In late 2016, I began wondering if perhaps I would like to get back into mastering. Of course, If I was going to do it, I would have to do it from home, which seems like a no-no, but I figured, fuck it, I make records from home all the time.
So I set out figuring up what I would need to do this. And with all things, got a starting point. The starting point for this, would be the monitoring system, and ultimately, the two most important pieces, mastering console and monitoring controller, both of which I would need to custom build, both for factors of functionality, my own odd standards in both quality and aesthetic, and because I had much more time than money.
So I figured I’d write up both projects, as they were fun and perhaps could help others.
First off, two parts…
1) The Mastering console
2) The Monitoring controller
DIY Mastering Console
When I considered getting back into mastering in late 2016, I decided I would definetely need a mastering desk. Having previously founded and run a mastering studio, I had learned a lot in what I need and don’t need. And the first thing I don’t need is a big fucking desk.
At MortarWorks, I had gotten a huge desk. It was one of the first things I bought. I guess I figure you gotta start somewhere. But acoustically, it was always in the way.
At first I had my monitor speakers behind it, which was OK, but the room didn’t develop it’s bass till farther back, so I’d have to keep moving around during sessions for an accurate picture of the audio. Eventually, I ended up doing something completely and seemingly rediculous. I put the desk facing the wall, and the speakers behind me, with nothing between my back and speakers. And then I’d just swivel my chair around, listen, then swivel around, and adjust the equipment. It seemed ludicrous, but i’d learn later, this is precisely the way the famous Mastering Lab in LA had their room setup. Unfortunately, it confused the fuck out of clients, and probably made them question my sanity. All studio speakers go behind the desk, right? Nope.
Anyhow, after that experience, I realized I’d actually prefer to have a small desk, off to the side of my listening position, ala the way Bob Katz works. And as I’d probably be mastering from my living room at, it would be at chair height.
So I designed a little desk to house everything I need.
And what do I need?
- A high quality controller for the audio
- The Digital To Analog converter
- Space for an EQ and a compressor
- And some analog meters to get an idea of levels
So I designed a very simple rack. Essentially a turret from my old desk, but at chair height.
So I sat in my chair, made some measurements, then got to work.
I wanted to have it accomodate all the equipment. As my most used piece would be the monitor controller, that would need to be the closest. The second closest would be the DAC, as it would have the volume control for my headphones. And the third would be for the Eq and compressor/limiter, as I would use those less, they could be farther away. But all the while, the meters would need to be visible.
So I came up with this design….
Which would give me the proper height, a 2 ru space at the top for the meters, and 8 spaces for the equipment. As I had decided on a 3ru for the monitor controller, that gave me 1 ru for the DAC, and 4 ru for an eq and compressor, which meant either 2 ru size units, or I could put in a 500 series rack and have a space left over.
I decided to use plywood for the design as it would need to be small, yet durable. And as I lack dovetailing skills, would also need to be friendly to accepting screws.
From my earleier measurements, I freehanded the design of the side of it on one of the pieces of plywood, then refined it a bit till I had it good.
Then I cut it out with a jigsaw.
Then I traced that side to the other, and did it again.
And it basically matched up.
At that point, I attached the rack rails to both sides, and put in a couple blank panels, making sure to space things out with a little wiggle room, then I measured the exact width it was, then cut cross bracings and the top and bottom panels at those dimensions.
Then I just screwed it all together, countersinking the screws, covered the screw ends in wood filler, waited a few hours, then sanded everything, dusted and vacummed it, then sprayed it with varnish, and then again a few hours later.
The finish worked pretty good. It’s a light coat. You can still see some of the screw ends, and the wood filler discoloration, but I’d rather see that and have a lighter stain color, than have a big dark thing sitting next to me.
It turned out great! Here it is minus equipment…
And here it is, with the monitoring controller and meters installed.
Fun little desk!
– – –
DIY Monitoring Controller
The late winter of 2016, I started entertaining the idea of getting back into audio mastering. Now, the last studio I had, MortarWorks, was a learning experience. And in that experience, I grew tremendously in my knowledge of what is actually important in a studio. And in a mastering studio, and really, ANY recording studio, your montioring chain is the MOST important. Why? Because all decisions are based off of what you hear. So hearing it right, is the most important thing.
Accordingly, I knew I would need a mastering desk, and accordingly, it would have a monitoring controller.
In the years since I shut down Mortarworks, I had imagined a dozen or so controllers. From time to time, they’d randomly pop into my head, like, ‘oh I could build one like this’. Or, ‘this would be a good feature’. So in designing it, I had a few things to draw from.
I guess the importance of the controller needs to be explained. You want the best speakers you can get in your studio, and you want them driven by the cleanest amp. You want the cleanest path possible. But most pro-amps run at 100%, there is no volume control. So you need to adjust the volume through something. A project studio may do it with the monitor control on their audio interface. One big problem here, there could be active electronics in the circuit that will color the sound. Another problem, the potentiometer in the box may actually skew the stereo balance. Most potentiometers have a variance between left and right. So if your doing critical monitoring, you may adjust the audio you are working on to compensate for your monitors. That would be a bad idea.
At Mortarworks, the controller was simple. A shallco stepped attenuator, mounted on a panel. It’s really about all you need. Stepped switch design, so the left / right balance remains pretty damn stable, and since it’s stepped, pretty easy to know when your at the volume you like to monitor at. Just mark the panel at that step position.
But, there were some issues. I’d have liked to have the ability to mute the signal. I’d also have like the ability to switch it into mono to see what was going on when summed as such. And I would have liked to be able to switch both inputs, and outputs between varying sources.
Parameter 1: Basic Requirments
So I set those things mentioned above as my basic parameters and got to work. They were:
- Volume control
- Mute signal
- Mono signal
- Output selection
- Input selection
Parameter 2: Balanced or Unbalanced circuit
The first big design issue that needs to be decided is are we going to work balanced or unbalanced. Balanced, like most pro audio gear is more resistant to noise from nearby sources that may interfere in the signal. Unbalanced, like most home stereo equipment, is super subjective to such things. But, to have a balanced circuit, you have to double everything. So were already doing two channels of audio, stereo, and now we need to double that. This has drawbacks. One would be cost, but probably the most important, would be complexity, because now any variance in the audio can occur on four different paths. Also, sourcing parts will become much more difficult, as balanced audio is the rarity of the world compared to unbalanced, and so accordingly, price will increase.
So I decided to build unbalanced.
I also decided to not drop $300 on a stepped attenuator right out of the gate, as if I decided to NOT start mastering, it would be pretty useless to me. So for the build, I sourced a much less expensive potentiometer to build with, one that I could easily swap for a stepped part if I decided to go forward.
Parameter 3: Passive vs. Active circuit
The other thing we need to consider, is whether the circuit should be active or passive: IE, are there any amplifiers in the circuit.
For one thing, active means you could adjust the levels between sources with less loss. It also means you could interface with other equipment easier.
With a passive design, your basically designing for the system you have. And there is no makeup gain. So any losses in the circuit must be dealt with. The upside is, if done right, you won’t get any coloration from line amplifiers in the circuit. If done wrong though, you will totally fuck up your audio. Impedance imbalances will cause, effectively, frequency loss, and distortion. And what will raise impedance? Too long of lines on the output would, so you need to consider that. Also, anything you fuck up.
For this, I knew what I was working with, so I decided to go passive.
Parameter 4: Form factor
Since I’d be putting in a stepped attenuator, and possibly one that was a shallco, I decided I need to leave room for that. Also, I have fucking working in tight spaces. This thing is basically gonna be a living prototype, and I’d sureley be resoldering things, so I decided to go with a 3 rack unit size unit. Any smaller would be a pain, any bigger rediculous.
Basic Parameters List
So at this point I had my basic parameters…
- Passive Circuit
- Volume control
- Mute signal
- Mono signal
- Output selection
- 3 RU unit
- Input selection (to add at some point, but not a priority)
And yeah, off to work!
One little Issue
There is however, one little issue. I have no clue what I’m doing. Like for real, no fucking clue. But being an artist, if theres one thing I’ve learned, its that if you want to do something that you have no idea HOW to do, it’s possible, so long as you define what you NEED it to do, and move along.
So I ordered drew up a basic schematic. Got my google-fu ready. And set to work.
Like I mentioned earlier, I have no clue what I’m doing. But being passive, I know one thing: The impedance of the volume poteniometer, and how that interacts with the output of what I’m feeding the box, and the amp I’m feeding it, will be paramount.
So I dug into my memory.
When I last did this, I asked the kind folks that made my DAC (the digital to analog converter I fed the attenuator at MortarWorks) what would be a good impedance. They told me, I think, 2.5k. So I looked around the electronics parts suppier sites, and ordered two. Neither of which were 2.5k. One was a 50k, which would be pretty standard for a home stereo, and the other, a 5k. I also have no test equipment, so determing distrotion levels would be a ‘by the ear’ affair. Also, I’d learn what the wrong one sounded like. A learning experience!
My plan was to just get a three rack unit panel, and attach it all to the back. So I ordered one, and it was great. Super thick. Sturdy.
When the poteniometer arrived, I drilled the panel with my drill, and installed it. Then ran the lines.
Wow, the 50k was not what I needed. It would track ‘weirdly’, things getting louder then softer, also it sounded weird. Also, I didn’t really know how to hook up a potenitometer properly, so that played a factor. Finally I figured it out.
So I pulled out the 5k. It worked much better.
At this point I should also mention, my brain just quits when I look at schematics. Like, if I’m looking at a hookup diagram, even a fucking picture, I have to go back and forth like ten times comparing the two, and even then I will fucking wire it backwards. Inhale solder while doing this to make experience even worse. Also add bad posture, and doing it while sitting on the floor or in a chair hunched over, and yeah, it’s fucking hell.
But I must do this, cuz fucking a, I want to.
At this point, I had basically created what I had at MortarWorks, albeit, this was a far inferior pot. But now, all further steps are uncharted territory.
So I need to switch the audio between sources. Also, to mono, and mute. No idea how I’m going to do any of that. But as it’s all two channel, I figured I could probably do it with a switch. So I started looking at switches. Again, the schematic stupidness that I am came into play. After hours of studying schematics, which probably shouldn’t have took that long, I decided I needed a DPDT switch.
So I looked those up in the electronics catalog. I would probably need one for mute, mono, and output switching. I guess I could have used a rotary switch, but the truth is, I missed those glowing switches from the old console I worked at in a recording studio. No idea what they were called though. So google.
And I found them. And they were like $35-$60 a piece. Fuck That.
But now seeing the parameters, I kind of figured out what I needed and found some cheaper ones made by a different company.
After about eight hours debating what color each switch should be, I realized that I should just get white.
I ordered one.
It showed up a few days later, and was round. I wanted square. (need to pay attention to spec sheets closer). I also realized I should try and make this thing not ugly, so I printed little mockups of the various sizes and shapes I could order, and placed them around the panel. Evenutally I decided one the size, and shape, and I ordered three of the square ones. They showed up. Awesome, exciting. I drilled the faceplate. Awesome!
Now, this was supposed to be super simple, but wanting those square buttons, and wanting them to glow, I had inevitably made a requirment for power. As I have no clue what I am doing, but having an electrical engineer for a father, I knew that I should probably keep the voltage as low as possible. (Yeah, I know it’s amperage that kills you but, fucking let me know when someone gets killed by a 5v circuit.) So I decided on 5 volts, and decided to skip the ‘lets build a power supply!’ party and just use a wall wart. After all, no audio would be driven from this thing’s power, it’s all just going to illumination.
However, it was at this point that I realized ‘Hey, Gary, You have a three year old daughter! And a cat!”, and I realized, yeah, maybe having all this just dangling off the back of a fucking faceplate is a bad idea. So, god, I need a fucking chassis.
So I google the chassis, and decided, cuz I’m poor, to just buy a aluminum box, cut one side of it out, and then attach it to the faceplate I already have. So I ordered it, and when it showed up, did it. I cut the chassis with aviation snips, which resulted in a shark-tooth line, that was ugly as hell, but inside the box. Also when I attached it, I attached it a bit crooked. Luckily, I ordered it smaller than the faceplate so you can’t tell.
I wired the switches for power, and carefully, connected the power up to figure out what is what. It worked. And it didn’t shock me when I touched it. Success.
Now, to build the circuit.
The first step to designing my circuit, I figured, was to wire it all togtehter, then insert the functions as I needed them. So I painstaking wired up all the switches, connected them together, got audio to pass through it, and then got to work on each function.
I started here, because it seemed the easiest. When the DPDT switch was pressed I wanted the audio to mute. I could have had it DIM (IE, drop 20db) as I have no clue what I’m doing, I worried I might just build a low pass filter by accident, so I decided to kill the audio completely. And from googling, I knew that the simplest way was to send all the audio to ground.
The real pain in the ass of this box was deicding grounding. Like, how? Do I run a brass bar around the unit and just connect everything to that? Or do I just solder it all to the chassis at one point like a fucking guitar pickguard does? Could work, or it will cause hum if things don’t interact well, especially If I’m switching between sources and outputs.
Playing it safe would mean keeping everything as seperate as possible. For the time being though, I decided to just design the circuit so that it would short to the ground on the input jack. Or was it the output jack, I can’t really rememeber, but it worked beautifully.
My inital idea was to switch the left and right positive leads of the outputs to the amp, and let the ground just hang there. And that worked too!
Except… it caused the main amplifier to hum (as in the way it would with a short) when I switched to the second output. So that is an issue.
I realized I’d need to switch grounds as well. This problem plagued me for a few days though, because I was thinking about it backwards, almost as if the outputs were inputs, for some reason. Eventually I realized it was just a matter of grounding the grounds, rather than lifting them.
But to do this would require switching three pairs, and not just a pair. Meaning, my DPDT switch is only a third of what I would need.
And then it hit me. Relays.
Theoretically, I know that relays are things that will remote switch for you. And I assumed they did it when voltage was applied. Accordingly, If I switch the voltage on, it will trigger the relay, and if I cut it off, it will untrigger it. As I already had some power going on in this box for lighted switches, I figured I may be able to run relays as well. And again, the power wouldn’t be touching the audio, it would be triggering a switch, so the wall wart would be fine.
But there are many different kinds of relays. After about a day of googling, I realized I needed ‘low signal relays’, and found a few suggested models that are good for audio.
I ordered six of them that would work at 5 volts.
At this time, I googled for a circuit design to wire them, anything remotely close to what I need it to do. I ended up in all sorts of forums, and eventually, found a schematic that kind of did what I needed it to do. So I redrew it to work with what I am working with, and voila! Wired them together.
At this point, I should be on a breadboard. But fuck all that, I just wired them in there hanging on the wires. It’s ‘handwired’. It’s ‘botique’. Fuck it.
So again, I carefully plugged it in, and hit the switch, expecting to hear a bunch of clicks. No clicks. Damn. But then I cut the audio on and realized it was switching the audio.
But this circuit didn’t work exactly the way I wanted it to. I wired it right, it worked fine when not bypassed, so theoretically it should work when switched. But now it was switching, but the left side, somewhat attenuated would still come through a little on output one, when I switched to output two. Hmm.
After a day of fucking around with it, I came up with a solution and wired it in. Woah. It did not work, but it DID send the signal to MONO!
So I looked at what I had done, realized I could recreate the same circuit with the DPDT switch, and wired it up. It worked perfectly.
Yay feature #3 accomplished! By accident..
At that point, I got tired of trying to fuck with the output switching, and just decided to leave it there for a bit. I’d accomplished 2/3rds of it’s functionatliy, which to me was fucking amazing, as like I said, I have no idea what I’m doing.
And now came the hard part. Well, the possibly hard to accept part. Does it fuck up the sound?
So I began testing it, in circuit, and out of circuit. And no, it didn’t fuck up the sound at all. In fact, it actually tracked better than the potentiometer in my DAC, (which is a very decent part). So accordingly, it was better than using the straight DAC, and it gave me two additional features.
After living with it about a month, I decided i’d like to be able to run my turntable through it to my speakers as well, so I added a simple rotary switch to switch between DAC and turntable. I settled on a four-pole, 3 positions switch (one for turntable, one for DAC, and one for future expandison), and just used three of the poles (left positive, right positive, both negatives tied together).
Also, the switch I ordered was WAY smaller than I expected, so soldering it was fucking crazy difficult, but doable.
Because of the difference in impedance and the incoming voltages, this means that at 9’o clock the volume is good from the DAC for listening, but the turntable needs to get cranked to 12 o’clock to match volume.
This is where a line amp would be good.
And that is where I am at now.
Possibly add a line amp? Also, needing to get that channel switching ironed out. I know the issue is in how I’m grounding it, and specifically, I bet it’s how I’m grounding the attenuator.
But, I’ve had colds and been kind of depressed and also working on tons of other stuff so it’s just a very low priority right now.
I also considered running a line from the controller directly to the VU meters in the console, (right now they run off a second set of outputs on the DAC, so they remain true despite any attenuation at the monitor controller) but then I realized, as all my work would be through the DAC, it wouldn’t actually benefit me in any shape or form.
But it does what I need it to do, wonderfully!
And it’s really nice interacting with something that I BUILT! I’m such a stickler for quality, and it really does work great.
So I feel a bit proud whenever I look over at it!