cmoorevt, like many other transformerless condenser microphones, the KM 184 doesn't polarize its capsule directly from the DC of the phantom power supply; it uses a DC/DC converter to step that voltage up--in this case, to around 60 Volts--for greater sensitivity and dynamic range. (A microphone drawing 3.2 mA would get only around 37 Volts; that capsule was designed for 60.)
Since transformers can't step DC up directly, there's a little oscillator in the amplifier circuit which provides an AC signal (normally well above the audible range), which can then be run through a tiny transformer and converted back to DC again. What you're hearing is leakage from that oscillator which is unable to get up to speed, so to speak. In most cases this indicates not that the microphone has a problem, but that your phantom power supply isn't putting out enough current for the microphone.
Your signature lists an Edirol interface. In any of their portable recording interfaces that I've ever tested (I'm the obnoxious guy who brings a set of phantom power testers to the Audio Engineering Society convention and tries them out in all the exhibit booths if people will let me), the phantom powering has been marginal at best. It may be around 48 Volts when nothing is connected to it, but plug in a microphone and the supply itself begins to sag--I don't mean the normal voltage drop across the 6.8 kOhm resistors; it's more than that.
Many modern, transformerless condenser microphones require from 2 to 6 mA apiece. Your KM 184 only wants to draw 3.2 mA (2.3 if it's the older model, but by now yours probably isn't), though during startup it may need a little more, and it seems as if that wasn't available. Check this microphone out with a known good phantom power supply, and the chances are very good that the microphone will prove to be OK.
--best regards