Mixer help - level vs gain vs st out

I have a yamaha mg10/2 mixer and need help on using it. I have it hooked up to my sound card and that is working but there are several level controls on it.
1) Level and gain on each channel
2) ST out

and then I have the level on my Omega sound card. to factor in. How do I set these controls for recording and which meters do i read for cliping. They all appear to change the level but what do i set them at to get the best recordings for a mic. I been changing them but don't understand what the gain is suppose to do or the channel vs the st out level as well as the level on my sound card.

There has to be some sort of logic behind this...

-Tom

Oh, Jeez. Ignore Ohhey. He's a nice guy and generally very helpful but right now I think he's either drunk or mocking you.

Welcome to the wooly world of professional audio where you have hundreds and thousands of knobs and faders and switches and blinking lights and meters and LEDs and the ultimate purpose of 99% of them is to make things louder, or to tell you how loud they are.

I am not going to tell you where to set your knobs and faders, but I will tell you some of the principles behind them. There is a very real method to this madness, and if you are not willing or able to wrap your head around it then it is better to get someone else to do the recording, seriously.

A central principle of audio engineering is "gain staging" a.k.a. "gain structure."

"Gain" is a word that refers technically to signal amplification but that is broadly used to refer to all kinds of making things louder or quieter, i.e. adjusting signal levels.

Gain *staging* refers to the art and science of "staging" your signal levels so that you achieve the best overall sound quality-- i.e. you minimize noise and undesireable distortion artifacts, and you maximize the subjective and aesthetic qualities of the sound.

There are several stages of gain in most audio transmissions. For example, let us imagine a singer in a band whose voice is being amplified through a PA system (bear with me here). The singer sings into the microphone, and this causes the mic "diaphragm" to vibrate (the "diaphragm" is typically a suspended piece of cardboard or or some such that moves in and out much like a speaker cone, except instead of generating sound waves, it responds to them). The mic diaphragm is attached to a coil of wire that is suspended around a magnet. When the singer's voice creates alternating positive and negative pressure, the mic diaphragm moves in and out, in sympathetic resonance. The coil of wire gets down and dirty, sliding up and down over the magnet. The magnet attracts the electrons in the coil, and those electrons start to move back and forth as the coil moves up and down over the magnet. These electrons being pushed and pulled generate a very tiny amount of current, and the microphone, in effect, becomes a very small AC generator, sending small pulses of alternating positive and negative charge across that coiled-up wire.

That coil of wire that is pushing lustily up and down over the magnet is then connected through a series of connectors and a mic cable back to a "microphone pre-amplifier," or "preamp" for short. Most mixers have built-in preamps, as I suppose yours does. The purpose of the preamp is to "amplify" that super-miniscule current being generated by the mic coil to a level that is usable and relatively noise-resistant. This is done with a transformer. I am not going to get into the mechanics of how a transformer works here except to say that it is somewhat similar to yet another electrical generator. Basically, the transformer runs a certain electrical charge along one side (the secondary or "output" side, let's call it), and *modulates* that signal acording to electromagnetic impulses from another electrical charge (the primary or "input" side). In this way, the transformer can take a very weak electrical signal, and sort of "map" the modulations onto a more powerful current or a higher voltage. This process is called "signal amplification."

At this point we should pause and talk about why we need to amplify signals, even though we are nowhere close to pushing a speaker yet. You're going to have to bear with me as I continue to alternate between different real-world examples and theoretical concepts, but believe me, this is going somewhere...

All electrical signal is susceptible to noise. Even a plain copper wire at any given time has a certain amount of electrons just sort of drifting randomly around, willy-nilly. If you were to amplify this random current, it would sound like hiss, or "white noise," as it's called. Similarly, through the exact same principles that make transformers and mic coils work, stuff like radio signals and power lines and refrigerators generate "electrical noise." That is to say, electrical currents will create a small magnetic field, and the magnetism from that field will slightly push and pull the electrons in surrounding wires, and this creates electrical current in those wires the same way that the magnet in the mic coil does. 60Hz guitar hum is probably the most common artifact of this phenomenon, but it happens everywhere and affects everything. So all the electrical stuff that is all around us is generating noise that is trying to infect our signal wires, and even if it wasn't, the wires themselves have a certain amount of noise.

So in any electrical circuit, there is a certain minimum level of noise that is always going to be present. It varies somewhat from one system to another, and from one physical location to another, and can be affected by a lot of factors, but it is always there. This bare minimum threshold is known as... the Noise Floor. Remember that. The Noise Floor. This is an important concept. You are about to experience it in all it's glory, because we are going to conduct a little experiment.

Take some kind of speakers or headphones that are readily available and turn them all the way up with no signal playing-- could be a home stereo or a guitar amp or your studio monitors or whatever. Provided that the system is loud enough, you will hear noise. There should be some hiss (random electrons bouncing around the wires), maybe some hum (induced noise from the AC power in your walls or whatever), might be a little radio interference from cell phones or radio towers, maybe some weird space monkeys from gamma rays or sunspots or whatever, but the fact that there is zero signal DOES NOT mean that there is silence. There is no such thing as silence in audio. There is only... The Noise Floor.

Let me say that again: There is no such thing as silence. There is only The Noise Floor.

The reason that we do not usually *hear* the noise floor when listening to recorded music is because we are typically listening to well-recorded music that has been produced in such a way so that the noise floor is much quieter than the "program material" (i.e., the recorded music). In other words, when we set the music at a comfortable listening level using the volume control on our playback system, the noise floor is *so much* quieter than the program material that we cannot hear it, or at the very least, our ears have "adjusted" to the much higher volume level of the music so that our brain kind of filters out the noise unless we are listening for it.

It may seem like I am belaboring the point but this is a very important one to understand and internalize. Think about it and listen for it. This is just the beginning, but it's important. More to come.

Cheers.

I am following so far Yep and always wondered why i got hiss in my headphones..,,thanks....so keep going and also what is headroom...

-ToI

We are closing in on having covered the preamp, and once you understand the gain implications of the first stage, then rest of it starts to fall into place, so bear with me once again.

We have talked about the danger of a too-low signal (the hissy infernos of the noise floor). Now the problems of a too-strong signal. I have so far avoided the use of detailed electrical or mathematical stuff but I reserve the right use them if my ability to translate breaks down.

"Saturation" is a cool word and it makes you sound smart if you use it intelligently regarding audio, and it's the best nonjudgemental word I can think of to cover all the bases of an overloaded audio signal. A signal that is driven past the point where the circuit can handle it it with full fidelity can be described as distorted, overloaded, clipped, "hot," rectified, compressed, and so on. Sonically, it may be good and desireable (guitar distortion, "in your face" vintage vocals, explosive snare drums) or it can be very ugly (digital overs/clipping, farty speakers, fuzzed-out signal).

All of these things are the product of saturated audio circuits. Like a sponge that's full of water and can't hold no more, an electrical circuit can become overloaded to the point where the wires and components are carrying more current or delivering higher voltages than they are really designed for. The conductors get physically hot, the voltage levels lag and flatten out, the audio flattens and fattens, new harmonics are created from the electrical ringing and resonance, and so on. In all cases, at the very least, the dynamics are compressed.

Some devices and components (such as good analog tape) saturate in ways that can be quite pleasing and flattering, and can give a sense of fire and fullness, and a crisp airy articulation to the sound. Some devices (such as digital converters) sound awful when pushed beyond their limits and just immediately turn into some kind of secret government weapon designed to give diarrhea and headaches to whoever hears it. Some stuff saturates well for some kinds of material and poorly for others. Sometimes you may want a fat, firey sound and other times you may want a crystal-clear, smooth and transparent sound. This is the "art" of studio engineering, and how you make these decisions is part of what makes your recordings sound different from mine, even if we both use the same equipment. It's aesthetic preference and judgement calls.

Most studio-quality analog devices such as preamps have a fairly wide and somewhat fungible and frequency- and dynamics-dependent range between where they are completely operating within spec and where they are grossly overloaded and farting out. Indeed, many highly-sought-after "vintage" components are prized for the natural, musical, and organic ways in which they saturate and compress the signal when pushed hard.

So there is often a fair amount of discretion as to where the "best" setting is for any given gain stage. The lower you go, the closer you get to the noise floor, and the higher you go, the more introduce saturation artifacts up to and including ugly distortion and clipping. This is made even trickier by the fact that sometimes the "desireable" effects of saturation in say the preamp stage might actually be stuff that would be better accomplished with say a dedicated compressor later in the signal chain, and by some very insidious and decieving "loudness effects" that I will talk about in more detail later.

For now it is enough to understand that the higher you push your signal levels, the more they will tend to distort and flatten in ways that may be good, bad, or debatable.

You may be wondering why they don't simply make devices that are noise-free and also totally linear and undistorted at all signal levels. Well, the short answer is that they can't (although some high-end stuff makes a very good effort).

A typical VU meter on an analog mixing console might ordinarily be calibrated so that the "average" 0dB level equals a 1 volt audio signal (I am oversimplifying for purposes of illustration, so audio nerds: start a new thread if you want to debate VU calibrations or whatever). Because dB are logarithmic, this means that every 6dB increase in signal level equals a doubling of voltage. So +6dB=2V, +12dB=4V, +18dB=8V,+24dB=16V. That gives about the typical headroom on a good analog mixing console, and 1~16 volts is fairly reasonable range. So far so good.

Here's the problem: let's say we want that console to handle "average" signals of greater than 0dB/1V. Let's say for the sake of illustration that you want to for instance handle a wider range of say 0dB to +12dB (or 4V) with equal fidelity and aplomb, to account for the fact that your singer has no mic technique and his/her levels jump all over the place. So now, in order for our preamp to have the same headroom and fidelity, it has to handle signals as low as 1V (0dB) and as high as 128 volts (12dB plus 24dB headroom) equally well! (remember the logarithmic scale) This is greater than the difference in voltage between a AAA battery and US wall outlet voltage! Moreover, 12dB isn't really even all that wide of a dynamic range for real-world signals.

If we wanted to say create a circuit that handle say a 24dB real-world range of average level, then it would have to have absolute fidelity from 1-512 volts! We're talking about huge wire guages and power supplies and massive industrial components and all of that stuff creates more noise! So we are back to where we started and pushing ridiculously hot levels just to get above the noise floor from all the high-powered stuff that is surrounding our signal path. Even if money and size constraints are no object we still run up against the laws of physics.

From preamp design to construction to engineering practices in the studio, we are constantly threading the needle between the noise floor and signal saturation. There is no free ride, although using higher-quality equipment can give us a little more wiggle room. More to come.

Cheers.

Tom,
Just some notes specific to your hardware while we await the next episode of Yep's excellent dissertation.
btw, yep, I will never look at a mic the same way again.
"That coil of wire that is pushing lustily up and down over the magnet "
On the MG10/2. This mixer has decent, and very quiet Pre-amps. I really like mine.
Typically, setting the preamp Gain to the 12 to 3 o-clock position is a good place to start.
The Level knob should be comfortable around the arrow at the 3 o-clock position. If it's not, adjust the gain. Always watch your meters, and use your ears.
The ST knob controls the main output level. Here again, 3 o-clock should be pretty prime, but adjust as necessary.
The REC OUT and ST OUT send the same signal, at different levels.
I use the REC OUT because my particular sound card(E-MU0404) will accept an unbalanced -10 input.
Your Omega should be ok with the ST OUT +4 balanced output from the mixer. You want TRS cables for this.
Don't get too carried away with the EQ on the Yamaha. It can be somewhat agressive, and it is very easy to overdo it.
Make sure to turn any unused inputs to zero, to minimize stray noise.
Now back to our regularly scheduled program.....