires extending the depth in the forward end as shown in blue dotted line. Long calculations using standard Jasper and Heller and some class rules also points out to dramatic increase in pressure above the waterline if plotted at a distance away from CoG, indicated by orange dotted line.

    Another rule (not in ISO) is that if the transverse is greater than 150 degree, treat it as one and use bottom design pressure. That is why I said, use the bottom pressure.
     

    Attached Files:

  13. Pammie
    Joined: Dec 2015
    Posts: 87
    Likes: 5, Points: 8, Legacy Rep: 10
    Location: Eindhoven, Netherlands

    Pammie Junior Member

    A short answer:
    where do you see WR at 91 N/mm2? In Table C4b I see 277 N/mm2 at mass content 0,6
    The CLT spreadsheet looks great!
    ISO does use a slamming factor in Ksls which influences the bottombaseload. And, as side pressure also depends on bottompressure also this will be higher in the forward section. In fact side pressure goes up on the forward half of the hull.
    Station 7: ISO says: bottom is below waterline, so you should take the average. The main difference between side and bottom is that the first one has less slamming. When slamming is regarded as proportional to projected area, the projected area of station 7 is small and almost equally divided over bottom and sides. But will take your advice in account!
     
  14. Pammie
    Joined: Dec 2015
    Posts: 87
    Likes: 5, Points: 8, Legacy Rep: 10
    Location: Eindhoven, Netherlands

    Pammie Junior Member

    Just to be sure:
    When I incorporate formula's H2 and H3 in Annex H I should also transform the coordinate system of sigma c/t of column 9. Is this correct??
     

  15. rxcomposite
    Joined: Jan 2005
    Posts: 1,812
    Likes: 122, Points: 63, Legacy Rep: 1110
    Location: Philippines

    rxcomposite Senior Member

    Formula H2 and H3 are formulas used to predict the behavior or Elastic Modulus of a laminate and works mostly for Unidirectional fibers. You need at least 4 principal properties as inputs which correspond to the type of material you are using. From the elastic modulus, you will be able to predict the Poisson's ratio. If not Unidirectional, the type of weave (Stitched, Woven, Twill) will also influence the outcome as there is a coefficient of influence.

    The matrix method of CLT is the best tool.

    I suggest you stay away from this if you are just chasing the properties of a biax. Just use any ISO or any published data available.
     

    Attached Files:

    Last edited: Sep 17, 2018
Loading...
Similar Threads
  1. Peter Vella
    Replies:
    5
    Views:
    498
  2. richardmg9
    Replies:
    21
    Views:
    525
  3. Kkuzmicki14
    Replies:
    8
    Views:
    560
  4. Mely
    Replies:
    26
    Views:
    1,008
  5. useragentseven
    Replies:
    13
    Views:
    853
  6. Joe M
    Replies:
    12
    Views:
    1,029
  7. JohnMarc
    Replies:
    53
    Views:
    2,341
  8. hotwd92
    Replies:
    1
    Views:
    632
  9. Rabah
    Replies:
    11
    Views:
    1,906
  10. the brain
    Replies:
    4
    Views:
    842
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.