Design Temperature Definition 2

[athomix]

I would like to understand the meaning of design temperature for a pipe sizing:

- is it the maximum operating temperature?
or
- is it the over temperature, defined by the code, which the piping will support, calculated from a lower known maximum operating temperature? (i.e. it's the safety margin oover the maximum operating temperature defined by a design code).

If it is the 2nd condition, that temperature can happen that never will be achieved during expected operating conditions and can occur under extraordinary unexpected operating condition.

The problem is that I have a contract definig design temperature for condensate return of 185ºC and the maximum condensate return temperature in the contract operating scenarios is 140ºC.

Thks in advance.

 

[JLSeagull]

I don't set the design temperature. I think that a common practice is to set the design temperature about 50 degree F above the highest expected operating temperature. Based upon your last sentence this looks about right.

The metalurgy, design temperature and design pressure establish the flange class for valves and piping components. One common mistake is to establish the PSV setting at the maximum pressure and temperature for carbon steel. Then when instrumentation selects a stainless steel instrument flange we have a non specification higher class flange for the instrument.

 

[insult2injury]

Typically, the maximum operating temperature refers to the maximum CONTINUOUS operating temperature when everything is operating as planned, not for relief scenarios, etc. Depending on what code the piping is designed, there is typically an allowance for abnormal temperature and pressure excursions as long as it doesn't exceed the yield point and doesn't occur for longer than the interval in code. If abnormal operation may exceed these operating scenarios, then they must be considered in the design conditions. Often times an arbitrary 'CYA' margin is applied to the maximum continuous operating temperature to cover minor excursions which are more frequent (exceed the code-maximum frequency/interval). Other people apply the margin because they don't know enough about the process. And still others apply the margin because they saw someone else do it and don't know any different. In most cases, it probably doesn't change the piping, valves, or flanges so it's a mute point. In the end, it's better to set your design temperature too high and pay a little more than have it end up too low.

I2I

 

[BigInch]

Design temperature for pipe and fittings is most often set by the piping material and its allowable pressure calculated form its allowable yield stress as stated in the piping design code covering the installation. The design codes, such as the ASME B31.x series have certain materials that are permitted for use and list allowable yield stresses at various temperatures. So, normally you would know the maximum pressure you need to contain the fluid in the pipe, and you would select a material, then iterate to set a pipe wall thickness,

looking up one of its yield stresses at one of several of the given temperatures,

calculate the wall thickness, and then you know you can operate that pipe up to that temperature.

If that's not a high enough temperature, you move to the yield stress allowable for the temperature you need and

recalculate the wall thickness for the pressure required.

The given temperature for whichever yield stress you wind up using is the design temperature for the pipe.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

 

[insult2injury]

The values included in the tables (such as B31.1) contain allowable stresses not yield stresses. The allowable stresses are typically 67% or 85% of the yield stress (depending on the level of testing, etc.).

Biginch: Are you saying that it's common practice to use the maximum allowable temperature for the piping as the design temperature? I have not often seen that done and imagine that it would result in a higher total cost than necessary or practical in many cases.

I2I

 

[BigInch]

Actually the ASME pipeline codes use minimum allowable YIELD stresses as the allowed stress varies with area classification in which the pipelines are installed, the non-pipeline codes use allowed stress.

No, I was thinking about this from a pipe design computer program perspective for some reason, so starting at the highest allowable stress (lowest temperature), would give you the thinnest wall possible and keep you closer to the economic curve, as overall cost of material, transportation of same and construction usually runs closer to steel weight than it does to yield or allowable stress. Once that is set, you take a look at the temperature allowable to see if that works for ya. If not, go to the next lower allowable stress (a higher temperature).

No reason you couldn't work it backwards, that just appeared to be the best systematic attack algorithm.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

 

[SJones]

Why is it a problem? If you are a designer - you use 185 deg C to design your piping as specified in the contract. If you are an operator, you use 140 deg C to set your process controls. Whether that is efficient or not is another matter. Design temperature is defined in 301.3 of ASME B31.3. The tendency is for piping to be 'selected' rather than designed and hence some wild differences between 'design' temperature and maximum operating temperature dependent upon pressures required. It also gets confusing from an integrity point of view when assessing degradation mechanisms from a design temperature perspective to find that the defined design temperature isn't actually the design temperature; it's just the maximum temperature for the pressure rating.

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[BigInch]

Yes max is max and operating is operating. Might be totally different values.

Selecting the pipe procedure is more than efficiency. You pretty much have to do it that way, even if you fially wind up making a mill order for pipe with the exact wall thickness required.

If you design exactly according to the procedure suggested by the 31.3 code, yes, effectively you do select the pipe and subsequently find the "design temperature". 301.3 immediately refers you to back to 301.2 Design Pressure.

301.3
"The design temperature of each component in a
piping system is the temperature at which, under the
coincident pressure, the greatest thickness or highest
component rating is required in accordance with para.
301.2. (To satisfy the requirements of para. 301.2,different components in the same piping system may
have different design temperatures.)
In establishing design temperatures, consider at least
the fluid temperatures, ambient temperatures, solar radiation, heating or cooling medium temperatures, and the
applicable provisions of paras. 301.3.2, 301.3.3, and
301.3.4."

Taking the "301.2 Design Pressure" road,

"(b) The most severe condition is that which results
in the greatest required component thickness and the
highest component rating."

That route leads you to determining the wall thickness of each component and then selecting the design temperature as the lowest from the possibly several "different design temperatures", just as they say.

As far as integrity evaluations, etc. those should always be based on actual operating temperature and pressure, so the design temperature would be just as irrevelant as would the design pressure. On a pipeline for example, you might have one place where you reach the design pressure and the design temperature might occur at a completely different location 1000 + miles away, so an evaluation based on design temperature-pressure would never be realistic.

I think what's important is not setting some arbitrary temperature (or pressure) today based on today's process conditions when you might want to run at a different pressure and temperature X years from now, or tomorrow. If you arrived at the true max P & T limits from the most conservative of all components and rated the system as such, you could turn up the dial without getting all the engineers together to agree to refile for higher operating parameters and possibly be delayed for a long time while you get regulatory and safety approvals first.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

 

[SJones]

For integrity: not quite true when considering non-age related degradation mechanisms that may be dependent upon partial pressures of components within the pipe. Age related mechanisms - yes use operating conditions; non age related - use design conditions. Therefore, it's still a case of setting sensible margins between process control and design parameters. Apart from that, on the management of change issue, it's down to the process engineers to get it right in terms of pressures and temperatures over the design life. Like I said, piping tends to be 'selected', pipelines tend to get 'designed' - that's why you will see a lot of piping with design temperatures quoted as -30 to +120 deg C for example, even if the process engineering design has it at 0 to 80 deg C, because the rated temperatures for the pressure class are quoted.

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[gerhardl]

Thank you to SJones and BigInch for enlightment on the practical issues of design and operating temperature.

In my opinion, the trap to avoid is to select too low pressure class on economical reasons. The classical example is arguing going down one pressure class on flanges or valves, because : 'the next higher (correct) class is exceptionally costly, and only a slight bit of the allowed temperature/pressure curve exceeds what is allowed for the other class.' Or: 'If we calculate the pressure peaks with safety margins half the usual, we will come just under the limits allowed for the lower pressure class'.

As a supplier we have on repeated occasions refused to supply components to a lower pressure/temperature class than allowed.

 

[BillBirch]

I'm with gerhardl on this one. Years ago, operators were not aware of the clause in B31.3 that quantifies the allowable over-pressure excursions in piping systems, above those pressure cases anticipated. Now they want you to invoke that clause to justify known operating conditions above the piping class limits, even when they cannot guarantee the duration of the events.

Thus, they are requesting a design pressure less than an anticipated maximum case. My interpretation of the Code is that the design pressure should be selected on the basis of the anticipated maximum operating pressure, and the over-pressure clause left for unanticipated maximum pressures.
Regards,
Bill

Sorry to stray off track a bit, but Gerhard reminded me of the argument I had a few years ago about this.