AMSE 31.3 Over Pressure from Water Hammer

[AliBy]

I require some perspective on interpereting pragraph 302.2.4 which states
"Allowances for pressure and Temperature Variations. Occasional variations of pressure and/or temperature may occur in a piping system. Such variations shall be considered in selecting design pressure (para. 301.2) and design temperature (para. 301.3). The most severe coincident pressure and temperature shall determine the design conditions unless all of the following criteria are met."

We use Flow Master to assess the worst case scenario, usually an uncomtrolled pump trip (similar to slamming a valve shut on full flow) which results in water hammer.

Sub Paragraph (d) requires that "The total number of pressure-temperature variations above the design conditions shall not exceed 1000 during the life of the piping system."

For a life of a pipeline of 20yrs this implies that only 50 "over-pressure" events are permitted per year.

I interperate this to be addressing fatigue. In light of this it would imply that any pressure peak above the Maximum Allowable Working Pressure (MAWP) is considered an over pressure event.

In the mining environment we are operating at ambient temperatures and in most cases pressures not higher than 200 bar.

As soon as the percentage overpressure reaches about 5% the number of peaks predicted by the software exceeds about 20 as the damping is not high. This would imply that you cannot have more than 2 trips per year?

Is that a correct deduction? What other standards may be better to use in this type of application and design assessment?

Looking forward to some response.
Alistair

 

[bvi]

The limit of 1,000 cycles is to exclude fatigue due to pressure cycling from consideration, so your conclusion is correct, although not commonly considered in this fashion.

An alternative to consider is doing a more rigorous analysis, as permitted in paragraph 300c3. That more rigorous analysis would explicity consider fatigue due to pressure cycling. As there are no rules in ASME B31.3 for consideration of this, you need to go based on appropriate engineering practice. For example, you can look to Section VIII, Div 2, Appendix 5 for guidance. You would use the more rigorous analysis to justify exceeding 1,000 cycles of the variation.

 

[unclesyd]

Adding to cb4's reply I would like to add one additional consideration. One of the biggest problems I seen on industry concerning pressure events is not the pipe itself but the hangers and supports.

 

[jay165]

CB4, repectfully:

Could he not follow the rules of Chapter IX, High Pressure Piping and thus still conform to B31.3?

 

[bvi]

Chapter IX does not permit variations. Also, it carries with it many additional requirements that you would not want to apply to a typical piping installation. Further, Chapter IX does not actually include detailed fatigue assessment rules, but instead refers to Section VIII, Div 3.

If the question is to use selected portions of Chapter IX, that does not comply with B31.3.

Note that the method I suggested above does comply with ASME B31.3, as permitted in para. 300c3,

 

[LSThill]

Team Member & cb4 (Mechanical)

K304.8.6 Extended Fatigue Life. The design fatigue life of piping components may be extended beyond that determined by the Section VIII, Division 2, Appendix 5 fatigue curves by the use of one of the methods listed below, provided that the component is qualified in accordance with para.

LST

 

[AliBy]

In terms of more rigorous analysis, we use both FlowMaster to determine more acurate Transient Response during a pump trip, and Caesar to establish pipe stresses (and anchor/support stress - as appropriately mentioned by <b>unclesyd</b>) in the system.

What design criteria could one apply in this case? Section VIII still requires conformance to 302!

Aside:
I am not trying to get around the code, but rather trying to find a code or paragraph that addresses this scenario. I "was" under the impression that a high temperature - high pressure combination for extended durations being repeated is far worse than water hammer peaks!? The code seems to address the former and not necessarily the latter. There again I have visited a gold mine where a Ø400 steel pipe (32mm wall) had been ripped open due to water hammer. Perhaps water hammer it is as bad as the code makes it out to be???