3 channel passive mixer going crazy!

[fatecasino]

well, I have a small box with 3 different preamps. Each one has a unique particular sound. Channel 3 has a delay chip integrated in its circuit.
I have applied the simplest passive mixer idea:



There's something though that makes me go crazy, channel 1 and channel 2 have both a faint distorted delay sound!!

I have tried to increase the resistor/pot of channel 3 but I ALWAYS get some distorted low volume echoes.
The only way to get rid off the delay/echo sounds in channel 1 and 2 was to increase the channel 3 resistor to 1M or more. Then there would be no more echoes but no channel 3 as well. Its sound would be very low volume and poor quality.
I am obviously missing out some theoretical elements.
Do you have any suggestions?

PS: I tried an active mixer circuit, but I didn't like it for various reasons (power consumption, extra noise, etc)

[Jdansti]

I don't have an answer, but would like to know too. Maybe you invented a new effect!   

[PBE6]

Some useful info on signal mixing:

http://sound.westhost.com/articles/audio-mixing.htm

[PBE6]

It's surprising that the active mixer circuit you tried was noisier than a passive mixer. I've used the active mixing circuit shown in Figure 4 of the article a few times with no problems or excess noise concerns, maybe give it a try? One suggestion, try adding a buffer to each signal before the mixer circuitry. It may not always be necessary, but it will help prevent surprises.

[fatecasino]

thank you very much for your quick response.
i'll give active mixing a try and reply soon!
Anyway, I am really curious how the signal insists getting into channel 3. Through ground??

[fatecasino]

there's one thing I don't understand in Fig.4

do I need bipolar power supply or just give 9v to pin 8 of tl072?
(i saw several tl072 schematics on the internet and I am a bit confused)

[PBE6]

No you don't need a bipolar supply. If using a single 9V battery, connect pin 8 to 9V+ and pin 4 to ground as always. Instead of connecting the non-inverting (+) input to ground as shown in the diagram, connect it directly to Vref = 4.5V+ instead. This will bias the opamp mixer properly.

[samhay]

No you don't need a bipolar supply. If using a single 9V battery, connect pin 8 to 9V+ and pin 4 to ground as always. Instead of connecting the non-inverting (+) input to ground as shown in the diagram, connect it directly to Vref = 4.5V+ instead. This will bias the opamp mixer properly.

and add a fairly large (1u-ish) capacitor between the 10k resistors and the inverting op-amp input.

[PBE6]

Wait, don't you mean a cap from Vref to ground?

[fatecasino]

i used this voltage divider

everything works but all 3 channels sound always distorterd/overdriven (great quality though), even if i have turned down most of the volumes
What did i do wrong with the tl 072?

[PBE6]

That voltage divider is correct. Can you post a schematic of your circuit as-built? The signals should not be overdriving the opamp.

What kind of signals are you feeding in? I assumed they were unaffected guitar signals. If that is the case, I would recommend adding buffers on the front end of each signal. If you have a Boss stompbox lying around you can use that buffer as a test to see if the signal gets cleaned up.

[fatecasino]

I used this kind of preamp:
http://construyasuvideorockola.com/proy_reverb.php
In passive mixing I didn't have such problems. I 'll give it a try with a boss
Should I try changing the values] of the resistor R4?

[PBE6]

That's a microphone preamp. Are you plugging microphones into it? Or something else, like guitars or keyboards?

First thing I would do would be to test the preamp output on its own. If that's fine, try playing your guitar through the Boss pedal into the TL072 mixer. If that works, then there's a problem where the two pieces come together. Could be the output impedance of the mic preamp, could be the level it's sending, or something else.

See if you can isolate the problem and we can try to provide more help as necessary.

[fatecasino]

yes I am plugging an XLR mic there, to mix it with my guitar playing in the other channels (1 or 2)
The best this is to isolate and debug the whole thing.

Facts:
*I haven't tried to add the Boss pedal, but even with tl072 distorted sound, I still get echoes from channel 3 while I am playing in channel 1 or 2.
*These delay echoes are strangely distorted (in both active and passive mixing) even if the original sound is clean
*In both passive & active mixer I have a 100k trimpot at the output of the mic preamp and the only way to get rid of the echoes is to actually kill the output of the mic preamp by giving the trimpot its maximum value

Questions:
*Is it possible to have that signal leakage via power supply?
*How do I test the output impedance of the mic/delay preamp? can you please suggest a link to improve my understanding on input/output resistor?

[PBE6]

First things first:

1. Does your mic preamp work by itself? Plug it into something other than the mixer and see. Is it distorted? Is it clean?

2. What kind of output is coming from the back of the mic preamp, XLR or 1/4"? If it's XLR, what kind of circuitry did you use to connect it to the mixer?

3. Trace your mixer circuit and generate an as-built schematic. Don't skip this step, often times if there is a wrong connection you'll find it here. Show it to us, right now we're just guessing.

4. Does the mixer work with just 1 guitar being sent to it? Does it distort with the channel volume on full? Can you kill the signal completely when the channel volume is off? How about with 2 guitars?

5. When you plug in the mic preamp, does anything happen to the other signals, other than the echo? Does the connection cause the signal to distort?

Once you've done those tests and provided a schematic, you'll be in a better position to track down the problem.

With regard to your questions, I believe it is possible for AC to leak through a power supply as it presents a low impedance path, but there are people here better able to answer that question. As for the impedance, just google "measuring output impedance". Here's one example that came up:

http://www.zen22142.zen.co.uk/Theory/inzoz.htm

[PRR]

> 3 different preamps
> simplest passive mixer idea

You say 3 inputs but show a 2 input mixer?

> channel 1 and channel 2 have both a faint distorted delay sound!!

Bad pots or bad grounding.

Adding active stuff won't fix that.

Check that your pot wipers go to ZERO Ohms, and that all ground paths are ZERO Ohms. In a 10K mixer this means well under 10 Ohms, preferably nearer 1 Ohm, though good real 10K pots may not go that low.

> this kind of preamp: http://construyasuvideorockola.com/proy_reverb.php

The diagram shows a technical problem.

http://construyasuvideorockola.com/imagenes/impresos/diagrama.jpg

There is no DC bias to the + inputs of the TL072 amps. Ideally the amp does not know what DC level to be at. It "may work" depending on stray PCB, chip, and cap leakages, or if you play VERY LOUD, but is not reliable.

[fatecasino]

First things first:

1. Does your mic preamp work by itself? Plug it into something other than the mixer and see. Is it distorted? Is it clean?

2. What kind of output is coming from the back of the mic preamp, XLR or 1/4"? If it's XLR, what kind of circuitry did you use to connect it to the mixer?

3. Trace your mixer circuit and generate an as-built schematic. Don't skip this step, often times if there is a wrong connection you'll find it here. Show it to us, right now we're just guessing.

4. Does the mixer work with just 1 guitar being sent to it? Does it distort with the channel volume on full? Can you kill the signal completely when the channel volume is off? How about with 2 guitars?

5. When you plug in the mic preamp, does anything happen to the other signals, other than the echo? Does the connection cause the signal to distort?

Once you've done those tests and provided a schematic, you'll be in a better position to track down the problem.

With regard to your questions, I believe it is possible for AC to leak through a power supply as it presents a low impedance path, but there are people here better able to answer that question. As for the impedance, just google "measuring output impedance". Here's one example that came up:

http://www.zen22142.zen.co.uk/Theory/inzoz.htm

Hi!
i spent some time doing some tests..no results yet


1. Mic preamp works very well, clean and loud. I tried it directly to another guitar amp
2.The original preamp input is a mono jack. I have connected an XLR following this circuitry

3. At this point I cant provide the schematic, I 'll try some program to generate it. It is exactly as shown in the initial post, instead of 2 inputs I used 3 inputs
4. All 3 channels play really well. Seperately each of them + simultaneously. The sound of each of the channels is clean and loud, but when I turn on the echo volume in channel 3, channel 1 and 2 obtain echoes which actually can be controlled by the pots of channel 3 (echo time, echo vol).
5. No, everything works fine when I plug in the XLR or when all 3 channels are being used.

The point is that indeed "debugging" from distance is really hard to do. I tried unscrewing the channel 3 pots from the box and many other tricks, but somehow signal from channels 1 and 2 ALWAYS sneak into channel 3 preamp.
The only way to get rid off the distorted echoes in channel 1 and 2 is to actually cut the wire that connects channel 3 with the passive/active mixer.

[fatecasino]

> 3 different preamps
> simplest passive mixer idea

You say 3 inputs but show a 2 input mixer?

> channel 1 and channel 2 have both a faint distorted delay sound!!

Bad pots or bad grounding.

Adding active stuff won't fix that.

Check that your pot wipers go to ZERO Ohms, and that all ground paths are ZERO Ohms. In a 10K mixer this means well under 10 Ohms, preferably nearer 1 Ohm, though good real 10K pots may not go that low.

> this kind of preamp: http://construyasuvideorockola.com/proy_reverb.php

The diagram shows a technical problem.

http://construyasuvideorockola.com/imagenes/impresos/diagrama.jpg

There is no DC bias to the + inputs of the TL072 amps. Ideally the amp does not know what DC level to be at. It "may work" depending on stray PCB, chip, and cap leakages, or if you play VERY LOUD, but is not reliable.

a. I cloned the schematic of the passive mixer and added one more channel. All three channels have exactly the same input
b. The pots are a bit old/used, but I tried unscrewing from the box and nothing changes. The signal from channel 1 and 2  leaks into channel 3 and I get low volume, distorted echoes which can actually be controlled by preamp 3 pots (echo time, echo vol, volume, bass, treble)
c. I am not that good in electronics, I realize that there must be some design errors, but at any stage this preamp plays clear and loud.

It's true that it is almost impossible to get "debugging" help from distance. The only thing I would like to know is some debugging directions. Supposing you have 3 preamps that get connected to a passive/active and then to tiny giant. Which are the possible ways that the signal of 1 and 2 can sneak into channel 3? In passive mixer, the only way to stop the echoes sounds is to increase the input resistance that much that you actually hear almost nothing from channel 3 itself.

[fatecasino]

***************UPDATE**********
while posting a new test came across my mind. The fact that the echoes are low volume distorted drove me to the idea to test the output of the passive mixer.
I soldered a jack female plug parallely to the output of the passive mixer. Something like a LINE out. I connected this line out to a guitar amp and guess....the echos DISAPPEARED
which means that the signal is getting in channel 3 AFTER the preamp/mixing stage.Somewhere around the tiny giant amp (input/output to speaker). Tomorrow I'll do some further testing.
One question,
If I want to keep this LINE OUT, should i leave it as it is or I should add some output resistor?