Condenser mic capsule noise

I was surprised to find noise in an Oktava MK-102 mic that appeared to be the result of the capsule body connections.  The front and back screw off, and simply adjusting the tightness of the screw threads made broadband noise drop by up to 15 dB.  I never opened it all the way, and never touched anything internal.  Worth knowing if you encounter broadband noise in a condenser head. 

Here's an averaged noise measurement comparison of a bunch of Oktava capsules on the same body, with the white line showing the noisy capsule.  Purple at the bottom is the system noise floor.   

contact resistance? Did the gain stay the same?

Gain never moved, and I really can't tell if anything on the inside would have touched something inadvertently.  Seems only like contact resistance around the body, which I guess is the return path. 

not surprised.

Concur on body as return path for power, contact resistance issue there as cause? 

Doug, I'm impressed.

Where is the mike when you do these noise measurements?

Have you got an anechoic or quiet room?

Don't believe the 'calibration', it's a reference to 0 dBFS, and likely shoddy at that.  I took no particular care to isolate ambient noise at all, just a quick comparison of multiple capsules under similar conditions.  The mics are laying on a wooden desk in my control room, and I'm doing my best not to move or breathe while the averaging settles out.  The area is not particularly quiet, there was just no one else around and no cars passing at the time. 

I see this typically with old C24 mics, so seeing it on a newer low quality mic is not a surprise. You are on the right path.