Pipe rack forces

[CQT3]

Hello all, the engineering firm I work for has a structural and piping department and we design pipe racks. The piping department gave me really low seismic forces at the directional anchors. I did a quick hand calculation taking the tributary weight for the pipe for each directional anchor and multiplied it by the same seismic coefficient and the results are 4 times larger than what piping gave me. Is this normal? I understand that piping is not straight forward as taking the weight and multiplying it by the seismic coefficient but 4x seems a lot. I am just the junior engineer and I informed the engineer of record but he says to trust the pipers since they are the professionals in their field. I asked the piper about the discrepancy but he is a really old guy and essentially just said it is complicated. The software he uses is CAESAR. Are there any pipers out there that can explain to me how CAESAR works and potentially how the forces are so low? Should I take the piping forces as is or should I modify it for structural purposes?

Thanks!

 

[GC_Hopi]

The design principles behind the analysis are the same. Some people just don't like pulling back the curtain. The main difference with pipe analysis is the restraint conditions, internal ie pipe joints and external ie pipe supports. Meaning the load path may be a little different since it is dependent on the restraint conditions.

The seismic coefficient can be a little different based on the exact equation used. But in the end when you look at the global free body diagram the forces should be in equalibrium so the structure does not move.

Here is a brief description Link

 

[JoshPlumSE]

There is the weight of the pipe and then the weight of the contents. Often these pipes are nearly empty with merely steam or gases passing through them. However, some gravity design loads will be very high because of hydrostatic testings (i.e. when the pipe is full of pressurized water). So, for lateral loading, you are usually talking about nearly empty pipes.

Lastly, some of these pipe systems have a lot of curves and bends in them to control thermal expansion. If so, the piping guys may look at it as a much more "flexible / ductile" system that absorbs seismic load.

Note: I'm no expert, but when I first started my career I worked for a company that did a lot of pipe racks for petro-chem facilities. I spent a number of lunchtimes listing to one of our pipe stress engineers giving a "cross training" class so that we could better understand their issues. That was probably 25 years ago, though. So, any knowledge I gained has been thoroughly blurred in the haze of time and cannot be considered authoritative.

 

[JLNJ]

Don't be shy about asking questions of the piping guy. My experience is that sometimes they understand what they are doing and sometimes they don't. Ask what code they are using. As Josh indicates, ask about full versus empty pipe. If they are using an ASME code, the results might be different than those from ASCE 7.

The local forces seen by the pipe won't necessarily align with the global forces for which you will design your rack; however, you are right to be concerned that you are getting 4x the force.

 

[ThomasH]

I can think of several resons why the results differ, full or empty pipes are one possibility. But ultimately, you are dealing with the same dynamics regardless of if it is a piping system or a space frame. I don't work with piping but I have designed a few pipe supports and I have seen piping software that was based on a beam analysis method. Then the connections and details was designed based on a piping specific code.

Have you provided the piping department with properties for the piping supports? That is, the stiffness of your pipe rack, or have they assumed a stiffness? That is one obvious possible difference.
Another is, could there be a difference in the definition on the seismic load depending on different codes?

You write that the piping guy is "really old", in a positive context that can mean very experienced, in a negative context it can mean something else .

Thomas

 

[phamENG]

Where is your seismic coefficient coming from? Is the Cs for the SFRS? If so, and your SFRS is, say, Ordinary concrete moment frames (R=3 per ASCE 7 Table 12.2-1) and the piping is "Piping in accordance with ASME B31...joinst made by welding or brazing" (R=12 pre Table 13.6-1), you could end up with a 4x discrepancy. Maybe? I don't deal with seismic design a lot but it could be worth looking into. Somebody please feel free to tell me I'm wrong.

 

[bones206]

Agree with others that the effective seismic weight (empty vs. full vs. operating) and the R factor for the piping system have a major influence on the seismic load and are easy enough to check. I would check those two things first.

I am just the junior engineer and I informed the engineer of record but he says to trust the pipers since they are the professionals in their field.

A questioning attitude is healthy and should be encouraged. The fact that the little voice in your head was telling you something doesn't add up is good thing that will serve you well. Always trust but verify... verifying assumptions/inputs shouldn't be left to faith and we can all help keep each other out of trouble through verification. In my experience, pipe stress engineers tend to be weaker on seismic requirements relative to the other aspects of pipe stress analysis.

I asked the piper about the discrepancy but he is a really old guy and essentially just said it is complicated.

Sounds like the same (lack of) mentorship I experienced early on in my career. I've found that those who know what they are doing tend to enjoy passing down their knowledge and explaining things, while those who have just managed to skate by tend to be defensive and stingy with answers. But Eng-Tip people are the former type, so you've come to the right place. Don't quit until you've resolved the discrepancy and satisfied your doubts. Feel free to post your assumptions and seismic coefficient calculation here so we can provide more specific feedback.

 

[CQT3]

Hello everyone, thanks for the responses. I was able to back calculate the CS that the piper gave me and it seems to be based on an R of 12 for nonstructural component and z/h=1. I calculated the CS of the pipe rack lateral system assuming an R of 3.5 which gives a similar CS to the non-structural CS for this unique case. However, can someone confirm that in the case the two CS are different, I should scale the seismic piping loads to correspond to the lateral system CS when checking the lateral system?

Back to the issue at hand, for my sanity check of the seismic pipe loads, I used the exact same CS the piper provided so it shouldn't be the problem. I also used the operating loads the piper provided (I summed up all the operating loads at the nodes closest to each directional anchor) and multiplied it by their CS so the weight should not be the issue since it is based on what the pipers provide (This weight is specifically labeled as "operating load" btw). I also went ahead and assumed a 40 psf loading as the dead load of the pipe and multiplied it by the CS they gave and all results suggests the forces they provided are too low.

ThomasH, I wrote that the guy is "really old" because I think he perceives me as really "green". I found a discrepancy in his load locations before and privately emailed him about it with a marked up PDF and he just responded saying he doesn't have time to make my change request. This implied that he didn't even read my email and just assumed I was asking him to move the anchors around. I had to talk to his supervisor that time to get it fixed.

I have brought this issue up to my structural colleagues and they agreed something is off but the engineer of record says that as long as it is all documented that we got these loads from piping, structural is fine but I am not sure how that works. To make things more interesting, the pipe is going on an existing pipe rack that fails almost everywhere per the current building code (even before the new pipe is added). Therefore the 5/10% leeway for existing structure is being used at a lot of locations which is another story in itself.

At this point, I was just hoping someone can inform me of a good reason why using these lower than expected anchor forces are acceptable. It is difficult for me to get to the bottom of this since I am new, the EOR has given the green light to proceed with these forces, and the deadlines are coming fast.

-Thanks

 

[phamENG]

CQT3 - have you every read up on what the R value is and what it means? If not, I encourage you to do so. In a nutshell, it's a measure of a structure and/or system's ductility and flexibility. The more flexible a system, the more energy it can dissipate under dynamic loading and the resulting dynamic response at the anchorage is much lower as a result. An R=12 means it is VERY flexible, which is why you're getting such low loading. A welded piping system can flex all over the place and deform without ripping off its supports. If you had a brick sewer pipe, on the other hand...well the slightest movement and it would crumble so you'd design it for a higher force to keep it from moving. Make sense?

You don't directly scale the results for non-structural items to the overall lateral force. You will, however, include the weight of the system in the Total Seismic Weight of the structure, and that in turn will be multiplied by the Cs based on R=3.5. So, the U-bolt holding the pipe to the rack need only be designed with R=12, but the bracing and anchor bolts to the floor would be designed with the R=3.5 including the total weight of the piping system (assuming the R=3.5 is correct).

(Again, I'm not in a high seismic zone - if anyone from California or the other more seismically active regions/countries want to chime in and correct me I welcome the critique!).

As for the personnel issues, I don't envy you. It sounds like the EOR is more interested in covering his backside than providing a suitable structure. At the end of the day, the legal liability is on him and his neck is on the line. If you truly believe his design is going to create an unsafe condition that will place people and/or property in harm's way, I believe you have a duty to do something about it. Be prepared, though - it could have a short term negative impact on you. Good on you for speaking to the other engineer personally before calling him out - too bad he couldn't be bothered.

 

[RPMG]

It is as straightforward as multiplying the weight by the seismic coefficient, except it gets a 0.7 factor in ASD.

In my first experience with a piping engineer, I was told to review a 200 page submittal of piping analysis. The last node, representing equipment included a reaction force Rx = 20,000 lb. The piping engineer described it as, "Yes, that's what the analysis says."

 

[bones206]

phamENG's explanation is a good one. The design of the connections and local elements would utilize the R value of the piping system, but the global seismic load on the rack structure utilizes the R value of the pipe rack system:

Local elements of the structure, including connections, shall be designed and constructed for the component forces where they control the design of the elements or their connections. The component forces shall be those determined in Section 13.3.1.

I would just add that for some load combinations you would use the operating weight of the pipes, but for uplift combinations it's more conservative to use the empty weight of the pipes. For some load combinations the hydrotest weight should be considered, but definitely not in combination with seismic effects.

 

[RPMG]

CQT3, proceed per your EOR's direction. It's your EOR's decisions, not yours. It is your job to inform the EOR of the facts. You do not make judgement calls. Your EOR does based on their experience.

Put a note on the drawing that references the old guy's document with design forces.

 

[phamENG]

While I technically agree with RPMG (I don't think anything he said is wrong), I wouldn't recommend using that as your modus operandi. As I said before, it's the EOR's neck on the line and his call - but you start developing your own judgement day one. If you don't think something is right, push back until you prove yourself right or they prove you wrong. If your employer can't be bothered to take the time to help you develop your engineering knowledge, skill, and judgement, they don't deserve the profit they are (or will, depending on how junior you are) be making from your time. Find somebody who will help you become a good engineer, not just somebody who makes pretty drawings of buildings.