TCRM 11 discussed using an overhead microphone pair to pick up cymbals as well as to capture an overall picture of the kit. Now we’ll explore some of the concepts and techniques for close-miking the entire kit. Then, a few pointers on recording vocalists….
Drum Kits - Overview
The drum set can be a difficult instrument to record since it is a collection of so many sonically distinct parts - numerous instruments being played by a single performer. Complicating things further is the fact that the specific tonal character of any particular piece in the set can vary greatly from kit to kit. This is based on its dimensions, materials, heads, tuning, beaters used, and performance techniques. In addition, these instruments are jammed together closely, yet some may be as far apart as six to eight feet. Collectively, they cover a solo frequency range rivaled only by the piano. These factors can make both microphone choice and placement complicated and time-consuming.
The simplest way to get a good image of the kit quickly in the studio is by using a stereo overhead pair, or a set of room mics (as discussed in TCRM 11). Both of these methods require precise placement and a bit of space from the drum kit to achieve the best overall balance, stereo image, and sense of acoustic space. Finding a compromise mic position to suit each of these criteria often yields a good recording, but severely limits the tonal palette and degree of control during mixdown. To add flexibility in these areas, close positioned accent mics are often used in addition to the overhead pair. If accent mics are added to most of the drums and hi-hats, the overhead pair can be used to concentrate specifically on the cymbals and for capturing more of the acoustic signature of the room.
Adding microphones to individual instruments (kick, snare, hi-hat…) can make it much easier to balance these elements of the kit in the final mix as well as shape their individual sounds. Those who desire the greatest control during mixdown often use one or more microphones on each piece. As more microphones are added, however, the setup becomes more time consuming, complex, and costly (requiring more tracks, microphones, preamps, and cables). Phase issues and redundant signals also increase with the addition of each microphone. This can cause wreak havoc on both tonal color and localization within the stereo field while making balancing the individual instruments quite difficult.
Minimizing comb filtering and leakage requires well-placed microphones chosen specifically for their polar patterns and frequency response. All three of these factors - placement, pattern, frequency response - should be utilized in an effort get the strongest signal from the intended instrument, while limiting the pickup of other pieces of the set.
Also, keep in mind that leakage can also occur between the kit and other instruments playing at the same time. An effort should be made to isolate the sound of the kit by putting it in a separate room, surrounding it with movable barriers (called gobos), isolating each of the other instruments and/or by recording them directly using DIs.
Finally, before placing any microphones, you should already be aware of the type of tonal qualities and image of the kit you’re looking to achieve. As stated in TCRM 11, it’s important to consider the perspective you want of the kit. There’s left to right from the drummer’s side of things, or from the front (audience) perspective. Choose one and keep track of the stereo location of the mics. If you do not, the image of the kit will be unclear, and the tone compromised.
Getting a good kick sound comes down to capturing the right balance of its three main elements: beater, head(s), and shell resonance. While a little of each of these are always needed, the exact mix required for a particular session is determined by musical style, microphone choice/placement, and a heavy dose of personal preference.
A large diaphragm dynamic microphone placed inside the drum can be used to find a number of differing tones. To do this, it is necessary to have either a hole in the front head (the head farthest from the drummer) or, preferably, for the front head to be removed completely. With the front head off, it is a lot easier to position a blanket or other dampening material properly within the shell. This technique is often needed to tighten up the sound by reducing the duration of the modal resonances. It is also easier to make adjustments to mic positioning with the rear head removed.
While requiring a bit of attention as to exact mic placement, this flexible technique also offers the greatest isolation from the other instruments. After dampening and mic placement is complete, a blanket can be draped over the front of the shell, creating a sort of kick drum mini-booth. This both helps keep other sounds from entering the kick mic, and the kick sound from getting out to the other mics.
The exact position and angle of the mic sculpts the drum's recorded sound. As the mic is moved closer to the head, it picks up more of the beater and accentuates the sharp attack of the drum. This lends clarity to the beat. As the mic is moved closer to the head, however, it becomes more sensitive to the particular resonances of that spot along the head. Very small changes in position towards or away from center of the head can change the tone significantly. This is also true of moving the mic around in a circular fashion while keeping the distance from the center (the radius) the same. Generally speaking, a spot around 1/3 to 1/4 of the way in from the sides will offer the most balanced skin tone. A position closer towards the center will accentuate the head's lowest frequencies. Towards the sides, the head will have greater strength in its highest partials (both harmonic and non-harmonic frequencies).
Moving the microphone towards or away from the center of the shell also has a great effect on the pickup of the modes inside the drum. A dynamic microphone placed exactly in the center will only accentuate half of them, while placement closer to the edges will be less discriminating. Again, placing the mic around 1/4 of the way either from the sides or off-center may give more balanced results.
When a kick drum has a hole in the front head, and the drummer wants to keep that head on, a microphone can still be placed within the shell through the opening. Some engineers place the mic slightly outside the hole. In this case, care must be taken so that the rush of air from the hole does not blow the mic's diaphragm. Solutions to this can be found by angling the mic slightly, by moving it further away or off-center, or by using a windscreen or pop filter.
If there is no hole and the drummer does not wish to take off the front head (or if you are hunting for a particular tone) a single mic placed close to the front head can still yield some useable tones - positioning is extremely influential on tone here. If not enough beater clarity can be found, a second microphone can be used near the beater impact area on the rear head (the beater/drummer side). Mix these two tracks together as needed.
Be aware that phase issues between these two mics can cause noticeable interference and frequency cancellations. A simple phase invert on one of the channels may help this, but a more precise solution is to use a very short delay. Without any feedback settings and the mix at 100% wet (this is not a guitar pedal effect), slowly sweep the delay time on the beater channel from 0 to 3 milliseconds until you find the best tonal quality.
The sonic components of the snare drum can be broken down into the stick, head, and shell parts of the sound, and, of course, the snares themselves. These crooked metal strands vibrate against the bottom head and give the instrument its distinctive sound. An appropriate combination of these individual aspects is needed for successful snare drum recording.
The most common method for miking a snare is to place a dynamic microphone about an inch above the head, roughly an inch inside the rim. The angle and polar pattern of the microphone should be used to minimize pickup of other instruments (especially the hi-hat). Since the mic is so close to the drumhead in this technique, the sonic color can be affected greatly by minor adjustments in positioning. When choosing a placement for the microphone, be sure to keep it out of the way of the performer. There’s nothing like a few wacks from the drumsticks to ruin a track (and maybe a mic).
Since the rattle of the snares against the bottom head is so influential in the character of the drum, some engineers also place a mic below. When this is mixed with the top microphone, it is best to flip the phase of one channel. As with the double-miked kick drum, a very short delay on the top head can also do the trick. In this case the delay time would be even shorter, generally less than one millisecond.
In some circumstances, especially on touring drum kits, mics have been placed inside the shell through a tiny hole. These mics can be used in the studio as well and offer great isolation from the other instruments. Since adjusting the exact placement of the mic element is difficult to impossible, the tonal possibilities are more limited. This technique, however, can still yield great results.
To capture the sharp attack and high frequency energy of the hi-hat, a small diaphragm condenser works well. Placing the mic 4 to 6 inches above the outer edge (furthest from the snare), facing straight down, can capture a great sound. This spacing gives the instrument room to move and offers a more balanced tone. If placed too close to the top cymbal (esp. above the interior), a more bell or gong-like tone is sometimes obtained.
The hardest thing about recording the hi-hat is reducing the leakage from the snare. Again, this is where polar pattern and angling can be invaluable. Hypercardioid designs can be especially handy when miking from above as they can be tilted away from the snare at around 30 to 45 degrees, placing it within its weakest pickup angle.
Another successful technique is to place a microphone horizontally just outside the edge of the top cymbal. Care must be taken here to avoid rumble from the rush of wind escaping from between the cymbals they close. Pop filters and windscreen come in handy here. Angling the microphone slightly up or down can also help. Again, use angles and polar patterns to reduce the amount of snare. But don’t drive yourself (and others) crazy trying to be too picky here… there will always be a fair amount of snare sound in the hat mic!
As with the snare, the most common method for recording toms is by placing a dynamic mic close to the top head just inside the rim (and furthest from the performer). Sometimes, a single microphone is used to record two neighboring rack toms by placing it halfway between the rims several inches above the level of the top heads. This is usually done to save tracks and reduce the number of mics needed. Be aware, however, that doubling up in this manner causes you to lose the ability to pan and EQ each drum separately in the mix.
An alternate method for recording floor and rack toms (without a bottom head) is to place the microphone up inside the shell. As stated in the earlier section on kick drum, this will accentuate the lower resonances of the space within the shell. Exactly which of these modal resonances will be stressed is determined by the exact placement of the microphone. Adjusting the mic's position by only an inch or so can change the tone. This method also has the added benefit of offering greater isolation from the other instruments.
Stands, clips, others
It is not always necessary to spot-mike each drum in a kit. Very good results can be obtained with just three mics - two overheads and a kick. This generally doesn't work unless the room sounds very good and the overall setup provides good isolation for the drummer's position. Record a sample snippet and listen back - three mics may be all you need. Sure, you won't have endless mixing and panning options, but if the kit sounds great and natural and each drum speaks clearly, go with it. Of course, the decision to use this method is also be dependent on musical style and the sound desired.
The more spot mics you decide to use, the crazier the tangle of mic stands and cabling gets. You may want to look at the clip-on mic holders that many companies make. Some clip-ons are suitable for a variety of mics, others are available only with an integrated mic that cannot be swapped out for another model. Another draw-back is that that clip-ons don't usually provide much variety for the final positioning of the mic capsule. Also, you have to be careful to be certain that unwanted vibrations are not translated from the drum through the clip into the mic; generally, you want only the vibration that is translated through the air.
Similarly, gooseneck stands can be handy on the drum kit, and especially on the kick, but you must be aware that they can create audible creaking sounds sometimes and/or droop under too much weight and vibration from the loud drum kit.
An overall approach
OK, now that we’ve covered issues of mic placement, mic choice, polar pattern, leakage, phase, comb filtering, and drum shell acoustics it’s time to take a step back. The best way to a good drum recording is by starting with a good kit and performer. While there may be little you can do about the drummer’s skills, you should be sure to tune the kit and lube both the kick drum and hat pedals as needed. Squeaky pedals are not welcome influences in a recording.
As for an underlying thought on exactly how to position mics: a good engineer should know what tone he or she is looking for and be willing to spend time moving mics around to find the best spot to capture that tone. Let your accumulated experience and knowledge guide you, but not get you thinking in terms of hard-and-fast rules. Such rules may lull you into thinking there is a perfect mic choice and position for recording kick drums. Since not all drums, heads, beaters, performers, performances, or musical styles are the same there cannot be a single, exact miking convention.
Get over it.
The basics of miking a vocalist seem much simpler after having miked up a drum kit. For starters, most recordings of singers are made using a single large-diaphragm microphone. Generally the performer is also very close to the mic. Ideally, with a professional singer, you set up a good mic at the right height, with a pop screen, and off you go.
The only variable, with a professional, is the compatibility of the voice and mic. If you don't already know which microphone works best on this singer, set up a loop of the accompaniment track for playback. Have the singer warm up to it and record trial snippets, then swap out microphones as needed. Once you're satisfied, start recording for real.
Professional vocalists will have what's called "mic technique" (also called "working the mic"), meaning they will back off slightly when hitting the loudest notes, and will control their voice in other ways, also. Those are the things that make them true pros. But, life not always like that. So what’s to think about other than which mic to select? Plenty….
The dynamic range of the human voice is much greater than most people are generally aware of. Placing someone in close proximity to a mic brings out this obvious fact very quickly. Not only do singers vary the volume of different parts of the song, even individual words, for dramatic effect, but they also have certain ranges and individual notes which are stronger than others. This can cause the mic, preamp and/or A/D converters to distort when the singer inevitably gets louder. Be aware, singers are notorious for getting much louder as they get more comfortable and inspired by the music. This is a good thing, except when you don’t have the headroom to accommodate it!
There are a few methods for addressing this problem (other than a Valium for the singer). First, be sure to leave the singer plenty of headroom on the preamp and converters. For some microphones this may require applying the pad on the mic and/or the preamp. If the dynamic range of the performance is so great that a large amount of headroom causes the noise floor to be too great for quiet er passages, a compressor may be necessary. (see TCRM 6 on setting levels and TCRM 17 & 18 on compressors). A third approach is to have the singer move further away from the microphone for louder passages, but don't count on them doing this correctly if they aren't already blessed with good mic technique. Having to suddenly concentrate on moving closer and further from the mic, for the first time in their career, will probably distract them so badly that a good intuitive performance will not likely happen.
This does bring up another common issue with vocalists: they move. Once again, the inspiration bug gets them and they start swaying or jumping around. This movement can cause level differentials of up to 17 dB! Even when just moving their head from side to side, the difference can be greater than 6 dB. This can make mixing extremely tricky, to say the least.
There are two major fixes for this problem. First, you could let the singer hold the mic. Then when they move, the mic may stay around the same distance from their mouth. There are a few possible problems with this method. One of the most common is handling noise. The hand and its movement on the mic can create unwanted noises in the audio. Then there’s the idea that this uncontrollably gyrating singer will pay enough attention to where or how the mic is held. This method may create more problems than it solves. The best solution? Get the singer to stand still!
Threads and Bling-Bling
A commonly overlooked aspect of vocal recording sessions is the attire of the performer. Though in our daily lives we have become accustomed to ignoring the sound generated by what we wear, when a singer’s in an acoustically treated isolation booth in close proximity to a sensitive microphone, these sounds can seem to leap out of the speakers. Many otherwise great tracks have been haunted by the swishing sounds of rubbing fabric or the clinking or jingling of jewelry. Be aware of what the singer is wearing. Have them remove jackets and jewelry, spurs, and (of course) cell phones! Leather looks cool, but it creaks.
When a singer gets close to a directional microphone (all types except omnidirectional), bass boost becomes evident. Some microphones have switchable high-pass filters built in to address this. Often, these filters are also found on preamps. It is also useful to note that using these filters will remove some of the low-end mechanical rumble common in rooms, as well as wind noise, handling noise, and minor bumps to the mic stand. They may also aid in setting more appropriate levels.
So why not use an omnidirectional mic? You could, and it will offer a flatter bass response, but be aware that more room tone will come along with it.
Stop Breathing so much!
When someone sings, they naturally expel air… sometimes lots of it! Certain sounds (called plosives) such as “p” tend to push more air than others. When the vocalist is close to the mic, such a strong rush of air can actually blow the mic diaphragm, causing a rumble or pop. While the filters mentioned above can help reduce this noise, it is much better to avoid them in the first place.
Windscreen and pop-filters do a fairly good job of this, but may not always work 100 percent. Angling the microphone slightly may help the air travel around the diaphragm better by making it a bit more aerodynamic. If all of these methods fail, the singer can simply be moved a bit further back.
Another side effect of recording a singer with a closely positioned mic is an accentuation of the more noisy sounds like “s”, “sh”, and “f”. These are known as sibilants. If these are a problem, they can be reduced by:
- using a microphone with less accentuation of the high end or slower transient response;
- moving the singer further away from the mic;
- using a device known as a de-esser. The de-esser is a frequency-dependent compressor which reduces the level of the audio whenever sibilants are present. (see TCRM 18)
Besides the obvious nearby surfaces like walls, ceilings, floors and windows that reflect sound, surfaces like a music stand can reflect sons and cause comb filtering. Absorptive materials on the wall or between the wall and singer can solve that problem. Also, you could try draping a towel over the mic stand and angling the stand and paper so that it doesn't directly reflect sound into the mic.
Avoid the need for page turning during a take. If possible, hang the music on the end of a boom or string a clothes line, high enough so that the singer doesn't have to look down but raises the head up slightly and opens the throat. Of course it is often easier to get a better performance and recording if the singer has memorized the music and lyrics, but that does not always happen.
Microphone choice is, again, a matter of both style and taste. The biggest factors here are frequency response, transient response, and polar pattern. The various frequency responses of different microphone models will either accentuate or reduce different areas of a singer’s voice.
Of particular interest is usually the area known as the presence peak. This is a boost somewhere between 2 and 8 kHz. As well as being an area where some of the more influential harmonics lie, this range contains frequencies called formants that are naturally accented by most voices in vowel production, and which make words more intelligible while lending each person's voice its distinctive tone.
The exact frequencies in question depend on physiological factors such as gender, stature, and vocal technique and the ear is quite sensitive to them. For example, recognizing formant structure is how we can tell the difference between a man and a woman singing at the same pitch.
Transient response will also affect the timbre of the recorded voice. Longer transient times are sometimes said to sound “smoother” while shorter ones may seem more “accurate” or “immediate.”
Polar patterns are chosen to reduce reflections as well as capture certain amounts of the acoustic context of the room. For example, if a closer sound is desired, but a bit more room tone still called for, a figure eight can bring the best of both worlds. It will also still minimize reflections from the side walls.
For positioning and comfort, consider side-address versus front-address microphones if all else is equal. For sight-reading, place the sheet music or lyrics behind a front-address mic since it's easy for the singer to look up and over the top of the mic; if it's cardioid, you'll have the further benefit of minimizing reflections from the paper.
Remember, the right mic is the one that captures the sound you desire.
John Shirley is a recording engineer, composer, programmer and producer. He’s also a Professor in the Sound Recording Technology program at the University of Massachusetts Lowell and chairman of their music department. Check out his wacky electronic music CD, Sonic Ninjutsu, at http://cycling74.com/products/c74music/c74009/
Supplemental Media Examples
Following up on the stereo recording techniques used to record drum kits discussed last time, TCRM 12 looks at spot (or accent) miking and overall miking/recording techniques on drums. This is followed by suggestions for recording vocals. Here are some sound files (and see the pictures as well!) to illustrate what we discussed in the column. You may want to import them into your own DAW to make it easier to jump back and forth between examples to hear the differences.
Special thanks to Connor Smith for helping run the sessions, Kevin Martin who played the drums, and Rachael Bigelow for providing vocal stylings.
Here, a small drum kit was recorded using a variety of microphones and techniques. The kick was recorded with both a Shure SM7B and an AKG D112 on the front head (there was no hole in the head). The snare was recorded using four microphones: Shure SM57s on the top and bottom heads, an Audio Technica Pro37 on the top head, and a Shure SM81 on the bottom head. THe high hat was covered by a Shure SM81 over the top and a Neumann KM184 on the side. Each of the two toms had it's own SM57. There were two stereo microphone setups for overheads using Neumann TLM103s: an XY pair and a spaced pair. One final mic, a dpa4006, was placed further out in the room to capture a mono room tone.
First, let's listen to what all of the microphones at unity sound like: TCRM12_1.wav
By taking a subset of these, the imaging and tone can clear up quickly. Here, the kick is the D112, the hat is the SM81 over the top, the overheads are the spaced pair, the toms are in, and the snare is a mix of the Pro37 on the top and the SM81 on bottom: TCRM12_2.wav
Now, let's make an attempt to balance the instruments in the last example (TCRM12_2.wav) by simply using gain (volume) changes: TCRM12_3.wav
The snare tone can be shaped by inverting the phase of the mic on the bottom. This is done in an effort to phase align the two microphones: TCRM12_4.wav
And if we bring the room mic up another 6dB?: TCRM12_8.wav
Now let's listen to what the other set of microphones sounds like. Here we have the SM7B on the kick, the KM184 on the side of the hat, and SM57s on both the top and bottom of the snare. The overheads are TLM103s in XY and the toms are still SM57s: TCRM12_9.wav
By inverting the polarity (phase) on the bottom snare phase we get: TCRM12_10.wav
Sometimes just a few mics can get a good result on the drums. This is what just the spaced pair and the D112 on kick sounds like: TCRM12_11.wav
Now, the XY pair plus the SM7B on the kick: TCRM12_12.wav
Now, let's try an opposite approach: using just the accent or spot mics.
Now, let's try isolating the snare microphones by using the the same setup as before, but swapping the active snare microphone for the Pro37 placed over the snare: TCRM12_14.wav
Again, the same setup but trying the SM81 under the snare: TCRM12_15.wav
The final snare variation uses the SM57 placed below the snare: TCRM12_16.wav
Mixing the SM57 on top of the snare with the SM81 below the snare: TCRM12_17.wav
Swapping the kick mic for the SM7B: TCRM12_18.wav
Finally, we'll go back to the D112 on the kick and try the KM184 on the side of the hat: TCRM12_19.wav
So that you can get a sense of each mic and technique by itself, here are the individual tracks. Feel free to import and mix them yourself once you've become comfortable with how each sounds individually….
The D112 in front of the kick: TCRM12_22.wav
The SM7B in front of the kick: TCRM12_23.wav
An SM57 on top of the snare: TCRM12_24.wav
A Pro37 on top of the snare: TCRM12_25.wav
The SM81 under the snare: TCRM12_26.wav
The SM57 under the snare: TCRM12_27.wav
An SM81 over the hi hat: TCRM12_28.wav
A KM184 to the side of the hi hat: TCRM12_29.wav
An SM57 over the hi tom: TCRM12_30.wav
An SM57 over the low tom: TCRM12_31.wav
Finally, a dpa4006 used as a mono room mic: TCRM12_32.wav
Now that we've explored drum mixing a bit, let's try vocals. Each of these was recorded on-axis with a pop-filter unless otherwise noted.
Now the M147 again, but this time without the pop-filter. Listen carefully to the sound of the "p" on "Promise me": TCRM12_37.wav
Now, we'll simply back the singer away from the mics...
And even further back:
The Shure KSM44 large-diaphragm condenser, now at 6 feet away. Listen for the difference in tone and the increase in the room sound (now sounding more like reverb instead of simply early reflections): TCRM12_42.wav