The blowdown condensors are part of the relief path for drum PSVs (if drum PSVs are routed to BD system). When the condensors foul (and can you think of any coker heat exchanger more prone to fouling?), they partially restrict the relief path. Very much like partially closing a block valve on the PSV discharge. In the case of block valves on PSV discharge, we always LOCK THEM OPEN because we need to be sure there is NO RESTRICTION. But in the case of BD condensors, we just let them foul until some arbitrary measure (usually when drum cool time is impact) is used to call for cleaning. I submit that the BD condensors need to be looked at as a block valve on the PSV discharge. Unlike a block valve where you can see if the stem is partially closed, the degree of fouling (restriction) in the condensors is hard to quantify. But understanding the degree of fouling in the BD condensors is essential to know your true PSV relief capacity. Obviously with severly fouled BD condensors, the PSV capacity will be reduced, and coker feed should be reduced to stay within the capacity of the PSV.
For new cokers, it would seem wise to design for at least 20% overcapacity so that when one drum is at maximum cooling (i.e. maximum load on the BD condensors), and an adjacent drum has vapor line blocked (PSV controlling case), the BD condensors will still provide adequate relief path EVEN WITH 20% FOULING. But you still need good monitoring to know when you’re at 20% fouling so you can schedule cleaning. Alternative relief paths include main fractionator (everybody groan) or a separate quench drum (my preference for best process safety).