Temperature de-rating

Easing back into writing after a long period of zero inspiration has been helped by some advice from Phil Venton in response to a query he received about the temperature de-rating rules in AS 2885.1:

I was going through the pipeline temperature de-rating section in AS2885.1.  Can you please explain the data/reasoning behind this equation and why there is a different maximum temperature which requires de-rating for the pipelines and flanges to AS2885 and flanges to ASME B16.5?

Very reasonable question!  The relevant part of AS 2885.1 is Clause 3.4.3.  I’ve touched on this before but in a vague sort of way.  Phil’s response is much clearer and went like this (slightly edited):

The matter relates to the basis by which pipes and flanges are designed, and the change in properties of materials in response to changes in temperature.

The yield strength of steel declines more or less continuously with temperature, whereas the tensile strength does not start to degrade until perhaps 200°C.

Pipe is designed using yield strength.  In principle the yield strength should be reduced for any temperature higher than the reference temperature, however given the complexity that this introduces to typical pipelines, most of which have a design temperature less than 60°C, it seemed reasonable that the nameplate yield strength should be used for design at temperatures <65°C, while for temperatures ≥65°C the yield strength should be reduced at the nominated rate, assuming a starting temperature of 23°C.  (The note at the end of Clause 3.4.3 points out that 65°C is a convenient temperature for this purpose since it covers the discharge temperature from most gas pipeline compressor stations.)

Flanges are designed to a pressure equipment standard (ASME VIII or AS 1210).  These standards require the use of the lesser of the tensile strength/3.5 (/2.35 in some instances) or the yield strength/1.5.  It happens that the factored tensile strength governs the flange design to about 120°C.  Above this temperature (approximately) the factored yield strength is less than the factored tensile strength.

Consequently the basis for each temperature de-rating requirement is entirely rational provided you appreciate the principles, and recognise that the authors of the Standard have made rationalisations to provide practical rules for design.

(Thanks to Phil Venton for this clear and useful explanation of rules that at face value can appear arbitrary.)

This entry was posted in Pipeline design, Standards. Bookmark the permalink.

4 Responses to Temperature de-rating

  1. Ross Cochrane says:

    Clause 3.4.3 Strength de-rating
    With reference to carbon steel valves and flanges this clause creates much discussion with the expectation during the preliminary stages of a project that meeting all the requirements will be without issue but unfortunately when the detail work is required to be carried out this is not the case. It is not unusual to have to change to full compliance to the listed flange and valve codes, especially for station work. I feel that in reality it does not value add and is just an increase in the risk. If full compliance was understood as a requirement at the start of a project there would be no issue. I think it is now relevant to delete this clause with reference to flanges and valves and just comply with the referenced standards. I do not believe that this issue will ever go away without changing to conform to the referenced standards. This is particularly important in the stations. Pipeline and station piping in general use ASME B16.5/B16.47 flanges and ASME B16.34 valves. All the piping connection flanges on vessels must meet the B16.5/B16.47 ratings. These standards do not allow the de-rating. In general we only install flanges in systems at valves and vessels (equipment) so at the interface it is always an issue.
    This clause goes against the basic way engineers carry out work. A project is carried out by complying with numerous nominated standards. As noted in AS2885.1 Clause 3.2.2 numerous standards are listed. Engineers apply the standard as they cannot be expected to design all these components from scratch and the standards provide an acceptable level of risk. By trying to apply this de-rating clause the risk is obviously increased and I am not convinced that this appropriate. I understand that if a flange is designed from first principles the strength is basically unchanged to 120 degrees C but in the projects we work on a flange is never designed from first principles and if carried out the flange design would be larger than the standard B16.5/B16.47 flanges.
    Please find my comment on the notes below:
    Note 1. “Reference ASME B31.3, ASME VIII, and MSS SP44 – At temperatures up to 120°C flange
    designs are based on (a constant) ultimate tensile strength resulting in no strength derating
    Comment ASME B31.3 & ASME VIII do not allow ASME B16.5 flanges to be operated continuously above the ASME B16.5 ratings. I do not understand how these codes are referred to in note 1. It implies that these codes accept the de-rating of ASME B16.5 flanges which is not the case. Agreed MSS SP 44 flanges are not derated but this standard specifically states that the valves will not meet the flange rating. It is also noted that AS1210 has deleted MSS SP44 from its standard components which comply with the code.
    Note 2. “The temperature limit for flanged valves applies only to the flanges. Assurance should be
    sought from the valve manufacturer that the valve body and seals are suitable for the required
    service conditions.”
    Comment: This is the clause which creates the largest discussion and is always an area of confusion. By reading this note it would be expected that the valve manufacturer would be able to provide assurance that the body and seals are suitable for the service as a matter of course. This unfortunately is never the case and there a no widely circulated document from any valve manufacturer stating this. I have concluded providing assurance is not possible for the valve manufacturer due to QA requirements. All the common valves that are utilised on projects meet the ASME B16.34 requirements, which again refers to ASME B16.5 for valve ratings. For QA reasons the valve manufacturer cannot state that his valve complies with B16.34 and give the assurance required. The only way I see a valve manufacturer comply is meeting B16.34 requirements by either offering a stronger material (different B16.5 rating class) or a higher flange rating or if the valve is special class (welded end valve only). All options meet B16.34 but obviously have a cost impact. i.e. The valve manufacturer just comply with ASME 16.5.
    Note 4. “This permission does not currently apply to vessels designed in accordance with AS 1210 (e.g. filter vessels). In these cases, a design check in accordance with AS 1210 or ASME Boiler and Pressure Vessel Code (BPVC), Section VIII, should be considered.”
    Comment : This note implies that permission will be forthcoming. I just do not see this ever happening . ASME B16.5/B16.47 flanges do not comply with AS1210/ASME VIII if calculated from first principles so any design check is a waste of time.

    • petertuft says:

      The working group reviewing AS 2885.1 will be looking at this and taking your comments on board. It’s hard for me to comment on the specifics because my own expertise in this area is rather shallow. However I am aware of a few projects that have been able to take advantage of the de-rating dispensation, apparently without great difficulty. The alternatives to the dispensation are either to specify flanges of the next pressure class (Class 1500 for a Class 900 pipeline – enormous) or de-rate the whole pipeline to the pressure rating of the flange at the station design temperature (as was done for the Amadeus Basin to Darwin Pipeline, mentioned in my original post on this subject, https://pipelinesoz.wordpress.com/2010/10/06/un-de-rating-flanges/). Neither is at all palatable.

  2. Chris Hughes says:

    One of the factors which can cause problems with flange ratings is designers ignoring the requirement which first appeared in AS2885.1 in 2007:
    “A number of design temperatures and their associated design pressures shall be determined”. The SEAGas pipeline, which was designed in 2002-3 before this requirement first appeared, followed the usual practice of stating “Max. design pressure 15MPa, max design temperature 55degC” without any comment as to where these temperatures and pressures might co-exist – at the compressor station discharge point where the max. temp and max. pressure may coexist the AS2885 derating allowed the use of Class 900 flanges but because the same coexistance was applied to the outlet meter station design it ended up with AS1210 filter vessels fitted with Class 1500 flanges – totally unnecessary as the gas temperature entering the station would never exceed maximum soil temperature which in Adelaide is unlikely to be much above 30degC. It would have been cheaper to erect a shade sail over the filters to minimise solar heating than to use the Class 1500 flanges!

    I am surprised that so few people seem to consider the use of Class 1.2 materials for the flanges to get round the temperature derating problem – I would expect a Class 1.2 #900 flange to be cheaper than a Class 1.1 #1500 one.

  3. Anonymous says:

    I have been involved with several pipelines with this non-derating applied and have had no problem with valve suppliers complying with the higher temperatures. Of course, the designer should properly specify appropriate and applicable seals rather than selection from a catalog.

    We also have used an AS2885 over-rider (thus permitting higher temperatures on flanged joints) over B31.3/AS4041 pipework citing AS2885.1 clause 6.3.1 “Design of station pipework shall comply with AS 4041 or ASME B31.3, except as provided in Clauses 3.2 and 3.4.3. The use of any other Standard shall be approved. AS 2885.1 remains the governing Standard and the design shall comply with its other requirements.”, 3.4.3 being the operative clause.

    I understand (perhaps erroneously) that the temperature limits paper has gone across to the AS4041 committee for potential inclusion in a future revision of AS4041.

    In the bad old days prior to this clause, using MSS SP-44 flanges, I did notice a flange where “ASME B16.5” had been ground off, and “MSS SP-44” had been stamped into the flange, presumably as a result of examination of mill certs, and re-badging of the flange.

    To my knowledge, there has never been any problem created by this approach. Perhaps someone could suggest an incident where this has been a problem? I do not believe it is sufficient to simply state “risk is obviously increased” without some basis.

    Fundamentally, the weakest portion of a flanged joint is the gasket, followed by the bolting, and then the body of the flange. Might I suggest that the relatively low (38C) temperature where B16.5 commences de-rating be a function of pre-spiralwound gaskets????

    I strongly would counsel against the deletion of this clause.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s