Heat Exchanger's tube rupture two-thirds rule and location 3

[koba78]

Dear members:

For exchanger where the low pressure side design pressure is
at least two-third(or 10/13 from our senior engineer
) of the high pressure side design pressure,
the tube rupture case can be ignored if the upstream and downstream
equipment in the pressure system is capable of handling the high side pressure.
And, relief device may not be installed on the low pressure side of the heat exchanger.

For example
low pressure side (L.P) : design pressure(D.P.)=180psig
high pressure side(H.P.): design pressure(D.P.)=300psig

D.P of L.P/D.P. of H.P=180/300=0.6 < 2/3 and tube rupture is
a valid case.

My question is 10/13 is a little bit larger than 2/3, and is
it a valid factor ?
And sometimes it is hard to decide the the PSV located in the upstream
or downstream.If located on the downstream side form the source of the relief,
the PSV set pressure may be reduced by the pressure loss through the exchanger
at relief conditions. So is it preferable to located the PSV on the upstream side?



Thank you very much.

 

[don1980]

koba,

ASME changed the hydrotest pressure for Sec VIII pressure vessel in the late 90's. The earlier version mandated a test pressure at 150% of MAWP. The later (current) version has different allowable stress values, and the required test pressure changed to 130% of MAWP.

The rule exempting tube rupture as a credible scenario is based on the hydro-test pressure of the low-pressure side (LPS) of the exchanger. That rule says that the tube rupture scenario can be neglected when the LPS test pressure is greater than or equal to the MAWP of the high pressure side. That makes sense. The risk of over-stressing the LPS due to tube failure is pretty low if the LPS has already been stress tested at a value GTE the high pressure side MAWP.

So, for exchangers designed and fabricated according to the old version, you use a 2/3 rule (100/150). For exchangers designed and fabricated according to the current code, you use a 10/13 rule (100/130). To determine which rule applies, pull the U-1 report for the exchanger, or check the nameplate, to see which hydro-test pressure was used. The apply the applicable rule.

 

[jamesbanda]

Don and Koba,

I've not heard this rule before, can you advise of the source. Would this be a company manual, API design code / pratices.. Engineering best pratices.. our internal designs .

In the quote.does it apply if The low pressure side can handle the normal pressure but not the high side relief pressure or both..?

This is merely my curisty to understand the detail and reference..

e.g. regardless of pressures if the low pressure side (operating) that is.. can handle the high pressure side relief pressure the you would not need a relief valve right?

 

[Duwe6]

Rule is from ASME Sect. VIII Div 1
PRV is still required for Fire Case, or if there are valves that can isolate the low-pressure side.

 

[moltenmetal]

don1980 has it reasonably well, except I don't think the intention is to avoid installing a relief valve on the shell side. Rather, it is to ensure that such a relief device is able to be effective.

Please correct me if I'm wrong, as I'm going from dusty memory of a colleague's point of view on the issue and I could well be mis-remembering: but as I understand it the issue is the potential for guillotine tube rupture which may or may not be proceeded by a detectable leak (i.e. giving time for depressurization and shutdown prior to failure). In the case of guillotine tube rupture, a shell-side relief device of any practical, reasonable size may not be able to keep up with the flow from the two broken ends of the tube- indeed, the shock wave generated by the initial rupture may be enough to rupture the shell if it is rated greatly lower than 10/13 (formerly 2/3) of the tubeside rating- even with a relief device fitted.

By selecting the shell for the hydrotest pressure, which is basically putting the limiting material in the shell to the onset of yield (ie. because SAS is often 2/3 of yield), you may still have a damaged exchanger after a guilllotine tube rupture, but you are very unlikely to rupture the shell and hence lose containment of the tubeside material.

 

[don1980]

James...this "rule" is found in API 521, along with lots of other practical relief design guidance. In the latest edition (5th ed) of API 521, refer to 5.19.2.

I regard API 521 as an in indispensable reference document for relief designers. Whereas ASME Sec VIII states the rules, API documents provide engineers with plain-language guidance on how to comply with those rules.

Can this 2/3 or 10/13 rule be used to neglect the risks of complete tube rupture (guillotine break of tube) and slow leakage from a hole in the tube? YES, this rule applies to all modes of tube failure. Read the section in API 521 covering this rule. It does not distinguish between slow leak and sudden tube failure.

Does this automatically mean that no relief device is needed? No, we're talking about tube rupture. The question of when a relief device can be omitted is a whole other question.

So, are there any cases in which a relief device can be completely omitted from the coolant side of an exchanger? Yes, if liquid thermal expansion is the only remaining scenario. Refer to API 521 5.14.1.

What about fire exposure? Aren't we required to size the PSV for fire exposure? No. The decision of whether or not to consider fire as a cause of overpressure is up to the user. The only exception is low pressure storage tanks containing flammable or combustible liquids with a flash point < 200F, and the tank is located in the USA. That's an OSHA requirement (OSHA 1920.106). In all other cases, the decision of when to size for fire exposure is based on the user's assessment of the risk of a flammable liquids pooling under or very close to the vessel. Again, refer to API 521 for further details and guidance related to fire exposure.

 

[don1980]

Whoops....should have proofed before posting.

My OSHA reference should say 1910.106 rather than 1920.106.

 

[pleckner]

Unfortunately, it is a little more complicated than what has been discussed above. The rule given by API 521, 5th ed. January 2007 can be found in section 5.19.2. It does not say one can neglect tube rupture as a relief sizing case when the LPS test pressure is greater than or equal to the MAWP of the high pressure side. It says, "Loss of containment of the low-pressure side to atmosphere is unlikely to result from a tube rupture where the pressure in the low-pressure side (including upstream and downstream systems) during the tube rupture does not exceed the corrected hydrotest pressure (see 3.21 and 4.3.2)." We have to interpret this as meaning the corrected hydrotest pressure of the low pressure side. The corrected hydrotest pressure is defined in section 3.21 and an example given in 4.3.2.

API also gives the engineer some additional wiggle room to boot so you need to read the section.

So it no longer has anything to do with 2/3 rules, 10/13 rules nor is it compared to the MAWP of the high pressure side. However, the closer the design pressure of the low pressure side is to that of the high pressure side, the more likelyhood you will fall within this criterium and be able to eliminate tube rupture as a relief sizing case.

===============================================================

(3.21) Corrected hydrotest pressure: hydrostatic test pressure multiplied by the ratio of stress value at design temperature to the stress value at test temperature.

(4.3.2) gives an example. Don't be confused by the fact that section 4.3.2. is titled, "Closed outlets on vessels."

 

[don1980]

Thanks pleckner. I should have referred to corrected hydrotest pressure which incorporated the temperature component of stress.

It's common practice to base this decision on the MAWP, and that's OK as long as the process side operating temperature is below the MAWT of the LPS. In my experience that's usually the case, but it's not always so. If it is, then one can use the 2/3's or 10/13's. Those ratios can be increased, based on this temperature correction, if the hot side temp is lower than the the LPS MAWT. But, if the hot side temp is greather than the LPS MAWT, then one needs to calculate the corrected lower ratio before applying this rule.