• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join free Log in
    Want zero display ads or expanded classifieds tools? Compare plans.

Passive Mixers

Sir Edward V

Sir Edward V

Not Actually Knighted... Yet!
Dec 11, 2006
3,062
5
Massachusetts
#1
edit: Topic changed, I'm now interested in making an active mixer.
Read on.

old original post:

simple question: How well do they work?

I know several people have used them here, I have heard stories about them not always working. Why wouldn't it work well?

Informed responses only please. :D

Also, how hard would it be to make an active mixer? I know passive mixers are easy, but an active one would be more complicated methinks
 
#2
Sir Edward V said:
simple question: How well do they work?
Click to expand...
Depends on the application. Suffice to say they have a propensity to cause undesired interaction between signals and circuits if not used with care.

Why wouldn't it work well?
Click to expand...
Because they're passive. They essentially provide no isolation between circuits attached to their inputs and outputs. This can mean that two circuits attached to two outputs can interact, or they might cause the circuit attached at the input to behave strangely. (Trying to stay away from engineering descriptions here...)

Also, how hard would it be to make an active mixer?
Click to expand...
Yes, it's more complicated, but, simple mixers are among the most simple op amp circuits out there. If you can solder and read a simple circuit diagram you should have no trouble making a simple summing mixer.
 
#3
niftydog said:
Depends on the application. Suffice to say they have a propensity to cause undesired interaction between signals and circuits if not used with care.


Because they're passive. They essentially provide no isolation between circuits attached to their inputs and outputs. This can mean that two circuits attached to two outputs can interact, or they might cause the circuit attached at the input to behave strangely. (Trying to stay away from engineering descriptions here...)


Yes, it's more complicated, but, simple mixers are among the most simple op amp circuits out there. If you can solder and read a simple circuit diagram you should have no trouble making a simple summing mixer.
Click to expand...

you can go into engineering descriptions if it can help, I am learned to some degree in this, I am done with a year of computer engineering at my college, and I am interested in these things and have investigated into them somewhat. I can probably grasp most of what you would say about it.

I made a booster pedal, I would imagine a mixer is similar to it. hell I even still have parts left over, including three dual channel op-amps, maybe I'll make an active one...
 
#4
You only need one dual-opamp. You put a resistor on the hot side of each input jack. Tie the ground side of the jacks to the ground (or vref) attached to the non-inverting input of the opamp. All the resistors are tied together into the inverting input of the opamp. The second opamp is used to invert the signal back to normal, you could leave it out if you can live with an inverted output. It is important that the summing opamp be inverting.

IIRC this is called a virtual ground summing opamp configuration.

From here, you can go wild. Passive or active gain controls on the inputs. Phase inversion switches (in case somebody else cheated and inverted an output ;) The inverting opamp can provide volume boost. And on and on.
 
#5
Plenty of info on google - summing amplifier is a good phrase. Personally I wouldn't bother about un-inverting the signal. (If you have phase issues in an unbalanced system a 180 degree phase shift is unlikely to solve them.)

A booster is likely very similar, save for a few extra resistors and a change to the gain.

To get a bit more in-depth... a passive mixer puts simple resistance between all inputs and the output. So, an input device like a passive bass pickup suddenly "sees" not just the input impedance of the next device in the signal path, but also the potentiometer resistances and the output impedance(s) of the other input sources - all in parallel. The result is a reduced input impedance for the pickup which results in a loss of high end.

Some devices (like simple fuzzes and boosters - with relatively poor buffering and isolation) can be sensitive to signals going into their outputs. Some people report feedback and even interaction between pedals. Futz with the tone pot on one pedal and it may cause another pedal to do something unexpected.

Furthermore, mix a well buffered pedal with a lot of gain with a poorly buffered pedal with not-so-much gain and you'll find the strong pedal will over-power the weak one and you'll find it hard to match the volume levels.

So yeah, passive is simple, cheap and needs no power, but you get what you pay for.
 
#7
seanm said:
You only need one dual-opamp. You put a resistor on the hot side of each input jack. Tie the ground side of the jacks to the ground (or vref) attached to the non-inverting input of the opamp. All the resistors are tied together into the inverting input of the opamp. The second opamp is used to invert the signal back to normal, you could leave it out if you can live with an inverted output. It is important that the summing opamp be inverting.

IIRC this is called a virtual ground summing opamp configuration.

From here, you can go wild. Passive or active gain controls on the inputs. Phase inversion switches (in case somebody else cheated and inverted an output ;) The inverting opamp can provide volume boost. And on and on.
Click to expand...

sounds pretty simple, actually. Could the resistors be turned into pots for passive control? What would be the easiest way to get a volume boost out of it?

also, +1 to the good descriptions
 
#8
You could put a pot to ground in front of the resistor. It is important that the resistors are the same value, so you cannot make them pots.

Since the opamp is just a simple inverting opamp, you can add boost to it just like any inverting opamp. The simplest would be to make the opamp always have gain, then slam a volume control on the output. Just don't go too crazy with the gain or you could have clipping problems with a 9V battery.
 
#9
ok i get it: make the mixer with pot to ground volumes and a total volume control afterwards in the form of a booster. The boost circuit I used for my booster was complicated enough that I don't feel like making another, is there a simple way to get a small boost? actually, will the mixer change volume at all? (other than what I do with my passive volume pots)

Also do you have a suggestion for resistor value?
 
#10
Plenty of info to be found on the net, try googling "summing amplifier" and you'll find heaps.

The values of the resistors are important in terms of creating the desired input impedance, make them high like 10k or more.

Worth looking into this a bit because there's a nice little trap or two for young players - the main one being that the summing action of the resistors causes all the signal amplitudes to be halved which means you need at least a gain of 2 just to get back to where you started.
 
#11
Here is a quick schematic. I have *not* tested it! But I am fairly confident it will work.

UPDATE: I added the DC voltage divider to the schematic.

Link Removed

It is actually a spice model, so there are some "funnies" to get the model to work right. The circles with a sine wave in them are the AC inputs. You can ignore them. R1 and R2 are the summing resistors. You can add as many as like, but they should all be the same value. The only problem is, the more inputs the more chance you will start clipping. 9V is not a lot of voltage.

If you add the input gain pots, they go *before* R1 and R2.

The gain is defined as R10 / R1. I have given it a small gain of 2.2. C10 just cuts down on noise.

R11 is the output pot. The spice program I used cannot handle dynamic values, so I just show a resistor. Put a pot to ground with the center tap going to C11. C11 is the output decoupling cap. Rload is required for the spice model to work. You would not actually put a resistor there.

A note about the DC. Opamps are designed to run off dual +/- supplies. I have shown a single ended supply, like a battery. Vcc is the positive battery terminal, Vcc- is the negative(ground) battery terminal. The ground is usually labelled Vref or Vr. The ground sits at 4.5V but is considered 0V for the circuit.

So, don't connect the battery or power supply ground to the ground shown in the schematic.

This circuit can be run with up to 30V. Basically the more voltage, the more headroom.
 
#12
seanm said:
Here is a quick schematic. I have *not* tested it! But I am fairly confident it will work.

Link Removed

It is actually a spice model, so there are some "funnies" to get the model to work right. The circles with a sine wave in them are the AC inputs. You can ignore them. R1 and R2 are the summing resistors. You can add as many as like, but they should all be the same value. The only problem is, the more inputs the more chance you will start clipping. 9V is not a lot of voltage.

If you add the input gain pots, they go *before* R1 and R2.

The gain is defined as R10 / R1. I have given it a small gain of 2.2. C10 just cuts down on noise.

R11 is the output pot. The spice program I used cannot handle dynamic values, so I just show a resistor. Put a pot to ground with the center tap going to C11. C11 is the output decoupling cap. Rload is required for the spice model to work. You would not actually put a resistor there.

A note about the DC. Opamps are designed to run off dual +/- supplies. I have shown a single ended supply, like a battery. Vcc is the positive battery terminal, Vcc- is the negative(ground) battery terminal. The ground is usually labelled Vref or Vr. The ground sits at 4.5V but is considered 0V for the circuit.

So, don't connect the battery or power supply ground to the ground shown in the schematic.

This circuit can be run with up to 30V. Basically the more voltage, the more headroom.
Click to expand...

wait wait wait... DON'T connect the neg. to the ground???? I don't get it... I thought that was what made things work... the opamp booster I made has the battery connected to the ground...

I assume I can make R10 a pot, right? I would like that as the overall gain pot.
 
#13
Sir Edward V said:
wait wait wait... DON'T connect the neg. to the ground???? I don't get it... I thought that was what made things work... the opamp booster I made has the battery connected to the ground...
Click to expand...
Right, don't connect the neg to ground. Opamp circuits that use a dual supply show ground as I have show. A lot of single ended supply opamp circuits show ground as negative and use Vref for where I show ground. I assume the booster chose that.

I updated the schematic to show the DC part. I should have done that from day one, but I used a spice "equivalent" that I thought would confuse non-spice users.

Sir Edward V said:
I assume I can make R10 a pot, right? I would like that as the overall gain pot.
Click to expand...
Well, ok, I guess I will let you do that ;)
 
#14
seanm said:
Right, don't connect the neg to ground. Opamp circuits that use a dual supply show ground as I have show. A lot of single ended supply opamp circuits show ground as negative and use Vref for where I show ground. I assume the booster chose that.

I updated the schematic to show the DC part. I should have done that from day one, but I used a spice "equivalent" that I thought would confuse non-spice users.


Well, ok, I guess I will let you do that ;)
Click to expand...

lol

i think I get it now

once I get the rest of the parts I need I'll try it out
 
#15
Well I did it today, and it works great! one problem: its grounded differently than my other effects so I can't daisy chain it!!!

Gah! THANKS A LOT SEANM!!! :mad: :mad: :mad:

Seriously though, thanks a lot for the help so far, but do you have any advice on how to change the circuit so it is grounded with neg. straight to the ground so I can daisy chain it?

I'm going to cross-reference your schem. with the one I used for the booster to see the differences and I might be able to adjust myself... I'll let you know
 
#16
ok I think I figured it out... it looks like in the booster the ground is the same as your -Vcc and the booster's V.R. is the same as your ground, I think the only difference is that you have what would be the booster's V.R. grounded and the booster has what would be your -Vcc grounded

So I'm going to try grounding the -Vcc in your schem. instead of your ground

I'll be back in 5 if it works :hmm:
 
#18
It won't work if you do that, depending on what it is you're actually doing!

What you have to do is remove the ground symbols from the op amp, and also remove the one at top right of the schematic. Then, call both those points "VCC/2" and join them together, but DON'T join them to ground. Now, change all the "-VCC" bits to ground symbols and connect them appropriately.
 
#19
What exactly are you connecting up? And by daisy chaining, do you mean the power?

The inputs are sitting with a DC offset, but most things should have an output cap..... so I wouldn't expect that to be a problem. You could try attaching input caps, but I don't think this is the problem.

And if you do connect a power supply, connect it to +Vcc and -Vcc, *not* to ground. If you have ground attached to a metal box, the power in must be isolated from the box. Much like the +ve grounds used by Boss.

But fdecks suggestion should also work.
 
#20
Well I did exactly what nifty said (its basically the same thing I said, just said more explicitly), and it works... kinda. The only problem is that the pot I have only lets the signal through in a very short range of its span.

And daisy chaining is with the power supply, it didn't work before, and it works now, so I must have done something "right". Like I said the only problem now is the output pot. but I actually used it today with my band, so all is well. I still need to drill holes in the box for the pots and wire up the switches, but it works. Would audio (not-linear) pots have been better to use in this situation? I got all linear ones, and it seams like the controls would work better... I got two 100k pots for the inputs and a 50k pot for the output. the input pots are wired to ground on the outside lug and the output pot isn't.

Any suggestions on my problem now?
 

Latest posts

Top Bottom
Back