Case Study: “Last Mile” in the Analogue Signal Chain

Right now, I’m making up my mind how to assemble the last few steps in the signal chain for More than Moore, my album project with an all-analogue signal chain.

Of course, there’s a recorder (analogue) somewhere, and typically you’ll also have some kind of dynamic processing and equalization on the bus coming out of your console. In my special case, there is nothing like a mixdown, i.e. the audio goes directly to the master tape. (Why? Tape generations, of course!)

This article is not about the choice of effects or other components, rather how the effects (and other components) are wired.

Ancillary Conditions

There’s a few things that are decided or nearly decided.

First, the console is a Mackie 32*8 console.

The recording medium will be cassette tape, for that using a Pioneer CT-95 deck. And maybe (for a second copy in the first generation) also a TEAC deck of roughly the same feature set and quality. Which also means both of them are a) three-head decks and b) use RCA connectors.

To simplify the discussion, I will right now assume I’m only using one tape deck.

There will be a mastering chain which will consist of one compressor and one EQ. Optional devices are one of that old psychoacoustic things (Vitalizer, Aural Exciter) and a limiter. If there’s a limiter, it will be my trusted Aphex Dominator II. Why is there that psychoacoustic thing? I’m aiming for a certain sound.

Ah, listening. This will go to my 2.1 Dynaudio combo.

Correct Order

The tape goes to the end. The limiter goes right before that. If we neglect the psychoacoustic thing for now (I may very well not use that), that leaves compressor and EQ, which can go in either order.

How is that connected to the console?

Console Connectivity

If we look at the options the Mackie gives us, we get:20171221_205346_Roßtalerweg

  • Main inserts (pair of jacks, which means unbalanced),
  • Main mix outputs (1/4 unbalanced and XLR balanced),
  • Pair of unbalanced basic inputs (“External”, “2-Track”) that can be directly sent to the control room → good for returns from the master recorder,
  • Control room and studio outputs (unbalanced).

The Control Room goes to the monitors, that’s a given. And the tape return goes to 2-Track.

Pre-Fader or Post-Fader

The big question seems to be: what goes before the console’s master fader (i.e. patched via the Main Inserts), and what goes after it (i.e. fed from the Main outs)?

From my understanding, there’s a few things to trade/consider here:

  1. Any connectivity that is via the inserts is unbalanced from and to the console. Things after the main out can be hit balanced (but only up to before the tape deck, because that’s unbalanced).
  2. If something is in the inserts, I can control the output level of that with the master fader (but controlling the input level is only possible by using all channel/group faders).
  3. If something is fed from the main out, I can control the input level of that with the master fader (but controlling the output level at the end needs to be done with the tape deck’s recording level, if not accomplished with something in the chain).
  4. Positions in the signal path I can monitor: after the master fader (and as a result, after the inserts), at the tape deck’s output fed from the input, the signal from the tape’s “third head”. There’s no easy way to listen to what goes out through the inserts.

The Individual Trades

Balanced vs. Unbalanced

While some engineers would immediately (or blindly, or should I say “deafly”?) pick balanced connections over unbalanced ones, for our application (that being: line-level signals running at max a few metres), unbalanced connections do not have real deficits, and if the components are designed to be able to run in an unbalanced fashion, the result is actually better. My hero Bob Katz says so, too.

So with that out of the way, the kind of connection (balanced from one of the main outs vs. unbalanced from the main inserts) will not play a role in any trades.

Level Control

It’s of course, about efficient, precise and reproducible level control.20171221_194905_Roßtalerweg

The level that goes onto the tape is set with the recording level knob on the tape deck. Which is not exactly calibrated or uses units that mean something. So for that reason, it’s obviously best to leave it at a sensible position once and for all.

The Dominator has both input and output level. There’s also a so-called “density” knob, which sets the distance between the output level (which is the level the thing will never go above. Ever) and the threshold of the (non-brickwall multiband) limiter. All of the things are ganged. The output level can be set over a 24dB range in steps of 0.2dB. That is sufficient (but not good for fades, but for that, you don’t use something on the limiter, anyway).

20171221_194929_Roßtalerweg.jpg
Every precise level setting thing should be like this!

 

Input level is set over a 30dB range in steps of 3dB. That’s a little bit coarse. Rationale: as you typically use a limiter for gain reduction in the region of up to 4dB, you may up in a situation where you are up to 1.5dB off due to that stepping – and that is nearly half of your full limiting range. So I will need something finer in front of that.20171221_195017_Roßtalerweg

The console’s master fader is a standard 100mm fader. Which should be good enough. It’s also one fader for the stereo sum, so that would be a good place to do fades.

We don’t know the details about the other components on the bus yet. However, those considered for the job often have some calibrated trims for gain (or threshold in the case of compressors).  With one example, I found that the threshold setting also only has a granularity of 2dB. The argument I used earlier why 3dB is too coarse works here the other way around: for a typical master bus ratio of 1.8, being 1dB off (to either side) results in a little more than 0.5dB deviation in the gain reduction – in relation to something like 6dB or more. I consider that acceptable. The EQ, finally, is not that much of a problem. I intend to run it in its linear region, and if it has a kind of makeup trim, that’s even better.

20171221_194952_Warthofstraße.jpg
No calibrated thingies on this BSS 944

There’s two kinds of level changes I’d typically consider during a mixing run. One (driving the compressor more or less) is tricky to do with just two hands. The other, a fast or gradual fadeout, shall always happen before the limiter, and in an ideal world, after the compressor.

 

With that, the result looks some like this:

The limiter shall go directly before the tape deck (meaning: after the master fader). The compressor should go before the master fader (and the EQ best stays in its vicinity).

Signals Taps for Monitoring

Here’s how reality looks:

  1. There is a tap for monitoring the console’s main out by selecting “Main L/R”. With that, we get everything after the master fader (including anything patched into the main inserts).
  2. There is a tap at the input of the tape deck by selecting “2-Track” and set “Monitor” on the tape deck accordingly. With that, we get everything that is sent to the tape deck (including things between console and tape deck).
  3. Same goes for a tap from the third head of the tape. With that, we get the true end result.
  4. There is no tap before the main inserts, so no way to directly listen to the signal before it gets processed through the inserts.

20171221_205326_RoßtalerwegThere’s workarounds, of course: by switching individual processors into bypass, I can act “as if”, but (and that is important) not while I’m printing a master tape.

Looking at the DAW experience, this isn’t that bad. You typically only get one point on the 2bus, and that is the end of it. If you want to listen to something else, you need to bypass individual elements. The only exception I’m aware of is WaveLab 9 where you have the possibility to tap the signal anywhere in the chain. Still, I’m gonna turn the degree of freedom we have into a conditional weak requirement:

If it is considered crucial to be able to monitor the influence of specific processors also during the actual printing, they should go after the mains.

Summary

We will now trade the position of the limiter and the EQ/compressor combo independently (the limiter always goes after the other two), namely either in the inserts or between main out and tape, based on the results above.

Trade Item Comp/EQ: Insert Comp/EQ: After Limiter: Insert Limiter: After
Bal/unbal O O O O
Lvl Control + ++
Signal Taps (-) (+) O O

Decision

The limiter will go after the main out.

The other things will at first go into the main inserts.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s