DIY rotary DJ mixer

[efinque]

Yes, basically, if you do passive summing, the degree of isolation between channels is determined by the ratio of the potentiometer value by the value of the summing resistor. So if you use a relatively low value potentiometer, eg. 10k, and a relativelt high summing resistor value.. eg, 100-200k or higher.. you will have very little interaction.

You mean crosstalk? I was fully aware of it when starting the build. My first project had 10kOhm summing resistors but I switched them to 220R because it was too quiet, didn't help much though.. it even had a master preamp.

The problem being that if you put let's say a typical load on the output.. let's say 47K, you will experience a noticeable drop in level. If you put a low impedance load such as headphones you'll drop the output to almost nothing. The solution is to put an amplifier with a high impedance input and low impedance output.. and then from there you connect your various loads. With an output impedance of let's say 100ohms from your buffer, and let's say your headphone amp has an input impedance much higher than that.. let's say 47k as the absolute minimum.. then you can switch the headphone amp in and out without it being noticed.

Here I was thinking whether an output transformer would fix the impedance matching/clicks&pops.. too bad the stereo line preamp modules I used are no longer in production.

[efinque]

Also, could this sagging occur with extension cords as well?

Is it measurable with a DMM using program material (ie. line level music) and turning the knobs/plugging loads?

EDIT : tested my smartphone output with a DMM and it showed 0.1V peak.. no readings when I plugged the RCA into the mixer (and measured AC from the master outs with/without headphones).. I'd probably need an oscilloscope.

[light-o-matic]

Also, could this sagging occur with extension cords as well?

Is it measurable with a DMM using program material (ie. line level music) and turning the knobs/plugging loads?

EDIT : tested my smartphone output with a DMM and it showed 0.1V peak.. no readings when I plugged the RCA into the mixer (and measured AC from the master outs with/without headphones).. I'd probably need an oscilloscope.

No, if you want to measure the effects you should use a steady tone, I use two android apps.. one is a simple tone generator called PA Tone, the other is called Audio Tool and it has an RTA as well as a tone generator. Then you can measure with your DMM or scope, or really anything with a VU meter will work, but you need something with enough sensitivity to really show the levels you are working with.


But basically yea

[light-o-matic]

Extension cords no, they won't have any effect in thos situation.

[efinque]

No, if you want to measure the effects you should use a steady tone, I use two android apps.. one is a simple tone generator called PA Tone, the other is called Audio Tool and it has an RTA as well as a tone generator. Then you can measure with your DMM or scope, or really anything with a VU meter will work, but you need something with enough sensitivity to really show the levels you are working with.


But basically yea

Yes.. I've calibrated a sound system once using white noise so I know program material shouldn't be used for that. Sine waves, white noise, pink noise..

I have a large segment LED VU meter somewhere too but I can't be bothered to use it or install an app.. I was thinking of recording to another smartphone while twiddling the knobs and plugging/unplugging headphones but I'm too lazy.

One thing I noticed the channel cue (pre-fader) gets louder when the master volume is turned down.. is there a reason for this?

[light-o-matic]

One thing I noticed the channel cue (pre-fader) gets louder when the master volume is turned down.. is there a reason for this?

It really depends upon your circuit. Do you have a schematic?
But yea if you put a varying load on the output and your summing resistors are low values (which I think you said they were).. then absolutely you will have interactions. The way to prevent this is an amp stage between the summing and the master.

[efinque]

It really depends upon your circuit. Do you have a schematic?

Here's a schematic I sketched on my smartphone :

But yea if you put a varying load on the output and your summing resistors are low values (which I think you said they were).. then absolutely you will have interactions. The way to prevent this is an amp stage between the summing and the master.

Yeah I need to source a preamp module/kit or build my own.

EDIT : I'm becoming more and more convinced building my own is the way to go..

[light-o-matic]

Yea the way you have it, with the series potentiometers, is going to cause huge interactions, not to mention odd effects such as, when you have both channels turned up.. it won't matter which channel you have selected in cue.. you'll hear both. That's just not going to work well at all. If you actually use this mixer to mix a set and you record the result, you are going to hear a lot of weirdness.

I don't have time to draw a schematic right now, but if you go to this site, Elliot Sound Products, he has got a ton of useful circuits there, with schematics. You can buy some of his circuit boards to make the kits.. but even if you don't, he's giving away the circuit diagrams and you can build them yourself or just learn from them absolutely for free.

https://sound-au.com/

[efinque]

Yea the way you have it, with the series potentiometers, is going to cause huge interactions, not to mention odd effects such as, when you have both channels turned up.. it won't matter which channel you have selected in cue.. you'll hear both. That's just not going to work well at all. If you actually use this mixer to mix a set and you record the result, you are going to hear a lot of weirdness.

Yes that's crosstalk.. but the way it's intended to be used is that you cue with the channel vol down and listen to the main outputs. No separate monitor out so it's only good for home use I think.

About the outputs interacting there's no solution afaik, you just have to live with it (a booth out would make things even more complicated, and there's no space in the front panel)

But all in all I'm sort of content with the project and I think I built a solid mixer with my limited knowledge and tools.

I don't have time to draw a schematic right now, but if you go to this site, Elliot Sound Products, he has got a ton of useful circuits there, with schematics. You can buy some of his circuit boards to make the kits.. but even if you don't, he's giving away the circuit diagrams and you can build them yourself or just learn from them absolutely for free.

I'm familiar with ESP.. I've built one of his projects.

EDIT : regarding the mixer schematic; there's a volume(?) trimmer in the mic pre module which I completely forgot.. it's a prebuilt SMD module based around LM386 for which I have no schematics

EDIT 2 : regarding the outputs; could a DPDT relay be used so that when the pgm cue is on the main out goes directly through the relay (say, a 5-12V control voltage from the rotary switch) and when it's not in use it goes via a series resistor to mimic the headphones being in a parallel load? Aren't relays noisy (I know there are relay-based cue systems) and would it introduce clicks/pops in the output? Use solid-state relays instead to prevent arcing?

[light-o-matic]

Well.. if you're happy with your mixer that's all that matters I guess.

But it's not true that doing it better is beyond your tools or your ability. Not true at all.

The circuit for a simple audio mixer that actually works right is VERY simple and has not changed for nearly 100 years. It's just a potentiometer (for each channel), a resistor (for each channel) and a simple amplifier.. the amplifier can be as simple as 1 transistor, 1 resistor, 1 capacitor. That's ALL you need to make a mixer that works perfectly with none of the problems you will have with this mixer circuit.