Electronics > Projects, Designs, and Technical Stuff

Summing a stereo signal

(1/3) > >>

richlooker:
[ Specified attachment is not available ]I am trying to find the best/correct solution to a well-known problem for which there exist several more or less working solutions, but none that are proven by theory.

There is a range of small, affordable hardware synthesizers - the Korg Volca range - exemplified here by the Volca Bass:



With a built-in sequencer, ribbon touch keyboard and a headphone output, these can be used standalone, but the real fun starts when you combine several of these units (9 different ones exist,) and use external MIDI controllers (eg. sequencers, keyboards and drum pads.) This requires that you hook them up to a mixer, and therein lies the problem. 6 of these 9 Volca synths produce a mono signal, but they all use the same stereo headphone amplifier chip (TI TPA6111A2) and a 1/8" stereo TRS socket (which is the only available signal output.) Here is the schematic of the headphone circuit:



Mixer channels are a limited (and expensive) resource, so using two channels for what is essentially a mono signal is not an attractive option. As most mixers have a balanced mono 1/4" TRS inputs, these are some of the different approaches/attempts to connect these synths to mixers I have seen;


* 1/8" TRS to 1/4" TRS - seems logical on the face of it, but this means the R and L channels are routed to the positive and negative rail of the balanced input, so what you get is the difference between L and R. Probably because of component tolerances, there is some difference, but the effective signal is very weak, and mixer gain controls must be turned way up before you can hear anything
* 1/8" TS to 1/4" TS - also logical as we have a mono signal, but this shorts one of the output channels to ground, resulting in attenuation, compression and distortion of the other channel
* 1/8" TRS to 1/4" TS with only one channel connected - works, but you lose some signal level (6dB?) compared to using both channels and a stereo input and compared to the true stereo Volcas
* 1/8" TRS to 1/4" TS with a passive summing resistor network - my hypothesis is that this is the optimal solution, if the resistor values are chosen correctly
I have found this generic solution, which does not account for output and input impedances:



I want to establish; how do I passively sum the two channels, without straining the Volca output, while getting the highest possible signal level? (The output impedance of the headphone amp chip is not stated in the data sheet. The input impedance of the mixer is 10K between each of +/- and ground.)

Cheers, Richard

moffy:
Since you have 2 outputs which are the same mono signal, you only need to use one of them, you don't need to mix them. :)

bateau020:
How about making the output differential?
Insert an inverting opamp circuit before or after C114, or (not better but a lot easier) connect the left side of C114 to IC16A pin 1? For the latter, IC16A has no gain (well, it has -1), so that should work.
Then you have a balanced signal between T and R, and you gain 3db compared to if you would connect only 1 side (T to S or R to S), and it would be as the balanced mono 1/4" TRS destination inputs expect.

edit:
By the way, seen the input impedance, I would expect no difference in signal level between these 2:

--- Quote from: richlooker on June 25, 2020, 10:47:53 am ---
* 1/8" TRS to 1/4" TS with only one channel connected - works, but you lose some signal level (6dB?) compared to using both channels and a stereo input and compared to the true stereo Volcas
* 1/8" TRS to 1/4" TS with a passive summing resistor network - my hypothesis is that this is the optimal solution, if the resistor values are chosen correctly
--- End quote ---

richlooker:

--- Quote from: moffy on June 25, 2020, 11:30:20 am ---Since you have 2 outputs which are the same mono signal, you only need to use one of them, you don't need to mix them. :)

--- End quote ---

I know, but I want the signal level to match that of the true stereo Volcas, such that the signals are equally loud when the mixer faders line up.

Cheers, Richard

richlooker:

--- Quote from: bateau020 on June 25, 2020, 11:36:33 am ---How about making the output differential?

--- End quote ---

I could do that, but then the signal would sound really strange when I use the output for headphones. Plus, I don't want to mod the synth unless I have to.

Cheers, Richard

Navigation

[0] Message Index

[#] Next page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod