You must be able to relieve the expanded volume. Spacing is therefore a function of expanded volume, which is a function of blocked-in volume and temperature change. Normal practice as I know it is to calculate the volume that can be blocked-in, such as volume actually between adjacent block valves, calculate the expanded volume at the high temperature and the time to raise to that temperature, then size the valve for [Δ]V/[Δ]t.
Trying to size multiple relief valves in any one blocked-in segment spaced a certain distance apart has the potential to complicate the problem considerably, as it becomes a problem with possible multiple flow paths being set up within the same pipe, even if you don't consider elevation differences between relief points, or a nonuniformtemperature distribution along the pipe segment.
What would you be doing, if you knew that you could not fail?