I presume the 37kA is the fault level upstream of the circuit breaker. Depending on the breaker type it may or may not provide some degree of energy limiting - by cutting the prospective fault current off quickly enough, a fast acting fuse or breaker can reduce the effect of the fault current downstream of it, although a normal PSSC meter which uses the voltage drop on  a small test current will not see this effect and will over estimate the potential for damage.

Without 'let through' or PSSC curves for the breaker this will be hard to verify.

If the limiting is not enough, or cannot be verified then the simplest fix will probably be a fuse with a suitable rupture capacity in the live side and in close proximity to the SPD.

As above, SPDs can fail dead short, especially if over-zapped during a thunderstorm or similar, and it is important that a blown SPD is disconnected gracefully, and does not generate a more serious problem than the original.

This is for the wrong sort of fuse, but it shows an example of the sort of limiting action clearly enough, where the let through can be quite a bit 

less than the fault current upstream at higher fault currents.


Note that these lines  are more or less straight on log-log axes, but more nearly  proportional to 1/ (I squared t) to the right and proportional to I to the left.

I presume the 37kA is the fault level upstream of the circuit breaker. Depending on the breaker type it may or may not provide some degree of energy limiting - by cutting the prospective fault current off quickly enough, a fast acting fuse or breaker can reduce the effect of the fault current downstream of it, although a normal PSSC meter which uses the voltage drop on  a small test current will not see this effect and will over estimate the potential for damage.

Without 'let through' or PSSC curves for the breaker this will be hard to verify.

If the limiting is not enough, or cannot be verified then the simplest fix will probably be a fuse with a suitable rupture capacity in the live side and in close proximity to the SPD.

As above, SPDs can fail dead short, especially if over-zapped during a thunderstorm or similar, and it is important that a blown SPD is disconnected gracefully, and does not generate a more serious problem than the original.

This is for the wrong sort of fuse, but it shows an example of the sort of limiting action clearly enough, where the let through can be quite a bit 

less than the fault current upstream at higher fault currents.


Note that these lines  are more or less straight on log-log axes, but more nearly  proportional to 1/ (I squared t) to the right and proportional to I to the left.