5 replies to this topic

[Nirav]

Hello,

What I'm going to ask might be observed by many of you, but I do not understand the reason of it that's why I'm putting it across.

In many instances of cooling water exchanger, it is possible that process side (mostly shell side) design pressure is significantly larger than the cooling water system.

In most of such cases, based on API & ASME, we increase the tube side design pressure to meet 10/13 rule and avoid any pressure relief valve on tube side.

For example,
If Shell side (process) design pressure = 50 barg
Tube side (Cooling water) design pressure = 10/13 x 50 = 38.5 barg.
Cooling water system (piping) design pressure = 16 barg.

In general, design pressure of equipment will always be extended up to at least first isolation valves at inlet and outlet of equipment to maintain integrity of system depending on location of safety valve.

However, in above case, the design pressure of 38.5 barg is NOT extended beyond the heat exchanger flanges on cooling water inlet & outlet side. Meaning the isolation valves on cooling water side are not designed at 38.5 barg, but at 16 barg. Similar is true of piping between these isolation valves and heat exchanger.
(There will be a small thermal relief valve on piping. But that can't be considered for any scenario other than thermal relief.)

So, I do not understand that --->
Why the piping between inlet/outlet isolation valves and heat exchanger is not designed at 38.5 barg?
I came across such situation at many times but NOT ALWAYS. Sometimes, piping is designed at tube side design pressure, and sometimes, it is not. I do not understand it.

What could be the reason for such different philosophies adopted by different owners of process plants?
Can anyone have any idea?

Warm regards,

[pleckner]

There is no philospophy in the works here. It is just plain wrong or an oversight by the process designer or a lack of understanding.

[gvdlans]

I agree with Phil Leckner, but may be able to give some historical background.

Up to the 1997 revision (4th edition), the background of the 2/3 rule was not clear. There were basically two interpretations:
1) When 2/3 rule is followed likelihood of a tube rupture is so small that it can be neglected.
2) When 2/3 rule is followed, there is a possibility of a tube rupture but this is unlikely to result in loss of containment from the low pressure side to atmosphere.

When you follow the 1st interpretation, only the exchanger itself should be designed according to 2/3 rule. When you follow the 2nd interpretation, not only the exchanger but also the upstream and downstream systems should be designed according to the 2/3 rule.

In the 90's, there has been an offical question asked to API on what interpretation should be followed. API then confirmed that the 2nd interpretation was correct and in the 4th edition this is now also indicated in section 3.18.2 of API RP 521.

[gvdlans]

After some digging I found the question and official API Interpretation based on the 1990 issue of API RP 521:

"General Edition: Third Edition, November 1990
Question: Our interest is in the basis for the ‘two-thirds rule’ given in API 521, and the way in which this should be applied. The code states that,
‘For relatively low pressure equipment, complete tube failure is not a viable contingency when the design pressure of the low pressure side is equal to greater than two-thirds the design pressure of the high pressure side.’
This ruling is often used to avoid fitting a relief valve on the low pressure side by raising the design pressure to two-thirds of the high pressure side. Does ‘low pressure side’ refer to just the exchanger itself, or to the low pressure system as a whole? There would seem to be two explanations for the code’s guidance. The first is that by increasing the design pressure of the low pressure side, the exchanger construction becomes so robust that tube failure is no longer credible. The second is that if the low pressure side is designed for two-thirds it will have a test pressure (1.5 times design) equal to the design pressure of the high pressure side, and therefore is extremely unlikely to be damaged by any pressure rise caused by internal leakage. The first explanation would require only the low pressure side of the exchanger to be designed for two-thirds, whereas the second would require the whole of the low pressure side to be designed for this.

Reply: Increasing the low pressure side design pressure will have little impact on the likelihood of a tube rupture. However, by raising the low pressure side design pressure to a point where the low pressure side test pressure is equivalent to the high pressure side design pressure will ensure that damage to the exchanger due to a tube rupture is extremely unlikely. In most situations, this may involve raising the low pressure side design pressure to two-thirds that of the high pressure side. To effectively assess the consequences of a tube leak or rupture, the entire low pressure system into which the high pressure side can flow should be evaluated. The latest edition of RP 521 (the Fourth Edition, March 1997) has an expanded and clarified write-up on the impact of tube ruptures. You might find this information helpful to your understanding of the issues involved. "

[Nirav]

Mr.Guido,

Thank you very much for providing the above communication.
It helped me a lot to clarify myself about the situation I was getting confused with. I would like to quote following from API official reply.

To effectively assess the consequences of a tube leak or rupture, the entire low pressure system into which the high pressure side can flow should be evaluated. The latest edition of RP 521 (the Fourth Edition, March 1997) has an expanded and clarified write-up on the impact of tube ruptures. You might find this information helpful to your understanding of the issues involved.

Based on that, I referred API RP 521 (the Fourth Edition, March 1997). I would like to refer 3 sentences from it.

Clause 3.18.2, Pg 23
Para 1 : "The use of maximum possible system pressure instead of design pressure may be considered as the design pressure of the high-pressure side on a case-by-case basis where there is a substantial difference in the design and operating pressures for the high-pressure side of the exchanger."

Para 2 : "Pressure relief for tube rupture is not required where the low pressure exchanger side (including upstream and downstream systems) is designed at or above this two-third criteria"

Para 3 : "For new installations, increasing the design pressure of the low-pressure side may reduce risk. Upstream and downstream piping and equipment systems must be thoroughly evaluated when this containment approach is taken"

Interpretation of Para-2 clearly says that upstream and downstream system must be designed as same as heat exchanger.
However, based on para 1 & 3, there's a scope of study of system based on actual system evaluation on "case-by-case" basis depending on actual maximum operating pressure vis-a-vis design pressure.

Based on above, I think in certain cases (having significant difference in high pressure side "operating" & "design" pressure), based on study results, the exchanger low pressure side is designed for 2/3 or 10/13 criteria, but it was NOT extended to piping up to first isolations at inlet and outlet. This is particularly followed for cooling water system due to following.
[1] During normal plant operation, it is very much unlikely to get it pressurized due to the fact that it is "open to atmosphere" at one end (cooling tower) of it.
[2] If isolation valves are closed, it is again very much unlikely that process side is operating.

It is always better to follow API 521 and consider piping system also for higher design pressure to avoid any confrontations. However, API 521 is "recommended practice" - RP and not the "standard". Therefore, with proper analysis and characteristic of system like above, deviations could be acceptable from 'recommended practice'. Isn't it acceptable?

I again thank you for providing those "API official communications" as reference.
I would be able to think in more appropriate direction while coming across such situations next time, since safety can not be ignored in any case.

Warm regards,

[gvdlans]

Nirav,

I guess the wordings in API RP 521 still leave much to be desired. One thing about standards and recommended practices is that they should not be blindly followed. They contain a lot of useful knowledge and information but in the end it is you as a chemical engineer who has to make sure that the installation is safe. This can be translated as "the risk involved with the installation is tolerable and as low as reasonably practicable (ALARP)".

With regards to the heat exchanger design, API clearly adopted a risk based approach. Risk is a combination of likelihood of an unwanted occurance and its consequences. The logic that API is following is that the likelihood of a tube rupture is considered very small, and that if the low pressure side is hydrotested at a pressure that is at least equal to the high pressure side design pressure, the likelihood of loss of containment to atmosphere can be considered extremely small. Therefore, risk of loss of containment to atmosphere is considered acceptable, even though the consequences of loss of containment to atmosphere can be (very) high. Examples of the consequences can be formation of a flammable or toxic cloud, fires and explosions, resulting in injuries or even fatalities.

If you can demonstrate that this risk is still acceptable when using the high pressure maximum operating pressure instead of design pressure, or by not raising the design pressure of the pipe to the cooling tower, you actually demonstrate that the design is safe. This clearly can only be done on a case by case basis, so no general statements can be made in a recommended practice such as API RP 521.