PC mixer

[Analog_Fan]

hi there,
For experimenting i wanna build a PC powered audio mixer and sourcing the 12V power from the ATX psu in from the pc.

What's your thoughs about this schematic?
I have done in the past for CV mixing in the modular synthesizer and worked perfect, but AC coupling was not needed.

 

[Tubetec]

Powering audio devices from a desktop PC isnt a good idea at all , you'll almost certain have noise issues in the analog signal path .
A small standalone mixer with USB connection to the PC will do exactly what you want .

 

[Analog_Fan]

Powering audio devices from a desktop PC isnt a good idea at all , you'll almost certain have noise issues in the analog signal path .
A small standalone mixer with USB connection to the PC will do exactly what you want .

Well Thnx, i was thinking about adding a USB B connector, but i though the ATX psu would do the same job.
Should i opt for the USB B type connector?

 

[Analog_Fan]

OPA1642 or OPA1678?
For witch should i decide?

OPA1642 seems to be very power efficient.

 

[ruffrecords]

I would advise you to change all the 1K pots and resistors to 10K.


Cheers

Ian

 

[Piedwagtail]

Check ebay for '

5V TO 12V -12V +12V Mini DC-DC converter board'​

for your experiment.
Agree with Tubetec - noise issues will need to be solved.

R

 

[Analog_Fan]

I would advise you to change all the 1K pots and resistors to 10K.


Cheers

Ian

the values where Falstad standard, the actual values of the resistor in the circuit is 10K, i only have 100K(A) dual gang pots.

 

[Analog_Fan]

Check ebay for '

5V TO 12V -12V +12V Mini DC-DC converter board'​

for your experiment.
Agree with Tubetec - noise issues will need to be solved.

R

5V should be enough, splitting them, results in +2.5V/-2.5V witch leaves me about 1V headroom over line audio.
It works in phones, co players, etc.

 

[Piedwagtail]

?
USB 5V > tacked on converter board gives you +12 and -12 for your opamps.

The one I use has all the ins/outs in a row such that you can solder in a pin array and have it as a plug in on your pcb.


R

 

[ccaudle]

results in +2.5V/-2.5V witch leaves me about 1V headroom over line audio.
It works in phones, co players, etc.

CD standard is 2V RMS maximum output, which is just a little more than +2.8V/-2.8V.

 

[Analog_Fan]

CD standard is 2V RMS maximum output, which is just a little more than +2.8V/-2.8V.

You made me scope the output of my pc.
I played a track on my pc on full loudness with the scope on 1V/div and i measured 2 V.p.p.
I have "faders" on the inputs it should work, it doesn't have to be a hi-end thing, but just a thing that allows me to hear audio experiments made by a MCU along the pc audio output.

 

[ruffrecords]

the values where Falstad standard, the actual values of the resistor in the circuit is 10K, i only have 100K(A) dual gang pots.

In which case I would advise you to change all the resistor values to 100K. SInce this is a non-critical application a little extra noise will not be an issue.

Cheers

Ian

 

[Analog_Fan]

Powering audio devices from a desktop PC isnt a good idea at all , you'll almost certain have noise issues in the analog signal path .
A small standalone mixer with USB connection to the PC will do exactly what you want .

What if i put a 78L10 voltage regulator on the 12 V (Molex) connector?
Than use 100nF and 100uF capacitors after the voltage reg. to filter the noise from the ATX psu?
Would that work?

 

[gyraf]

..10nF is too small for smoothing your virtual-ground reference - better make that at least 10uF..

 

[Analog_Fan]

..10nF is too small for smoothing your virtual-ground reference - better make that at least 10uF..

Well, i saw Don Buchla do it on the Buchla 259 complex wave generator and copied the value from that "ever" since.
But many thnx for the suggestion.

 

[Analog_Fan]

i ordered some Kemet SU10VD 80 uH common shokes witch is advised by Mean Well in their datasheet to some DC-DC converters.



Now I'm thinking on including their EMC circuit in my pc mixer schematic, witch I'm finalizing.


to witch i solder a 4 pin molex connector to power the mixer (12V), i have the pc side panel never mounted, right or wrong?

 

[soulsonic]

You don't need the extra driver transistors for the virtual ground, a single TL071 can source plenty of current for this circuit.

 

[Analog_Fan]

You don't need the extra driver transistors for the virtual ground, a single TL071 can source plenty of current for this circuit.

The circuit contains 5 quad and 1 dual opamp, i got a pcb witch has 3 microchip dual rail 2 rail opamps (mcp6002) powered like this and the opamp barely provides enough power as long as i reduce the amplitude with a trimmer to less than line level, the waves are passing through, scaling back up again at the end of the circuit, all is well.
if i increase the audio level, sound turns in to noise.

When i started the circuit, i wanted 4 channels, but i reduced it to 2 with 2 outputs and 2 "neutrik" balanced outputs.

The 2 inputs have 2 pole Sallen Key filters on 48 kHz each, that alone is 2 quad opamps.



the transistors should give me 100 ma, rather than some 25.

 

[soulsonic]

If you used higher voltage devices, your current requirements would be very much lower (only 6v absolute maximum for an MCP6002.) I don't see any practical reason to be using low voltage opamps for this. You're needlessly cutting your headroom to a fraction of what it could be, which also wrecks signal-to-noise.

I also don't understand the purpose of an active 48kHz filter on the inputs. These frequecies are already filtered by whatever DAC the computer is connected to, be it an interface, sound card, whatever. It's also very easy to make sure any frequencies above [whatever you choose] aren't amplified in the summing stage.

 

[Analog_Fan]

If you used higher voltage devices, your current requirements would be very much lower (only 6v absolute maximum for an MCP6002.) I don't see any practical reason to be using low voltage opamps for this. You're needlessly cutting your headroom to a fraction of what it could be, which also wrecks signal-to-noise.

when i bought the mcp6002, i bought them for being rail 2 rail but skipped the part of just being 6 volt in the datasheet, rail 2 rail are somewhat uncommon in "local" electronics shops, i got them from mouser.
I wasn't aware of "standard" rail 2 rail opamps, like TL07x is.
they do actually run on +/-4V in my circuit and still work till this day.

This Imsa guy on YouTube told me a idle, not connected to anything tl-072 consumes 2 ma.

I also don't understand the purpose of an active 48kHz filter on the inputs. These frequecies are already filtered by whatever DAC the computer is connected to, be it an interface, sound card, whatever. It's also very easy to make sure any frequencies above [whatever you choose] aren't amplified in the summing stage.

I'm making this mixer for breadboard audio experiments while still being able to listen to the pc (onboard audio).
Someone mentioned it would be wise to protect the summing core in some online article.

When i disconnect the jack/cable from the cheap speakers from the pc mainboard, they are completely silent, but plugged there is some humming. maybe this will help
: )

Now i after reading a datasheet from Mean Well (NSD15-D-spec.pdf) DC-DC converters i also decided to add a EMC filter. to get the best possible hobby mixer. i know very little about common shokes, did watch some YouTube video's on it though (but don't own a >10.000€ scope to verify) and ordered the value of inductance Mean Well mentioned. maybe it will "shoke" the possible noise coming form the pc psu where it's gonna be connected when needed.

a few weeks ago i found this page and also decided to add 2 XLR outputs to connect the akai mpc x and headphone (+6 opamps). http://www.all-electric.com/schematic/mix-eq.htm