Considered temperature for thermal stress analysis 1

[Fran67]

Hello everybody,
I work on a piping stress analysis and I would like your point of view regarding the couple pressure/temperature to use for a piping stress analysis, especially when sudying effects of thermal growth.
I used to consider the expected OPERATING conditions when I study the effects of thermal growth on pipe supports and displacement range, I mean I don't use the DESIGN conditions which is in my case only used to calculate hoop stress and select pipe thickness.
Do you think it is the good way?
DESIGN and OPERATING conditions are data given by the end user, I don't know if the DESIGN value is given by a failure case, or the Operating with margin, or the max value given by the rating, it is only a contractual data coming from end user.
Do you think it would be better to make the analysis using DESIGN value? Is it not too much conservative?
Thank you
F67

 

[1503-44]

For pipe Stress analysis, operating conditions are not very important, as the pipe and support system do not have an optimum performance point. IMO you must ensure no failures occur between minimum design temperature and maximum design temperature.

Pipe stress does not have an optimum design point, at least in the sense that pumps, boilers and other process equipment, or pipe hydraulics do. Pumps must be sized for optimum performance at operating conditions, such that the pump must have its highest efficiency at the operating flow rate, even though it still must be designed to reach max pressure and temperature at design conditions which would normally be somewhat higher. Hydraulic operating flowrate would determine the most efficient diameter to select for the pipeline, then at that diameter the pipe should still be able to flow at the design flow rate, perhaps again somewhat higher than operating flow rate, but not necessarily at highest efficiency.

If you can envision some case where the Stress and support design would be significantly different at operating temperature then at design temperature, you might have a case there that would warrant special study, but I would say that is probably going to be quite unusual.

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[0707]

IMO all your calculations shall be based according design conditions, becouse during operation you can have, up-sets which in principle should fall within design conditions, allowing you to work inside a safe margin.

 

[Snickster]

By design conditions are you referring to the actual design pressure and temperature of the piping as listed on the Piping Line List? Or are you referring to the maximum allowable pressure and temperature of the Piping Material Specification sheets?

The stress analysis should always prove that the stated design pressure and temperature of the piping system as indicated on the Line List can be met regardless of what you think the actual operating conditions will be. The actual design pressure and temperature as shown on the Line List should be derived from the maximum expected operating pressure and temperature with additional margin added in accordance with most extreme expected operating upset conditions and with good engineering practice.

The piping material may actually have a higher MAWP since a standard piping material specification sheet could be based on 150#, 300#, 600# etc. flange rating, but the stress analysis would only have to be performed based on the actual stated design pressure/temperature indicated on the Piping Line List. In a lot of cases you would not want to design for the full flange rating pressure and temperature of the Piping Material Specification sheets if operating at a very low pressure since you may need to have some low pressure rated compoonents and equipment in the line. So if you stated the full pressure temperature flange rating then that would exclude using lower rated components.

There are other non-operating cases that should be looked at such as steam out cleaning of lines for instance. In this case the temerature of the line could be much higher although the pressure could be very low if done atmospherically while line is not in operation. Still you would have to check the large thermal movements and that they do not overstress the piping or put loads on the supports that they cannot handle and cause damage.

Also there are other scenarios that should be looked at. For instance say you have a set of three pumps acting in parallel with one of the pumps a spare in a very hot piping system. I would look at separate case that analyzes the stress with either one of the pumps acting as the spare with the other two operating. In each of the three cases different stresses will result depending on which pump piping is cold versus which one is hot.

Try to think of any other scenarios that would cause the stress and expansion in the piping to be significantly different.

 

[LittleInch]

Not s simple as it might look.

Assuming this is B 31.3, the code in section 319.3.1 says

"Values for Stress Range. Values of thermal displacements to be used in determining total displacement
strains for computing the stress range shall be determined
from Appendix C as the algebraic difference between the
value at maximum metal temperature and that at the
minimum metal temperature for the thermal cycle
under analysis"

Now frankly this could mean anything and it is up to you to come up with different scenarios.

So you tend to get OPerating stress ad well as design stress and also for expansion purposes sometimes empty stress is worst, especially if there is less weight and more sun.

Pipelines specifically say operating temperatures when they do the stress analysis.

If it passes the design temp check then anything else will be lower, so it's always a good first pass in my opinion.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Snickster]

Yes you would be designing to B31.3 likely so you would actually need to look at the required B31.3 load cases of sustained, displacement range, and occasional (sustained with wind and earthquake). Not including wind and earthquake (OCC) the following load cases would be analyzed:

Basis load cases:
L1 WW+HP (HYD)
L2 W+T1+P1 (OPE)
L3 W+T2+P1 (OPE)
L4 W+T3+P1 (OPE)
L5 W+P1 (SUS)
L6 L2-L5 (EXP)
L7 L3-L5 (EXP)
L8 L4-L5 (EXP)
L9 L2-L3 (EXP) (Total stress range)

L = Individual load case
W = Weight of pipe with service fluid
WW = Weight of pipe with water during hydrotest
P1 = Design pressure
HP = Hydrostatic test pressure
T1 = Maximum design temperature
T2 = Minimum design temperature
T3 = Other temperatures of concern during operation or service (such as steam out)

Note that the operating (OPE) load cases are used to determine loads on supports and other piping general design requirements as these are the worst case loadings and what is actually present.

The B31.3 cases are sustained (SUS) and expansion (EXP) which must pass Code allowables where L2-L3 is the actual maximum displacement stress range for design maximum to design minimum temperatures which usually give the maximum displacement stresses but max/min/other temperatures should be looked at independently.

Desgin pressure would be normally based on the relief valve set pressure.

Maximum design temperature would be normally based on maximum fluid temperature in operation.

Minimum design temperature would normally be based on lowest ambient temperature.

Note that design pressure, and max/min design temperature would be noted on the official project Piping Line List for record.

Now if you have a speicial temperature case such as steam out while line is depressurized I am not sure how this would be handled per B31.3, e'g., would this then be considered the maximum design temperature? I never analyzed a system with non-operating temperature greater than in-service temperatures. Refer to the Code for short term abnormal temperature and pressure excursions for allowables.

 

[Fran67]

Hello,
Thank you for your feedback,
I understand that non-operating cases (short duration, failures..etc) are generally not considered to determine the min and max DESIGN temperatures, therefore the thermal stress analysis is based on DESIGN temperature which are values equal or close to OPERATING values.
In my experience some industrials (oil & gas) do not consider the chapter 302.2.4 of B31.3 and consider occasionnal variations as DESIGN conditions, it generates very large range of DESIGN temperatures, the thermal stress analysis give very high stress values and finally makes no sens (too much conservativ approach). Anyone has already met similar problem?
F

 

[Snickster]

If it is an occasional variation of temperature and pressure then IMO it should be considered in setting the design pressure and temperature as it may only have to happen one time to cause damage or failure of piping, and if you are in the oil and gas, refining, chemical plants etc., industry dealing with high pressure fluids and gases any failure could be catastropic. Now if you are dealing with low pressure non-toxic or harmless liquids and water you might not want to design for absolute worst case that may only occur once or twice in a lifetime and you would just accept the piping or pipe support damage and fix it rather than spend all the money up front to mitigate damage from it.

I don't have latest B31.3 but I believe you are referring to occasional loads such as wind and earthquake where allowables are increase by 33%.

Never do anyting that will put saftey of the public or your engineering liscense in jeopardy no matter what the Codes say or anyone tells you is OK or they have done it like that before.

 

[LittleInch]

What we see sometimes for extreme events is that you need to look at them and then see what levels of strain or movement you would actually get in a particular element and then see if this is acceptable.

A small amount of yielding can relieve a lot of force, but if you end up snapping a branch off then it's a different story.

Often you find for piping the empty case in summer / black body temps gives you the highest amount of movement and you don't want your pipe to fall off the rack before they've even been commissioned....

"makes no sense" and "too much conservatism" are judgements which can be made, but need data to show the "what if" scenarios.

Even if a system only sees a condition once in its operating life, you need to do your best to ensure that that foreseeable condition doesn't result in failure.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[1503-44]

[ignore][/ignore]If you are still confused, reread the code. Word by word.
"Occasional variations" are an operating concern, not a design condition.
We may be able to operate something at 10% above design pressure, if the time that occurs is less than 24h in a year, but that does not change the design conditions.
If the system would have to be operated for 25h in a year at 10% over design pressure, then the design pressure would need to be rated to a higher pressure, such that the higher pressure would not be exceeded for more than 24h. That provision should have been known to the engineer setting the design pressure. After that design pressure is assumed to have been set in accordance with the code's provisions and one should design to that pressure without further consideration of excursions from it, unless we think it needs some further discussion with the engineer determining those design conditions.

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[RVAmeche]

IMO its a bit like relief valve piping needing to be designed for the nameplate capacity of the valve. This way, even if your worst case relieving scenario is less than the relief capacity, the piping can handle the device capacity and prevents people from changing process parameters in the future and incorrectly thinking they're protected.

In a similar vein for pipe stress, I prefer to use the design temperature of the system. This is the absolute worst case and if operations wants to go above the system design temp, its obvious major considerations are needed. The issue is lots of companies have generic specs that apply to numerous systems without different design conditions. Just an "A106B Class 150 spec" that has a single design temp/pressure. Here it's a bit of a judgement call and it's not unreasonable to use the maximum expected operating temperature in my opinion. So if this generic A106B spec was good up to 600F, but this is XYZ process that cannot get above 450F, it may be reasonable to use 450F. Again, this leaves the possibility that someone down the line will replace equipment, try to get the system to 600F since "that's the spec" and may run into issues.

 

[1503-44]

Pipe line spec does not specify design conditions, but rather the other way around. This is especially true for small diameter, low pressure systems, where "mechanical strength" considerations
May easily override a wall thickness selected for pressure. Many company standards will have a line spec of SCH 180 for 2" diameter, but use SCH 40 for 6", or even require a 900 ANSI RTJ Flange, for a 1000 psig system, just because its offshore. In other words, there is no direct relationship between line spec and actual design pressure or temperature. Its simply that the design conditions must be within the capability of what is stated on the line spec applied.

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."