# Attached plate effective width

Discussion in 'Class Societies' started by Roni, May 18, 2020.

1. Joined: May 2020
Posts: 30
Likes: 0, Points: 6
Location: Brazil

### RoniJunior Member

Hi there,

I am working the scantlings on a day charter catamaran in accordance to DNVGL HSLC rules. Don't worry, this is not a design that will go into production, just a exercise on my part. I guess that this whole quarantine has finally given me the free time study boat design, an old wish of mine.

I've sketched ou a typical srtuctural section. The bottom is solid fiberglass and the sides are cored. There are 3 stiffeneres in the bottom, plus the lower deck, dividing up the panels in the bottom. My question is how should I treat the attached plate for the stiffeners calculation that are between the solid bottom and cored sides? Should I calculate de effective witdth of the attached plate with for each side (solid bottom and cored side) separately, and use half that width with the respective laminate to calculate the stiffener section properties?

Thanks a lot

#### Attached Files:

• ###### section sketch.jpg
File size:
69.1 KB
Views:
197
2. Joined: Oct 2008
Posts: 7,323
Likes: 1,216, Points: 113, Legacy Rep: 2488
Location: Japan

Hi Roni,

Welcome to the forum.

If you look at the rule in Pt.3 Ch.4 Sec.6, 4.2.4 where you calculate the effective breadth, that "B" is "...distance between the stiffeners or girders in adjacent fields.."

So in your example, it is this:

But for your cored region, that is subtly different. The distance is "tehncially" the same, that being the distance from the upper most Long.t to the next, or in this case appears to be the deck.
However, as noted in the blue circle

You have no continuity of structure....so this region is the weakest part.

For your cored sides to work 'effectively', you need to take the core to the long.t

This is the same as we did for a Catamaran CTV some years ago:

The bottom was single skin, and the sides - sandwich core.

So in the blue region, for this sandwich core to work and be effective, the "support" is the long.t.
You need to also make sure that the long.t's strength is calculated as being supported between the WTBs too.

Good luck

Last edited: May 18, 2020
Roni, gonzo and RAraujo like this.
3. Joined: Sep 2011
Posts: 7,002
Likes: 563, Points: 123, Legacy Rep: 300
Location: Spain

### TANSLSenior Member

The lower deck, which I suppose is resistant, would be the limit of the plate associated with the reinforcement. The separation between reinforcements, for that specific area, can be taken as the distance from that deck to the side reinforcement. The associated plate, in addition to this, will depend on what the CS says about how it should be taken and what width. Some rules give a maximum amount for that width, regardless of the spacing between reinforcements.
The transition from monolithic laminate to sandwich must be corrected. As it appears in your scheme, it cannot be done, should you extend it to the reinforcement, or, if not possible, put a new reinforcement in the transition.

4. Joined: May 2020
Posts: 30
Likes: 0, Points: 6
Location: Brazil

### RoniJunior Member

Thanks a lot.

I was very much in doubt on how to do the transition from the solid bottom to the cored side, and the placement of the stiffener. If I understant correctly I should bring the cored laminate all the way to the stiffener, to keep continuity in the stucture and eliminate weak points.

The maximum effective width is, in accordance to the rules, equal to the load width.

And yes the cored side is supported by the top-hat stiffener in the lower end and the deck in the upper end.

Concerning the bottom panels if possible it would be best if the lowerdeck could count as a support also, reducing the width of the panel substantially. How can I guarantee that the lower deck is stiff enough for that. Can I treat it as a flat bar stiffener and run the calcutions? What effective height for the flat bar could I consider?

Once again thanks a lot for the replies, structural enginnering is fascinating. I really appreciate the knoweladge and experience shared.

#### Attached Files:

File size:
53.9 KB
Views:
152
File size:
57 KB
Views:
170
• ###### solid bottom - cored side transition detail sketch.jpg
File size:
89.1 KB
Views:
159
5. Joined: Sep 2011
Posts: 7,002
Likes: 563, Points: 123, Legacy Rep: 300
Location: Spain

### TANSLSenior Member

The lower deck must be considered as if it were a deck, that is, panels with reinforcement on the lower part. These reinforcements can be profiles or, given their proximity to the bottom, longitudinal girders and/or floors.

6. Joined: May 2020
Posts: 30
Likes: 0, Points: 6
Location: Brazil

### RoniJunior Member

Hi TANSL,

The idea is to support the lower deck with floors between the WTB. If the lowerdeck is dimensioned as a deck, divided up in panels with load as a pressure normal to the surface, how does that guarantee I that is stiff enough to be one the boundries of the bottom panels? At first glance it does seem that it would be, but I wouldn't trust just my feelings on this.

7. Joined: Sep 2011
Posts: 7,002
Likes: 563, Points: 123, Legacy Rep: 300
Location: Spain

### TANSLSenior Member

Forget de lower dock and think about the main deck. How do you guarantee that it is stiff enough?. You use the same method with the lower deck.

8. Joined: May 2020
Posts: 30
Likes: 0, Points: 6
Location: Brazil

### RoniJunior Member

That does make a lot of sense

9. Joined: Sep 2011
Posts: 7,002
Likes: 563, Points: 123, Legacy Rep: 300
Location: Spain

### TANSLSenior Member

I don't understand your question either. I do not know why in this case the doubt assails you and instead on the upper deck, or on any other deck that is not called "lower", you do not need to check if "it is stiff enough". Perhaps it is that in this case there is something special that I am not able to understand. Could you or someone who understands explain what the problem is so that even I can understand it? I do not pretend to doubt anything or want to offend you, honestly, I do not understand your problem. If the deck does not resist the load that the CS says it must bear, then you reinforce the deck.

10. Joined: May 2020
Posts: 30
Likes: 0, Points: 6
Location: Brazil

### RoniJunior Member

No offense, I was just saying that what you said makes sense. No irony. My doubt is wheter the lower deck could be considered as the edge of the panel for the bottom panels calculations, if it gives it enough support. Correct me if I am wrong as far as I understand the panels edges have to be stiff enough to act as a support for the panel. I was just wondering if the lowerdeck meets this criteria, when it is not checked as the stiffeners for the load normal to the hull bottom.

As you metioned the deck does support the side plate, and is not analyzed for the pressure on the hull side. I guess its a satisfactory assumption that the main deck will be stiff enough to support the sides, so that check is not needed. I am assuming that assumption is reasonable for the lower deck to.

Does that make sense?

11. Joined: Sep 2011
Posts: 7,002
Likes: 563, Points: 123, Legacy Rep: 300
Location: Spain

### TANSLSenior Member

12. Joined: Oct 2008
Posts: 7,323
Likes: 1,216, Points: 113, Legacy Rep: 2488
Location: Japan

That is correct.
The 'weak point' is that the core does not attach to the the support. If the core does not attach itself to the Long.t, it means it gets no support structurally at the ends and therefore that core panel must be treated as if a cantilever, not a built-in supported beam.

Yes, this is commonly done on structures. It also leads into your orginal question above too. Since if you understand structural theory then structural design becomes much easier.
The support for the core panel, is that long.t stiffener. BUT, and this is the important part to understand, if the long.t stiff was just a simple flat bar, would that be stiff enough?...well by inspection no. Ok... if it were a huge TEE hot hat stiffener, would that be enough?... one could say by inspection yes. But where is that cross over from not stiff enough to - stiff enough? Well, that is provided here:

So, you assume a span, the span being from WTB to WTB. The effective breadth of that Tee Long.t stiffener is as per rule alliances (you can usually use 50t of the web as a rough rule of thumb too - where 't' is web thickness).

So back to your can the lower deck (sole) be used, yes and if you understand how to make the Long.t stiffener stiff enough to be a structural support for the core panel to be a simply-supported or a fully built-in support (rather than a cantilever) , then this is no different. You treat is exactly the same way, like this:

The deck becomes the "web". The sides shown in red, the flanges, are taken as an effective breadth of the attachment. There is a discussion about effective breadth is in this link HERE, where the poster asking the question doesn't understand how to apply the rules - most likely from also not understanding basic structural theory. So, this is where understanding structural theory answers all you questions - don't over think things, just break the structure down into simple parts to be analysed as a simple beam.

So, now you have a web + 2 flanges. This = a simple I-Beam.

So the I-beam is now being supported from WTB to WTB.
And since the I-beam is slender, in the sense that its web thickness to depth ratio is high, it is likely prone to buckling. And in this case buckling is a major structural failure. So you need to check.

So using the I-beam, you can easily calculate its modulus. And then from the applied hydrodynamic/hydrostatic sea pressure, you have a load and therefore, you can calculate the stress, one assumes it to be low. It is by inspection to be very low, as it is a huge I-Beam. BUT... you need to calculate the stress to apply the load to the web. To check if the web will buckle. If it does, when you establish what stiffness of panel is required to pass. This is done by either increasing the depth and skins of the deck and/or, add small transverse anti-buckling stiffeners or small mini-frames.

Sorry - EDIT - forgot to add.
Thus your spans (using the deck as an I-beam for support) now become this:

That's it.

Good luck.

Last edited: May 19, 2020
Roni and DCockey like this.
13. Joined: May 2020
Posts: 30
Likes: 0, Points: 6
Location: Brazil

### RoniJunior Member

Thanks a lot for the clarifications. Since you've been very generous with your replies I'll throw another question.

If I understood correctly in order to check if the sole is adequate as support I would consider it as the web of an I beam, with the flanges being the attached plate, and check for buckling of the web. In case of the DNVGL rules as per section Pt3 - Chap4 - section 6.4 Bucklinf of beams, girders, stringers.

However, in order to provide hatchs for access to the bilge, the idea is to have cut-outs in the sole, such in the skecth below. In that case it seems to me that the whole I beam sole as web model is no longer valid. Would treating the sole as a inverted t bar (with the attached plate as lower flange) with a height equal to the part of the sole before the cut-out be reasonable?

14. Joined: Oct 2008
Posts: 7,323
Likes: 1,216, Points: 113, Legacy Rep: 2488
Location: Japan

Correct.

Why would you consider this?
I-beams have holes in them too - called lightening holes.

Whether the web of an I-beam, or in this case, the sole, have a hole in it, makes no difference. It is treated exactly the same.

And in your example, all you need to do, is add anti-buckling members to to do just that - prevent buckling.
Like this:

The red lines being either simple members like a FB which can be sniped at the ends as it is not necessary to fix their ends, or the red lines can coincide with a frame, and thus be a frame too!
Make sense?

Roni likes this.

15. Joined: May 2020
Posts: 30
Likes: 0, Points: 6
Location: Brazil

### RoniJunior Member

I thought that as the hole is large in comparison to the overall dimension of the web that they would have to be treated differently. Concerning the sole and its treatment I was left with a doubt. In this picture you have drawn the load in blue acting at one side, thus treanting the sole as the web of an I beam in bending.

However, since that load is hydrostatic pressure wouldn't it act on the other side to? In that case wouldn't the sole act as plate under compression, and my check buckling of said plate under compression.

In any case if I understand correclty adding the extra frames or flat bars as you have suggested you prevent the bucking in either case, correct? With such anti-buckling members can no consider the dimension of the web perpendicular to the load in the buckling calculations as the distance between frames or between frame and WTB, correct?

Thanks a lot for the free lessons on strucural theory.

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.