PD:
Thanks for the additional basic data and operational information. Now we can start dealing with specifics. However, there are some points that need clarification or confirmation from you:
You don’t say it, but I have to assume you are normally maintaining a N2 vapor blanket on top of the hot HMD stored in the storage tank. Am I correct?
14.5 psig = 401.35862 inch H2O. Therefore, the pressure relief set point on the 6” conservation vent on your storage tank can be set as high as 400 inches of WC instead of the indicated 20 inches of WC. Am I correct? If not, why is the storage tank an ASME Section VIII vessel instead of an API 650 design? This seems to be in conflict and may need clarification.
What you describe as possible scenarios involves an unsteady state heat and mass transfer operation – a process engineer’s nightmare of calculations. However, as I previously stated, what safety design guidelines dictate is that we define the worse-case scenario and design to relieve it – both on the pressure side as well as any possible partial vacuum situation. If we can accept this simplified criteria, then:
Case #1:
The worst-case scenario for in-breathing is that the tank is 100% full of only 170 oC (N2 + HMD) vapor –with no reasonable amount of hot liquid HMD present - and the maximum flowrate of cold feed is sprayed into the vapor space causing the HMD vapor to cool down to it’s dew point and causing a drop in the overall vapor space pressure due to the loss of the HMD’s partial pressure. However, you have to establish that the N2 supply would fail to operate and maintain the required N2. Additionally, there is a positive effect contributed by the maximum cold HMD flowrate into the tank: it increases the vapor space pressure by vapor displacement due to liquid addition.
In my opinion the possible failure of the N2 blanket feed supply is a credible possibility and this does not involve a double-jeopardy situation. Therefore you have a credible situation where you might pull a partial vacuum on the tank beyond the rated 1.5 inches of WC, and you must check to make sure you have enough inbreathing capacity through the vacuum relief. The solution of finding the HMD vapor condensation rate then is to establish the cold feed’s capacity to cool and condense the vapor and subtract the cold feed’s liquid flowrate. In order to condense the HMD vapor, the cold HMD liquid must sensibly cool the vapor to the dewpoint and then condense the vapor by removing the latent heat of vaporization. To obtain a complete solution, you strictly must apply simultaneous heat and mass transfer, but I would only employ the unsteady state heat transfer effect, assuming that the mass transfer rate is negligible. If not, then I would apply a conservative safety factor in the final vacuum flowrate requirement.
Case #2:
By working backwards, I assume you mean to say that you expect to be flashing the hot (180 oC) into the 80 oF storage tank. I would not take any credit for condensing any of the presumed flashing vapors. The amount of vapor generated by the flashing is determined by an adiabatic flash calculation of the hot HMD liquid (under presumed column pressure). However, one question arises here: isn’t this taking place 100% of the time, whether the storage tank is heated or not? If so, then the existing 6” relief valve should already be rated for the worse-case pressure scenario – which is the flashing taking place when the tank is operating hot, not cold. Clarification of this point may be needed to clearly understand the existing situation and potential for over-pressurization.
I believe I’ve interpreted your basic data and information correctly; if I haven’t please indicate so. I hope my comments are of some help.
Art Montemayor
Spring, TX