Points Loads to Distributed Loads 4

[SubseaDeep]

Hi,

I am in the offshore oil and gas industry. We do analysis on offshore structures to be placed on oil platforms, rigs and boats in RISA 3D and Visual Analysis. The deck loads are then given to the client in the form of point loads/reactions at the base of each column of the structure.
As we have points loads as data, can someone please advise what is the best way to convert a point load into a distributed load on the deck? This is because all decks are rated in lbs/ft^2 or t/m^2. Will putting a thick plate at the reaction point help in distributing the point load. I assume that if the point load is 25 Kips and we put a 12X65 W-Beam, 25" long, then the distributed load will be 25 kips divided by 25x12 sq.in. Will it practically be a correct way to distribute the load on the steel deck?

Regards,

SubseaDeep

 

[steellion]

A) What type of deck is it? Steel, wood, concrete? It could have an effect on the answer. The simple answer is that converting a point load to a distributed load to compare it to the "rated" PSF capacity is not a good idea. It would be unconservative. The deck spans between supports, and needs to be strong enough to support the load between the supports. Then the beams go to columns, columns to foundations....

B) Hire a structural engineer (you're going to get that comment anyway, might as well be me)

 

[CTW]

A column reaction is a point load and should not be converted to an equivalent distributed load. If you complete your example of a 25 kip point load with a 25" long by 12" wide beam, then the resulting distributed load will be 12 kips/sf. This is not practical and unrealistic to compare to a rated deck capacity. It's possible that the platform deck may be able to support the 25 kip point load but certainly not the 12 KSF load.

 

[Ussuri]

I would say you cannot do what you are trying to do. The rating of deck is typically there as an indication for carrying cargo which is spread over an area, not applied locally.

If you are putting something on a deck in terms of local point loading you must do an assessment to the deck capacity specific to the loading you are imposing.

For example of you place your column on the deck arbitrarily it may rest on deck plate, above a bulb flat, on a girder or over a bulkhead. The load carrying capacity of a bulkhead is far greater than just 11mm steel plate. So if you just reduce your column load to a UDL how do you know the local capacity of where you are placing it?

 

[msquared48]

I agree with the others here. Just put a steel beam under the point load large enough to carry it. Use the appropriate beam equation in the AISC manual to compute the shears and moments.

Don't try to make the deck do something for which it was not designed.

Mike McCann
MMC Engineering

 

[weab]

I am assuming that your columns are not going to hit the web of a deck beam. If not, then do like others are saying. Try to add an intercostal beam beneath the deck plate if possible. If hot work is not possible, then add a beam on top of the deck which will span over 2 deck beams. You can bolt this to the deck.

Though you could bolt to the beams below also, if hot work cannot be done, then you can't seal weld around the top flange at the deck plate. At least if it is above the deck, you can see the corrosion when it occurs in the future.

AFTER RE-READING YOUR POST:

Are you asking how to determine if you are overstressing the beams? Though typically the decks may be rated for 500psf or something similar, you should check nearby affected beams when actual loads are known.

To the best of your knowledge, check beams with actual loads and apply the rated load in open areas.

 

[BAretired]

You said:

I assume that if the point load is 25 Kips and we put a 12X65 W-Beam, 25" long, then the distributed load will be 25 kips divided by 25x12 sq.in. Will it practically be a correct way to distribute the load on the steel deck?

The answer to your question is no. 12" is the depth of the beam and has nothing to do with the load distribution on the deck. Provide a sketch of the layout and someone may be able to offer some help.

BA

 

[paddingtongreen]

SubseaDeep, If you meant 25'-0" long x 1'-0" gives you 25 sqft, you are correct, but the beam won't distribute the load evenly. It might carry it to the slab's supporting beams, but it might not.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[BAretired]

I think we need a sketch to figure out what is going on.

BA

 

[SubseaDeep]

I appreciate everybody's help here, and the time you have taken. BAretired, as per your query, I attach a sketch. Maybe it will make more sense as per attached sketch.

 

[Ussuri]

Is your entire structure resting on deck plate alone, in between the under deck girders, is this plate stiffened?

If the deck is stiffened you can put a beam under the columns to spread the load between multiple stiffeners, you need to make sure the beam is stiffer than the deck and so able to spread the load.

If you cannot get the deck plate to work on its own then you need to transfer the load back to something that has the capacity to carry it. Can you span beams between the under deck steel and rest your three columns on top of it? How far is that? Can you change the geometry so your columns line up with the deck steel and increase the size of the beams accordingly?

 

[paddingtongreen]

If I was designing that support, I would start with the columns above the deck beams and build the vessel support level from there. The tail is wagging the dog in your scenario.

The only way I see to support this structure, as designed, is to put long beams, an inch or so clear of the deck, spanning from deck beam to deck beam. The inch could be filled with a baseplate and/or grout. (I assume it is a reinforced concrete deck).

Are there any distributed loads on this deck?

The only other way that I can think of is to analyze the deck for actual loading, using the point loads instead of the uniform load, but it has to be on a case by case basis using the actual load positions.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[BAretired]

Why don't you turn the beams the other way and forget about the deck? The two green lines in the attached sketch represent two new beams just above the deck.

BA

 

[MarineConstruction]

Agree with BA. Easiest solution based on what you've provided.

 

[weab]

I agree, I think that everybody is on the right track. I'll add one thing.

If you are using the deck below your structure, you might find that these addtional beams may be an obstruction depending on length. You might need a skid, that is, with beams in both directions. Add deck plate and a step if necessary, then you could place equipment or storage.

Good luck.

 

[SubseaDeep]

paddingtongreen, it is a steel deck on oil rigs and offshore supply vessels, and there are no distributed loads on the deck.

As per jsdpe25684 to put in a skid, i agree with that. So would need to calculate what load the skid (say for example made up of 4 beams in longitudinal and transverse directions) will put on the deck.

Thanks.

 

[paddingtongreen]

I think that jsdpe25684 means that the skid would span from deck beam to deck beam, as shown in BAretired's markup of your sketch, his suggestion is that if the beams are in the way, you might need a floordeck on the top of your skid, and possibly a step for access.

I would put bearing plates at the reaction points to keep the beam an inch or so above the existing deck so it doesn't hold water through capillary action

When you have the reactions, you need to check that the deck beams will not receive more load from the skid than they would have from the design uniform load. This will probably be obvious but I have no idea of the proportions of the deck relative to the vessel and skid.

Michael.
Timing has a lot to do with the outcome of a rain dance.