DIY Analog Mixing Console

frederic

New member
First, some necessary things...

  • Thank you to those who expressed an interest, both publically and privately. While I'm more than happy to do this, time is also limited so I'm certainly happy I'm not wasting my time.

  • This is offered for "informational purposes only". While I have assembled each building block (more on that as this progresses), I offer ZERO warranty and accept ZERO liability for what you do with this information. Use at your own risk.

  • I am happy to discuss, and help out as is convienent for me. This might become inconvienent tomorrow, and I am not providing technical support. Use at your own risk.

  • This is not a project for the meek. While it's not necessary to have a solid understanding of electronics, it definately helps. Further, if you cannot solder consistantly, and reliably, don't waste your time and money.

  • Some of these building blocks are of my own design, some are from chip manufacturer reference designs, some are "leeched" off schematics in the back of instruction manuals for pro-gear, some are "leeched" off the web. I apologize up front to anyone concerned about copyrights, but alas I'm not able to make a reference for each building block simply because I don't recall where everything came from. I'm just disorganized. Since this is for educational purposes anyway, I see no real issue with this at all.

Okay, enough of the disclaimer crap. Lets build a mixing console.

First, lets look at what a mixing console is. Here is the simpliest mixing console we can make, as it's the foundation of all mixers regardless of size. Input circuitry, connected to a common buss, which drives and output. In case it's not obvious, technically a "Y" cord combining the output of two synthesizers (or other) to an input on something else, is a simple mixer. Not a good one mind you, but it illustrates the point.

mixer1_raster.gif


So keeping this simplistic drawing in mind, we all know that a "real" mixing console will have many inputs, many busses, and many outputs, with some control "stuff" in between all three components. Buss select, Volume, Pan, EQ, and so forth. Before we design that aspect, let me point out two things that really irk me about modern day, affordable mixing consoles, things that some of the pro consoles I've had the pleasure of using in the past certainly have.

  • "Flip" Switches - Pet Peeve #1

    Most consumer grade analog mixing consoles force you to utilize some of your input channels as tape returns. Look at the illustration below.

    mixer3_raster.gif


    In this example I've drawn the typical wiring scenario for a 32 input, 8 buss mixing console with an 8 track recorder. You can see that the recorder's eight outputs "steal" eight input channels, leaving you with only 24 input channels. The second "problem" is with the eight busses, two of them have to be output busses for monitoring, thus leaving you with six busses to feed the recorder. If you have two eight track recorders, the problem gets worse. Wait, yes, I know that most mixers also have direct outs (often used to feed recorders) as well as a stereo buss. But, wouldn't it be nice to be able to use the direct out for something else, rather than using it to feed your tape recorder? Plus what happens if you want to record input channel 1 to track 14. A patch bay will solve this problem, but so will what Neve, SSL, Amek and Trident (to name a few) did in their professional consoles - they designed in a real "flip" switch. Essentially what they do is design their input channels to be two circuits. One circuit is fed the usual mic, synth, guitar stuff, and has pan, fader, EQ, compressor (etc), and the other circuit is really a minimalistic thing - fader and pan - which feeds different busses. This way, while you're multi-tracking, you feed the tape returns to the simple circuit, adjusting fader and pan, while simultaniously recording on that input channel through the more complicated circuit, and not mixing your tape returns with the new material. Multi-tracking 101! The "flip" switch reverse the two inputs, meaning that the tape return is now fed into the complicated circuit (with compressor, EQ and so forth), and the mic/line input goes into the simple, fader/pan only circuit. There are instances where you'd want to do that, and when your console can't do it all you do is lose more input channels and spend time at the patch bay. Aggrevating. Some manufacturers (Fostex in particular - the old 450 series for example), had a cheesy version of this, where a switch selected between mic/line and tape return, to feed into the only input circuit. You'll see how this turns out and why later on. This is not a feature that is necessary by any means, but it's huge for me.

  • Input Channel Compressors - Pet Peave #2

    In the "old days", mixing consoles had built-in compressors in theiir input channels. Nowadays, you have to buy external outboards and attach them to your "insert" jacks. This works okay I guess for most people, however if recording a large drumkit for example, micing each drum individually, you may 12 mics and need 5-6 compressors to get the mix right. Maybe compressing the whole thing is okay with some people, but I'd much rather avoid that if I can. Might make certain drums sound better, but will "ruin" others, and cymbals, for example. So you'd either have to have a rack full of patchbays and compressors, or have them built into the console. I'm not talking very fancy, digitally controlled compressors, but rather a few parts just to take the edge off and soften, or limit, certain signals as necessary. This was one of the reasons why when I made the step into digital recording, I bought the Tascam TM-D4000 mixing console - it had assignable compressors that could be used on "all" channels (per original brochure). What they don't tell you in the brochure is that there are only eight compressors that you can assign, and only enough horsepower in the machine to run five if you have any other effects going. Six, if compression is all you are doing (no EQ). So that to me defeats the purpose. Anyway, the design here will have an individual compressor (albiet simple) for each channel. If you don't want this feature, you don't have to make it. Simply jumper across.

So now, with my two pet peeves out in the open, lets look at a diagram which somewhat represents what I've designed for. Obviously you can add or subtract to this as you wish. Each box is a "building block", and over time (more like as I have time!), we'll put the design for each "block".

mixer2_raster.gif


Now some of my above paragraphs might make more sense. You can see there are two circuits per input channel, one for the mic/synth/whatever, and one for the tape return for the recording/multi-tracking phase. Flip switch, then the tape return goes through the main circuit and whatever is on the mic/line jacks go through the simple circuit.

In some "real" mixing consoles, the compressor "block" and the "EQ" block are reversed. This is preference. I prefer it this way, because I can then tweak the EQ to eliminate as much "huffing" from the compressor should that ever be a problem.

In the upper right I've identified some busses. These are left up to you, as to the quantity of each that you want to spend money on parts, time in soldering, and surface area in knob/switch placement. Obviously the more "stuff" you add, the more flexibility you have, but at the same time if you don't need 48 busses, why waste the time and money. Same for aux busses and so on.

Main busses are fairly obvious, and often used for two purposes. During recording, they are used to feed inputs on the recorders, and for mixing, they are used for surround sound. If you have six busses, you can mix 5.1, if you have eight busses, you can do 7.1. Generally most studios can get away with eight busses, though I've been involved in projects where 12 or 16 busses were absolutely ideal, but that isn't the common place.

Aux busses generally are used as "Aux Sends", for feeding outboards (delays, reverb, etc). On consoles that don't have more than one stereo buss for the console room, you can also use say, "Aux 6" as foldback, providing a seperate "mix" for the vocalist, musician, or whomever, seperate from the console room mix. This is useful when your vocalist needs to hear his, or her voice significantly louder than the actual music, or they wander off in pitch like a shrieking cat.

Pre Fade Listen means that you hear the signal of the selected channel alone, picked off from a point before the fader, so that the fader level has no influence on the PFL level. On some consoles, the PFL signal is routed equally to left and right speakers, on others the position of the pan control is retained. PFL is often used for setting the gain control: on each channel, one at a time, press the PFL button and set the gain so that the main meter reads a good strong level without going into the red. Solo or AFL: Solo, or After Fade Listen, is as above, but the position of the fader is taken into account. It's common for PFL and Solo busses to be mono, as they are more for troubleshooting than anything else.

Foldback Buss is what I would use to feed musicians (instead of wasting Aux sends). They are similar to the control room mix, except tailored for the musicians. The more of these busses you have (mono or stereo), the more unique simple mixes you can provide for musicians. If you record one musician at a time, you can have one of these (or a pair), or use busses if you want to "spend" them on foldback. If you record four musicians at a time through headsets (i.e., a 4 piece band, or just a rhythm section for example), this might be nice flexibility that not everyone needs.
Not shown in the diagram are some other obvious things we'd need. Like sub-mixing of the main busses to the control room busses, return aux sends to the main busses and possibly the foldback busses, and so on. I left them off the diagram intentionally to make it easier to read. As we build each building block, we'll discuss what goes where and why. And again, if you are building someting like this, you can add things, or hack off things you dislike, don't want, don't need.
 
Last edited:
technominds said:
Im impressed that you are even going to attempt this, are you going to make one yourself?

I've made each of the building blocks over the years... in fact some of them found homes in things I have used, and will continue to use in my studio going forward. For a while I toyed with the idea of building a large format modular analog console, but decided against it because of the time and energy involved. Someday I may do so, if I ever have the time to work out what to interface the digital control aspect *to* (a PC? Midi?). I believe a hybrid console - analog in the signal path for warmth, digitally controlled for automation and repeatability, is the best combination. This is opinion, not fact of course.

The idea was to build an expandable, warm analog console, with the ability to "save" every knob, fader, and switch setting via digital "stuff". This way, after spending 4 hours getting the settings correct for a particular flutist with a particular silver flute, I could recall those settings on any input channel as a starting point for the next time I'd record that flutist. Kinda like a library of console settings, to drag and drop from. I figured out how to digitally control everything, and make those settings controllable with knobs, switches and faders, and available for "something" to download, save, upload and restore. I just haven't had the time in the last two years to figure out the "something" part.
 
frederic...

Of course you are gonna give old Velvet Elvis your prototype right!?!?! :p

How's the kiddo? Last time I talked with you, you were expecting one!

Velvet
 
So you like in-line consoles with compressors, eh? Another SSL user, I take it?

Yeah, you are deffinatly trying to break my bank. I like it.



Light

"Cowards can never be moral."
M.K. Gandhi
 
Velvet Elvis said:
Of course you are gonna give old Velvet Elvis your prototype right!?!?! :p

:D You can make a few of he pieces modular, make a huge frame to put them in, then add as you want to. That's the idea behind modular!

Velvet Elvis said:
How's the kiddo? Last time I talked with you, you were expecting one!

Born Feb 28th 2005.

_im001439.jpg
.
_im001291.jpg
 
Light said:
So you like in-line consoles with compressors, eh? Another SSL user, I take it?

Oh yeah.

Light said:
Yeah, you are deffinatly trying to break my bank. I like it.

Here is what I intended to do as far as the modularity. I'll talk more about this later on as the building blocks are defined, drawn, and available for all. But in a nutshell, one would make say, 16 input channels, and 8 buss channels. Need two more buss channels for aux sends, you solder them up, and mount them in the chassis next to the main busses. More input channels? Same thing. The only thing that is critical is to decide, up front, how many busses of each kind you want, because the input channel switching circuitry has to be deteremined before you make the input channels. Then duplicate as necessary. The one input channel I made selects between 16 busses, eight aux sends, solo, PFL, and three stereo control room pairs. More than enough.

I'll be posting parts of the input channel starting next week.
 
Frederic,

Thumbs up on the digitally controlled analog idea. You are in good company there, that's for sure. Klark Teknik, Jim Gamble, etc. If you aren't familiar with Gamble consoles, you should do some Google work on the guy and his stuff, notably the DCX console. Cortek has a software package (downloadable demo) to do the control/recall part, and Mixed Logic makes some cool control surfaces, worth a look for anyone.

http://www.corteksoft.com/DCXV2.htm

http://www.corteksoft.com/main.asp

This should be lots of fun, thanks for the efforts. I don't know how far I wil take it, but I really like the idea of building my own console.
 
frederic said:
Born Feb 28th 2005.

_im001439.jpg
.
_im001291.jpg

Outstanding... Maybe I knew that, as my latest addition to our family was born last May..

I dig the console idea that you are putting together... I'd LOVE to have something that is analog, but fully capable of being automated.

I sold my Mackie 24*8 to buy a Mackie D8b (I bought Mackie because I could afford it)... and I have to say that I much prefer the analog side... not worrying so much about latencies etc... plus the analog just sounded better to my ears... But I truly do love the automation aspects, so I'm torn.

Velvet
 
Velvet Elvis said:
Outstanding... Maybe I knew that, as my latest addition to our family was born last May..

Congrats! We did briefly talk about both our better halves being pregnant, but didn't get to chat since unfortunately.

Velvet Elvis said:
I dig the console idea that you are putting together... I'd LOVE to have something that is analog, but fully capable of being automated.

What I'm putting together for this forum is totally analog. What I may put together for myself someday is a two piece console. A large box with tons of jacks, and a long "cord" to a large control surface. The control surface looks like a mixer, feels like a mixer, and acts like a mixer. Except, all it does is tell the "box" what to do. This gives the best of both worlds I think, AND allows the audio "box" to be very close to the audio gear. I'm trying not to get too far into that on this forum, as to keep this project as simple as is realistic.

Velvet Elvis said:
I sold my Mackie 24*8 to buy a Mackie D8b (I bought Mackie because I could afford it)... and I have to say that I much prefer the analog side... not worrying so much about latencies etc... plus the analog just sounded better to my ears... But I truly do love the automation aspects, so I'm torn.Velvet

One of my pro studios years ago had the 32*8 with several sidecars. I liked it, yet hated it. It didn't sound anywhere near as good as the pro studio before that, which had an antique large format Neve. But I'm comparing a chevy to a rolls royce here, and I have to admit I got more than I paid for with the Mackie. It was a really nice console, considering how dirt cheap it is.

I did consider the Mackie D8B when it first came out, and at the time I couldn't get squat for under $10K initial investment. While I've spent more that on my tascam "daisy chain" of digital mixers, I got in with just the TMD-4000 (used, less than $2500), and every month or two I added a little TMD-1000 toy mixer off ebay for about $300 or so plus shipping until I had enough input channels. This is why I like modularity - buy what you must have, then add what you can use later on. This analog console design follows that mentality as well.

BTW, since you have the D8b already, and do enjoy the animation it provides (which is darn good BTW), you might consider assembling some of these building blocks to enhance it. That's partly why I'm putting the effort into break things out into building blocks. If you solder up one of my pre-amps, test it and and like it, you might consider building a few of them, and connecting the outputs to your "line in" jacks on the D8b. I do this with my Tascam TMD4000 actually. I have in my producer's desk a homemade vacuum tube pre-amplifier I made (with no knobs!) which the mic from the vocal booth will eventualyl go to, which in turn will feed one of the line ins on the digital mixer. Not a perfect solution, but it's acceptable. The tube pre-amp is getting old however, and takes a while to warm up to where I like the sound (about 20 minutes, now). It's somewhat of a copy of the old tube-based Neve pre-amps, kinda sorta.

Anyway, as things get posted here, you might find pieces you like and may consider building as an add-on. Or, maybe you'll just enjoy the giant posts LMAO.
 
Velvet Elvis said:
I sold my Mackie 24*8 to buy a Mackie D8b (I bought Mackie because I could afford it)... and I have to say that I much prefer the analog side... not worrying so much about latencies etc... plus the analog just sounded better to my ears... But I truly do love the automation aspects, so I'm torn.

Velvet

I had to replace my Mackie 32x8 this past fall due to some water damage, and I decided to go with a hybrid mixing setup for just those reasons. I have a small good quality Soundcraft analog console on the input side and for monitoring purposes, and I also purchased a Tascam 1082 control surface/firewire interface so I can have tactile control of my Cubase software since I have commited to mixing in the box at this point. So far so good.

Frederic - I can't wait to see the evolution of this thread. I've thought about doing just a mic preamp section, but I like the idea of a complete channel strip better, so I'm definitely going to keep up with this baby. Thanks for the time (in advance) that this will take from you. It is of benefit to a lot of us who are tinkerers at heart and just haven't crossed that bridge yet into building something this complex. I have actually built a headphone distribution amp based on one of the Paia designs, so now I'm ready for something a bit bigger!!

Darryl.....
 
Time to get this moving.

If you scroll up to my first or second post, you'll see the block diagram of the mixing console. I've completed drawing the first few blocks. Click on the pictures here to open a new browser with a much larger version.

Microphone preamplifier:



This building block is based on Texas Instruments' INA163 operational amplifier. This is a very similar design to TI's reference design, with some minor changes as well as the addition of the phantom power. Note that this is designed for low impedence microphones. It's not designed for cheap higher impedance dynamic microphones. Use the line preamplifier module (below) instead and crank the gain.

It's an interesting circuit because of the low parts count. The $5 US INA163 is more expensive than other op amps, but for $5 you won't believe how warm this sounds. I've built mic preamps using the LM5532/34 series and they were "okay" whereas the IN163 sounds so much better.

R6 (2k) is the gain control. R8 is 5 ohms, but if you can't find that value feel free to put two 10 ohm resistors in parallel, that's actually how I made mine. The 2K is a linear pot, even though this is an audio circuit. The INA163 expects this and compensates internally. SW1 provides phantom power to the XLR jack through a pair of 6.8K resistors. C3 and C7 act as blocking capacitors, allowing the microphone signal through but blocking the +48V phantom power to the INA163. When you flip on, or off phantom power there typically is a "click" or a "pop" in the circuit, and the four 1N4148 diodes help cut that down to a mininum. Because they are backwards, anything over 15.6V is passed to the supply rails preventing an overvoltage condition to the INA163. Don't put these in backwards!

Line/Tape preamplifier:

.

The above two circuits are identical, with the exception of the part designations. The reason for two seperate diagrams is my schematic software is "stupid" and cannot number parts correctly if they aren't actually on a diagram. But they are the same circuit through and through. The 10K variable resistor in the center of the diagrams is the gain control. At first this might look more complicated than is necessary, and in some senses it is... but by having these three op-amps wired as they are... you can plug into the TRS jack both balanced devices, and unbalanced devices, and the gain control will still represent infinity to 0db regardless of what is plugged into it. The NE5534 have good enough specs to be used in this application with minimal noise. They make great buffers, line amps, and so forth. You'll find them slathered all over professional gear actually.

Next up will be the Mic/Line switching, and Tape "flip" switching I spoke about initially. Give me another week or so. Holiday season and all :D
 
Hey MsHilarious... it could be a small-format console too! Just make less modules!

Scale up, scale down!
 
frederic said:
Hey MsHilarious... it could be a small-format console too! Just make less modules!

Scale up, scale down!

I'm happy with my modded A&H and ITB mixes :o I'll just admire from a distance :)
 
mshilarious said:
I'm happy with my modded A&H and ITB mixes :o I'll just admire from a distance :)


Not me. I'm making this thing.


I'm already sourcing parts.





Light

"Cowards can never be moral."
M.K. Gandhi
 
Two questions on the mic pre Fredric: what about a polarity switch? Just a DPDT swaping pins 2 and 3 on the input? And L1 and L2: do you have a part number?


Light

"Cowards can never be moral."
M.K. Gandhi
 
Well, so far the mic pre is less than $12 a piece. Of course, I still need that inductor part number, but I would say we are still looking at under $15 for the mic pre. This is also, of course, assuming that one is buying parts for a whole console, which saves a fair bit of cash (buying one resistor for $0.09 is much less attractive than buying 200 for $0.02, and similar breaks can be had on some caps and diodes. 200 may be a lot of resistors, but if you do a large console, it only takes 45 of them to make the break worth while). Now, this is without the power supply, but that certainly will not be that pricey. And of course, that does not include the circuit boards, and when I do this I would be getting some very nice, double sided, plated-through hole circuit boards with a nice, thick solder mask. If I REALLY wanted to be forward thinking (and I had the room), I might even mount the electrolytic caps on turret lugs on the circuit board, so that WHEN they need replacing (in 10-15 years time), I could do so without damaging the board. But of course, I am getting ahead of myself here. We gotta see the rest of the designs first, of course. I might prototype this mic pre (and some of the other blocks) when I’m done with a couple projects, though.


So, we figure the mic pres on a 48 channel desk are under a thousand dollars. If, as Fredric has said, the mic pre opamp is going to be the most expensive opamp in the unit, this thing is going to be quite reasonable, for what you get.


Myself, what I'm thinking right now is to do a very small format version of this as a test, once we have the parts, to use for all the live corporate crap I spend SO much time doing. I have been using crappy boards (Mackies and such) so much that I have, for some time, been thinking of making a decent board designed specifically for the kind of gig I do so frequently. A great EQ is a MUST for me, and a solid buss compressor is also a big deal for me. And of course, the ability to slam the mix buss if I have to without it falling to pieces. So that is probably what I will build first, when the time comes. Probably 8-12 mic channels, 4 line channels (I don't need a in-line console for this stuff), a few (4-6) aux busses, and very good EQ with the decent compressors Fredric talked about. I am assuming those compressors will be along the lines of the SSL 4000 compressors, which I am quite familiar with, and I would LOVE to have available on for live work. I have to imagine that I would want a separate buss compressor (don't know what you have planed in that regard, Fredric), but RNCs are so bloody cheap that doesn't bother me at all. It seems to me like it would, at the very least, make a good test of the concept.


Light

"Cowards can never be moral."
M.K. Gandhi
 
Light said:
Well, so far the mic pre is less than $12 a piece. Of course, I still need that inductor part number

He used an inductor component, but labeled it as a bead. I think that's just a ferrite bead?

What about the rating of some of those caps? C3 and C7 are awfully close to their rating, I believe C4 is too.
 
I'm getting ahead of you here, but what's the approach to designing the power supply? That's trickier to do in a modular fashion, at least the transformer has to large enough to handle the max intended design. Or would you just spec different parts for maybe 1A, 2A, and 4A designs?
 
Back
Top