17/06/2026
Throwback to 2006. I needed a dual mic pre for bootlegging concerts on a mini disc recorder using XLR condenser mics (just 9V phantom needed). So I built this little thing around a INA2126 chip and a TL074 quad opamp. The Pan pots allowed me to mix two mics to mono if needed. After sitting unused for at least 18 years I plug it in and it's fine (battery full, woohoo!). Except for one issue. In this instance I am feeding the unbalanced outputs into little isolating transformers in order to send the signals about 100m through ethernet cable. Only a severely high passed signal comes through. The reason is explained in the last image/scribble; I hadn't grounded the ring contacts on the output jacks in the little preamp. This doesn't matter when connecting to electronic differential inputs as the + & - are referenced to ground, but when connected to a transformer the primary of the transformer needs a return path for the signal, otherwise it has no reference and no current can flow! And the reason why you will get heavy high-pass effect is due to the capacitive coupling through the transformer windings, allowing only the very highest frequencies to leak through. If you wanna be really safe then you should ground the ring (or XLR pin 3) on unbalanced I/O with a 47Ω resistor. The reason for this is that some early electronically balanced output stages did not like it when one side was accidentally shorted to ground. Damage could occur, or at least some distortion in the other signal. Modern designs always (!?) have build-out resistors to protect against this.
