3

I have several voice-radios with amplified mono outputs. The radios are designed to connect directly to a 4Ω speaker. I would like to connect two (or more) of them to the same speaker.

The radio transmits human voice (think walkie-talkies), so the signal is relatively narrow at 300-3kHz. Each radio has an independent volume knob.

Naively mixing two (or perhaps more) audio signals is relevant, but the solutions suggested are active. In my case, my signals are already amplified, so in theory I don't need further amplification. Furthermore, an active solution could have an undesired impact on a certification effort.

Using resistors is likely to substantially alter the frequency response, as explained in https://electronics.stackexchange.com/a/29871/248767, so that looks like another dead end.

Question: can these amplified signals be mixed together passively, maybe via some intelligent usage of decoupling capacitors and/or transformers, or is the simplest solution just to run each radio to its own independent speaker and call it a day?

Kenn Sebesta
  • 786
  • 5
  • 13
  • Mixing without mixing controls? – Andy aka Aug 15 '23 at 16:21
  • @Andyaka yes, so it basically becomes a summer. – Kenn Sebesta Aug 15 '23 at 16:25
  • Simple way is to connect the two radio's GND together... then connect one end of a speaker to one radio audio-hot and the other end of the speaker to the other radio audio-hot. Only works for 2 radios. – glen_geek Aug 15 '23 at 16:25
  • @glen_geek that is simple! Very nice. Question: would independent volume control still work, and would it still work if one radio were simply off? – Kenn Sebesta Aug 15 '23 at 16:38
  • @KennSebesta You need to be careful with this, many two-way radios use balanced amplifiers and will NOT like having their speaker terminals grounded or connected to another radio. Can you tell us what model radios you plan to use? – GodJihyo Aug 15 '23 at 16:59
  • Good questions - independent volume controls should work. I'd guess when you turn one radio off, volume would go down for the other radio...hard to say how much volume you lose without seeing audio power amplifier circuit. Do ensure that both radios have one of its speaker terminals going to GND,as @GodJihyo has cautioned. There's another scenario where simple radio-to-radio GND connection may need a slight change...(transformer-coupled speaker). – glen_geek Aug 15 '23 at 17:13
  • @GodJihyo radios is TBD, but it's definitely something I would thoroughly investigate before hooking up. My hunch is to presume they are balanced amps. – Kenn Sebesta Aug 15 '23 at 17:45

3 Answers3

5

Question: can these amplified signals be mixed together passively, maybe via some intelligent usage of decoupling capacitors and/or transformers

Connect each radio output to a transformer (for isolation) and connect both transformer secondaries in series to achieve a summed output. No mixing controls or anything special.

Andy aka
  • 456,226
  • 28
  • 367
  • 807
  • I lpve the simplicity. Do you have a reference I can read up on that? I wasn't able to find much on people doing this with high-power, although https://gearspace.com/board/showpost.php?p=666410&postcount=6 sounds very pessimistic about the technique. – Kenn Sebesta Aug 15 '23 at 19:44
  • 1
    No reference; just 50 years of experience and electrical common-sense on transformers @KennSebesta – Andy aka Aug 15 '23 at 20:00
  • Hey, sounds like you are the reference! :D How sensitive is this to transformer type? If I grab two random transformers from my spare parts bin, is it likely to work well enough to give a taste, or is this sensitive to proper parts matching, winding turns, magnetic core, etc... – Kenn Sebesta Aug 15 '23 at 20:14
  • 1
    Just try it and see but, realistically for perfection, you need to use the same transformer model and, that needs to be frequency spectrum rated for audio (not power) and ideally both will step down 2:1. – Andy aka Aug 15 '23 at 20:22
  • Excellent advice, thanks. Is the 2:1 ratio because of two inputs, so for three inputs it would be 3:1? – Kenn Sebesta Aug 15 '23 at 21:04
  • Yes, and just to avoid overloading a single speaker. – Andy aka Aug 15 '23 at 21:10
  • if both radios outputs are near perfect voltage sources this will work, if not you'll get exacerbated crossover distortion and intermodulation – Jasen Слава Україні Aug 16 '23 at 00:31
  • 1
    @JasenСлаваУкраїні please justify your concerns by writing an answer (comments aren't enough). Try and show how cross-over distortion would occur (given that the OP may be using class A amplifiers). How close to perfection is "near-perfect". I see your comment as an unwarranted and unjustified "spoiler" at the moment so clarity is needed (not guesswork or speculation). Sure, what I propose is rough and ready but, that's the whole point. Your answer would be seen as acceptable if it tells the OP the passive idea is badly flawed because of this, that and the other etc.. – Andy aka Aug 16 '23 at 07:43
  • @Andyaka 1) I did some digging and it seems at least one of the amplifiers will be class D. Not sure if or how that affects sound quality. 2) That's good advice about the speaker power, I'll rate it to > the sum of the inputs divided by the transformer ratio. – Kenn Sebesta Aug 16 '23 at 11:29
  • Also, is there not a concern that the amplifiers will be back-driving each other via the transformers, possibly being overloaded and damaged? – Kenn Sebesta Aug 16 '23 at 11:57
  • I think there are quite a few variables here so why not try it out on a bench first. – Andy aka Aug 16 '23 at 12:43
  • @Andyaka what would be a suggested way to try it out on the bench and what would I be looking for? I don't have any equivalent hardware except the real things, nor a fundamental understanding of what good and bad look like. If the concern is damaging the amps, mightn't that be more of a long term heat problem than a short term functionality test? – Kenn Sebesta Aug 16 '23 at 12:52
  • 1
    I'm sorry Kenn but I think I have gone past the point from advising you to solving problems that can only be done with a bench test; sometimes you just have to try things out carefully at first then, see what happens. These are proprietary radios so I suggest contacting the supplier now. – Andy aka Aug 16 '23 at 13:09
  • @KennSebesta The reference: transformerless audio output stages are a rather “new” thing. Back when transistors had poor current handling, transistorized output stages drove speakers via step-down transformers. Inherently high-impedance vacuum tube output stages did so as well. Audio began with a transformer driving the relatively low-impedance speaker. Open up most any transistor radio from the 60s and there’s an output transformer I’d bet. – Kuba hasn't forgotten Monica Aug 16 '23 at 13:54
  • @Andyaka no prob, I understand. Thanks a lot for taking me this far. – Kenn Sebesta Aug 16 '23 at 14:16
2

The two receivers have internal amplifiers that drive the loudspeakers.

It's hard to be sure how amplifiers will behave when they have another amplifier connected to their output terminals,

One possible way to isolate them from one another is to use a hybrid transformer

schematic

simulate this circuit – Schematic created using CircuitLab

If galvanic isolation between the two sources is not needed (only power isolation is needed) a simpler circuit can be used. Here L1 is a 1:1 (centre-tapped) autotransformer. This can be constructed from a 1:1 transformer by connecting together one leg from each winding. Or just using a centre-tapped winding on a transformer with other windings.

schematic

simulate this circuit

Kenn Sebesta
  • 786
  • 5
  • 13
Jasen Слава Україні
  • 33,738
  • 1
  • 33
  • 70
0

One speaker per radio is the simplest solution. A multi-gang rheostat can be used for output volume control. It won’t affect the frequency response enough to matter. Each radio’s own volume knob is then like a level knob in a mixer: the relative volume of each radio in the “mix” is adjustable. The neutral volume knob position for each radio would be at 50%-75% depending on the desired maximum volume.

Kuba hasn't forgotten Monica
  • 40,960
  • 1
  • 39
  • 120
  • I don't fully understand. Did you mean one resistor per radio? If it's one speaker per radio, then wouldn't the rest of the answer not be applicable since the radios are isolated at this point? – Kenn Sebesta Aug 16 '23 at 14:30