The following is a log and writeup of a recording and mixing project I completed for my college course, hopefully it will be helpful to those wishing to understand the processes and techniques behind this often badly and scarcely documented task.


I had 8 hours of time in the studio to record the two songs my band has planned. To ensure I achieve this to the best of my ability I will need to carefully plan my time.

Planning can help highlight potential problems and bottlenecks that can hamper a good recording session. An example of this would be a plan detailing which microphones are to be used so that they can all be retrieved in one run to the cupboard.

Decisions such as artist and microphone placement will affect the recorded sound so this should be considered and recorded if the recording session is to be spread over multiple days. This will allow recording conditions to be re-created for subsequent sessions.

Room and Equipment Setup

Before the band arrived I started to ensure the recording studio was tidy and that we had all of the equipment listed in my plan. I first began to prepare the drum set up, this consisted of removing the cymbals as they were to be replaced with the drummers’ own, as the ones supplied by college are very battered and actually had a section missing from it. The drummer also had his own double kick pedal which was installed.

We removed the tom drum that was not to be used and used one of the band member’s jumpers to slightly dampen the kick drum. When the drummer actually started playing I noticed the snare was very resonant and ringing which was fixed by applying a small amount of electrical tape to the side of the drum skin.

In the studio we moved the monitors forward from their stands so that we were using for their intended use; near field monitoring compared to the midfield they were originally set up for. This was to help ensure I could hear the recording session optimally.

After this I started to set up the microphones, beginning with the kick drum. This was because I had planned to set up the mics in the same order to where they would be plugged into the stagebox. For the kick drum I was using two microphones, the Sennheiser E901 inside and the 52A for outside on the vent. The E901 was positioned closer to the beater to catch the click and it’s XLR cable was threaded out of the vent on the front on the drum. The 52A was placed in the hole to catch the “woof” of the kick; we used a special, extremely short mic stand for this in order to position it in the correct place and keep things tidy.

Once the microphones were in place and plugged into the stage panel in the correct order I retreated to the studio with my track sheet to make sure everything was in order. After checking the audio interface was set up correctly and opening Pro Tools I named all of the drum tracks started to set the gain structure by getting the drummer to play some patterns and adjusting the preamps accordingly. I ensured he had played the loudest he intended to and was not clipping, and noticed that the Tom drum mics were the wrong way around. I switched this in software through Pro Tools.

Next I set up the keys using a Yamaha Electric Piano and a DI box. The bass guitar I put through a Neve preamp for a warm and high quality sound, then into a DI box and straight into the desk. The signal was also routed in parallel to an amp so that he could hear himself play. The Neve preamp uses a tube which adds subtle harmonics to the sound and is highly sought after by many studio engineers.

The guitarist also used a DI box routed to both the mixing desk and an amp. We routed the signal from these to the drummer to help him play along with his band members who were in the control room. This was to ensure minimal bleed from the instruments and keep each track of the recording in a much cleaner way. Another thing we used for the drummer and pianist’s benefit was a click track; this was to ensure the piano at the start of the song would start in time and so would the drums.

Recording Process

After we had the keys for the intro recorded I setup automation to fade out the click shortly after the drummer started playing, this was to ensure the track still felt natural with enough swing.

In this one session we recorded drums, bass, keys and guitar for two tracks. Each instrument took a couple of attempts to get the perfect take and by the end of the day we had nearly everything recorded except for the synthesisers for one track and the vocals.

This included the overdubs, for which I mostly used the punch in technique. To do this I get the player to play along with the track and hit the record arm key just before it gets to the area they need to record from or improve. This ensures a clean and smooth transition, normally sounding a lot cleaner than cutting and pasting. One of the overdubs for a bass guitar “drop” was spliced in on another channel, which sat slightly higher in the mix to give it emphasis as it is just after a crescendo.

To record the vocals we used a different room, M101 “The Temple of Sound”. This room has lots of acoustic treatment on the walls such as angled foam blocks in the corners and padded and carpeted walls. It also has a number of large acoustic panels that can be arranged around a vocalist to further isolate them from the rest of the room. The room is built without any parallel walls in order to avoid standing waves and phase issues.

We used the Peluso microphone and recorded three takes. Using these takes we went through and chose the best parts of each in order to get a perfect vocal track. This is called vocal comping, where you have multiple takes and choose a compilation of the best parts of each for the final master. I did this for just one track, as in the other I was to use selective multi-tracking when mixing it. The synthesiser parts were also recorded in this session using a midi piano and a synthesiser in Pro Tools. We made the decision to use MIDI so that the sound would be far more adjustable and adaptable later when mixing. This is a more recent feature in Pro Tools and is very useful when adapting sound and testing different concepts and ideas.

Microphone Information and Choices

Types of Microphone: Dynamic Vs Condenser

Dynamic Microphones directly covert sound waves into electrical energy using electromagnetic induction. It uses the same principle as a loudspeaker but in reverse with the electrical coil generating current when the diaphragm and attached magnet vibrate. This allows them to have a more closed and robust design and means that they are far more resilient to high pressure levels and moisture levels than condenser microphones. One disadvantage of this type of microphone is that they generally have worse transient and frequency responses, but their advantages more than make up for these flaws when used in close micing and live audio settings. For example their resilience to damage and lower sensitivity makes them ideal for jobs such as mic'ing the inside of a kick drum or for a loud powerful vocalist on stage.

Condenser and Ribbon Microphones measure changes in voltage and magnetic field on and around a small electrically charged conductor which is called a diaphragm. These require more complicated and delicate capsule assemblies to hold the delicate diaphragm in place. These types of microphones contain complex electronics in order to control aspects phase alignment, polar patterns and amplification circuits. In order to drive these often complex electronics, many preamps will supply phantom power at -48v. Certain high end microphones use their own power supplies and cables such as the Pelsuo P12 with 12 polar patterns and a 7 pin XLR used to transfer power on completely separate wires. This helps to minimise any risks of interference from the switching circuitry, transformer hum or power spikes through the diaphragm material affecting the audio.

Polar Patterns of Micrphones

Most of the microphones we used have switchable patterns which were left set to omnidirectional, as they were distanced from the only instrument, the drumkit, enough to be left to record to sound naturally.

A unidirectional polar repose records sound equally from all directions and is good for capturing a sound in a distant area of the room or an entire instrument. Some of the polar response patterns on the close microphones were much tighter and were part of the reason they were chosen. A “tight” polar pattern refers to a cardioid or hyper cardioid response pattern where mostly sound coming from in front on the microphone. This gives directionality and allows different parts of the kit be isolated with certain mics. It is a major influence when deciding which mics to use for the snare, hi hat and outer kick.

Microphone List and Reasons for Choices

Kick Drum: Sennheiser E901 and Sure Beta 52A

The first microphone, the E901, was for inside of the kick drum as this microphone is specifically designed for this use being able to withstand very high SPLs; such as those found inside of a drum. It is of a type of microphone known as a Boundary Microphone, also know as a “Pressure Zone Mic” which uses a microphone capsule mounted very close to a fixed surface to reduce comb filtering effects caused by phase cancelation by delayed signals. This design causes the shifting the comb filtering reflections into the ultrasonic range, far beyond our hearing. It also has a built in high quality preamplifier and therefore requires 48v phantom power. For the outside of the drum, in the hole, I used a Sure Beta 52A. This microphone can deal with extremely high sound pressure levels. It is a dynamic microphone with a specially tailored frequency response designed for kick drums and other bass heavy and high SPL instruments. It has a fast attack and 'punch' and features a modified supercardioid pattern, reaching throughout its frequency range to insure maximal rejection of unwanted sound. This means it should pick up sound from only within the drum and is part of the reason this mic was chosen.

Snare Drum: Beyer M201 TG MG201 and Sure SM57

For the top of the snare I used a Beyer M201. It is a small dynamic microphone with a tight hypercardioid pattern with a very smooth response from 40Hz-18kHz. The tight hypercardioid pick up pattern is due to its’ small capsule design, which enables one to easily isolate individual components of the kit.

For the bottom of the snare I used a Sure SM57. It is a cardioid dynamic microphone with a contoured frequency response and presence rise, which helps the sound cut better through the mix.

Hi Hat: Sennheiser E614

The Sennheiser E614's is a small diaphragm condenser with a super-cardioid pattern, with a fast transient-response. This in combination with a neutral frequency response and moderate sensitivity, insures optimum isolation from other parts of the kit and this is why I chose it for the job.

Overheads: Neumann KM184

The Neumann KM184 is a pressure transducer microphone with an approximately flat frequency response featuring very smooth frequency responses on both the 0° axis and the lateral (off-axis) sound incidence. Often, there is no coloration of sound over their wide pickup angle and that is makes them suitable to recording a large part of the drum kit. The set we used is a matched pair, which means the frequency response is matched or adjusted to be identical at the factory, allowing them to provide an accurate stereo image.

Toms: Sennheiser 421

The Sennheiser 421 is a dynamic cardioid mic with an extremely flat response from 100hz to 1000hz, This along with their ability to deal with reasonably high SPL’s makes them perfect for capturing the Tom drums accurately.

Room: AKG C414 and Peluso P12

The AKG C414s are switchable polar pattern, large diaphragm condenser microphones with two pressure gradient capsules each. This, along with some clever onboard hardware allows for 5 different polar patterns, but I will be using the omnidirectional setting. They are also a matched pair in order to record an accurate stereo image.

Other Microphones

For the mono room and centre overhead we used Peluso P12 microphones. I also used one of thes microphones later for vocals. The extremely high dynamic range of these microphones and high sensitivity and accuracy make them perfect for recording an overall sound of the drums. They are condenser tube-amplified microphones, giving them a warm and accurate sound. This is due to the specific harmonics and overtones that the vacuum tube-based amplifier imparts on the sound. I also used this microphone to record the vocals in a later session as aI feel it gives a nice “warm” aesthetic to the sound helping to reinforce to vocal. As it is such an expensive and sensitive microphone it has its own holding cradle to keep it safe and isolated from vibrations travelling through the stand. We also used much sturdier, counterbalanced microphone stands.

These will become especially useful channels when it comes to doing parallel compression and help give the kit an overall sense of space.

Mixing Process

Computer Setup

I used Pro Tools to mix and edit these multitrack recordings. Most of this was done with high quality Dynaudio Acoustics monitors allowing us to hear the sound clearly and faithful to the original recording. This is due to them having a reasonably flat frequency response, this means that they reproduce the sound honestly, without any severe boosts or cuts along the frequency range. This allows us to master the songs without the speakers influencing the sound too much.

First I had to check that the settings on Pro Tools were set correctly for the mix session. This consisted of setting the Delay Compensation Engine’s “Playback” setting to “Long”. This allows Pro Tools to read further ahead when rendering the track and can cause a reduction in phase and timing issues. Another setting we changed was the “Hardware Buffer” which we set to a high setting – 512 samples. These settings both increase latency which is problematic when recording, but when mixing it does not cause any problems and they help to provide an increase in audio quality and accuracy.

Effects, Equalisation and Track Summing

note; this works and looks much better formatted into A4 sheets, feel free to edit and reformat. Nevertheless I will being going back and doing this myself in the next few days.

I set up two Aux tracks for the time based effects. One of these was for reverb and one for delay. I used a nice drum plate reverb with the decay reduced slightly on the reverb, and customised settings on the delay which were based around a vocal setting.

The first section of the tracks I decided to master was the drums. Both track’s drums were recorded in the same session so I mixed them both together then made small adjustments for the second track. This saved time and helped to provide a more unified feeling between the tracks.

At this stage I muted all the instruments except for the drums. This was so that I could hear them more clearly when editing and mixing them.

Because the pairs of microphones on both the kick and snare were spaced apart from in each other, they had not recorded their respective drum hits at precisely the same time. These sorts of tiny errors in the timings of transients in sounds can have serious implication for the final audio including phase issues such as selective frequency cancellation and comb filtering. In order to bring them in time I used the Triple Click feature which selects the entire track of audio. In doing this I ensured that the entire length of each section of the racks of drums was edited equally in precise timing and in phase with its’ counterpart.

After ensuring the kick drum tracks were hitting at the same time, I inverted the signal from the lower snare microphone. This ensured that it would not be out of phase, due to being underneath the snare drum and recording the lower skin noise. It was also slightly out of time with the top snare track due to the physical distance between the microphones. I fixed this using the Triple Click to select the entire channels’ worth of audio then using the Drag tool to carefully move it into time. It only required a tiny change. Much like with the kick drum.

I used this opportunity to ensure all of the drum kit was in phase. I needed to realign to audio for the overheads as they were slightly out of phase. I also attenuated the signal drum the right one by 3db to ensure they were of similar volume. I then moved the separate recording from their two channels onto a newly create stereo track. I also did this for the room wall mics but did not change any of the timing as they were just to be used to add space and room ambience to the drum track. This allows me to control both sets of microphones better in the mix and apply effects to both Left and Right channels of equally. I used this opportunity to apply a high pass filter to both channels at 150hz.

Next I put an EQ on each part of the kick recording. On the Inner kick channel I applied a high pass filter in order the catch the click of the beater on the drumhead as this microphone was placed very close to it. The Outer kick channel I used a low pass filter with the cutoff set at the same frequency. Both EQ’s had slight cuts at around 300hz, this was to remove boxiness and muddiness. In effect this combination of filters and EQs worked like a frequency crossover on a speaker unit with negative cutoff resonance and with the signal travelling in the opposite direction – back to tape rather than through loudspeakers. I wanted to combine these two kick tracks into one so I created a new audio track with its input as Bus 11 and set it to Record Armed. I set the Output of both kick drum channels to Bus 11 and press Record and Play from the beginning of the track. This combined the phase aligned and selectively filtered kick channels onto one track. Once this completed and these drums had been recorded onto my sum track I used an EQ to make a small cut around 400hz. This was to make the kick sound slightly more hollow, giving a “scooped” sound and removing the boominess often associated with badly recorded drums. I also added a high shelf boost of 5.9db at 2.22khz to emphasise the click of the kick track.

I used a similar method for the snare but with much milder cuts in the EQs without actually filtering the sound. I did this as I was not using as much signal from the secondary microphone’s recording, as it sounded rather harsh and scratchy. Due to the high quality of the top snare microphone it managed to record a good amount of the brightness on it’s own, through the drum head. After the snare microphones had been summed together I muted the two original channels. After I had the kick and snare summed I made the original tracks inactive and disabled them in track view in order to remove the clutter and give myself more room to work. On the sum channel for the snare I added an EQ with a high pass filter set at 150hz to remove low frequency rumble.

The next thing to do was gate the kick. This is necessary as the spill that kick drum microphones pick up from other parts of the kit is often highly muffled and rumbling. This type of spill can reduce to overall clarity of a production and is harder to remove once any other effects have been added, as well as being difficult to identify in a final mix. I used a short attack of 24ms in order to catch the fast transient and entire amplitude of the kick’s click. I experimented with a shorter release time but this made the end of the kick tone audibly “duck” so I increased it until an adequate setting was found. This setting was 676ms and it sounded better as the longer release allowed the gate to stay open for the entire length of the kick drum’s sustain.

Different Attack and Release settings (parts of the Envelope) can completely alter the overall sound of a kit depending on how they are used. Long attacks and fast releases can be used for an extremely muted sound and the transients can altered in interesting ways using these parameters when the situation calls for it.

The next effect I used on the kick drum was a compressor. I used this so that the sound had more overall loudness, from the start to end of every beat. The compressor allows this by smoothing and reducing the peaks in the sound. With the peaks reduced a higher overall level can be achieved. The smoothing is controlled by the Knee and the reduction in dynamic ratio is a direct translation of the Ratio once the Threshold is met and exceeded.

The only processor I am going to use on the snare is a High Shelf boost using an EQ in order to give the top end of the drum a little more sparkle. This contrasts with the high hat channel, to which we applied a Hi Pass Filter at 286hz to remove any rumble and unwanted low frequency noise In addition to a very high frequency shelf boost of 1 db at 4.4khz. This was to emphasise the airy detail and fast transients of this part of the kit. I then Panned it slightly to the left – this then sounds like you are sat in the drummers perpective relative to the kit when listening back. I chose to do this as non-percussionists are unlikely to notice it whereas percussuionists are likely to appreciate the gesture of realism from their point of view. I used the same approach when deciding which of the overheads and wall microphones was left, and which one was right.

After listening to the Mono overhead we decided it was too noisy and distorted due to the sensitive nature of the microphone and the loud nature of the drummer’s playing style. To reduce clutter in the session we deleted this channel.

For the Toms we used the edit tools to manually “Gate” the sound by deleting all of the areas except for the specific times when these drums were played. Some of this we did manually with highlighting and deleting the rest was done with the strip silence tool. This work similarly to a gate but it deletes the audio that does not make it through from the arrange window.

In order to avoid clicks and pops I used the Fade tool to fade out the Tom section smoothly, I had to adjust to curve on one of the rolls in order to make it a smoother and more natural fade. I used the Shortcut for this to save countless time in the dropdown menus. In order to do this I highlighed the area which I wished to apply the fade to and press Apple+F. This brings up a Dialog menu to help choose the type of fade to use and is highly useful for keeping a quick and smooth workflow.

The EQ for the Toms consisted of a major “scoop” out of the deep and full sounding middle frequencies between 100 and 1000hz, and a small presence boost at each drum’s identifying frequency range; 2.87khz for the upper Tom and 2.45khz for the lower one.

All of the drum except for the Kick and Toms that did not already have it were given a high pass filter set at 150hz. This was the decrease rumble and unnecessary and inaudible low frequency information.

Once the levels and processors had been set for the drums I decided to use a technique called parallel compression. This takes a copy of the drum track compressed to the limits of what is possible. This track is recorded to preserve it and is referred to as a squashed track. This squashed track can be included in the final mix at a very low level to add dynamics to parts of the song which are lacking.

I routed all of the signals for the drum kit through a single stereo channel in order to consolidate my signal flow. By doing this it allowed me to compress and EQ all of the drums at once, as well as enabling adjustments of the main drum level universally and independently to the mix between individual elements of the kit. I applied a small amount of compression that had a small ratio and reduced some of the peaks very slightly. The EQ I applied provided a small boost in the higher frequencies and emphasised the air of the drums. I also sent a small amount of signal to the reverb bus to give more atmospherics.

For the Bass Guitar and the normal Guitar I used Pro Tool’s Amp Simulator-Sansamp. This allowed me to get a warm sound out of both. The effect is less severe on the bass as I did not want to interfere with its’ low frequencies too much.

On the ordinary guitar I set up some automation towards the end of the first track in order to give it more presence. As some of the other instruments fade away I gave the guitar a large boost in volume and a subtle panning from side to side. This brings the guitar out from the mix as it is the only instrument to do this.

On the bass guitar I used a compressor with a slow attack. This allowed the twang of the string to strike through and allowed the gain to be turned up overall, causing the notes’ perceived hold length and power to be increased.

I set up the slight panning for the left and right piano channels. This was to emulate a real piano which produces stereo sound as each note is situated in a different place. I also set up a send on a bus in order to later add reverb to this instrument. For the vocals I used a compressor and peak limiter to prevent clipping. This was because the tracks were very dynamic, especially the comped one. The compressor was set with a high ratio of 6:1 to allow the gain to be increased more and the peak limiter helped to stop the track from clipping. I set up Sends to the Buses that I had put time based effects on. These were Delay and Reverb.

On the second track I used the multiple takes to selectively multitrack the vocal. This allowed me to emphasise certain builds ups and areas of the chorus and provide a unique effect. To do this I set the two extra channels of audio to automation write mode, then I used the controls on the desk to quickly switch them in and out.

On the other track I used larger, but highly varying amounts of delay and reverb. I achieved by automating the send levels on the vocal channel around the chorus and other areas of the song to switch things up and keep the atmosphere changing. This allowed me to change to sound of the vocal without changing the amount of reverb and delay applied to the drums, even though they shared a reverb unit.

Note: the difference between effects and processors

Effects apply time or modulation based processes to the sound, and should have a wet and dry mix control. The dry signal is the original sound and the wet is the result once it has passed through the effects unit. For most of these types of effect I have used Aux buses to reduce processing power requirements and ensure the reverb and psychoacoustic effects do not change suddenly during the song. Processors are all on the individual instrument channels. This is because their output is entirely processed – being non-time based it is not desirable to add to “dry” signal to the processor output. Because of this we use the insert points on the individual channels to affect the sound.


QR Code
QR Code band_recording_project (generated for current page)