Posted March 6, 2006, 10:33 a.m.
By Gary Bromham
Performing a lead vocal is arguably the toughest job in the recording studio. This in turn puts more emphasis on the importance of capturing and recording the vocal performance as perfectly as possible. Vocalists often tire easily and generally their early takes tend to be the best (before the thinking and over-analyzing takes over!)
Usually, and in a very short space of time, an engineer has to decide which mic, signal path (preamp, compressor eq etc) to use, set the correct level for recording and headphone balance, create the right atmosphere for singing and generally be subjected to, at best, minor grunts, at worst verbal abuse until the penny drops! Vocalists are a sensitive bunch and need nurturing, cuddling and whatever else it takes to make them feel like a supertar!
During this article I shall attempt to set out a strategy for accomplishing these goals and maybe throw in a tip or two I’ve picked up along the way to assist in capturing the perfect take.
Selecting a Microphone
Microphones come in all shapes and sizes but a basic understanding of how they work will help in any assessment of which one we choose.
All microphones work in a similar way. They have a diaphragm (or ribbon), which responds to changes in air pressure. This movement or vibration is in turn converted into an electrical signal, which is amplified to produce a sound. This is very simplistic but essentially the basic science behind making a sound with a mic.
There are three main types of microphone to choose from.- Dynamic, Ribbon and Condenser.
Dynamic microphones are generally used for more close miking purposes such as drums or guitar cabinets, their sound is usually more mid-range focused and they can cope with higher sound pressure levels (SPL’s)
Condenser, or Capacitor mics, as they are also known, are more sensitive to sound pressure changes. They also tend to have a greater frequency response or dynamic range than dynamic mics. For this reason they tend to be the de-facto choice for vocals. Condenser microphones require a power source, called phantom power, to function. This is needed to power the built in preamplifier and also to polarize (power) the capsule. However this may not always be the choice. Bono from U2 for example likes to use a Shure SM-58 dynamic mic as it allows him more freedom to move around or perform the vocal as if in a live environment! A condenser mic, due to its sensitivity, might prohibit being held in the hand due to noises or indeed it may have too greater frequency range!
Ribbon mics are the curved ball here, so to speak, as they are often richer in tone to both our alternatives. They are softer or more subtle; they tend not to have the hyped top-end of condenser mics and unlike dynamic mics are very sensitive to SPL changes. For this reason they have to be treated with care as the ribbon will not tolerate excessive movement from either loud sound sources or being thrown around! They also generally have a much lower output level to the other two and subsequently need more gain from a pre-amp.
Here’s a short snippet of a vocal I’m currently working on recorded with a Shure SM58:
The same recorded with a Neumann U-87:
And with a Coles 4038 ribbon mic:
When comparing models there are a number of important specifications we need to consider.
When we look at the frequency response of the vocal mic we select will it sound flat or natural or will it boost certain frequencies? It is often preferable, particularly with vocals, for a microphone to enhance or accentuate certain frequencies, which suit a particular singer. Check out www.microphone-data.com for a detailed look at different mics specs. I love this site and spend hours trawling through the pages…does this mean I’m sad and need to get a life?
Sound Pressure Level (SPL)
How much dynamic range or level can the microphone cope with? This is the difference between the maximum sound pressure level and the noise floor or in basic terms the range of usable volume without distortion at high level or noise at low level. Dynamic mics are generally much better at dealing with loud source material than the condenser or ribbon variety.
Noise Floor or Noise level
How loud is the background noise created by the microphone itself? Obviously for someone who sings rock music this is less of an issue than some body who sings ambient jazz. As a rule of thumb Capacitor mics are more adept at capturing subtleties and nuances than dynamic mics.
Scientifically this is a measure of how efficiently a microphone converts sound pressure changes to control voltage or electrical signals. This basically is how loud the microphone is capable of being. You remember I mentioned earlier that Ribbon mic require a preamplifier with lots of gain to get the correct level to feed the mixer or recorder-in our case ‘Reason’.
Don’t worry…nothing to do with global warming!
Our final consideration in choosing a microphone is the pickup pattern or as it is more commonly known the Polar Pattern. On a circular graph this is a representation of which direction a mic picks up sound. The diagram illustrates how this works.
Fig 2. The diagram shows three basic polar patterns. All other patterns are variations of these. The blue circle is an Omni pattern, the red circles show a figure of eight and the green line shows the cardiod.
There are essentially three basic patterns for us to consider when understanding where the mic will pickup sound:
As its name suggests, the microphone will pickup sound equally from all directions. Useful if you want to record all the ambience or space around the source.
Otherwise known as (and as its name suggests) ‘heart-shaped’ Picks up the source mainly from the front, while rejecting most sound from the sides and rear. The advantage here is that the microphone captures only the source that it is pointing at. Hyper-cardiod is similar and often cast in the same category. These simply have a narrower field of pick up than the normal cardiod and are very well suited for singers where more isolation is required or else where feedback is a problem.
Also known as ‘figure of eight’. Here sound is picked up equally from the front and rear, whilst signal from the sides is rejected.
Generally speaking there is no rule as to which type of microphone or which pattern we should use when recording vocals although most engineers tend to veer towards a condenser mic and use a cardiod pattern. There is good argument to suggest that an omni pattern is the ultimate setting but this poses a further question relating to the recording environment, which I shall touch on, later in this article.
In summary our checklist when choosing a suitable microphone should look something like this:
- Consider the frequency response. Is it flat and will it therefore produce a more natural result or does it boost particular frequencies and thus enhance our vocal sound?
- Check the polar pattern. Does it have the pick up pattern we require?
- Check the sensitivity. How much gain will we need on our preamp to get the required level for recording?
- Check the dynamic range and the noise level. Can the mic handle the softest and loudest levels for capturing the vocal performance?
Practical tips and further considerations
It is generally beneficial to use a shock-mount when recording vocals. This prevents low frequency sounds from getting in to the microphone.
Fig 3. Hi-pass filter in the Reason eq section, set to take out any unwanted noises below 80Hz.
It is also a good idea to use a pop shield. Often when a singer stands too close to the microphone sudden puffs of air such as ‘p’ and ‘f’ sounds produce unwanted noises. Pop filters can be bought ready-made to prevent this but the more resourceful amongst us have sometimes resorted to the DIY approach and stolen a pair of our wife’s (or husbands!!!) stockings and stretched them over a wire coat hanger to achieve a similar result.
To help with both of the above problems we might also try a hi-pass filter. If the microphone does not have a dedicated switch then we could use the filter in the eq section such as the one in the channel strip in Reason.
The Proximity Effect
Nothing to do with over-use of garlic in cooking and need for breath fresheners!
As we get nearer or further away from a microphone the bass frequencies increase or decrease accordingly. Typically a cardiod microphone will boost or cut frequencies around 100Hz by as much as 10-15dB if we move from 25cm to 5cm. and back again. This phenomenon is known as the ‘Proximity Effect’. We can use this to our advantage if the singers mic technique is good, producing a richer, deeper and often more powerful sound that pop or rock singers really like! Radio DJ’s or announcers have been using this technique for years, particularly on the late night luuuurrrrv show!
Unfortunately the use of this effect requires the vocalist to maintain a fairly consistent distance from the mic and for this reason it is often more desirable to select a mic which has less of this proximity effect. As a generalisation condenser microphones are better at this than dynamic mics. Here are a few audio examples demonstrating this principle.
Vocal recorded using a Neumann U-87 recorded at a distance of 3cm:
At a distance of 18cm:
Finally at 60cm:
The Tube Effect
A ‘tube’ or ‘valve’ microphone uses a valve as the preamplifier for gain rather than a conventional solid-state (usually FET) circuit. Most early condenser microphones such as the Neumann U-47 or the AKG C-12 employed this circuitry, at least until transistors were invented. The tonal characteristic is often warmer and more pleasing to the ear, the sound is however coloured and not always suitable for every singer. In reality the tube is adding a small level of distortion and if overused can sound muddy and unfocused!
The sound of the classic AKG C-12 Valve microphone:
Something people often overlook and underestimate is the effect of the room or recording environment on the sound of the vocal. Indeed you can be using the best microphone in the world and still obtain an awful sound if the acoustic space is reflective or badly treated. To an extent our ears are able to block out or ignore deficient room acoustics whereas the microphone only records what’s there! An omni-directional microphone will accentuate this whereas a hyper-cardiod will, to an extent, minimise it. Unfortunately, not all of us are blessed with perfect recording environments all of the time and often have to adapt or improvise with the set of conditions we have.
Reflection filters have become very trendy these days with the advent of bedroom studios. The SE reflection filter is one I use personally at home and highly recommend. Failing on this, duvets, carpets or anything absorbent, will help in alleviating the situation. In summary it is often not your mic that is the problem but the environment in which it is being recorded which is the real problem. Save some money, get down to Ikea, and buy a couple of new duvets before spending another $1000 on a bespoke microphone!
The headphone mix, after microphone selection, is probably the most critical part of the recording process for a vocal. We can save ourselves hours, not to mention several tantrums, if the balance is good for our singer. Some vocalists also like to sing with one side of their headphones off. In my experience this can be an indication that their headphone mix isn’t completely right. Singers also tend to have their ‘cans’ too loud because they believe they can’t hear themselves when in reality they are not relaxed or are hearing themselves incorrectly. When pitching is flat this can also be a tell tale sign of headphones being too loud, the reverse being true when the singer is performing sharp! (‘ I should point out at this stage that ‘cans’ is on the list of banned word in certain studios!)
It is often a good idea to setup a separate headphone mix or, cue mix, as it is sometimes called for the singer. This is useful for a number of reasons. If we want to tweak the vocal whilst the singer is singing, but without them hearing us do so, we need to set up a separate balance from the one we are listening to in the control room. For this we simply use one of the sends in the mixer in Reason. We route the output from one of the sends in the master section to the hardware interface, which in turn feeds our headphones via dedicated outputs on our soundcard. Provided we have the Pre button engaged, to the left of the Send knob, when we solo a channel in the mixer the vocalist will still hear only their individual cue mix.
Fig 4. Shows Send 8 being used as a cue send for the vocalists headphone mix. Note the send is sent Pre-fade so that whatever changes we make to the channel (level, solo mute etc) do not affect what is heard in the cans. Also note that in the Master section we can monitor the FX send via the control room out.
Fig 5. Here we see the back of the Master section where send 8 is routed out of Outputs 3-4 of our soundcard via Reasons hardware interface.
Time is of the essence
Capture that take before it’s too late! Singers have a tendency to over-analyse or be over critical of their performance. Often the first thing they sing, before the inhibitions set in, are the best things they sing. My strategy is to record everything. It is after all much easier to repair a less than perfect vocal sonically, even if the compressor and pre-amp weren’t set up perfectly, than it is get the vocalist to repeat that amazing performance.
Singers often like to sing with reverb. This is in itself okay but not if it’s at the expense of pitching and timing. It’s harder to find the pitch of a note if all you’re hearing is a wash of reflective sound. Isn’t that what we spent all that time trying to eliminate when we created the recording environment? Not exactly, no, sometimes vibe is an important factor but there is a happy medium here!
Selecting a preamplifier, and if necessary a compressor, is often almost as difficult as choosing the right mic. Whether you are using a Neve 1073 or an Apogee Duet as your preamplifier the principle of setting up and recording a vocal is the same. Increase the gain on the pre-amp until you start to hear a small amount of audible distortion or see slight clipping and then reduce the level by 5-10dB.
I have often heard it said that if we select a valve microphone for our vocal then the preamp and compressor might be better suited to being solid-state. Too many valves in the chain can often add too much colour! Personally, in my chain, I like a valve mic with solid-state preamp followed by a valve compressor.
We only really need a compressor (or more likely a limiter) if we have a particularly dynamic vocalist. An LA-2A opto or 1176 FET style is ideal if the budget allows! Be very sparing as compression if set incorrectly cannot be undone! The M-Class compressor can be set up to behave in a subtle way, conducive to controlling level fluctuations but not squashing the sound.
Fig 6. A typical compressor set-up for recording vocals. Moderate attack and release settings ensure a relatively inaudible effect on the input when recording a vocal.
The Hardness Factor!
An interesting exercise when evaluating different microphones for different vocalists is to rate them on a scale of 1-10 on a hardness factor. A Shure SM-58 dynamic mic might get an 8 while a Rode NT2 condenser mic might get a 4. When selecting the microphone, we also give our singer a rating as well. A hard sounding voice gets a softer microphone whilst a more subtle vocal may require a harder sounding mic.
Whilst writing this article I have been conscious of not being too preachy! These are only guidelines to recording a vocal and often the great thing with recording is breaking rules. A basic understanding of how the microphone works is helpful but the single most important thing is getting the atmosphere right for the vocal to happen in the first place. Many great vocal performances have been captured with strange microphone selections and singers insisting on stripping off in the vocal booth to get the vibe! Don’t question it, just remember…record everything!
Gary Bromham is a writer/producer/engineer from the London area. He has worked with many well-known artists such as Sheryl Crowe, Editors and Graham Coxon. His favorite Reason feature? “The emulation of the SSL 9000 K console in 'Reason' is simply amazing, the new benchmark against which all others will be judged!”