Attenuation issue with mini mono passive mixer?

[Jasonmatthew911]

Hey, I just built a 4 input to 1 output passive mixer using 10K pots for each input to control them separately, as well as 10K resistors for isolation...It all works just fine, but the signal of my guitar,  little synth, and oscillators I have get attenuated, and I have to raise the volume on my amp a lot more....Should I be using different value pots and resistors for this application???...Is there anything I can do without having to add a pre-amp, so that my signals don't get attenuated?....Any help will be appreciated...

[R.G.]

I just built a 4 input to 1 output passive mixer

Passive always implies mixing loss of some amount. An active gain stage will be needed to bring the mixed inputs back up to unity.  This can be in your amp or in an active stage you put in the mixer, or in bigger outputs from the things that go into the inputs. It's the nature of the beast.

[Jasonmatthew911]

I just built a 4 input to 1 output passive mixer

Passive always implies mixing loss of some amount. An active gain stage will be needed to bring the mixed inputs back up to unity.  This can be in your amp or in an active stage you put in the mixer, or in bigger outputs from the things that go into the inputs. It's the nature of the beast.

Can you give me an idea of the simplest active gain stage you can think of?...I just didn't want to have to do too much extra wiring...I'd like to keep it as simple as possible since all this is going mono into my tube amp.

[Jasonmatthew911]

I just built a 4 input to 1 output passive mixer

Passive always implies mixing loss of some amount. An active gain stage will be needed to bring the mixed inputs back up to unity.  This can be in your amp or in an active stage you put in the mixer, or in bigger outputs from the things that go into the inputs. It's the nature of the beast.

Would a simple JFET Clean Boost cicuit, with J201 after my 10K isolation resistors do the trick to boost my Volume?.......If not that, what about a simple LM386 gain stage?...I thought that the LM386 might dirty my signals a bit...What do you think about using a simple J201 Clean Boost circuit Vs. an LM386 circuit for a gain stage....Also, my 4 inputs can be fed into the same gain stage right?

[R.G.]

Would a simple JFET Clean Boost cicuit, with J201 after my 10K isolation resistors do the trick to boost my Volume?

JFET amplifiers are not as simple as they seem, don't give much gain (although maybe enough for this one) and are hardly ever clean.

If not that, what about a simple LM386 gain stage?...I thought that the LM386 might dirty my signals a bit...

The LM386 is pretty clean as long as it is not driven outside it's power supply limits. It is generally used at higher gains for distortions. As it comes out of the tube,
the LM386 has a gain of 20. With a 9Vdc supply, that means the biggest signal you could reasonably put into its input to have it have a clean output is something like 400mV. So you'll need to ensure the incoming signals are *small* enough if you want it to be clean. The inputs both have 50K resistors to ground inside the chip, so you could mix with, say, 220K isolation resistors. That would cut the signal by 50K/(220K +50K) = 0.185, which would then be amplified by 20 to an overall gain of 3.7 times. Might work OK. Each output would have a gain from input jack to output jack of 0 to 3.7x.

What do you think about using a simple J201 Clean Boost circuit Vs. an LM386 circuit for a gain stage....Also, my 4 inputs can be fed into the same gain stage right?

A "clean boost", capital letters or not, is rare with single JFETs. The LM386 is a better bet, I think. Yes, with the LM386, you can use four 10K pots for input pads, each wiper going through a series 220K resistor to one of the LM386 inputs, all four to the same input. The unused input should be grounded. The output will have a DC level of half the total power supply on it, so you must use a capacitor in series with the output to eliminate this DC level to avoid upsetting whatever comes after it.

So far, so good.

However - a guitar going right into this thing will be severely loaded by the 10K input level pot. Guitars don't like driving resistances below about 100K and most people don't like the treble losses from loads much below 1M. The output of a pedal driven by the guitar is probably OK, but if you are going right into this from a guitar, it really needs a buffer. The synth has an output level of -??? some synths have 10V signals out. I don't know the level of your oscillators.

[Jasonmatthew911]

Would a simple JFET Clean Boost cicuit, with J201 after my 10K isolation resistors do the trick to boost my Volume?

JFET amplifiers are not as simple as they seem, don't give much gain (although maybe enough for this one) and are hardly ever clean.

If not that, what about a simple LM386 gain stage?...I thought that the LM386 might dirty my signals a bit...

The LM386 is pretty clean as long as it is not driven outside it's power supply limits. It is generally used at higher gains for distortions. As it comes out of the tube,
the LM386 has a gain of 20. With a 9Vdc supply, that means the biggest signal you could reasonably put into its input to have it have a clean output is something like 400mV. So you'll need to ensure the incoming signals are *small* enough if you want it to be clean. The inputs both have 50K resistors to ground inside the chip, so you could mix with, say, 220K isolation resistors. That would cut the signal by 50K/(220K +50K) = 0.185, which would then be amplified by 20 to an overall gain of 3.7 times. Might work OK. Each output would have a gain from input jack to output jack of 0 to 3.7x.

What do you think about using a simple J201 Clean Boost circuit Vs. an LM386 circuit for a gain stage....Also, my 4 inputs can be fed into the same gain stage right?

A "clean boost", capital letters or not, is rare with single JFETs. The LM386 is a better bet, I think. Yes, with the LM386, you can use four 10K pots for input pads, each wiper going through a series 220K resistor to one of the LM386 inputs, all four to the same input. The unused input should be grounded. The output will have a DC level of half the total power supply on it, so you must use a capacitor in series with the output to eliminate this DC level to avoid upsetting whatever comes after it.

So far, so good.

However - a guitar going right into this thing will be severely loaded by the 10K input level pot. Guitars don't like driving resistances below about 100K and most people don't like the treble losses from loads much below 1M. The output of a pedal driven by the guitar is probably OK, but if you are going right into this from a guitar, it really needs a buffer. The synth has an output level of -??? some synths have 10V signals out. I don't know the level of your oscillators.

Ok, I'm building the LM386 circuit...On the output of LM386 (Pin 5), should I have the 10 ohm resistor & .047uf capacitor going to GND and then a large electrolytic Cap, between 100uf - 470uf going to the output jack?...I know this is done when using the LM386 as a headphone amp or mini amp to drive a speaker....In this case, should I apply this as well, or would it be enough with a 1uf capacitor going from pin 5 directly to output???

[Perrow]

Search for "4 channel mini mixer", it's exactly what you have now, but with an opamp added. I do have the schematic but I'm on my phone (and on the bus) right now, can post it later.

[Jasonmatthew911]

Would a simple JFET Clean Boost cicuit, with J201 after my 10K isolation resistors do the trick to boost my Volume?

JFET amplifiers are not as simple as they seem, don't give much gain (although maybe enough for this one) and are hardly ever clean.

If not that, what about a simple LM386 gain stage?...I thought that the LM386 might dirty my signals a bit...

The LM386 is pretty clean as long as it is not driven outside it's power supply limits. It is generally used at higher gains for distortions. As it comes out of the tube,
the LM386 has a gain of 20. With a 9Vdc supply, that means the biggest signal you could reasonably put into its input to have it have a clean output is something like 400mV. So you'll need to ensure the incoming signals are *small* enough if you want it to be clean. The inputs both have 50K resistors to ground inside the chip, so you could mix with, say, 220K isolation resistors. That would cut the signal by 50K/(220K +50K) = 0.185, which would then be amplified by 20 to an overall gain of 3.7 times. Might work OK. Each output would have a gain from input jack to output jack of 0 to 3.7x.

What do you think about using a simple J201 Clean Boost circuit Vs. an LM386 circuit for a gain stage....Also, my 4 inputs can be fed into the same gain stage right?

A "clean boost", capital letters or not, is rare with single JFETs. The LM386 is a better bet, I think. Yes, with the LM386, you can use four 10K pots for input pads, each wiper going through a series 220K resistor to one of the LM386 inputs, all four to the same input. The unused input should be grounded. The output will have a DC level of half the total power supply on it, so you must use a capacitor in series with the output to eliminate this DC level to avoid upsetting whatever comes after it.

So far, so good.

However - a guitar going right into this thing will be severely loaded by the 10K input level pot. Guitars don't like driving resistances below about 100K and most people don't like the treble losses from loads much below 1M. The output of a pedal driven by the guitar is probably OK, but if you are going right into this from a guitar, it really needs a buffer. The synth has an output level of -??? some synths have 10V signals out. I don't know the level of your oscillators.

I added the LM386 circuit with .1uf input cap, and 220uf output cap...It works good, but of course when I put the volume full it starts distorting very lightly...I changed the pot for the guitar input to 100K (Audio) and the isolation resistor to 220K, it basically has a tiny bit more highs than the 10K with 10K combination....Any way to be able to have my volumes up all the way without cracking up you think, or not with this circuit?....

I saw that 4 channel mini amp schematic with TL072 op-amp I think....I guess an opamp would be an even better option for a much cleaner gain stage....Thanks a lot guys...

[R.G.]

The LM386 is an opamp, albeit one that's optimized for driving a speaker.

Any opamp in this circuit will be limited as to how much clean output it has. The limit is the power supply. If you're running from 9V, a different opamp from the LM386 will have similar limitations, although they may differ slightly in exactly where they start to limit. An opamp mixer with a +/-15V supply will offer much larger clean output range.

However, there is a concept lurking here. Having the volumes all the way up is a visual and conceptual thing, much like Nigel Tufnel's "but these go to 11" comment. Yes, there is a simple way to have your volumes all the way up without distortion. That is to put a resistor in series with the input signal so that the combination of the input resistor and input volume pot never can provide a signal to the LM386 that will drive it to distortion. It's exactly the same thing as turning the volume control down a little till it doesn't distort, but it provides the visual effect of letting you turn the knob all the way up.

The output signal is limited by how many volts of power supply the LM386 - or whatever other opamp you use - has to work with. By the way, a JFET used as an amplifier would have much less clean signal range with the same power supply as an LM386, as an aside. You have two options with an active circuit: either provide a big enough power supply so nothing can ever distort, or limit the inputs so it doesn't distort with the power supply it has.

[Jasonmatthew911]

The LM386 is an opamp, albeit one that's optimized for driving a speaker.

Any opamp in this circuit will be limited as to how much clean output it has. The limit is the power supply. If you're running from 9V, a different opamp from the LM386 will have similar limitations, although they may differ slightly in exactly where they start to limit. An opamp mixer with a +/-15V supply will offer much larger clean output range.

However, there is a concept lurking here. Having the volumes all the way up is a visual and conceptual thing, much like Nigel Tufnel's "but these go to 11" comment. Yes, there is a simple way to have your volumes all the way up without distortion. That is to put a resistor in series with the input signal so that the combination of the input resistor and input volume pot never can provide a signal to the LM386 that will drive it to distortion. It's exactly the same thing as turning the volume control down a little till it doesn't distort, but it provides the visual effect of letting you turn the knob all the way up.

The output signal is limited by how many volts of power supply the LM386 - or whatever other opamp you use - has to work with. By the way, a JFET used as an amplifier would have much less clean signal range with the same power supply as an LM386, as an aside. You have two options with an active circuit: either provide a big enough power supply so nothing can ever distort, or limit the inputs so it doesn't distort with the power supply it has.

Right on....So do you think if I used an LM386-4, add a clip on heat sink, an 18V zener diode and run it on 18V would make a nice difference for more headroom?

[R.G.]

Yes.

A couple of observations:
- You probably won't need a heat sink. A power amp (which is what the LM386 is) doesn't dissipate much power at all unless it's driving a load. This one won't be, so the power dissipated inside may be quite small.
- Sometimes power amps get funny with little or no loading. This might make you need a resistor of 1K to several K on the output. Or not.
- The only real advantage of the LM386 over a more generic opamp is that its inputs work fine at 0V DC. This eliminates two or three resistors and one capacitor. That may or may not be a big deal for you.
- A more generic opamp will withstand up to about 36V total in power supply voltage. The LM386 is less, as you have insightfully noted in thinking about the LM386-4 for 18V. But if 18V is enough - and it probably is - then there's not much difference.

[DavenPaget]

If you want 18V you can use a ICL7660ACPAZ

Use 100uf's instead of 10uf's