Hi all,

Firstly I would define true installation temperature to be the temperature of the piping when the final closure weld / joint is made for a section of piping between anchors or control points. As the piping fabrication and erection occurs at different points of time when the ambient temperatures are different we cannot precisely tell the true installation temperature for the piping of interest. Keeping this situation in mind we have to decide a single installation temperature for stress analysis calculations by considering the range of ambient temperatures from minimum ambient to maximum ambient temperature to which the piping is subjected during installation, that would ensure the maximum possible stresses and reaction forces are checked.

ASME B31.3 319.3.1(b) makes it mandatory to compute the values for reactions as algebraic difference between the value at maximum(or min) temperature for the thermal ccle under analysis and the value at the temperature expected during installation. However I have the correct maximum ambient and minimum ambient for the site location, and cannot decide the true installation temperature, as I do not know when the piping is going to be fabricated and erected.

Can anyone tell me how best the "installation temperature" as given in 319.3.1(b) of ASME B31.3 be approximated.

You could have two installation temperatures- the minimum ambient for the hot lines and the maximum ambient for the cold lines.

Hi S,

Ok, that seems to be too conservative. This approach has a problem particularly when checking the allowable loads on equipment nozzles.

You mean to say that the phrase "Temperature expected during installation" in 319.3.1 is to be read as "minimum(or maximum) temperature expected during installation", as a conservative approach. My doubt is whether the conservative approach suggested by you really represents the true intent of the code.

I would like to think so.

Does anybody have a different approach in arriving at the value of installation temperature for the purpose of stress calculations, given the maximum ambient temperature and the minimum ambient temperature for the site location ?

Thanks in advance.

Hi,

I've had this problem a lot of times too, and I have it everytime a new project starts.

Sometimes I´ve used the same conservative method as S. But as u say, is too conservative, and then u have problems with pump nozzles, very big reacctions, ...
In addition, also the spring designed are going to have more range of displacement, ... displacement that never is going to happen if the installation has been made in summer (even the spring can be out of range for take the minimum temp!).

I'm sorry but I haven't got a definitive answer for this. In my opinion (what I do) you should check the temperature range of the location, and see the differences between the average temperature, the minimum and the maximum. If you get a map of average temperature for each month better. Also, try to recover information about when (in which month) the installation is going to be made (I know this is not easy).

Also is important to know which kind of installation are we designing (is not the same a nuclear installation that a water pump system), and also which is the highest temperature of your piping system.

With all this I do a mix:

If the difference between the average temperature and the minimum is about 10-15 ºC and the highest temperature is more than 300ºC, then the error if you take the average temperature is low, and I think that is covered by the CODE. But if the highest temp of your system is 50ºC then the error is bigger and I would calculate it for the highest temp difference.

If the installation is a nuclear one no doubt! Always the worst temperature difference!!

To calculate it always by the conservative method sometimes for some locations is too conservative. Imagine a location with minus 40ºC temp as the lowest!. In this case try to guess when the installation is going to be made, and in 99% of the cases, then installation won't be made in pure winter with -40ºC outside!

I would like to know another opinions about this topic...

Another situation can be shutting down of the plant and restarting it in winter season by any chance. Here the temperature range is maximum. This probability also may be considered as a criteria with the above method.

Hi Walito,

I appreciate your opinion on the subject.

Dear Shiny Mathew,

Let me bring to your notice that once the installation is made at a temperature, the shutting down and restarting in winter/summer have no impact on the maximum reaction forces generated. The requirement of the code is to check the reaction forces generated when temperature rises from installation temperature to maximum operating temperature or the temperature falls from the installation temperature to the minimum operating temperature whichever is highest. Once we have checked these forces and designed the supports to withstand these,with the true installation temperature, all the cases of startup or shutdown during any season of the year are taken care of. The extreme conditins of ambient temperature are taken care of, but the reference point for measurement(installation temperature) of these extreme conditions of temperature (for the purpose of computation of reaction forces)is the topic of discussion.

First I am in agreement with Walito, which was a good post and i'd like to add my own thoughts/ramblings to his.

In my opinion the trouble starts with not knowing the completion month and for min/max amb. temp. [Ta.] differences 20 to 30°C i.e. a non-temperate climate.

@max temp of 300°C this is around 10% and therefore significant. (adjust accordingly to max)

I have several thoughts on the matter owing to my general ignorance. The general extreme of a min/max amb. condition would tend to be the min. as extreme cold conditions are less practicable to work in than extreme hot conditions. (apologies to those who work in said discriminatory conditions)
Therefore a site will work during the warm months up to their autumn and usually completion. There is always an urgency to complete the construction before winter months, I don't see the inverse in warmer climes, where the work can still be completed owing to workable parts of the day/night and orientates itself more around finance, supply & demand et al. rather than a schedule.

So where does that take us? If you cannot assess the completion month then use the min Ta? (conservative and more probable?)But what if the min. Ta. isn't realistic for completion? Do we take a limit on that? i.e. work can be done at -10°C but not -20°C? This becomes opinion engineering and ultimately the client must provide the details/scope for the answer.

Please correct/add your opinion to my own, this is a good thread.

I could not properly express what I meant in the previous post. Intension is to find out the appropriate value for installation temperature for calculation. I was adding one more criteria for selecting a suitable reference value while accepting that method. Few lines I missed is added here with the previous post.

“Another situation can be shutting down of the plant and restarting it in winter season by any chance.
((Where the plant was installed in ‘summer’ and the mean temperature as calculated above has been considered for installation temperature. Say the operating temp is higher than the mean value. As far as the plant is running the difference and effective loads is stable as it was started from summer temp and reached the operating temperature in reality. But consider a situation where this running plant is shut down in a cool winter season where the atmospheric / room temperature is below the mean temperature. Now the piping will cool down to that winter temperature which is lower than the mean installation temperature we used for Analysis. When it restarts the range is much higher than the first start up. So while selecting the mean reference temperature for analysis purpose the chances of frequent shutting downs in a low temperature also may be considered. Because in this scenario the range from minimum temperature to operating temperature is higher and the calculated load based on mean temperature can be much different from reality. So a waitage should be given for this chance. If the plant stops and restarts at same near by temperatures as that of the initial start up no any difference will be experienced. Obviously the extreme range is safest ))
Here the temperature range is maximum. This probability also may be considered as a criteria with the above method.”

Establish an installation temperature using a reasonable basis, document what you did and go with it.

You can always find a number of scenarios where it will not be correct.

A few years ago, I worked on a project in a northern European country with no guidance or specification from the client on an installation temp. I pulled climatic data from the internet for that region and used the yearly average daytime temperature. I assume most construction would be done during the day and it had a long construction schedule.

If your calculations cover from the installed (ambient temperature) up to the max. temp. and down to the min. temp. and the max. to min. range, your covered as well as you can be, what more can you do?

I'm OK with the discussion here but why point to 319.3.1(b)?
As I read it, this paragraph sets the coefficient of expansion for the old style T1 (thermal alone) evaluation rather than the comprehensive "L1-L2" approach recommended by CAESAR II. It's a minor point here.

Hi Dave,

Yes whatever you said is OK, ASME B 31.3 code is recommending the old method for reaction force evaluation. [color:#FF0000][color:#6666CC]I referenced 319.3.1(b) because this is the only paragraph in the entire code which talks about the installation temperature. We are using this temperature only for reaction force evaluation. [/color][/color]For stress check it dosent matter which installation temperature we choose, as our load cases will be based on the entire range of minimum metal temperature to maximum metal temperature (max delta T) using the L1-L2 method given by CAESAR-II. We inlude the minumum ambient temperature and maximum ambient temperature if they contribute to maximum delta T.
The problem definition is:

"WE NEED TO CHOOSE AN INSTALLATION TEMPERATURE, BASED ON COMPETENT ENGINEERING JUDGEMENT, THAT WOULD OPTIMIZE THE DESIGN ENSURING SAFETY"

I feel choosing the maximum ambient temperature as installation temperature for cold lines and minimum ambient temperature as installation temperature for hot lines is too conservative and will result in higher costs due to high reaction forces on equipment.

Check your local Occupational Health and Safety Regulatios which may specify a minimum outdoor temperature allowed for work outdoors. This can become you minumum install temperature. Also talk to the construction contractor that will be buliding it and ask them what temperature they will work down to.

Also remember to consider if the pipe is indoors or outdoors and what stage building construction will be at when your piping is installed. Buildings even on construction will usually be heated to above freezing.