Calibrating the signal path

Calibrating the signal path


For you to achieve an accurate mix, you need to calibrate your signal path to show a consistent value throughout the signal path.

The signal strength at the input stage should match, exactly, the signal strength at the output stage. This is called Unity Gain.

What this basically means is that if you input a value of, say, 3dB, then you should see and hear that same value right through your signal path to the output, 3dB in, 3dB out.

If you input that 3dB value as a synthesizer gain (synthesizer’s output level) through your mixer’s input/s, then through the master stereo outs (or sub groups) of the mixer, into the sound card, and finally through the master outs in your sequencing software, then you should see and hear the same value of 3dB.

This is what you would think, right?
Please read my tutorial on sound and it’s measurement.

It isn’t that simple I’m afraid, and the reason being is that we have different standards in the audio industry and as far as analogue signal levels are concerned, though, there are only two to worry about: +4dBu and -10dBV, respectively the professional and semi-professional standards.
But what do these levels actually represent?

The reference point in any decibel scale is always 0dB (please read my tutorial on sound and how it is measured) and a suffix letter is used to denote the chosen standard.

In an analogue mixer, a test level signal of 0VU on the output meters means that the main outputs should measure +4dBu. That is the pro level standard, which means we can align input and output levels to exhibit unity gain throughout a signal chain. In other words, you can pass signals between equipment and know that you won't overload anything or disappear into the noise floor.

The semi-pro level standard of -10dBV was adopted for unbalanced signal interfaces using much simpler (or cheaper) circuitry. The standard -10dBV level equates to about a quarter of the voltage of the professional +4dBu reference level, or almost 12dB lower.

Notice that I am avoiding using any math or physics to explain these measurements and standards. This tutorial has been written with the layman in mind, not the technician. If you want to mesmerize yourself and read the technical specifications and the math, then there are some excellent articles on this subject.

For the sake of our calibration chapter, I am only interested in supplying you with working figures.

Back on subject.

Most professional systems are designed to handle peak levels in the region of +22dBu. +28dBu is very good, while +18dB is pretty standard on budget equipment; therefore, working off the pro nominal signal value of +4dBu, with the maximum peak level being +22dBu, we have 18dB of headroom in the system.
This is simple to calculate, and always keep that simple calculation in mind when you buy a piece of gear, and the manufacturer boasts it’s spec claims. You now know how to calculate the headroom required on any system.

How does that equate to the reading on the mixer?

Well, you have to be aware that almost all analogue mixers, that follow a VU metering system, never show the ‘true’ scale peak value, but the average.
What this means in the real world, is that although the VU meters will show, say 6 dB above 0, the actual figure is well above that, usually by as much as 6 dB. This is because VU meters are not instant metering tools, and react to average values. For music hitting the +10 LED on peak-reading meters, the true signal peaks will be reaching the +20dBu mark (+14dBu plus 6dB overshoot), which is only 2dB below clipping in a typical system.

A mixer will often have enough headroom, and usually around the 10 dB mark, for the user to be able to ‘drive’ the signal past 0 dB and still have no distortion. Of course, gain pushing on analogue mixers is a technique used for getting more ‘warmth’ out of a mix; by driving past 0 dB. Another funky tip, but alas, shows how old school I am.

Now let’s look at the digital side of this.

Digital systems cannot record audio of greater amplitude than the maximum quantising level (please read my tutorial on the Digital Process). The digital signal reference point as at the top of the digital meter scale is 0dBFS, FS standing for 'full scale'.

Since analogue equipment provides around 18dB or more of headroom, it seems sensible to configure digital systems in the same way.

In the US, the adopted standard of setting the nominal analogue level is; 0dBu equals -20dBFS, thereby tolerating peaks of up to 20 dBu. In Europe, 0dBu equals to -18dBFS, thereby tolerating peaks of up to +18dBu.

This all sounds complicated but all you really need to be concerned with, as far as the digital world is concerned, is that we have a peak meter scale of 0 dBFS. Beyond this and you have clipping and distortion. Bearing in mind that the industry standard for 'Red Book' specification for audio CDs insists that material should peak above -4dBFS, then you can see why 0dBFS peaking is crucial.

In the digital domain, you can kiss gain pushing goodbye. 0 is max, that is Gospel.

The problem for most semi pro and project studios is that the vast majority of A/D converters are already adjusted to accommodate the headroom, as discussed earlier, according to the international standards. Using +4dBu as the standard, this will produce a -16dBFS digital signal. So, the analogue mixer’s peak levels will register +12dBu (+8VU), but will only achieve peak digital levels of about -8dBFS. Without any form of dynamic gain boost, this level will sound too quiet.

This is where calibration comes into the equation.

The accuracy of calibration from A-D is crucial, because any deviation either side of the equation, will have a negative affect on the accurate representation of the mix.

I can honestly say that out of every 10 studios I visit, home or semi-pro, 8 are not calibrated. How often have you created a mix, then played it in your car/home hi-fi system, only to find it is either too quiet or so loud that it distorts?

Another advantage of signal path calibration is that of ‘balance’. By going through the calibration process, you will invariably sort out any bias problems that you may be experiencing with the stereo imaging. Bias of either side of the stereo field is as damaging as an inaccurate signal path value.
What we are trying to achieve is a clean and strong audio signal, equal in value right throughout the signal path, and showing no bias to either side of the stereo field.

This may sound complicated but in practice it is actually quite simple.

The procedure.

If you have a mixer’s outputs connected directly to your sound card’s inputs, then you need to start right at the input stage of the mixer.
If you do not have a mixer and are going directly into the sound card’s inputs, then follow this procedure as well, but substituting the sound card’s inputs as the direct input stage.

You need to input a line level signal into one of the input channels on the mixer. I always recommend a constant, non fluctuating signal, like a sine test tone at 1 kHz, or any sound that is constant, sustaining and not dynamically fluctuating.
Avoid sounds that have variety in their waveforms, like drum loops or evolving pad sounds etc. We need a constant single level input signal, like a sustaining sine bass sound, or a constant raw waveform, but not noise.

But first, you need to ‘flatten’ (flatline) the mixer. This basically means that you turn all faders down, all pan pots to centre, all auxiliaries and inserts to off/zero, remove all EQs by depressing them or turning them down to centre where there is no cut or boost, depress phantom power, and finally, turn the input gain knob on the mixer’s channel, that you are inputting the tone through, to zero.

We now have a flatlined mixer.

Most mixer faders start at 0 for Unity, and can be moved down or up. We need to put the channel’s fader at 0 and the mixer’s main stereo outs faders to 0. We are trying to achieve a signal value of 0dB, because we know that equates to +4 dBu (pro standard).

Make sure your monitors (speakers) are connected to the mixer and on, and that your mixer’s monitor outs (control room or main mix or master outs etc.) are on and at a level that you can hear the monitors.

Now start to raise the input knob (trim) on the input channel until you have a level showing on the meter or LED. Your input signal should be bang on at 0, and the master outs metering should be bang on 0VU. Do not adjust the faders for the input channel and master outs. You need to set your input level using the input channel’s gain/trim knob.

This is called Unity Gain.

We now need to calibrate the sound card..

Check the control panel of the software that came with the sound card and check to see what the level is going into the computer. The control panel will probably have faders for controlling levels going into and out of the sound card. Select the ‘audio in’ fader and set this to 0, as you did with the mixer.

We know that in the digital domain the input level should be -16 dBFS, as discussed earlier.
You need to decide what level of adjustment you need to make.

I prefer to use a value close to commercial CD standard, of about -2 dB.

For me; I adjust the input level to around -14 dBFS, which allows me the headroom to peak at -2 dBFS. I do this because I know I can push the mixer to + 10 dB before I experience any distortion, and in the digital domain that would be my peak value of -2 dBFS. In other words, I am accommodating the headroom with a setting of -14 dBFS.

If you wanted bang on 0 dB, then you can adjust the input on the A/D (soundcard) to be exactly that, but you run the risk of clipping beyond 0 dBFS in the event that the material being input peaks beyond 0. You need to calibrate your system based on what you’re A/D is doing in the digital domain.

It’s always good practice to record and test a signal after you have completed the calibration. If you find that your recording is too loud or too quiet, then you need to adjust the input level accordingly.

Now we need to sort out the replay end of things. This is the D/A setting. We know that 0 dBFS in the digital domain equates to +20 dBu on the analogue mixer (using pro standard of +4 dBu). That is way too loud and will distort the signal. If you have control over the D/A output, then setting this to 0VU on the analogue mixer is important for proper calibration.

Commercially recorded music needs to be aligned by using the following signal: -8dBFS to align with +4dBu or 0VU.

Personally, I prefer to have the same input and output values right through the A/D D/A, so that unity is shown both at input and output.

If there is a discrepancy at the input or output stage, then adjust the sound card’s respective gain knobs, but checking with the control panel to make sure there is no clipping etc.

Once you have set up the right listening environment and calibrated your signal path, then you are in a position of strength.

Hearing exactly what is being mixed, right from input to final output, in a room with no bias towards certain frequencies, will allow for a much more accurate and precise mix. It will also make the whole mixing experience much more enjoyable.

I often find that mixing in the right environment, and with a calibrated signal path, affords me more creative ideas as I know I am mixing from a position of strength, and do not need to worry about the mix sounding one way in the room and another way on a hi-fi system, or in my car.
This confidence makes me more assertive in trying different things out. It also saves me so much time in having to correct bad sound representations that I allocate the time saved to trying out different versions of the same mix.

Excerpt taken from Mixing Simplified e-book.
Grab this e-book here.


Eddie Bazil (Zukan)