Super simple passive headphone mixer

Setting up a headphone mix for the talent is boring. So why not give them simple tactile control of the mic level, and track level in the cans. Apollo has 2 HP outs with 80mv into 600ohms, more power than I've ever needed in the cans, so I figure I can afford passive.

I made up a box roughly following this circuit http://www.epanorama.net/circuits/linemixer.html but figuring separation is not that important, and to keep attenuation minimal, I changed resistors to 560R. It's not great though, the taper on the pots is super weird. Very little happens in the first 80% of the pot travel, it's all in the last %20. Thought about changing to linear pots, but I'm thinking this is an indication of an inappropriate circuit.

Anyone recommend a better one?
1. Mix 2 stereo 80mv sources
2. Passive. Theres plenty of juice coming out of the apollo.
2. Full seperation not needed. 12db or so difference between sources should be plenty.
3. Minimal attenuation (considering point 2)

  I agree the talent must have something to play with instead of  *%#*^& the tech, and if in the way it takes something out of our hands better.

  I'm not sure what you mean by separation... X-talk doesn't make any sense as it's a single mix. Independence between channel A and B maybe?

  I had a  time where I  was about to build something similar to that, but higher level. A few power amps (8 or 10, 25W to say something) driving a few passive mixers for each guy on the room. IIRC I was thinking in something like 500Ω linear pots and 100Ω resistors, thinking in lower Z headphones but maybe some load to help behave with higher Z ones. The load on the pots from the resistor and the HP makes a linear closer to logarithmic. 500Ω pot and 100Ω load makes 13dB mid point, a bigger pot could be used as well, maybe better, 1k pot and 100Ω load has 17dB at mid point.

  If your headphones are 600Ω it might be ok to use 560Ω resistors, but 10k log pots for sure it's not. You could go for very low resistance log pots or a bit higher linear. 500Ω log should work, or 5k or 10k linear.

  More details for future answers... Are the HP you are aiming to use actually 600Ω? Is the maximum level enough in your set up?

JS

Thanks for the reply.

Yes crosstalk is what I'm referring to. Most passive mixer designs I've seen use high resistor values requiring make up gain, but crosstalk isn't an issue here. If low resistance values mean some signal will feed through even with a pot at '0' position, that's ok.

'80mv into 600ohms' is straight from specs for the Apollo HP outs, but generally it'll have much lower impedance cans connected to it, HD280s and such (64ohm), or the talents favourite pair they happen to bring with them.

It's much easier to swap in 10K linear than to source 500ohm log.

Would 10K lin and 100ohm resistors do the trick?

A passive mixer will also drive the headphones from higher impedance than an active amp.

Not generally a high fidelity approach.

JR

  I have 2 sets of HD380pro and they are quite neat to work in studio, not much leakage, very clear sound.

  X-talk makes no sense for a single mix, there is no X-talk, is not that it doesn't matter, you have nowhere to leak to.

JohnRoberts said:

A passive mixer will also drive the headphones from higher impedance than an active amp.

Not generally a high fidelity approach.

JR

Click to expand...

  In several HP amps discussed here, limited output voltage and a series resistor end being the way to go to avoid popping (or pooping?) the headphones or the ears, that resistor already means not so low impedance, making a ULZ passive mixer (not Mackie VLZ, even lower) ends with quite a low Z to drive directly the cans, provided you have an amp that can drive properly the mixer.

  For 64Ω cans, 500Ω lin pots (easier to get than lower values), sub 100Ω summing resistors and maybe, just maybe, if level is high enough, a shunt to lower the impedance, could be selectable or internal jumper if the option seems convenient.

JS

> Very little happens in the first 80% of the pot travel, it's all in the last %20

That's what happens when the pot resistance is higher than the load resistance.

You said you changed the resistors to ~~600r, but do not say you changed the pots to ~~600r. 500, 1K, either will work, and 10+ times smoother than 10K with low-Z load.

PRR said:

> Very little happens in the first 80% of the pot travel, it's all in the last %20

That's what happens when the pot resistance is higher than the load resistance.

You said you changed the resistors to ~~600r, but do not say you changed the pots to ~~600r. 500, 1K, either will work, and 10+ times smoother than 10K with low-Z load.

Click to expand...

Ok great, that explains it. The dual gang pots I could get locally start at 10K and go up.... looks like time for another mouser order.