depends on the differences of mine and your interpretation of whoop-whoop-whoop sound. Drive model JU-268A026C is what i used.
If you can record the sound it makes perhaps it may be relevant to someone here. Can you at least confirm if it comes from the drive itself, or definitely from some specific area of the PSU?
As for the drive, it is definitely NOT a stock 575. I presume now that you used the G3 drive in your 575 first?
According
to this article, the JU-268A026C is a manual inject floppy manufactured by Panasonic which came stock in the first G3. It is definitely a Superdrive and is otherwise compatible with all earlier 400K disks, software and hardware. I keep coming back to
Henry's claim that Apple switched to non-Superdrive 1.44MB drives (single speed PC drives) in late PPC & G3 Macs. However, this appears to be limited if at all. Perhaps, the Panasonic drives, while variable speed, were somehow electrically incompatible with a 512K. However, I find that highly unlikely considering it functions perfectly well on a 575 which is electrically compatible with earlier Macs. However, as the referenced article points out, the auto-inject floppies fail on the later OEM manual-inject systems. This suggests a likelihood that there is some minor difference that the extremely recent G3 floppy mechanism posses that could possibly cause a problem with your 512K Mac. Though to overload the PSU to such a degree seems to be an extremely unusual side-effect of an otherwise standardized interface. Surely such a recent drive has some electrical specifications online somewhere to compare.
One thing that comes to mind is that the Panasonic drive, being originally a PC manufactured drive, does not isolate pin 20 on the header to differentiate between the PWM and SWIM signal, i.e. it tries to interpret the PWM signal as the SWIM's +5V signal, resulting in a similar effect on an 800K drive, continually triggering the ejection mechanism. If that's the case, the ejection process is one of the highest loads the drive will put on the PSU, a continuous cycle will put a huge drain on the PSU and cause a weakened one to overload. One way to test this is to either brak the pin off the drive, or cut the associated wire on the ribbon cable (which at least allows you to repair it and return it to normal use).
But this really all comes back to you. Have you confirmed that your PSU is in good health and supplying proper voltages? Even under such a load, a healthy PSU, including a 128K/512K Mac's, would not react so violently to continuous ejection cycling. Putting non-standard equipment on a borderline Mac is just a recipe for disaster.