Pipes Resisting Pipe Rack Loads

[Althalus]

I've been designing pipe rack for a number of years now. I know that you're NOT supposed to depend on the friction of a pipe or the strength of a pipe for load resistance. It even says so in 4th Quarter 2010 of "Engineering Journal" magazine that you can't use it. And I had thought it was in a PIP standard (Structural Design Criteria). I can't find it.

Now a client is pushing to use it so we can avoid using stringers and x-bracing for longitudinal support against wind and seismic. I've given him this reasoning, but he is asking for the standard. I can't find it.

Does anyone have some more authoritative sources? I DO believe that if enough engineering type journals from different authors can be cited, then that may be enough. But it only takes one standard or code to convince them.

Anyone know?

 

[StressGuy]

Are you saying that the client wants you to consider the piping in your rack as effectively being the bracing for your rack? If so, that may be one of the craziest things I've heard in my career. Steel is there to support pipe, not the other way around.

I'm pretty sure ASCE 7 specifically says you can't count on friction to hold piping in place for wind and seismic considerations, that would probably be a good place to look for a citation.

Edward L. Klein
Pipe Stress Engineer
Houston, Texas

"All the world is a Spring"

All opinions expressed here are my own and not my company's.

 

[BigInch]

IMO it is a common assumption. The pipe crossing the rack beams can be used to provide their lateral support without horizontal in-plane bracing, up to the amount that they can provide by frictional force. I've done that for many a refinery pipe rack design. Whether you can do it or not is usually specified by the client's structural design specs. If not in the client specs, getting him to approve a design note stating the same is usually all that is required.

Independent events are seldomly independent.

 

[Althalus]

I was mistaken. It wasn't the client. It is the other engineers and the project manager.

I work for an offshore company. They're branching out into onshore. So, they hired me as the onshore expert. What I'm finding is that offshore uses the friction all the time. In my onshore experience, we comment on it all the time, but it isn't part of the structural analysis. It's basically an undefined factor of safety. Hence the clash in design philosophy.

BigInch, are you onshore or offshore? Do you perform the structural analysis of the rack?

StressGuy, I looked up ASCE 7. It says you can't use friction to resist SEISMIC forces. But it doesn't say anything about wind. It would make sense that you could make that extrapolation, but if someone has been using a method for years, nothing short of a definitive statement will suffice.

ALSO, I found out that "Engineering Journal" is actually a publication of AISC (at least that article came from them). So, I believe I can use that to set the standard for the office.

Thanks,

 

[BigInch]

ON and OFF shore experienced, since 1975. Structural engineering PE. Houston, New Orleans, Gulf Coast learning center based, including hurricane wind loads. My racks are still riding high at Exxon, Dow/Conoco/Enron/Badische and a multitude of other plants.

Seismic analysis can't assume friction is present because vertical accelerations within the seismic wave may cause a liftoff of the pipe at the exact wrong moment.

Independent events are seldomly independent.

 

[StressGuy]

Well, I've definitely learned something new.

Though, I can't say I've come across many racks around here that didn't have stringers connecting the main bents.

Edward L. Klein
Pipe Stress Engineer
Houston, Texas

"All the world is a Spring"

All opinions expressed here are my own and not my company's.

 

[BigInch]

StressGuy, It's a structural thing.
And yes, they all have stringers between individual bents, attached at the columns. Stringers are there to connect the racks at the columns (resisting about 10% of pipe weight as a net friction load and to keep the effective length of the columns reasonable, so as not to require the columns to have fixed base plates with their 2xL buckling length), but they do not provide lateral support to the beams actually holding up the pipes. Providing sufficient lateral support on the beam's compression flanges (against lateral bow-out failure) allows the beams to be designed to use a higher % of bending yield stress, say 24 ksi instead of 20 ksi or less.

Independent events are seldomly independent.