I would have thought a speed restriction for a Pacific would have solved it (less hammering) but maybe that's too complicated for those in charge.
That's easy to say when you're not the one having to sign off a loading as safe. In my professional life (construction not railways) I regularly have to assess structures and approve them for various loadings, and nothing concentrates the mind like having to put your signature on the approval, knowing that potentially lives are at stake if you've got it wrong.
Unless they have been superseded by a Eurocode (which I haven't used as my involvement with railway bridges was a few years ago), railway vehicle loadings are specified in what was British Standard BS5400, now forming Appendix A of BD37/01, itself being part 14 of the Design Manual for Roads and Bridges. The relevant loading for main line railways is RU loading to clause 8.2.1, with dynamic effects as clause 8.2.3. RU loading is explained in Appendix D of BS5400, which shows it as based variously on B-B and C-C locomotives, including double-headed, 4-axle bogie wagons, and exceptional wagons with more (up to 14) axles. There is no reference to steam locos that I have found, so unless there are documents from other sources covering them, it's possible that an engineer making assessments for them has to work outside the current codes. The high un-sprung reciprocating masses of a steam loco's motion may well produce dynamic loads well in excess of those from a diesel or electric loco, and it would be interesting to know how engineers carrying out bridge assessments for heritage lines determine their loadings.
The nominal static axle load in RU loading is 250kN = 25.5 tonnes, but the dynamic factors can increase the resulting forces by up to 2.0 times. The vehicle axle loads for the various types as listed above are between 20 and 25 tonnes per axle.