Here’s the last batch of tests, over all acoustic response of the K12 “true” condenser mics. This is really frequency response of two “RK-12s” from eBay. "Guaranteed exactly the same as Mic-Parts." Yup! They are a bit hot in the bass, wonky in the upper mids, and a bit hotter on top than the genuine ones I bought from M-P. But what the hey, they were half the price. They are not very well matched either. Now I know what a $70 eBay double side “RK-12” is like. You pays yer money and makes yer choice.


The OPA K12 with HPF is as Homero sent it, with added ground, and PCB cleaned.

Cardioid.



Omni.




OPA K12 has changed zener and 4.7uF non-polar output caps.

Cardioid.



Omni



… and the reference Dayton EMM6 mic. Just checking that the rest of the gear is OK.

There is no output from the Tannoy speaker below 40Hz, so that is random room noise in all graphs.



In short, these op-amp boards are the flattest and least distorted mic circuits I’ve ever tested. If you were to describe the “flavor” it would have to be something like “distilled water”. OTOH, like any mic, the main contribution to a “sound” is the capsule, and a good capsule will shine. A poor capsule will have no cover.

 











 

First, the electret and flat polarized circuit.



The slight rise at high frequency is probably a slight roll-off in the calibration channel.

The polarized circuit with LF filter.



Essentially a straight line above 40Hz.

Electret circuit distortion and noise.


Distortion was higher than expected, traced to a problem with grounds, which has been corrected with a new board layout.

Flat “true condenser” circuit distortion and noise.


This board tested more or less as expected with barely detectable distortion and noise.

High pass filter “true condenser” circuit distortion and noise.


There is of course “distortion” at and below the cutoff frequency of the high pass filter.

Transient response to a step impulse is clean and settles quickly with minimal overshoot. This is mainly a function of the output coupling capacitors, since the rest of the circuit is direct DC coupled.

The only other thing I could think of is replacing the regular electrolytic output coupling caps with non-polarized types, which usually show less distortion. So it turned out, replacing Homero’s 50uF polarized caps with 4.7uF/50V non-polar.