uBITX Electret Microphone

Started: 04/01/2019       Last Updated:


Circuit in the Box 1
Circuit View 1

        Circuit in the Box
Circuit View 2

Circuit in the Box 2
First Mic Hole, inner case

        Circuit in the Box
Finished, 2nd Mic hole, outer case

To view larger images click on the image you wish to see larger.


The circuit I used to build this microphone was taken from a recent article in QST with one modification: The uBITX mic element, is an electret mic element, not a dynamic element and it needs a biasing voltage to drive the mic element. I looked up other electret mic elements and how to bias them and found that in the past I'd used some from Radio Shack. Those needed a 1K-Ohm biasing resistor, so a 1 K-Ohm resistor was added.

You will also notice that I used a longer 1K-Ohm variable resistor for the transistor amplifier. There were two reasons for this: (1) this resistor is more accurate than a smaller pot would be, and (2) I have more of them than the smaller ones. I figured that if it didn't fit into the case I intended, I could always swap it out and use a smaller pot.

I considered that the 1K pot should probably be put at its center value, or about 500 Ohms for each side, and I would then adjust the gain while outside a case. Once satisfied put it into a case. As it turned out, this value seems to work rather well. While it may be able to be improved a bit, I thought the overall quality and volume was fine during my tests. This may change as I have SSB QSOs and get reports from others.

The image/schematic below comes from QST December 2018 Page 53 of Hints & Hacks, and is used with permission. The one change/addition which was needed was the 1 K-Ohm, R3, resistor. This resistor is needed to bias the electret mic element and, without it (as I discovered when I forgot to include it), the mic does not work at all. The original article was by Paul Dobosz, K8PD and, as mentioned above, was for a dynamic mic element.

Schematic With Phantom Power & Bias resistor

Listen to the mic: outside the case   &   inside the case


The recordings were made using the SDRPlay radio and receiving program. The radio was connected to a dummy load with a wire attached to the dummy load. The SDRPlay radio was also connected to a length of wire and placed close to the wire connected to the dummy load. Both lengths of wire were looped and the separation between the two loops of wire was about 12 inches.

SDRPlay and uBITX antennas


A whistle was noticed when testing with the mic inside the case. This whistle sounded like a noise heard on an AM radio near my battery desulfator circuit when it is working. Sure enough, a check showed that the desulfator was operating and was the cause of the whistle. A choke was tested on the power cord close to the radio's power input port. This test was done while the desulfator was still working and it sounded like the choke removed about 95% of the whistle so it was left on the power cord just-in-case the desulfator was not disconnected before operating the uBITX radio. A good article about how to build this circuit as well as the theory about deulfation is at: Battery Desulfator Concept and Design Explored


Phil Karras
04/01/2019

  Return to the: