This article covers the design and finally PCBs for one of my microphone upgrade designs.

This project dates back a couple of decades or so, to when I acquired a batch of small Shure boundary mics, MX391 or similar, without any accessories or converter etc. - at £10 each, if I remember right.

I wanted to use these with a "whole house" home automation system so I needed to design an interface to connect them with the XAP800 computer interfaces mixer I was using with the system. That had XLR mic inputs, my first experience with those.

The microphone capsules in the Shure casings were three wire electret, with both source and drain connections as well as ground. To minimise interference pickup I wanted to keep the signal connections balanced directly from the capsules. (I'm a radio amateur [ham] and quite active back then, so everything was exposed to quite strong RF at times from automated digital-modes gear).


These boards were the result, using equal value source and drain load resistors for the capsule and capacitive coupling from those to two [BC107C] common-emitter amp stages with emitter feedback and using the phantom power feed resistors as the collector load resistors. The output circuit was on top of a zener & capacitor shunt regulator to provide the low voltage feed for the mic capsule load resistors.

They worked well for their intended purpose and speech commands could be picked up throughout the house (using HAL2000 home automation software with integrated voice recognition & synthesis).



Later on, as high quality audio interfaces started to become available at reasonable prices secondhand, I got more involved with home studio type gear and started to pay more attention to things being advertised for this type of use.

Other than the long-ongoing "magic cable" scams, microphones seemed to be the biggest target for misinformation and false advertising, and I started studying the detailed technicalities of different types and versions being sold. 


I was surprised to discover that one of the commonest mic circuits, the "Schoeps" style design, was very similar to the one I came up with for my voice control mics!

The biggest difference is that I used output drivers with gain, as I was trying to approximate the signal level from the Shure XLR adapters for those mics (which have around 13dB gain) while everyone else uses emitter follower zero-gain output drivers.


The unbypassed emitter resistor in the output transistor configuration of my design makes these, in effect, modulated current sinks.

Current control is often used in preference to voltage drive in circuits to achieve better response - dating back to such as Teletypes and presently with eg. MIDI interconnects. It's not as fast response as an impedance matched and terminated transmission line can be, but very effective over ad-hoc and variable cable lengths any types.

Whether it is "better" in the context of microphones is so far guesswork, I have not done comparative tests - that's much lower priority that other projects I have in hand and queued up! If anyone with appropriate gear would like to do some tests, please let me know!


Back to the overall microphone stuff, I started looking for possible cheap-but-could-be-good mics to use and experiment with. Over a few years I added a few to my collection, as they appeared at low enough prices on ebay or in secondhand shops.

Some of these, such as the MXL V87, Shure SM58 MXL 603S pair are excellent as they are and remain untouched, though I may investigate the 603s one day.

Others such as the MXL 990 are pretty good but have some obvious flaws such as multilayer ceramic coupling capacitors, due to cost-cutting for manufacturing, and at the low end the scammy fake "Studio condenser" mics do not have much actually worth keeping inside the casing - but the casing alone is worth the cost, if you find them cheaply enough.


After doing major modifications to the 990 and one of the cheap mics, I started considering a full rebuild and bought a cheap large diaphragm condenser mic capsule on ebay. It was some time from that & various ideas considered and scrapped before I came up with a design I was happy enough with to go to the PCB stage.

That design allows for building it as either my design plus an optional phase splitter for two terminal electret capsules, a JFET input for condenser or bare electret capsules, the classic Schoeps circuit or numerous other variations including filtering and decoupling for the power to different sections.

I also designed a well filtered power inverter board to use with that, when a high voltage bias supply is needed.

The resulting  circuit: 

If it is built with NPN transistors and values similar to those shown as in the next image down, it works in a similar way to my original.

With PNP output transistors and appropriately changed resistor values as in the third image below, it becomes a Schoeps or one of the many variations of that style.


1: Generic fully adaptable circuit, used for the PCB design:


2: Example components for full build with NPN output, RJ style:



3: Example components for full build with PNP output, Schoeps style: