Post by thread:[EE] Simple op-amp mixer design

Hey all, I've been in a major "clean-up" mode lately and as part of this, I decided to have just one pair of speakers on my desk, but still use them from all 3 computers. So I need a *simple* 4-channel (1 extra for future use) stereo line-level mixer. I'd appreciate some help with this design... My initial (and hopefully final :-) thought is to do this with a single op-amp per channel, with a gain of greater than 1 (I'll just experiment to find the right level), and each input would be connected to this via a resistor and a decoupling capacitor. I'll also use a single-supply from a wall wart. However, how do I calculate the correct values of the input resistor and decoupling caps? I don't need individual volume controls, nor any switches etc. Finally, I intend to do this with parts I have laying around here (already have enclosures, stereo jacks, PCB's, resistors, caps, and op-amps), but which op-amp should I use? My choices are LM741, TL072, LF353. Thanks, -Neil. ____________________________________________

PicDude wrote: {Quote hidden}>Hey all, > >I've been in a major "clean-up" mode lately and as part of this, I decided to >have just one pair of speakers on my desk, but still use them from all 3 >computers. So I need a *simple* 4-channel (1 extra for future use) stereo >line-level mixer. I'd appreciate some help with this design... > >My initial (and hopefully final :-) thought is to do this with a single op-amp >per channel, with a gain of greater than 1 (I'll just experiment to find the >right level), and each input would be connected to this via a resistor and a >decoupling capacitor. I'll also use a single-supply from a wall wart. > >However, how do I calculate the correct values of the input resistor and >decoupling caps? I don't need individual volume controls, nor any switches >etc. > >Finally, I intend to do this with parts I have laying around here (already >have enclosures, stereo jacks, PCB's, resistors, caps, and op-amps), but >which op-amp should I use? My choices are LM741, TL072, LF353. > > Use the TL072 or LF353. 47K input resistors in series with a 100n capacitors go to the inverting input of the opamp. A 47K goes from inverting to output. This will make the gain unity. The non-inverting goes to Vcc/2. Use a couple of 10K resistors and a 10u cap to get your Vcc/2 point. Add a 47u output cap in series with a 100R resistor on the output. Run the whole thing from +12V. David... ____________________________________________

part 1 929 bytes content-type:text/plain; charset="iso-8859-1" (unknown type 8bit not decoded) > Hey all, > > I've been in a major "clean-up" mode lately and as part of this, I decided > to > have just one pair of speakers on my desk, but still use them from all 3 > computers. So I need a *simple* 4-channel (1 extra for future use) stereo > line-level mixer. I'd appreciate some help with this design... > > My initial (and hopefully final :-) thought is to do this with a single > op-amp > per channel... Hi Neil, How about just doing a zero-voltage summing bus? The quality should be fine(when compared to the ground loops you'll likely have with three computers), and you can do stereo with just a dual opamp, or use a quad if you want to uninvert the signal. Attached is a diagram of the left or right channel. If you run it single supply you'll need t provide vref and also use an output cap. Cheerful regards, Bob part 2 4485 bytes content-type:image/gif; name="mixer.gif" (decode) part 3 79 bytes content-type:text/plain; charset="us-ascii" (decoded 7bit) ____________________________________________

On Tuesday 19 October 2004 12:05 pm, Bob Blick scribbled: > Hi Neil, > > How about just doing a zero-voltage summing bus? The quality should be > fine(when compared to the ground loops you'll likely have with three > computers), and you can do stereo with just a dual opamp, or use a quad if > you want to uninvert the signal. Attached is a diagram of the left or > right channel. If you run it single supply you'll need t provide vref and > also use an output cap. > > Cheerful regards, > > Bob Hy Bob, I'm not familiar with the terminology "zero-voltage summing bus", but your circuit is very straightforward. I'm also wondering why you didn't use individual capacitors on each channel, so I'm assuming that's the point of the zero-voltage summing bus? And noticing that it's inverted, I'm not sure why I was adamant to make my circuit non-inverted. As long as both channels are inverted, all should sound fine. The non-inverted design also makes the calcs for the resistor values pretty simple, so I'm good there as well. Question: As for power, I was thinking earlier today, that I wanted to run it off of a wall wart, hence the single supply, but I thought of doing this instead for a split supply: Vin+ >-------'\/\/\,------o--------------> V+ to mixer | R `-, ^ Zener /_\ | o--------------> Gnd to mixer | `-, ^ Zener /_\ | Vin- >-------------------o--------------> V- to mixer I can't see why it won't work. I'll use 12V in from the wall wart, and 5.1V zeners for z +/- 5.1V supply. Of course, filter caps etc will be added. Thanks, -Neil. ____________________________________________

Hi Mike, Not sure exactly what you mean here. Since all 3 PC's are usually on at the same time, I want them all to be output from the same set of speakers simultaneously without changing connections. Cheers, -Neil. On Tuesday 19 October 2004 03:09 pm, Mike Singer scribbled: > Neil, > > What's wrong with handling the connectors manually? > > Mike. > ______________________________________________

On Tuesday 19 October 2004 06:22 pm, David Duffy scribbled: > Mike Singer wrote: > > <>Neil, > > > > What's wrong with handling the connectors manually? > > Well, he did ask for guidance on a circuit and we happily gave it. > Maybe he gets sick of friggin' around with those connectors! > David... Pretty much. But also because I am sick of having 2 sets of speakers on the desk, and having to listen to the laptop's crappy speakers, which are mostly muted since my palms rest on the speakers when I'm typing. Speaking of cleaning up, one of my computers is Mini-ITX based, and small enough that I mounted it on a keyboard tray under the desk. Nice stealth setup! Cheers, -Neil. ____________________________________________

Hi Neil, The reason for the zero-voltage summing bus, it means you get no (well, in a perfect world, but pretty much almost no) crosstalk between channels, and the gain of one channel doesn't change with another channel, you can even add or disconnect channels from the bus with no change to the other channels. > I'm also wondering why you didn't use > individual capacitors on each channel, so I'm assuming that's the point of > the zero-voltage summing bus? Just for simplicity, there's no need to use individual capacitors. If you run a dual supply and all your inputs have no DC on them, you can run with no capacitor at all. But make the capacitor plenty big, shoot for a 5 Hertz rolloff. > And noticing that it's inverted, I'm not sure why I was adamant to make my > circuit non-inverted. As long as both channels are inverted, all should > sound fine. The non-inverted design also makes the calcs for the resistor > values pretty simple, so I'm good there as well. Most consumer gear pays no attention to signal inversion anyway. For a split supply, lots of devices use an AC output wall wart, like this: *-------*------>|------* positive output | |------|<------* negative output *----------------------* ground Wall wart goes in at the left side, and of course you'll need filter caps. Cheerful regards, Bob ____________________________________________

David Duffy wrote: <<<<<<<<<<<<<<<< Well, he did ask for guidance on a circuit and we happily gave it. Maybe he gets sick of friggin' around with those connectors! David... >>>>>>>>>>>>>>>> Neil wrote: <<<<<<<<<<<<<<<< Not sure exactly what you mean here. Since all 3 PC's are usually on at the same time, I want them all to be output from the same set of speakers simultaneously without changing connections. >>>>>>>>>>>>>>>> Sean Alcorn wrote: > > What's wrong with handling the connectors manually? >Now that would not be much of a 'mixer' then, would it? Neil, You've got a mixer with each computer's sound system. Why not connect speakers to some central computer and other computers to its spare sound system's inputs (some inside computer case)? Regards, Mike. ____________________________________________

On Wed, 2004-10-20 at 16:03, Mike Singer wrote: > Neil, > You've got a mixer with each computer's sound system. > Why not connect speakers to some central computer and other computers > to its spare sound system's inputs (some inside computer case)? > > Regards, > Mike. Agreed, in fact that's exactly what I do, except I do it in a daisy chain fashion. Computer 1 output connected to computer 2 input. Computer 2 output connected to computer 3 input. Computer 3 output connected to home theatre. Works quite well. Only limitation is every machine down the chain has to be on (not a problem since I never turn those machines off). The multiple sound cards in the chain does lead to some extra noise, but I've never been bothered by it. I never use my PC for anything hi-fi anyways. TTYL ----------------------------- Herbert's PIC Stuff: http://repatch.dyndns.org:8383/pic_stuff/ ____________________________________________

You could also do a passive mixer. If, for example, you tie one end of three 10k resistors together and drive the other ends with voltage sources, the voltage at the common end will be the average of the three, or the sum of the three divided by three. You can compensate for the divided by three by turning up the gain of the amplifier. Harold (The ideal design has zero parts) -- FCC Rules Online at http://www.hallikainen.com ____________________________________________