Signal Chain Hierarchy

[Mark Nicholson]

Hey guys,

I'm a noob setting up my Model 12 for the first time. I want to be sure I'm setting things up optimally, so please check that this is correct or suggest a better way of doing it.

I'm starting off with one channel, and I'm using a BeatBuddy drum machine as my test source. It's a nice clean steady signal. It's giving me a boom-boom-boomboom-BOOM (1-2-3and4) with most volume on the last beat. I've taken the Left (mono) output from the BB and am inputting it to the Left channel of input 7/8.

I have set the BeatBuddy up to output 100% volume, and I have turned up the gain on channel 7 to just before where the signal LED turns red (the one right at the top of the strip). I see the signal LED be a solid green at about 4 o'clock, and any more gain causes the LED to turn red on the loudest beat.

On the Meter screen I can see that the peak volume is just above the dotted line on the screen. If I increase the gain til the LED turns red, the meter goes well above the dotted line. Where I have it is where the peak signal from the BB is just above the dotted line, and the LED is solid green. The peak is one resolution interval, maybe two, on the meter.

My theory is that this gives the cleanest signal, please confirm if I'm doing it right.

Then down by the channel fader I press the Main button to put the signal on the Main bus. I am listening on Headphone 1, and have set it to 50% so it is at 12 o'clock. I have set the MAIN fader to maximum, +10db Gain, and control the volume with the channel fader so that I just get the orange 6 dB LED to flash on the meter screen.

I am a bit unsure about how to use all the volume controls once I have my signal at a good level. I ave the channel fader, main fader, headphone out fader and gain on the headphones themselves. These last two are set at about 50% each and I'm getting a nice clean linear sound out of the phones with plenty of headroom to go louder or quieter.

Am I doing this right?

Thanks in advance,

- Mark

 

[Mark Richards]

This example in the DP-24/32/SD forum discusses gain staging. Basic concept, you don't want the output of the preceding stages to overload the input of the succeeding stages.

The Model 12 is 100% digital internally. With digital signals, the noise floor is so low that signal to noise ratio is not an issue:

• 16-bits (i.e.“CD quality sound”) can recreate everything between a typical quiet studio and a very loud symphony orchestra.
• 24 bits can store signals from an infinitesimally small noise floor (lower than a very quiet studio), all the way up to a roaring jet-engine volume.

So bit rate = headroom/dynamic range.

Clipping is what to watch out for, and inter-sample clipping (peaks that occur between samples) in particular because they won't show on the meters. If you work at -12dBFS peaks or less on your meters, then you should usually be good to go.

Your faders should always be set at or below Unity Gain (0 dbFS) to assure you always have adequate signal headroom and that the voltage output of your signal is at nominal level.

Also, from Post#11 in the "Tips and Tricks" sticky pinned at the top of this forum:
When tracking, the incoming Channel signal levels should be evaluated using the Channel SIG indicators and the Channel level meters on the Display Meter Screen.

• The SIG indicator will light green when a signal is input (-56dB or higher). If a SIG indicator stays lit red continuously, lower the GAIN knob.
• The Meter Screen display shows the actual levels of the signals being input to the unit and recorded.

The MAIN MIX L/R stereo meters should not be used to evaluate an incoming signal, as the MAIN stereo meters are not in the signal path to any of the MODE options ("Live", MTR, PC).

[While] [t]he MAIN MIX L/R stereo meters are dependent on the MAIN fader level position. The stereo meters only reflect the signal level going post-MAIN-fader to the XLR speaker outputs (as indicted in the Block Diagram).

 

[-mjk-]

My theory is that this gives the cleanest signal, please confirm if I'm doing it right.

What you are doing is more of an analog approach where you're not going for the cleanest signal, but the loudest signal before it clips. With digital recording, this will eventually get you into trouble because your metering can't always accurately tell you what's going on with the peaks.

When I bring signals into my analog console, I make sure that the channel clip lights do light up with peaks. When I bring signals into my digital console, I make sure that the channel clip lights never light up under any circumstances.

This was totally new for me, an old-school analog recording engineer, when moving into the digital realm. On analog tape, if you recorded too low the noise floor would creep up on you later down the line when you had to increase the gain in Mixing and Mastering. No so with digital. As long as the input signal is clean, you can safely make adjustments in the digital realm with no additional noise (I'll qualify that with an "if done properly").

With the Model 12, I would work with -12 as Mark suggested. But if you have any doubts about instantaneous peaks going over 0 dBFS, then lower the level to -15 or even lower. You can always safely normalize the track later.

I give you credit for establishing a good workflow up front. That is an excellent approach.

 

[Mark Nicholson]

Thank you gentlemen. You are right, I do have an analog mindset. I'll digest your replies and post again if I have more questions. Thanks again