The first question you’d have to ask is, “why would you want an audio mixer involved with a radio”. Looking on the web, there seem to be some people concerned with audio quality and wanting a large station all connected up through a single mixing desk. I don’t own enough radios for that, and I’d question the quality argument when the signal goes through a 2.4KHz SSB filter: so that’s not my reasoning.
My interest came about from planning for the SSB field day. We discussed that ideally the operators should be able to use a headset. If someone else is able to help with logging they’d also need to be able to hear the messages exchanges, and ideally the logger and operator should be able to talk to each other. Unfortunately most modern radios only have a single headphone output and that cuts off the internal speaker; “hear each other” means that the microphones would need to couple to the headphones. So some kind of dabbling with electronics would be needed.
In principle, simple audio amplifiers to arrange something aren’t difficult. Somewhere along the way, I thought “hang on – an audio mixer from Maplin can do all that”. I bought the second-cheapest mixer (£60) and set out to have a play. Hence the next two weeks of dabbling!
What I hadn’t anticipated is the world of “proper” audio. You can’t simply plug in a headset: more work would be needed. Most high quality audio signals are balanced feeds: think of a dipole fed with twin feeder, rather than a monopole fed with coax. Most “proper” microphones are dynamic mics, and have balanced connections using XLR connectors. It turns out that a mixer is covered in stereo jack sockets, but most of these only have mono signals – they’re also balanced. The “line” signals for everything else also have relatively high voltage levels – easily 4 volts peak-to-peak.
There are high performance headsets for amateur radio, with well-into-three-figure price tags. There are also high quality headets for PC gaming, at around £15-£20, but most of those use electret microphones. Electret microphones can’t connect directly to a mixer – they need a power feed. Although mixers can often provide 48v “phantom” power for semi-professional microphones, they can’t directly connect an electret. Given that electret microphones are widely used in the media and broadcast world, that’s a bit surprising.
I didn’t want to modify the mixer too much – certainly not to the point of stopping it connecting to “ordinary” microphones. I found a few circuits on the internet to derive a suitable low voltage power feed from the mixer, and tried a few. They all had the same performance, but the simplest is much easier to assemble. (It’s not even close to balanced, but it doesn’t seem to make a difference). I’ve been able to fit the components in a “rats nest” form into an XLR plug, with a 3.5mm stereo socket for the headet’s microphone in the top.
OK, two electret microphone adapters feed mixer channels 1 and 2. I now had audio into the mixer and out of the mixer from two headsets. I could set the volumes and adjust the tone balance with a 3 stage equaliser. (The electret microphones lack bass response, and the equaliser puts that back). Connecting RX audio from the radio was fairly straightforward: a simple stereo connection to mixer channels 3 & 4 that plugs into the headphone output. A set of PC speakers could also plug into the output, for any observers that don’t have a headset. It’s a stereo channel, so good for radios with two receivers driving the left & right ears.
A great little feature of the mixer is an ability to “send” audio to an external audio processor. The mixer has two rows of knobs to drive two external signals, and you can set the audio “mix” sent to each. This gives a way to choose which microphone connects to the radio – talk into the other and nothing is transmitted, but the operators can still hear it. So the radio TX feed was a mono feed from one of the “aux send” outputs. They’re at “line” level, so a simple resistive divider was used to drop to a sensible level for the radio mic input (probably 10mV peak-to-peak). There’s no provision for PTT in the mixer: that needs to be provided separately. I’ve created a little interface box with the potential divider, a push switch and a socket for an external switch using a phono socket. Alternatively, VOX can of course be used.
So how well does it work? We’ve established that RX and TX audio quality is good. It looks complicated - it needs an annotated photograph to show the required settings, in case something got nudged. What about RFI? On Chris M0KTT’s “inverted L” antenna, in a well sorted shack, we had no problems with RF pickup. On my FT450 with an end-fed and poorly grounded antenna, I get RF feedback into the headphone circuit (but not apparently into the microphone circuit). I’m pretty sure this is an antenna problem – RF in my shack is a well known issue - but I’m planning more work with some ferrite clamps to see how well it can be hardened.
So that’s where I’ve reached: it works, and requires little constructional effort. The audio quality hasn’t obviously suffered, and I can now use any of several microphones with the radio. In principle there are enough channels to couple up a second radio, if I could see a benefit. Alternatively a small & cheaper mixer might have worked too.
We acquired an FT450D a couple of months ago. This is an excellent radio: it copes with HF signal levels better than the FT817, and seems to have more "punch" in its audio that the 13dB increase in TX power would suggest. I wonder how much of that involves speech compression on TX - nevertheless the results are good.
Connecting it to a PC for CAT control should be straightforward: it does after all have an RS232 interface on the rear panel. It turns out the radio requires that RTS be asserted, or it won't respond (not unreasonable, seeing as that's the purpose of the pin). The cable I've ended up with is as follows:
Using that cable, Ham Radio Deluxe is able to take control - as expected.
On data modes, the radio appears to work well. Ham Radio Deluxe's data mode program "DM780" was able to demodulate several PSK signals using my FT817 data interface, although the signal level looked a lot lower than it should have been. Even the PC microphone picking up audio from the speaker was successful.