After replacing the capsules in a pair of MXL 990 mics with 35mm Alctron K67 clones, the sensitivity was lower than expected. The first thing I wondered about was the polarizing voltage. For a Chinese K67 style capsule, the bias voltage is generally between 40 and 60 volts, with more output and therefore better S/N with increasing bias. I looked at the MXL 603 schematic which is almost identical to the 990 and 991, posted online at XAudia.com (a great resource with schematics of many microphones and other info.)

http://www.xaudia.com/omnip/Mics/MicSchLib/MXL603S.jpg

(Schematic courtesy of xaudia.com)

Checking my 990s, I found that the Zener diode at D3 was 6.8V instead of 12V in all four of them. This resulted in a bias voltage at the hot end of C11 around 37V and a bit less at the capsule. My voltmeter has an input resistance of 1 megohm, so it can’t measure the voltage at C12 or the capsule itself. Oddly, the 991 pencil mics all have 10V Zeners, resulting in bias supply around 48V. My diode bin held some 12V 1W Zeners, so it’s swap time.

With a 12V Zener in place, the capsule bias jumped up to 64.4V and the headroom on the first stage FET is also increased. No side effects are apparent, so I converted all four, the ones with the modded MXL capsules and the ones with Alctron large diameter capsules. I left the 991s alone this go round, since they are already pretty close.

Will report later as the mics are tested in action. I THINK they are quieter and more sensitive and perhaps a little crisper, but of course that could be all in my head.  :^)