If you just need a new fan... why do you want to replace the fan driver as well?
Here's some data for the Papst 612 fan, courtesy of an early-2000s catalog (that series was marked at obsolete even then):
Throughput 40 m³/h
Static pressure 75-80 Pa (~ 8 mm H2O) @ 0 m³/h
5300/min
2.5 W
40 dB(A) - 5.1 bels
They still make the replacement 600N series, the closest model would probably be the 612NN (42 m³/h, ~65 Pa, 5100/min, 1.6 W, 35 dB(A) / 5.2 bels). Your Noctua would be more along the lines of a 612NML (25 m³/h, ~25 Pa, 3000/min, 0.8 W, 19 dB(A) / 3.9 bels).
I wouldn't just install a weaker fan. If you want to tweak something, the entire cooling concept should be examined, which in (vintage) test instruments can be rather crude (but might be somewhat more refined in a device that is pretty much design to dissipate power).
- At the very least, can the fan be shock-mounted (preferably without introducing air leaks in the process)? That used to help a lot with low-RPM Papst fans in the day, which have relatively few motor poles. Nothing more annoying than having this ticking amplified by the entire case. The designers of Siemens PBXs still haven't gotten the memo.
- Would there be ways of maybe fitting a second one in parallel, so a lower RPM model can be used?
- Is there sufficient intake area? With axial fans, obstructions on the output side tend to hinder airflow far less than a constricted input. (Oddly enough, the hot trend in 2018 PC cases was almost closed-off front panels with mere millimeters of clearance for front intake fans. )
Now as for the question of indutive surge... anything wrong with the classic parallel diode trick?