Blocked in overpressure protection requirement

[Nova N]

Dear Members,

I have a heat exchanger to heat Natural Gas (tube side) for combustion in 2 drum dryers. LP steam at 350kPa is on shell side. The NG inlet temp is 12 F and outlet is 68 F. I need to know if there is a need for a PSV (thermal case) on the tube side if the NG is blocked in. Most of the literature and API 520/521 refer to need for PSV (thermal case) for "liquids" that are blocked since liquids are incompressible.

 

[georgeverghese]

So what do you calculate to be the max possible blocked in pressure on the tubeside, and does this exceed tubeside design pressure. For the calc, use initial conditions for Pi to be max normal op press and 68degF.

Pf = Pi.(460+Ts)/(460+68), where Ts = sat temp for LP steam in degF.

Though the codes speak of overpressure due to liquid expansion, the same would apply to gas also. In any case, there are no exclusion clauses stated in the API for equipment that is gas filled.

 

[Nova N]

Yes, the calculated Pf is above the design pressure of the tube side. Design pressure is 770 psi and Pf is 1000 psi. So this would require a PSV for the thermal case (correct)?
So I have 2 questions:
1. Is the above equation available in API 520/521?
2. How do I size the PSV? Do I calculate the Thermal rate Q, and select appropriate PSV.

 

[Latexman]

A fire scenario for the tube side may be credible too.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[georgeverghese]

If you have access to a process simulator ( Hysis or Simsci - Pro II), you could easily model this blocked in heating scenario for the HX tubeside in the batch depressuring utility. These simulators will also show you the pressure rise profile with time before the RV actuates. The required relief orifice size is obtained by trial and error to maintain tubeside pressure at 770psig. Specify the external heat input to the tubeside to be the design case HX duty appropriate to the 770psig operating region.

The formula provided earlier can be derived from first principles - API521 provides general guidance and philosophies, and some formulas where derivation from first principles is not possible or is too involved.

Agreed, pls read up your company design guidelines on how to evaluate tubeside firecase scenario for the type of HX you have.

 

[Nova N]

Thanks for details. Much Appreciated.