Thermal Relief For Exchanger In Long Loop

[barnesed]

I have a heat exchanger in a water loop fed from a storage tank to a pump where it is circulated through the tube side of an exchanger and back to the pump suction (drops from the loop to users as-required). The exchanger shell side has steam to heat the water. Although the water valves at the exchanger are locked open, there are valves at the pump suction and pump discharge that are not locked open, and if closed, could block the cold side of the exchanger. Since this is a long loop, is a thermal relief valve in the loop required?

 

[georgeverghese]

Which design code is this exchanger built to - ASME, TEMA, DIN, JS or other? Look up the section on overpressure protection of the applicable code. What type of HX is this, by the way - shell and tube or gasketed plate frame or welded plate HX ?

 

[barnesed]

The exchanger is an ASME shell & tube. If the water inlet & outlet valves at the exchanger were not locked open, there is no issue. A thermal relief valve would be added between the valve and exchanger. The question here is that there is a long loop with valves not located at the exchanger that when closed, will block the cold side of the exchanger. (BTW, no other credible overpressure scenarios for the tube side.)

 

[georgeverghese]

At the outset, a TRV on the cold side would seem to be required, no matter where the isolation block valves on the tube side are - this is wrt the valves without lock open feature.
But I would suggest there is no need for a TRV here on the tubeside if the following conditions are met or are applicable:
a)A small hole is drilled on the pump discharge check valve.
AND
b)If you've got regular packed glands on the centrifugal pump inboard and outboard shaft seals, then these would act as a relief path since no packed gland is completely leak tight.
AND
c)All block valves between the HX tubeside and the pump shaft seals are locked open.

 

[mk3223]

IMO, a thermal PSV may be added at the "long" loop line near the tank outlet to protect the pipe. The discharge of PSV can be routed back to tank. The safe guide is simple and easy to implement.

 

[georgeverghese]

When you say there are no other credible overpressure scenarios, presume you mean the tubeside design pressure (and that for the pump casing and the entire loop piping) is equal to or exceeds that on the shellside?

This will then exclude tube rupture overpressure.

If there was no tube rupture ( and tubeside DP = > shellside DP) , the max boiling pressure of the tubeside contents cannot exceed the max normal supply pressure of the steam on the shellside, which would only be less than tubeside DP.

If the pump is stopped at the time of the blocked in scenario, could the tubeside operating pressure drop to atmospheric or some low pressure. At some low pressure, would the tubeside contents then boil up to