|Let's jump right in and start by defining a few of the terms commonly used when speaking audio.
Frequency Sound is created when an object is somehow excited (by being struck, or plucked or bowed, for example) into vibrating. These vibrations then radiate waves of pressure that are alternately above and below the ambient air pressure.
Take a look at figure 1. The horizontal line at 0 represents the ambient air pressure. One set of waves (one wave above ambient pressure, followed by one wave below ambient pressure) in one second is called a cycle, or Hertz (Hz.). Ten cycles in one second is 10 Hz. One thousand cycles in one second is 1,000 Hz or 1 kilohertz (kHz). The higher the frequency, the higher the pitch.
Most humans can hear sounds in the frequency ranges between 20 Hz and 20 kHz.
Most human voices produce sounds in the range of 100 Hz - 3 kHz, and an orchestra can produce sounds in the 40 Hz 12 kHz range. AC current in the U.S. is 60 Hz, (remember this, it could come in handy in post!).
Low frequency sound waves are longer and travel farther than high frequencies do. That's why you hear the bass in little Jimmys car stereo from farther away than you can the rest of the music. High frequency sound waves tend to bounce off obstacles rather than bend around them, and tend to dissipate their energy sooner.
Level Okay, so we have our pressure waves radiating into the atmosphere, and we know that the speed in which they alternate is called the frequency. How much the waves vary above and below the ambient pressure is called the sound pressure level, (SPL) and this is expressed in decibels (dB). Decibels can also describe the level of electrical signals as well as acoustical sound pressure levels.
If we look at the vertical, or amplitude, side of figure 1, we see that this sound wave peaks at 2 dB above and below the ambient pressure. The higher the peaks, the louder the sound. The threshold of hearing is essentially 0 dB. A quiet house would fall in the 2530 dB range. An average office, 5055 dB. A car, 8590 dB. Limp Bizkit in full bore, about 110 dB. The threshold of pain is about 125 dB. Most people can hear sounds over a 120 dB range without pain.
It should be obvious that a rock concert is much louder than a car, but because of the logarithmic nature of the decibel, a doubling of the actual sound pressure only results in an increase of 3 dB. Let's say a man speaking normally would produce a level of 60 dB. If he began to shout so that the sound pressure level actually doubled, then that would result in a reading of 63 dB. Confusing? You bet.
One more thing before we leave this quagmire. The inverse square law tells us that in a free field (outdoors) a doubling of distance from the sound source results in a 6 dB drop in level. You will see why this is important when we start talking about micing technique.
Gain structure It is very important to understand the concept of gain structure. Headroom can be loosely defined as the amount of gain over the normal operating levels before the system goes into distortion. Most semi-pro camcorders are notorious for having very little headroom, or room for error, so proper gain structure is critical. A gain stage can be a pre-amp, a fader, or an amplifier. When a gain stage is operating at unity gain, it is neither raising nor lowering the level of the incoming signal. A common mistake is to have one gain stage set too low, requiring another to be set too high. This gain mismatch introduces noise into the signal, usually in the form of hiss. A signal to noise ratio (S/N or SNR ratio) is the level of your signal compared to the level of noise inherent to your system, and is expressed in dB. When you have a low gain stage set early on in the signal chain, then your signal is not much louder than the noise. When you need to boost levels further on down the signal path, you will also be increasing the noise. This shows why it is important to keep the S/N as high as possible throughout the signal path.
Mic level vs. Line level Microphones produce a very low level signal, typically around -40 dB. This level is too low to be used by most gear, so we use a pre-amp to increase the gain to line level, which is typically +4 db. A rule of thumb is, if a signal is coming straight out of a microphone, it is mic level; if it is coming out of a mixer or a box, chances are it is line level. Exceptions to this are some wireless mic receivers and field mixers, which allow you to select line or mic level output.
Alright, now that we understand some of the basic audio terms, let's talk about tools. What will you need to get all those frequencies and decibels onto your tape and into your computer? The obvious place to start is with the...
Microphones Microphones most commonly used by videographers can be divided into two groups; dynamic mics and condenser mics.
Dynamic mics do not need a power source to operate. They are less sensitive, which means they can often be used to capture louder sounds without distorting. They are unable to capture as wide a range of frequencies as condensers. They tend to be very rugged and less expensive than condensers.
Condenser mics need to be powered to operate, either by +48V phantom power supplied by the camera or mixer, or a battery. They are more sensitive, which means they are more able to pick up softer sounds and are more accurate. They can capture a wider range of frequencies, are more fragile and susceptible to humidity and moisture, and can be much more expensive than dynamics.
Whether dynamic or condenser, mics are designed to pick up sound in different ways, which is determined by their pickup, or polar patterns. Omni-directional mics pick up sound equally from all directions. Directional mics, on the other hand, pick up sound from one direction better than others. This direction is called the axis of the mic.
Let's take a look at fig 2 below:
As mentioned earlier, the omni-directional mic (fig. 2a) picks up sound from all directions equally. Mic 2b picks up sound in a heart shaped (cardioid) pattern around the axis, and rejects sounds coming from the rear of the mic. The supercardioid mic (2c) picks up sound from the front as well as the cardioid, but rejects more of the sound from off axis. This better side rejection comes at the expense of a tail of sensitivity behind the mic. A hypercardioid mic is an exaggerated version of the supercardioid. It should be noted that these polar patterns will vary mic to mic, manufacturer to manufacturer, and according to what frequency the sound source is.
If you are still awake, let's talk about what form factors these mics are available in.
Handheld mics can be dynamic or condenser. They can have any polar pattern. Two common handheld mics are the Shure 58, which is a dynamic cardioid, and the EV 635, which is a dynamic omni-directional. These mics are intended to be used very close to the mouth, usually not more than a foot away.
Lavalier mics are very small mics that are usually clipped onto a persons clothing. They can be dynamic or condenser, but the condensers are much better. They can also be either omni-directional or cardioid, but the omni version is much more common, and sufficient for most applications. If the cardioid lavs shift just a few inches, your talent will be off-axis, and that ain't good. Sennheiser makes a very good omni condenser lavalier, the MKE-2. Sony makes the ECM-77B, also a good omni condenser lav.
Shotgun mics are a special animal. These are condenser mics, often supercardioid or hypercardioid elements combined with an interference tube that improves the directionality and reach over that of a standard cardioid mic. (See figure 2d) They can be mounted on a camera for natural sound recording, or placed on a boom stand, or a handheld boom pole operated by a, you guessed it, boom operator. Sennheiser makes the workhorse of shotguns, the ME66/K6. Audio-Technica makes a great inexpensive shotgun, the 835B.
Before we move on to cables and connectors, let's spend some time on wireless mic systems. All of the above microphones can be made to work wirelessly by connecting them to a wireless transmitter, which transmits the audio signal to a receiver, which is connected to the mixer or camera. Wireless can come in VHF or UHF frequency bands. UHF is less susceptible to RF interference. They also come in frequency-agile or fixed systems. Frequency-agile means that each transmitter/receiver has more than one frequency it is capable of using, a fixed system has one only. If you are getting RF interference on one frequency, you will like being able to change instantly. If you have a fixed system, you are dead. Another option is diversity or non-diversity. A diversity system has two discreet antenna systems in each receiver, and the receiver automatically sends the stronger of the two to the outputs, which means a more stable signal. Non-diversity systems have only one antenna system.
Nowhere is the phrase You get what you pay for more applicable than in the world of wireless mic systems. If you are a hobbyist, or live in a rural area, or just like living on the edge, then a fixed frequency, non-diversity VHF systems may work for you. But in general, a frequency-agile, diversity, UHF system is recommended, or at least as many of those three features as will fit in your budget. And that means bucks. In the world of pro field production wireless systems, Lectrosonics is king. Sennheiser makes quality wireless systems also.
But, you say, "I only have $500, I can't afford a Lectrosonics and I really must have a wireless system." Well, if you must, you must. Save your pennies for a little longer and then check out the Sennheiser Evolution series.
Now, there are probably some of you thinking to yourselves that your VHF, or non-diversity, or fixed frequency (insert favorite brand here) systems work great. If it works for you, wonderful, but Ebay is full of sub-$500 wireless systems that can't handle the requirements of a working professional. Avoid them if you can. End of rant.
cables and connectors. There is one important concept here that should be understood, and that is balanced vs. unbalanced signals. In an unbalanced signal path, there are two wires, or conductors. One carries the signal; the other is the path to ground. In a balanced signal path, there are three conductors, one for the hot (+) signal, one for the cold (-) and one for ground. The idea here is that the hot and cold signals are identical except that they are 180° out of phase, (see figure 3). Any noise introduced into the signal along the path will appear on both conductors equally in phase. At the other end of the cable, the cold signal is inverted so that the hot and cold signals are now in phase, making the signal twice as loud, and the noise on the two conductors is now out of phase, essentially canceling itself out. This is known as common mode rejection. Works like a charm.
Balanced or unbalanced, cables can be wired with a myriad of connectors, and shooters should keep a bag of adapters to connect to any number of devices. The four types of connectors most commonly used in videography are the Cannon XLR connector, the 1/4 inch TRS or TS phone plug, the 1/8 inch mini connector, and the RCA phono connector.
Let's start with the XLR connector. These connectors and cables are part of a balanced circuit and are used to connect mics and other devices. Most pro mics have XLR connectors, and XLR cables should be used when running a mic level signal more than 4 feet, because of its relative resistance to noise. They are also used to connect different devices in a professional environment.
1/4 inch phono plugs can be used in a balanced or unbalanced circuit and are used to connect devices together. They can be unbalanced mono (tip-sleeve) where the tip is the signal and the sleeve is the path to ground. They can be balanced mono (tip-ring-sleeve) where the tip is the hot, the ring is the cold, and the sleeve is the path to ground. They can also be unbalanced stereo, where the tip is the left (or right) and the ring is right (or left) and the sleeve is path to ground.
1/8 inch mini plugs come in all the forms that the phone plugs come in, and some shotgun mics that are designed to be mounted on video cameras come with these connectors, because many semi-pro video cameras use only mini connector mic jacks. Most pro cameras have XLR jacks.
RCA connectors are used to connect consumer gear like CD players to mixers and consumers amps. These are the same connectors you see on your home stereo.
There are a few other things that would be good to have in your bag of assorted adapters. You should have adapters that will allow you to connect anything to anything. Turnarounds are barrel connectors with the same sex connector on both ends and are used to change the sex of the connector at the device. In-line attenuators, or pads, are used to connect the line level output of a device, such as a mixer, to the mic level input jack of a video camera. I'm sure you recall from our earlier discussion that line level signals are much hotter than mic level signals, and thus need to be reduced to avoid distorting the mic pre-amp in the camera. A pad lowers the level of a signal by a set amount, which is usually marked on the pad itself. A 40 dB pad would suffice; a pad that lets you select from a few different settings of reduction is better. A tone generator can also be invaluable in troubleshooting circuits and setting up gain stages.
Okay, let's set this stuff up and see if we can get it to work. Oh, before we do that, though, we need to spend a few minutes talking about metering.
The two most common types of meters we might encounter are VU meters and PPM meters. Each type of meter is defined by its specific ballistic characteristics (say that three times fast), or the manner in which it operates.
Volume unit (VU) meters indicators react to changes in level very slowly, and are used to measure the average level of the signal present. The indicators in a PPM meter reflect changes in level much more quickly, and are better suited to measuring the peak levels. VU meters are adequate in the analog realm, because distortion occurs in a much more gradual manner, but in the digital world, we are much more concerned with the peak level of a signal, as distortion occurs as soon as the signal exceeds 0 dB and is very unpleasant.
The problem is many meters on prosumer or semipro gear do not adhere to the characteristics of either VU or PPM meters, which makes it difficult to know exactly what you are measuring when you look at a meter. That is why, even in a perfectly aligned system, wearing headphones and carefully listening for distortion is the way to go.
So, what is a good basic audio setup, you ask? One main objective is to get the highest possible level to tape without going into distortion, which will give you the best signal to noise ratio. One way to achieve this is to get your signal to line level as soon as possible. Another objective is to minimize noise, which means balanced cables.
Given these two objectives, connect a microphone to a mic input on a small mixer via an XLR cable (see fig. 3) Then two XLR cables are run from the line outputs left and right on the mixer into the line inputs 1 and 2 on the camera or recording device via XLR cables. Now that everything is connected, the gain structure must be set. Many mixers can produce a 1 kHz tone which is used to align the two devices. You must determine if your mixer has VU meters or PPM meters. If it has VU meters, set the tone level so the meters read 0dB. Then set your camera meters, which are usually PPM or similar, to read -12 dBFS (Full Scale). Why? Remember we talked about the differences in meters and what they measure. Your mixer will be measuring the average level, while your cameras meters will be measuring the peak levels. What you are doing is making up for the inability of the VU meter on the mixer to show accurate peak levels; more room for error, so to speak.
Because of the differences inherent to various types of mixers and cameras, this is just a suggested starting point. If you find that using these settings results in audio that is consistently going into the red on the camera and distorting, try setting the cameras inputs to -20 dB and see if you get better results. If the audio is consistently low, raise the input levels a little. You will have to experiment to find the best settings for your particular setup.
So this will work for one microphone, what if you have two people speaking. Easy. Just connect another mic to the mixer, pan one hard left and one hard right, and you have two mics isolated to two discrete tracks on your camera or recording device. If you have more than two people and only one recording device, then you'll need to gang two people on one track. Use as needed in your NLE.
Not all cameras accept line level inputs, you say? No problem. Get out the little bag of adapters you have and get the 40dB in-line pad. Connect that and any applicable adapter into the mic input. After a little testing and a few adjustments, you should be good to go.
But, what if a mixer isn't in the budget? Now you'll have to connect your mic directly to the mic inputs on your camera. This is going to result in noisier audio because the mic circuits in low to mid-level DV cameras still leave much to be desired. However, the goal is still the same
.to get the most level to tape without going into the red. Unfortunately, there is no magic setting that will work for everyone. You must spend time experimenting with your equipment to find the settings that will work best.
Now let's talk about micing technique. The main point to remember about micing is to get your mic as close to your sound source as possible. Why? Remember the inverse square law, the one that says that doubling the distance from a sound source results in a 6 dB drop in level? That's why. Let's look at an example.
Say you are asked to interview Warren Buffet, and the producer/director wants to do it sitting on a desk in the middle of his busy office. Now, remember that a busy office has an ambient noise level of 50-55 dB. So, you set up a boom with a shotgun on it as in fig. 4. The mic is just outside the frame line, and a foot or so in front of the subject and pointed toward the mouth. (I know, this guy doesn't look like Mr. Buffet, but bear with me.)
Let say that the distance from the mic to his mouth is 1 foot, and at that distance we are getting 65 dB of level from his voice. Now if we move the mic so that it is two feet away from his mouth, we see the level has dropped down to 59 dB, which is getting close to the same level as the general cacophony of the office. Move the mic away another 6 inches and his voice is no louder than the office noise as far as that mic is concerned.
On the Audio Professionals Forum here at the Cow, we get a lot of questions about why does my audio sound echo-ey. The majority of the time, this is why. The microphone is too far away and hears the desired sound from the subject no louder than the surrounding ambient noise. The next question usually is Well, then, how can I fix it. Unfortunately, the answer is, you really can't. This is one of the many things in audio that do not lend themselves to easily being fixed in post. So, this leaves two options
get the mic closer to the subject, or reduce the ambient noise of the room.
The inverse square law applies no matter what type of mic being used. Lets look at a few of the different types again.
Handhelds are good for run and gun, sports reporting, man on the street interviews, and news. They are the best choice in very loud environments because they are designed to be held close to the mouth and pick up more of the desired sound than background noise. A main drawback is that it has to be held, and inexperienced talent tends to wave the thing around like a wand, and that's not good.
Lavalier mics can be used in many situations. They can be placed on the talent leaving their hands free. They can be hidden is situations that require that, but sound much better when placed in the sweet spot, shown in red in fig 5.
If your talent is a guy in a suit, rejoice! Your job just got as easy as it's gonna get. Place the lav on the guys tie an inch below the knot, and you are good to go. If the mans head is going to be oriented towards one side or the other, then by all means put the mic on the lapel on that side, but you cant go wrong with the putting it on the tie.
Women can be more problematic, especially non-professionals, as they tend to wear clothing that is not microphone friendly. The woman in fig. 5b is not bad, but the woman in 5c could be difficult to mic because of the open area where the sweet spot should be is forcing us to put the mic on the very edge of the sweet spot. In addition, the necklace could hit the mic if she moves around a lot. Also, if she turns her head as if to speak to woman b and the mic is in the red spot, then she will sound off-axis. It is always a compromise.
Shotgun mics sound the most natural, can be used in almost all situations, and are invisible to the camera, if you keep the boom out of the shot. Some of the drawbacks are that you need a boom operator if the talent is moving around a lot, and you must be careful to keep the boom and its shadow out of the camera frame.
A word about AGC (automatic gain control). AGC is either the best thing since sliced bread or the bane of an audio professionals existence. AGC automatically rides the audio level of an incoming signal on a camera. Let's say you are recording an interview and the person is speaking and then pauses. During that pause, the AGC circuit is still trying to work, and automatically raises the gain looking for a signal. When the person begins speaking again, the level is much louder than when he paused until the AGC can react and reduce it, resulting in an uneven voice track, possibly even distorted, and you will hear noise and hiss rise and fall throughout your track. Rule of thumb
under controlled circumstances like an interview or voiceovers, turn it off if possible. If you are running and gunning, leave it on.
Sooner or later, most of us are going to be in a situation at a show or event where we are going to need to hook into the house PA system at a venue. Technically, it is the same as connecting the small mixer as we did a few paragraphs ago. Mixers outputs are line level, so connect the mixer outputs to your line level inputs. If you only have mic inputs, youll need the 40dB pad and any applicable adapters.
That is the easy part. Now you have to deal with the person operating the mixer. Keep in mind that this person is not there to help you. He/she is there to operate the PA mixer. The skill level will vary from raw amateurs who got stuck with the job to seasoned pros. The attitude will vary from helpful to get lost, pal, and every permutation in between. Given that scenario, be nice when you ask if you can hook up. Get to the event early. Try to record a rehearsal or two and listen critically to the tape, and ask if you can request changes in your mix. Depending upon which output is used to feed you the mix, it may not be possible without compromising the house mix. Throw the person a few bucks if you can, that makes people a lot more willing to help.
Well, that about covers the basics. The only thing left to do is share a few random tips gleaned over the years.
- It is always better to fix things at the shoot than in post. Listen carefully to the room you are recording in. Can you hear that clock ticking? So will the mic, unplug it. The refrigerator down the hall? The mic will hear it, unplug that also. Traffic noise leaking in through a window? Put blankets over the windows. A fluorescent light humming? Shut it off. Seriously, if you can hear it and you don't want to, deal with it now before you ever roll one frame of tape.
- Which leads us to
.always, always, ALWAYS wear headphones and monitor your audio. Distortion can occur without ever seeing the signal go into the red on the meters, and you will never know until you get into post, and then it is too late. Distortion, excess reverb, and off-mic audio are difficult if not impossible to correct in post, so listen and correct the problem before rolling.
- When running audio cables, never run them parallel to AC power cables, and cross them at right angles to prevent induction hum.
- Read Jay Roses book Producing Great Sound for Digital Video, it is a must read for everyone shooting video who wants it to sound good. And isn't that all of us?
Come visit us at the Audio Professionals Forum here on the Creative Cow if you have any questions, or just to say hi. We're friendly and love to talk about this stuff.
Audio Professionals Forum and
Broadcast Video Forum Leader