Attachment Ship Hull - FE Boundary Condition 1

[BatMan]

I'm looking into an attachment on the deck of ship/hull subjected to wind and wave loads (as it sits on the deck of the hull). If I know what sea-state(wave loads/accelerations) the ship has been designed to, is it necessary to model the complete ship model to know what's going on, on the attachment at the deck or would it be more/less make sense to just model the attachment with the specified acceleration at the support between the attachment and the deck?

 

[VoyageofDiscovery]

Sorry to be the bearer of bad news but you should model the entire ship, fatigue will affect other components further than you think, also without knowing the particulars, if it has a large buoyant effect, in combination with the design water wavelengths, this will have an impact on the entire structural system.

 

[jhardy1]

Wouldn't a better answer be: "It depends ..."?

Sure, if the attachment is "big enough" (mass / inertia / stiffness / dimensions) to potentially be significant as far as the ship's sea motion is concerned, then yes, you might need to analyse the whole ship / attachment system.

But for "small" attachments whose mass / inertia / stiffness / dimension is insignificant as far as the ship's motion is concerned, then you should be able to get the characteristic movements of the ship ignoring the attachment, and use these as base excitation inputs into a separate analysis model for the attachment. (You may need to model an appropriate sub-section of the ship's structure to ensure the load paths between the ship and the attachment are modelled reasonably.)

Would you analyse the complete ship / attachment system for a Heavy Lift Vessel with a 10,000 tonne off-shore oil rig as its deck cargo? Almost certainly!

Would you do the same to analyse a one-tonne deck crane? Probably not!

http://julianh72.blogspot.com

 

[BatMan]

Thank you both for your inputs. I think both have merit. I am dealing with "small" attachment installed on the deck of the ship. Relative mass/stiffness/dimension of the attachment compared to the hull is very small. Since I know what the ship has been designed to (and hopefully get the characteristic movement or acceleration at the deck at the point of attachment), I am hoping to run the ananlysis as a separate model. I'm not sure how the boundary condition will be at the attachment connection. Or maybe the mass of the attachment can be discretized and the acceleration can be applied at those points (??). Or will the boundary condition of the the attachment at the deck be more/less minimal-support-condition (3-2-1 I think it was called?).

 

[dwallace1971]

Start with a free-body of the bit you want to know more about. That should give you a good idea of the loads and constraints you want to apply. I can't imagine you'd need to model the entire ship. If you have masses attached to your hardware of interest, you might be able to model those as point masses connected to your hardware with RBEs. I've used that method to good effect with gun mounts. Anyway, first step is a good FBD. Good luck.

"On the human scale, the laws of Newtonian Physics are non-negotiable"

 

[racookpe1978]

At a minimum, if you are going to model the part + attachment + deck plate at all, then you HAVE TO model the deck plate and its underlaying frames around the attachment points of the part to the deck.

If you don't, bad things may happen.

1. The part (its attachment and its frame or casing) may fail because it is not strong enough to "hang on" to the deck plates and their frames due to the wave impact. Modeling the wave impact will be tough enough on its right!

2. Your attachments (bolting or welding) to the deck plate and its frames may be the weak point. The part is fine. But it ends up breaking free of the deck because the welds or bolts failed. Or pulled through the deck plate.

3. Your attachment is fine, the bolts and welds holding it on are fine, but the deck section underneath pulls out and rips off.

But the whole ship? Almost certainly not warranted.