I built 3 different DIY eurorack mixers, they were each quite different.

Everyone can use a little utility mixer, and they're easy to build, but how different are they? It turns out there are some pretty big differences in how they work. Let me show you.
All of the mixers were purchased by myself from AI Synthesis, they aren't a sponsor and this isn't paid placement. I just think they are good kits for beginners.
0:00:00 - Introduction
0:01:09 - The AI002 Audio/CV mixer (and other op-amp mixers)
0:04:39 - The AI022 "Harmonic Mixer" (and it's origins)
0:11:22 - The AI106 "West Coast Mixer" (and it's origins)
0:13:31 - Summary

[@unsoundmethodology]

Interesting that the "harmonic" mixer amplifier stage is the old Moog design. I first ran into the design (in DIY) when i was building a copy of the classic Cracklebox (or Krakadoos), an odd instrument from the '70s with six touchpads to control the sound. I think technically it's a push-pull amplifier made with Sziklai amplifiers. (Sziklai being each side, made of an NPN transistor and a complementary PNP transistor; you can make push-pull amplifiers with just one pair of transistors.) (I definitely need to buikd some of those mixers! Mixing in my DIY Eurorack system is currently limited to the three unattenuated inputs on David Halliant's Simple VCA and a passive/averaging mixer based on the North Coast Synthesis "Passive Multiples and Friends" design/panel.)

[@LeoPerantoni]

Your videos are SO INTERESTING. I absolutely love this format you make your videos in. It keeps me so focused through it all. Thanks for sharing :)

[@SoundVoltage]

Thank you so much!

[@fakshen1973]

It's fun being able to follow along without a lot of technical logic. Having the knowledge to know to ask is awesome.

[@depthbuffer]

An op-amp outputs the voltage difference between its two inputs, multiplied by a large amount. Normally this amount is too large to exert any sort of control: basically, a difference above zero will lead to maximum positive output, and below zero would lead to maximum negative. To bring this under control, a non-inverting amplifier takes an input signal into the positive input pin, and feeds its own output back to the negative input pin (as well as off to the final output resistor & output jack), creating a loop that stabilises when the output voltage matches the positive input voltage (because the output voltage is also the next negative input voltage, so when output matches positive input, negative input also matches positive input, and the difference between the inputs is zero). Effectively, with the feedback loop in place, a non-inverting amplifier "wants" to satisfy the equality negative input voltage = output voltage = positive input voltage. Positive input is our input signal, so it will swing its output voltage (and by extension the negative input) to track it. To make that into an amplifier, instead of just an input follower (AKA a buffer), you stick one resistor into the feedback loop, and another between the negative input pin and ground, creating a voltage divider. Let's say you divide the feedback voltage by two: the op-amp will now output double the positive input voltage before it stabilises. The negative input voltage is half the output voltage, so it has to push the output twice as high before the difference between the positive and negative inputs drops to zero. There you have it, a non-inverting amp with gain 2!

[@doublewah]

This is such a great video - so clear and demystifying and I love that you’re also clear about what you don’t know. So refreshing and it makes it feel like you don’t have to know everything to start digging into the electronics. Thanks! (One small bit of feedback that is a minor niggle… I would prefer it if you didn’t have the background music while you’re demoing the sounds. I thought it clashed a bit and made it harder to focus on what you were showing. Hope that’s useful feedback and not unwanted feedback!)

[@InflatableRobot]

From what I remember, using an inverting op-amp configuration allows you to attenuate down to zero whereas if you did the same with a non-inverting op-amp you can only go as low as unity with the incoming signal. Fine for a buffer but not great for a mixer.

[@omnidivergence9846]

Another excellent video! I like the inclusion of the schematics to help with understanding it. I recently bought the Erica 16 channel matrix mixer. The possibilities with it seem infinite.

[@xpump876]

Great explanations!

[@psynautic]

The easy answer to why use an inverting amplifier configuration with a unity gain secondary amplifier. a non-inverting amplifier config can not amplify less than unity gain. So it's a very strict amplifier, its minimum is unity gain. inverting config is the opposite its maximum is unity gain. Using just a non-inverting config in a mixer would result in adding all three input levels into clipping without a knob that could tame it.

[@SoundVoltage]

And there you go! Thank you very much Mr-someone-who-actually-knows-what-they-are-talking-about :)

[@Ether_Void]

The non-inverting circuit also can't sum multiple signals easily without the amplification changing depending on whether or not a input is connected. Because of the way the op-amp works the two inputs will always try to stay at the same voltage (with some minor difference) so all the inputs "see" is their input resistance (potentiometer and 100k to a virtual ground node) the other inputs are basically "invisible". This wouldn't be the case in a non-inverting topology because there is no virtual ground node, so a input providing 0V would actually sink some current while a not connected input wouldn't but ideally you want 0V and not connected input to behave the same way.

[@depthbuffer]

IIUC an inverting amplifier *can* go above unity gain (well, inverted but with magnitude greater than 1, e.g. gain of -2). You just need to make the feedback resistor larger than the input resistor. (I would link to sources, but every time I do that, KZbin deletes the reply because it thinks it's spam...)

[@psynautic]

@@depthbuffer right but non-inverting cannot go below unity gain, that's the main reason a double inverting setup is the way mixers are usually made.

[@depthbuffer]

@@psynautic I believe you're right that non-inverting can't go below unity, just wanted to correct the idea that inverting being "opposite" somehow means it can't go above unity. The reality isn't that simple because of the different architectures. I think the main reason for the double-inverting setup is because, as per @Ether_Void's reply, it's easier to sum multiple inputs with an inverting amplifier (because you can just keep adding more inputs with additional input resistors without interfering with the actual behaviour of the op-amp).

[@pugix]

Nice video! I learned about op amp mixers just around the time that the TL072 came out. I've used op amps to make all my own mixer designs. One thing you didn't mention was the pot taper. Mixers designed for audio use, guess what, audio taper pots. These give finer control over lower volume levels. Linear taper pots are used with DC-coupled mixers to keep the response even with respect to control voltage levels.

[@SoundVoltage]

I was going to do a separate small thing about taper, because this was getting long and -- from a DIY pov -- I just always use linear and then put a resistor across two of the pot legs. But you're right I should have mentioned it here.

[@terpenemusic]

Hi Mr. Sound + Voltage :) I really like Your videos and the nice graphics. The only thing that is difficult for me is the color that You use for highlighting the schematics because I am a bit color blind... Same thing with some Modules with green and red LEDs of course...sometimes its nearly impossible to have a clear visual feedback. Sure there are others who are also struggeling with that. :) Best regards Alex

[@SoundVoltage]

Thanks for watching. That's not something I had really thought of. I'm not sure what I can do, but at least now I will think about it. I appreciate the feedback.

[@michaelt_tamtam]

I want to try building DIY eurorack modules. I'm curious about the "AI Synthesis - Matrix Mixer". Would that be simple enough to build for a beginner?

[@SoundVoltage]

It's a little tight and fiddly, but not terribly complex. If you're not already comfortable with a soldering iron, I'd suggest the Mults that I built in the previous video. But the Matrix mixer definitely isn't super difficult.

[@agustinmuerto]

Hi! Does anyone knoe how good would the simplest OP-amp mixer work for line level audio? I'd like to use it for both purposes (10V CV and +4dBV signals) and I suppose I'd be fine / fine enough, but asking just in case! Thanks for the video, it's helping a lot to build some small utilites.

[@user-bb6tq4by8t]

3 blue 1 brown but modular. Yes.

[@SoundVoltage]

That is probably the single best compliment I've ever received on this channel. Thank you so much. It's not too surprising that I took a lot of inspiration from @3blue1brown. Grant is a fantastic educator.

[@TheDoctorM0]

I'm subscribed to both!