Skip to main content
shopping_basket Basket 0
Login

The Red Tin Audio Expander Part 2: Researching Active Matrix Mixers and Building an Op-Amp Circuit

Dave Ives
3


Researching and building an Audio Op Amp to be incorporated into an active matrix mixer

I previously built a passive matrix mixer that I named the Red Tin Audio Expander and the blog post attracted quite a few helpful comments, mainly suggesting that to be properly effective, it needed to be active design rather than passive.

This post takes a look at a first attempt at building an active matrix mixer with op-amps. I also decided to scale down this down slightly to 3x mono channels, plus a stereo in/out for the Red Tin send and return. This should make the build slightly more manageable.

Op-amps


Reading up on active matrix mixers and taking into account some of those comments on my passive design, I worked out that I was going to need an amp for each input. As DesignSpark member 44meurope put it, “an (op)amp for every input, and (summing) op-amp for every output”.

I have tried building audio op-amp circuits in the past and for some reason encountered a few issues, so I thought I should do some more research to get a better understanding of what I was dealing with.

Simply put, an operational amplifier (usually referred to simply as an op-amp) is a device to amplify voltage. My reading up taught me that, at their most basic, an op-amp takes a differential signal — the voltage difference between the V+ and V- pins — and outputs a voltage proportional to this difference via the Vs+ and Vs- power supply.

This “open-loop” operation typically results in a device gain (known as the open-loop gain or AOL) of 100,000 or more. Even a tiny difference in the voltage between the V+ and the V- pins works out to an output of nearly the supply voltage when the plus input voltage is greater than the minus input. This configuration acts as a comparator, turning a potentially varying input signal to a steady on/off output.

Op-amps are more usually used in a closed-loop configuration, with the output voltage feeding back into the inverting input to form a more controllable signal amplification. The simplest way to accomplish this is to use a buffer circuit, where the output feeds back into the inverting input with no resistors or other components.

I discovered that there are two op-amp golden rules:

  1. The output attempts to make the voltage difference between the inputs zero
  2. The inputs draw no current

So, to construct a closed-loop op-amp you need to feed the input voltage into the V+ input, connect the V- to the amplifier’s output and then the output should go to the same value as the + input to keep both equal.

This configuration can be used for weak signals that require an amplified output – which is what I wanted for my mixer and an Internet search turned up lots of schematics for audio op-amps that employed this principle.

Putting theory into practice

As mentioned above I had found quite a few schematics on the Internet, including a couple of plans for Eurorack matrix mixers (here and here) that I had come across when I was building my synthesiser.

The example schematic from York Modular which is shared here was particularly useful (I even ended up buying one of their PCBs to build one for my synthesiser).

I could see the TL082 (304-217) is used in many of these examples and so I decided that is what I would base my circuit on. The TL082 follows the common pin arrangement for dual op-amps:

After trying out various of the circuits on a breadboard, I ended up deciding upon the schematic I found here

Using this arrangement I managed to make a simple circuit for an inverting audio pre-amplifier.

Bearing in mind that I am in no way a professional analogue electronics engineer, after my research outlined above, my understanding is that R3 and R4 resistors act as a voltage divider to make 1/2 the supply voltage. For that reason, the two resistors need to be equal. The capacitor between the two resistors is there to filter out any ripples in the power supply and other noise.

The gain of this circuit is determined by resistors R1 and R2 and is calculated by the following equation:

Voltage Gain = R2 / R1

I also read various discussions and articles on the pros and cons of inverting vs non-inverting amplifiers and did not really come to any clear indication of which would be most suitable. In the end, I noted that non-inverting amps seem to be by far the most commonly used in all the mixer schematics I had looked at, so I decided to go with that. So I then modified my original circuit utilising the other half of my dual op-amp to produce a non-inverting amplifier.


You can see above the input (blue) compared with the output (red) with the potentiometer turned about halfway to full clockwise.

The circuit I arrived at uses the following components:

Resistors

Polyester capacitors

Electrolytic capacitors

Op Amp

  • 8pin DIL socket (674-2435)
  • TL082CP op-amp (304-217)


Next steps


My next job will be to transfer the amplifier circuit to stripboard and build enough of these for my mixer. I then need to work out how to connect them to my grid of potentiometers and inputs/outputs, before fitting it all into a suitable enclosure.

I currently look after production at AB Open. I have a background in the arts, environmental conservation and IT support. In my spare time I do a bit of DJing and I like making things.

Recommended Articles

Comments

Sue

February 10, 2020 12:21

If you have an op-amp with an inverting input the first gain resistor (R1 in your pic) becomes the input impedance. So as per your pic the inverting input impedance is 1K which is low for standard audio which tends to be 10K, however could be reduced to 600 ohms for profesional drive levels.

If you use a non-inverting op-amp with a single rail and a R+RC voltage divider be careful to have each op-amp with it's own divider, otherwise you will get an extra summing node mixer where the multiple op-amps each contribute to wiggling the bias point.

A much bigger capacitor is also recomended, say 15uF electrolytic with a 0.1uF foil (avoid ceramic capacitors in the audio path). As drawn with a 0.1uF cap (and ignoring the divider resistors) this circuit has a high-pass fc = 1592Hz; with a 15uF this then comes down to 15Hz.

Remember though that the phase shift of a sible R/C filter starts a decade above (or below) the f3 point, as keeping phase integrity is a very importent but often overlooked part of audio.

http://sim.okawa-denshi.jp/en/CRtool.php



Hope this helps :)

0 Votes

February 11, 2020 08:57

@Sue Good points, which I was also ready to make. The LF cutoff issue would be best resolved by increasing the input resistor value I feel - or better still - by using a non-inverting input stage. Having a (passband) gain of X10 in the input stage could also lead to overmodulation problems. The inverting stage would also not be ideal as the output amplifier - particularly due to the low impedance, following the potentiometers. I wouldn't be concerned about using a single bias network as the op-amps used have very high input impedance, so there should not be any modulation due to varying bias current. What I would be concerned about, using a single-ended supply, is the large amplitude shift as the DC conditions stabilise at power up. This could damage loudspeakers further down the line. Regards, Graham Booth BEng CEng MIET

February 17, 2020 23:08

David, while I admire your enterprise I'm sorry to say that your inexperience as an analogue designer does indeed show. The solution you are pursuing is wrong on many levels and I am concerned that others will be influenced by your advice. I would be happy to redirect you if you wish.

Kind regards, Graham Booth BEng CEng MIET

0 Votes
DesignSpark Electrical Logolinkedin