How can blocked cooling water outlet cause relief? 4

[BruceVan]

In relief valve analysis, one contingency is blocked outlets. This scenario consideres the outlet valve closed with process supply pressure on the inlet. If incoming pressure exceeds MAWP (or a lower pressure if other considerations require it), then relief protection must be provided. What I don't understand is the API 521 statement under section 3.5 Closed Outlets.

API states, "In the case of heat exchangers, a closed outlet can cause thermal expansion or possibly vapor generation." My understanding of heat exchanger operation, particularly on cooling water side, is that if only the outlet is closed, the inlet remains open and any thermal expansion will disperse through the open inlet. This would not produce a relief case.

When both the cold side inlet and outlet valves are blocked, a thermal expansion case is considered.

Can someone explain the reasoning behind the API 521 statement that closing the outlet valve with the inlet open can cause thermal expansion?

 

[Montemayor]

BruceVan:

In the case of a cooling water supply (CWS), particularly, you will find (if you trace the CWS back to to its source) that the CWS pumps have an inherent check valve on the discharge side of the CWS pumps. If you consider the CWS as also inherently incompressible (almost everybody does on a practical basis), then what you have under the API RP 521 example scenario is a situation where an incompressible fluid is trapped between a block valve upstream and a check valve downstream. You have, for practical purposes, a closed hydraulic chamber that is undergoing heating in the heat exchanger - with the customary liquid expansion coefficient contributing to the liquid's expansion and subsequent fast hydraulic pressure increase.

I hope you see the over-pressure concern from a safety viewpoint.

 

[svanels]

A closed outlet can generate steam if the temperature difference of the hot and cold liquids is big enough. Imagine fuel oil at 200 Celsius entering and heating up cooling water with 30 C. The effect will be shockwaves and steamhammer in the cooling lines creating dangerous conditions.

 

[EGT01]

BruceVan,

As you noted, closing only one valve on the cooling water side of a heat exchanger is not likely to cause a thermal liquid expansion relief case. In fact, in some cases it is usually considered sufficient to lock or seal open one valve as an administrative procedure to avoid adding a thermal relief valve. However, Art Montemayor gives one example where that may not suffice.

But, I believe you have missed the point of the section on closed outlets. Take another look at the first sentences in that section...
"To protect a vessel or system from overpressure when all outlets on the vessel or system are blocked, the capacity of the relief device must be at least as great as the capacity of the sources of pressure. If all outlets are not blocked, the capacity of the unblocked outlets may properly be considered."

Now you actually omitted a part of the sentence that you quoted from API. In its complete form...
"In the case of heat exchangers, a closed outlet can cause thermal expansion (see 3.14) or possibly vapor generation."
and you should refer to section 3.14 for further information on the subject.

By the way, maybe I've overlooked it, but I don't seem to find the API 521 statement "that closing the outlet valve with the inlet open can cause thermal expansion". There's a lot of information in API RP-521 and some of it is not always clear but try not to read more than what's there.