Pipe Rack Load Calculation 1

[Vikoll]

Hello All

As a new member of piping design team I have to compute and deliver loads imposed by piping on a T-shaped pipe rack of petrochemical plant. These loads will be used by structural for pipe rack design. Wind, seismic snow etc loads are not a concern in this case.

After I calculated load due to dead weight of water filled piping including insulation I need to determine horizontal loads due to friction and pipe deflection, acting on anchors and guides and space these restrains properly.
In particular I need to know how to calculate a force acting an anchor that is on a branch and approximately 40m away from the header and resulting from piping (header) deflection due to branch thermal expansion.
I am also wondering what percentage of friction force from one side of anchor should I take into account (there is counteracting friction force from other side of the same anchor)?

Should I consider any DLF?

Could anyone guide me through this process please?
Any helpful comments and references will be truly appreciated.

Thanking in advance
Vikoll

 

[BigInch]

The structural guys probably already do this part,

loads due to friction in any direction on non-anchored and non-guided pipes attached to supporting members are the weight * friction factor (usually considered to be 0.3)

(you should use as few as possible anchors and guides, NOT one at every support please) That means ONLY where you need them for stress control purposes. Its entirely normal to have some snaking, if its limited enough that adjoining pipes arn't touched.

The structural guys will take those loads, if they don't already know them (they usually just put about 40 psf on piperack spans and figure the uniformly distributed load from that to the beams), and figure and design for the resulting deflections.

The structural guys will not know any of the guided and anchor loads as those depend on your pipe config, op press and temps. You must advise them if any exceed the normal frictional loads W * f.

Pipes are space frames, so find a table of typical pipe configurations and resulting loads from temp change and calculate those yourself, or learn how to run Caesar, Triflex, some pipe stress program if you can't figure out the space frame formulas.

Most of the time you should take the entire anchor force without regard to what's happening on the other side. It will generally be conservative. Most of the time, if there is some counteracting force on the other side, you don't need the anchor located where you've got it. Why would you need it if forces are pretty much balanced, as there is no big differential to transfer to the anchor and underlying support? But, you as the stress engineer must know when and if it is conservative or not to do that. If you can't tell, run Caesar.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[doberdorks]

Most of the time, if there is some counteracting force on the other side, you don't need the anchor located where you've got it. Why would you need it if forces are pretty much balanced, as there is no big differential to transfer to the anchor and underlying support?

So that you're sure in 40 years the pipe will still be where you put it. It also makes life easier for the stress analyist.

Grin.

SLH

 

[Vikoll]

Thanks BigInch for your valuable input
I have to do this manually. That is why I am not running CAESAR. I also have to restrain the pipe 100 m long. That is why I expect some anchors and guides on it. Even ,if I put the anchor somewhere in a middle to distribute thermal movement both ways evenly, I still have to estimate the unbalanced horizontal load on anchor which will be present during different stages of plant operation (start-up, hot-water-flush, etc.)
If I just take all friction on one side of the anchor and split it between two bents of anchor bay, it will be probably huge and will not reflect real situation.

I need also to compute an unbalanced force due to header deflection and I mentioned above.

Could you please share your valuable experience to help to have this issue resolved.

Thanks once again
Vikoll

 

[BigInch]

What I was trying to say is, anchors are needed only to control deflections, especially into sensitive equipment. Anchoring pipe where you don't need it only replaces deflections with forces and stresses.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[Vikoll]

Thank you Gentlemen,

The only question remains is if there is a way to compute the force imposed on structure by anchors if the pipe has expansion loop between anchors and pipe thermal growth in each direction (loop deflection), loop dimentions and pipe properties are known.

Thanking in advance
Vikoll

 

[BigInch]

Yes its a frame made with "O-beams" and instead of a foundation it has a couple of horizontal "O" beam cantilevered supports holding up the frame. Maybe the cantilevers are guided or anchored or both. Oh and in addition to dead and live and wind and earthquake loads the frame expands and contracts with temperature and internal pressure. There are standard tables for these expansionn loops and many other typical loops and bendy pipe configurations in the "Piping Handbook" by Crocker and King, W.M. Kellogg Engineering Co. Pipe Stress Handbook and other refs. Need to find a copy of those if you're going to be doing this by hand or spreadsheet.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[BigInch]

PS don't anchor in the middle of the loop and allow growth towards the ends. That defeats the purpose of having a loop. Anchor the ends so expansion is towards the loop and the loop bends to convert axial load from the horizontal pipe to bending moments around the loop. Axial loads will propably be reduced to 10% of a straight pipe, or even less, if the loop can be made tall enough. If you can't make it tall enough, put another horizontal loop at the high level.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[RossABQ]

Anchoring pipe where you don't need it only replaces deflections with forces and stresses.

That is the most succinct explanation of support design I've seen! I am amazed at how many designers feel the need to control every inch of pipe, without regard for the consequences. Not to mention the cost.

ASME ran a great article on overly-constrained designs about 10 years ago.

 

[BigInch]

Right. Every time somebody puts in a guide, a stop or an anchor, stress goes UP. The "secret" is to direct longitudinal growth (as subtily as possible) away from equipment.

When you have equipment at both ends, a loop in the middle is A*E/L times better than anchors at both ends. Since E=Young's modulus, its a lot of times better. If end equipment is still overstressed, increase the height of the loop. If still overstressed, then add stops at both ends to force growth towards the loop.

A good desktop modeling tool is a piece of thick copper wire. Make a close-to-scale model of your pipe configuration using that piece of copper wire. Bend it into the shape of your pipe configuration and set it down on the desk. Push the ends together to see where the deflection goes and what bends the most. It can give you a real feel for what you want to try to do with the real pipe.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein