2/3 rule for heat exchanger to avoid tube rupture case

[thestephan]

Dear all,

we designed the low pressure side of reboilers with 2/3 the design pressure of the high pressure side. Doing that, we did not have to consider tube rupture (acc. to API).

Does this mean that tube rupture can not happen or does it mean it can happen, but it is no design case for the PSV?

In case it can happen, the question is, whether we have to apply the same design pressure for the system reboiler/tower (eg. reboiler 30/20 bar,g and tower 20, even though we would normally apply 6 bar,g for the tower)?

Who can help me out of that?

 

[TD2K]

A full tube rupture is pretty rare (leaks are another matter). As a result, some companies take this into effect when considering a tube rupture.

What I've seen discussed (and this is a matter for interpretation) is whether the low pressure side (associated piping and equipment in many cases as well as the exchanger itself) can be pressurized in excess of its hydrotest pressure (typically 1.5x design but with recent Code changes, it can be 1.3x design). Some companies use the normal operating pressure on the high pressure side to do the evaluation, other use the design pressure of the high pressure side. As long as the low pressure side hydrotest pressure is not exceeded, then it is not a design case for pressure relief and does not have to be considered (though in the analyses we note that it was looked at and why it is not a possible design case for the PSV).

As far as how far you have to look, that depends on the system. If a tube rupture will potentially overpressure downstream equipment and piping, you need to look at how far that overpressure can exceed, taking credit for what your practices allow you to do. It's not a cut and dry answer. If a tube rupture would potentially overpressure a downstream column, designed for 6 barg to 20 barg, you need to look at that. Do not just look at the exchanger and stop at the flanges.

 

[Sandeep1970]

The answer given by TD2K is correct. In your scenario if there is the probability of overpressuring the column do exist then you can look in to the option of switching the fluid sides i.e. fro shell side to tube side. Sometimes this can solve your problem easily. in the worst scenario this may be done at the expense of LMTD. Again if you provide more details then your problem can be addressed accordingly. Thanx

 

[WaltF]

Do you have access to API Recommended Practice 521, Fourth Edition, March 1997? Section 3.18.2 states...."...loss of containment of the low-pressure side to atmosphers, is unlikely to result from a tube rupture where the low pressure side (including upstream and downstream system) is designed for at least two-thirds of the design pressure of the high pressure side." This implies that you must consider the effect of tube rupture on associated equipment.

This section of API 521 has been modified in each edition which issues and will likely to be modified in the fifth edition.