Pipe stress analysis - criteria to assume zero friction factor 1

[plasticnoob]

Good evening all!

I am a project engineer who have recently been assign to do tie-ins a process line which runs from a furnace to a reactor.
Due to the high Temperature and pressure, I told my design contractor to run a pipe stress analysis to see if the nozzle would take the additional load we are adding onto to the line.
However my designer told me the nozzle loading on the furnace fail miserably and the main factor was because the previous designer assumed most of the nodes to have zero friction factor, which in my contractor point of view is incorrect.

Can i check with the stress analysis experts here on which conditions can we assume the friction factor to be zero?

If you guys need more details, please let me know.
Thank you.

 

[LittleInch]

When the node is in the air and no contact with anything else. If the node is a support, zero friction is not possible.

If you can post a layout that might be useful with nodes and loads.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[BigInch]

I doubt that your stress guy told you the whole truth. It would be unusual in my experience if an above ground pipe failed a nozzle because somebody forgot a friction factor. Friction factors being what they are, friction forces do not exceed the weight of pipe x 0.3 or less, and thermal forces are often 100's of times greater than friction and cause the pipe to move if friction is present or not. It is far more likely that the piping system does not have enough flexibility and axial thermal stresses within the pipe are causing end loads at the nozzle and subsequent overloads.

I hate Windowz 8!!!!

 

[saplanti]

plasticnoob,

A professional pipe stress engineer is responsible to consider all the cases to show the behavior of the pipe and the maximum effect on the nozzle with and without friction effect, and inform you about the consequence. During the start-up and shut down of the piping the pipe expand/contract. Depending on the pipe route and behavior of the supports, the support friction loads may need to be taken by the nozzle during these events and sometime during the operation.

In the delicate equipments this may cause problems if the exposed load exceeds the allowable which are given by manufacturer of the equipment. If your case is real you'd better talk to the equipment manufacturer for the available maximum loads and their cases,they might accept the loads or you may be forced to change the pipe route to make the loads under allowable.

 

[saplanti]

In case the supports are hangers, struts, sway braces and they do allow movement in restrained and/or unrestrained directions there is no need to apply the friction coefficient for these supports.

 

[StressGuy]

I have to agree with BigInch. Your contractor sounds very inexperienced.

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.

 

[saplanti]

I would not come a conclusion without seeing existing piping arrangement and the location and route of the tie-in. The support types and the route of the existing high temperature piping and additional force and displacements due to the tie-in piping are all effective in the discussion.

Sometime selecting the right tie-in location may eliminate the additional forces and moments on the existing piping. I would select the tie-in close to the guide, if available, which does not allow lateral movement. This may increase the support loads, and the loads on the nozzle of the tie-in.

However I would expect different type supports in the high temperature (?) piping due to large thermal movements in all directions, so selecting the right tie-in location and making the tie-in piping more flexible might be better solution for both tie-in and existing piping.
I guess plasticnoob will come back with more detailed information (pipe route, support types , pipe temperature etc...)to make this discussion more reliable.

 

[plasticnoob]

Hi all,

Thank you for your feedbacks.
I have try to attach a file but it seems like i need to upload my file to some file storage website before i can attached here.

Or is there a better way?

Btw the pipe design temperature and the design pressure is 380 degrees celsius and 51 barg respectively

 

[plasticnoob]

I have also managed to get a response from the previous designer and the following is his response:

"The treatment of friction in stress analysis is not a well-defined practice. Inclusion of friction is not an automatic input requirement. Friction is a transient effect, when applied close to equipment simulates primarily the startup case. In calculations where friction may give misleading results, the analyst will run a friction free analysis and also friction included analysis. Using his engineering judgment the analyst will run a variety of cases to simulate as closely as possible to real situation. This approach has been applied in this instance. Friction can restrict the vertical displacement when applied local to spring hangers, thus under sizing of hanger may occur.

The analysis has been executed in conjunction with the furnace vendor. The terminal conditions were transmitted to the vendor to apply as boundary condition, for his code compliance analysis. The vendor’s analysis was acceptable with these loads applied. The stresses local to terminal are nominal. Hence the terminal loads were acceptable."

As what saplanti has say, the previous designer did consider both cases before he applied his engineering judgement.
Now the question is more of if his engineering judgement was correct or not.
I will try to attached the pipe stress and friction factor from the previous designer once i figure out how to attach it.

 

[saplanti]

I was really surprised by the previous designer's response. I have added my thoughts in parenthesis.

"The treatment of friction in stress analysis is not a well-defined practice. Inclusion of friction is not an automatic input requirement. Friction is a transient effect, when applied close to equipment simulates primarily the startup case. (WHO GARANTIES THE PIPE TEMPERATURE IS TO BE EXACT 380 DEG C ALL THE TIME DURING THE OPERATION? DOES PROCESS HAVE THIS MUCH RELIABLE INSTRUMENTS NOT TO CAUSE THERMAL MOVEMENT DURING THE OPERATION?) In calculations where friction may give misleading (THERE IS NO SUCH A TERM. YOU CAN EXPECT OR UNEXPECT THE RESULT IN ACCORDANCE WITH THE PIPE ROUTE AND SUPPORT ORIENTATION AND TYPES) results, the analyst will run a friction free analysis and also friction included analysis (SOMETIMES FRICTION FREE ANALYSIS CAN INTRODUCE MORE NOZZLE LOADS BY DEPENDING ON WHERE THE OTHER RESTRAINTS ARE). Using his engineering judgment (SHOULD BE BASED ON THE PIPE ROUTE AND SUPPORT TYPES, AND OBVIOUSLY THE DIRECTIONS OF THE THERMAL EXPANSION/CONTRACTION)the analyst will run a variety of cases to simulate as closely as possible to real situation. This approach has been applied in this instance. Friction can restrict the vertical displacement (IS IT RELATED? WE DO NOT HAVE ANY DETAILS) when applied (I GUESS MISTAKENLY, IT IS NOT USUAL CASE) local to spring hangers , thus under sizing of hanger may occur.

The analysis has been executed in conjunction with the furnace vendor (THE TERMINAL POINT AND ITS BOUNDARY CONDITION -SUPPORT TYPES B/W PIPING AND FURNACE, AND MOVEMENT OF THE TERMINAL POINT, IF AVAILABLE- ARE ADDITIONAL FACTORS TO BE TAKEN INTO CONSIDERATION BOTH FURNACE DESIGNER AND PIPING STRESS ANALYST) The terminal conditions were transmitted to the vendor to apply as boundary condition, for his code compliance analysis. The vendor’s analysis was acceptable with these loads applied. The stresses local to terminal are nominal. Hence the terminal loads were acceptable."

I would expect a pipe stress analysis should have at least a brief report that contents the assumptions, design input, markups of supports on the isometric drawing(s) and support loadings. The treatment of tie-in should also be added for understanding the behavior and loading.
Sometimes, some regulations rule a third party verification to overcome this kind of problems. In this case you owned this verification job I guess. if you do not have background on the pipe stress analysis, I suggest you to give the job to a professional third party, and see what else they will ask. Providing the analysis result here is not going be sufficient for the court in case unexpected fatal accident happens.

 

[plasticnoob]

Sorry for the late reply.
After seeing your post, i decided to escalate the issue to my manager.
Anyway, i will attached the existing isometric for discussion and learning sake.
I have also attached the report for the friction factor used for your perusal.

I love to learn from you guys and appreciate your comments.
Thank you

 

[saplanti]

If you have problems with the allowable loads of furnace nozzle you cannot ignore the displacement of the nozzle while you allocate the supports. There is a two directional guide at node 240 on the last isometric drawing which we can identify it too close to the furnace since the nozzle is moving 3.5 mm in Z direction and the pipe is too stiff. I cannot say too much about this since I do not know the exact reason of this support other than assuming them the wind guide or stopper of the thermal expansion of the following piping to the vessel. However this support has big impact on the nozzle loads. The support location should be a neutral or close to neutral in terms of expansion relative to the furnace nozzle. This is the reason Mx seem high. If the isometrics are for the existing piping only, playing with supports on the existing piping can rectify all the problems. I would expect the original piping stress analysis should consider the planned piping attachment. Therefore they should not need additional pipe stress analysis for the future attachment. If this is an unplanned attachment for the original analysis you may have rights to modify the supports on the original piping to meet the requirement by consulting the owner.

The Caesar II input does not give us anything without Caesar II sketch and the connectivity with the isometrics provided.

I can see that you need a consultant who can reach all the information you have and ask more question on the furnace and piping with new attachment. You cannot solve this type problems by discussion on the forums unfortunately.

 

[DSB123]

Hi Plasticnoob,
Can you ask the contractor where he manages to obtain supports with such low friction values. I would love to be able to get "Actual" friction values as low as those noted without having to go for PTFE/SS units. Also guides and line-stops do not just have friction due to the vertical loading. Whereever the support/pipe contacts there is friction. I know a lot of pipe stress engineers use friction factors of 0.3 but if the supports are not well maintained/greased then in external environment friction factors of double those used actually occur. It's not the sliding friction value that matters it is the static friction factor before movement occurs which gives the largest reactions.

 

[racookpe1978]

Hard to justify using PTFE or assuming "well-greased" roller bearings right up next to a furnace at 380 degrees for long periods of (unattended and unmaintained) times ....

That "explanation" of assumptions from the contractor sounds like a lawyer re-wrote an old legalesed form from a generic contract. Has many words, but little engineering definitions specific to the furnace at hand.

 

[saplanti]

Today the engineering concept is too far different than what we used to implement. Most of the companies (probably owners or their representatives) split the job between the parties without a proper background or with lack of proper guidance on the parties (sometimes the guide is there, but that is all, none is to follow if the guide was implemented). So this type problems occur, and sometime ends up with vital accidents. (I do not say this is the case here, but it needs to be investigated adequately)

Engineers are most of the case under improper guidance for specialized works. Some cases they do not even have right to ask questions about the job. If you follow this forum requests you will see how we work in engineering. You know what I mean.

I cannot blame/support the stress analyst directly without hearing from him/her in this forum. The main problem is too far complex since engineers do not drive the company finance.