RECORDING YOUR OWN CD IN A HOME PROJECT STUDIO
PART 2 – GETTING THE SOUND YOU’RE LOOKING FOR

We ended last issue’s article with our home studio wired up in preparation for starting to record.  For this article I will assume that you will have at least eight input channels to get started with.  Let’s revisit the first pass channel assignments.

CHANNEL ASSIGNMENTS

1. Click Track (recorded separately)
2. Guitar/Keyboard
3. Bass
4. Vocal
5. Kick Drum
6. Snare Drum
7. Drum Mix Right
8. Drum Mix Left

 It looks a bit barren at this point in time but keep in mind this is only the first pass of many (but not too many we hope).  The click is there to hold everyone to the tempo.  If you really feel that you can’t hold to a click track, then I would recommend using that track for the drum high hat instead.  Remember – if there is not a click or drummer with near-perfect meter holding a uniform tempo throughout the song, your editing options at the end of the project are greatly reduced.  In higher end recording systems (e.g. Pro Tools HD) you can map the tempo and digitally pull the tracks into unified tempo.  A great option perhaps but I feel that tracking the music in tempo at the beginning of the session is much preferable.  

ON THE LEVEL
Let’s spend a moment on input levels.  There are basically two types of input levels, balanced and unbalanced.  A balanced input is preferred as this is the pro-level standard input.  Unbalanced signals (e.g. a guitar plugged in directly) work fine too but will give you a different signal level.  You should also have some level input options inside your recording software.

Using Pro Tools 6.4LE as the default recording software, we start by creating a new session.  This procedure will be similar regardless of the software you are using.  When creating a new session you will need to specify both the bit rate and the sampling rate.  The higher the sampling and bit rates, the higher the audio resolution (quality) is the rule of thumb. 

IMPORTANT NOTE:  You will need two hard drives installed on your computer to reliably record to multitrack sessions.  You may have a humongous system drive but you will have slow down and data degradation if you try to record the tracks to the operating system drive.  Also the higher the sampling and bit rate – the more hard drive space you will use.

 I have set up this session for the highest resolution my hardware and software will offer me 24-bit at 48 kHz.  Each track will take about 8.5 Megabytes for each minute of recording.  Even gobbling up hard drive space at this rate, a hard drive of 40 GIG should more than suffice for a whole CD project including mixes.

 In your software, create your first eight tracks.  In Pro Tools, it looks something like this:

 Now we should be looking at our first eight tracks in the computer mixer.

At the top of the display are the input channels your computer has assigned to each track, 1-8.  Notice at the bottom I have gone in and typed brief descriptions of each track.  Doing this before you start recording will name the actual files burned to your hard drive the name you have just given it.  If you wait to rename the track after recording has begun, your first guitar track’s actual file on the hard drive might be named Audio001_004.WAV instead of GTR1.WAV.  If you have to shuffle files around later, it is much easer to find the tracks if they are plainly named and labeled. 

You will probably have to make some modifications to the input gain.  On the DIGI there are two jacks in the front with a switch allowing you a -28db or +3db gain with a knob for fine tuning.  On the inputs in the back of the unit, you have to use the software itself.  In Pro Tools the input gain panel looks like this:

In this illustration, I have plugged an unbalanced signal from my keyboard (acting as drum machine) into Channel 1 of my DIGI 001. 

We’ll get back to recording an ensemble in a few paragraphs.  If you are going at this session alone instead of with a band, try this.  Sit down with your guitar (or whatever) and start playing to the drum machine.  Search for a pattern and tempo you can live with and play the song through with the machine.  When you get a groove going, make note of the pattern you are using and the tempo you are playing at.  If you know how to make a Nashville Number Chart (see sidebar), now would be the time to do it.

Now let’s take a look at each track separately.

First Pass – Track One - Click
This one is pretty much a no-brainer.  You plug the output of your drum machine into Channel One of your recorder.  On the DIGI the front panel of the unit has two multi-task input channels.  Both will accept a balanced or unbalanced XLR or ¼” input.  These two channels also supply phantom power to condenser microphones that have a powered capsule.   Enable recording that track by clicking on the record button.  This will give you a view of your input signal when you start the drum machine.  Now start the drum machine.  If there is a drum roll and cymbal crash in the tune, use that – the loudest part, as your highest input level.  Get the level as high as you can without going over 0db and into the red.  Depending on what kind of gear you have, going over the red line is bad.  On the DIGI there is a small db buffer area before the sound turns to noise, but not much. 

A good rule of thumb is to try and keep the floor (low noise) above -24 db and never let the ceiling exceed 0db.  In the old analog tape days you could slam the tape with more than a 0db signal.  The tape would be saturated and it would bring in a warm and fat distortion.  These days that trick is done by software.

IMPORTANT NOTE:  Another thing to keep in mind is that whenever possible – do not put any effects in the signal path until you get to the mix portion of the project.  Recording with effects being printed leaves you no out should you want to change the effect later.  There are many exceptions to this rule – in line compression on bass and drums come to mind.  I sometimes cut with effects when playing my guitar through a Line 6 Pod.  But I listen to the patch long and hard before I commit to it.  And if tracks are available, I will plug my guitar into a direct box, send the clean signal via the Cannon connection into the computer dry, then using the unbalanced pass through I plug that into the Pod and from the Pod in stereo to the computer.  Later I say to myself, “The performance is good but that effect sucks!”  Then just bus the clean signal you recorded out of the computer back into the Pod and back out of the Pod (with the new and improved setting) into the computer recording over the old effected tracks.  The clean signal lets you plug the guitar into effects, even different amps without the inconvenience of needing the guitarist present.

We’re almost ready to hit the red button for the first time but there is one more issue I would like to touch on before we start tracking; destructive recording versus Non-destructive recording.  A feature loudly and proudly touted by manufacturers are “virtual tracks” or “unlimited takes.”  Keep in mind “unlimited” generally means there is no theoretical limit to the number of tracks. In the real world though you will be limited by the amount of processing power and drive space.  As an example in Pro Tools the default is non-destructive recording.  This means that you are keeping every take – good or not.  It is the first “feature” I turn off because it is only confusing to have a dozen different takes of an instrument pass.  When I record a pass it falls into three categories, good, bad and maybe.  If it’s good I keep it.  If it’s bad, I try another pass over it.  If its maybe, I keep the track, create another adjacent to it and cut it again.  You can use parts of each take to composite or “comp” a track together.  This is not as desirable as cutting the whole track right once but sometimes you stumble on musical brilliance and even if there is a big mistake (clam) at the end of the passage, the first half is too killer to dump.  Save if for now and we will revisit it in the mix.   Always keep in mind that the fewer the tracks you have to work with at mixdown, the easier and faster the mix will go.  Just because you have 32 or “unlimited” tracks doesn’t mean you have to use them.   Too many tracks can just clutter a mix and make it hard for something to really stand out.  Less is more.  I have seen many mixing sessions get bogged down looking for the guitar take they want to use.  My rule of thumb is to use the good one.  Why keep bad or marginal ones?  If you play back a track and it doesn’t sound good enough, cut it again dude.  Pasting a part together one note or passage at a time is very labor intensive and the results are not always good.

So let’s push the red button for the first time.  I have set the software for destructive record.  I hit the record button and a second later start the drum machine.  Let two bars of drum machine go by before you actually start the song. This will serve as your count-in reference for the subsequent tracks.  Record enough of the click track to get all the way through the song and then hit stop.  You now have a click/drum track.  If you go to the waveform view in your software, it should look something like this:

You can see the valleys and peaks in the waveform.  This visual representation of the click track can be very helpful in cutting and pasting parts of songs later as it gives you a good view of beats to line up to.

TIP:  If your hardware/software will also record MIDI (Musical Instrument Digital Interface) tracks,  take the MIDI out of your drum machine and plug it into the MIDI in of your  hardware.  Enable a MIDI track and record it at the same time you put down the audio click track.  This gives you a real machine clock time that you can sync up to other devices later if you need to.  MIDI information requires very little drive space and can be a great help when syncing to other devices and projects.

Now that we have a drum/click track whacking away all by itself, its time to move to the next pass and track.  Let’s start the body of the song.  Again, we are talking about a solo recordist – we will touch on the whole band shortly. 

Pass Two – Track Two - Rhythm Instrument Track
Turn off the record enable on Track One or you will wind up recording over it by accident.  Record enable another track.  Plug your instrument (guitar/keyboard) into that input channel (you can use the same input bus you just used on the click on this new track).  Hit the instrument hard and adjust the input level in your software mixer until the level is as high as you can get it without clipping.  Rewind to the top of the song.  Hit the record button and start playing the song after the two bar count in has passed.  Play to the end of the song and hit stop.  Your waveform view should look something like this:

You are in business now.  You have the entire length of the song with tempo, meter and chord pattern into the computer.  You have a sketch of where the song is going.

Pass Three – Track Three – Scratch Vocal
A variation on this is to plug your guitar/keyboard into one channel and a vocal mic into another and record both the scratch rhythm part and the scratch vocal at the same time back on the second pass.  If you don’t go this way, record enable the next track and sing along to the tune.  Now we should have three tracks – click, rhythm instrument and vocal.  At this point, if you are going to be joined by other musicians on the project, you can burn CDs of just this rough sketch and they will be able to practice along with it before coming into the studio. 

The sketch grows as you add tracks and players.  Let’s leave the one-man-band scenario for a moment and look at a band recording.

Back To The Band
The most important thing to keep in mind on the first pass of the band’s session is to get the rhythm section sounding fat – the bass and drums are the core around which everything else revolves.

The first pass is usually the same with a band as it is for a solo recordist – recording the click track.  You can record it all by itself before tracking the band.  So now that we have a click track let’s start working on the band.  If we have eight inputs and the click is already in the machine, your second pass 8-channel assignment might look like this:

1. Bass
2. Guitar
3. Vocal
4. Snare
5. Kick
6. Hat
7. Drum Mix Right
8. Drum Mix Left

Bass Guitar
A big fat bass sound is sometimes one of the most elusive tones to get.  One way of getting a good bass sound is to use more than one channel on the bass.   I usually have the bass player with both a direct input and one with a mic on the bass cabinet.  We don’t have the luxury of two bass tracks on this sketch pass but we can come back it with a later pass using more channels.  For getting the sketch up, just adjust the input level again so that the loudest part is as close to 0db as you can get without clipping.

Guitar
The same applies for the guitar.  Tone is not that important on this sketch track.  What if you have two guitar players?  Well let the rhythm guitarist have the first pass and keep the soloist on the sidelines for now.  You can see the benefit of using a commercial studio here – you don’t have to cobble your rhythm tracks together like the one-man routine.  If you have a breakout box that adds channels like the Alesis ADAT converter mentioned in the previous chapter, your first pass would more likely look like this:

1. Click (recorded with band)
2. Bass (mic on cabinet)
3. Bass (direct)
4. Guitar One (direct if possible – see isolation note)
5. Guitar Two (direct if possible)
6. Vocal
7. Kick Drum (inside on head)
8. Kick Drum (outside at sweet spot)
9. Snare Drum (top)
10. Snare Drum (bottom)
11. Hi Hat
12. Wing Tom 1
13. Wing Tom 2
14. Floor Tom
15. Drum Overhead Right
16. Drum Overhead Left

 And we just ate up all sixteen tracks of the DIGI’s eight inputs and the eight inputs on the 8-channel but we are moving MUCH faster tracking than going at it one track at a time.   And the lead guitar player gets to join the fun.  Now we have to stop for a minute and consider isolation.

Instrument Isolation – Particularly Drums
The most powerful trait of multi-track recording is being able to adjust the volume of each instrument separately and place it in the mix.  Let’s say you have a killer guitar part that you want to bring up in volume.  You raise the volume slider on the guitar part and the drums get louder!  How can that be?  Simple.  You put a mic on the guitar cabinet in the same room as the drum kit.  The drums are bleeding over into the guitar amp mic.  The drums aren’t very loud in the guitar mic but loud enough that increasing the level of the guitar also noticeably raises the drum level.  How do you get around this?  Record all the instruments on the initial pass directly using no mics at all except on the drums and vocals.  Or, if you have the space, snake and headphone accessories, put the drums in one room, guitar one in another and so on.  You need a garage the size of the Astrodome to get this done.  The drum and bass tones are the most important tracks at this point in the project – they are the glue that will hold the whole thing together in later sessions.  They are the most important – once you have a kick butt rhythm section pumping the beat and bottom, you’re good to go.

Let’s take a brief look at some popular microphone placements and tracking tips.  We’ll start with the drums.

Remember that we are trying to do this recording on a very low budget.  Yes, I know there are better mics than the ones described here.  The mics mentioned here are low cost industry standards.

The Bass Drum
If I have mics and channels enough, I like to use two mics on the drum.  If there is a hole in the drum head I throw an SM57 (or other mic) in the hole on some buffering material (a blanket) and put the mic capsule about 2” from the drum head.  This is for my “thud” sound.  The other mic I place outside the drum head pointing at the hole if there is one.  If there is no hole on the drum head I point the mic at about 4 o’clock and three inches out.

The Snare Drum
Again if I have the resources available I use two mics on the snare drum too – top and bottom.  The top gives the booming punch and the bottom gets more of the snare rattle.  Top mic is placed coming over the front of the kick to about six inches from the hit hat and the bottom mic is pointing directly at and about 2 inches away from the snare wires.

Toms and Overheads
I mic the toms from the front of the kit and try to place them 2 inches off the head and as far into the drum head as I can go without getting in the drummer’s way.  Placing the mic too close will not get you the booming decay you might want from a tom and also might actually slap into the bouncing drum head during the performance so mic close but not too close.  There a different schools of overhead mic recording.   Some engineers point the overheads toward each other.  Others point the mics straight down.  Overhead mics should be at least 4 feet above the snare head and three above the cymbals.   I have heard great drum mixes using just one overhead but it was a killer microphone (Neuman KM184).  This is a “season to taste” call – try both ways, move the mics around and experiment.  After all this is YOUR sound we are looking for and your ears may hear it differently than mine or another engineer.

Mic Placement On Guitars and Amps
This one is pretty simple too.  On an amp get the mic about four inches from the speaker cone and a bit off axis – much like the placement of the outside kick drum mic.  For a fatter sound, channels permitting, throw up another mic pointing at the amp from about eight feet away – you can blend in the room sound if you so desire.  Also don’t forget that you can record a completely dry guitar channel for later experimentation with effects and amps.

Mic placement on an acoustic guitar is another “try before you buy” proposition.  I like the mic about 6-8 inches away from the guitar’s sound hole, a bit off axis and pointing toward the bridge if I’m looking for bright tone and more toward the guitar’s neck if I am shooting for a warm tone.   In praise of technology many new acoustic guitars sound awesome with their own electronics.  I have a Takamine that blows chunks when in front of a mic but totally rules when plugged in directly.  When possible again do both – mic the guitar and track it plugged in directly.

Big Bottom – The Bass Guitar
The low end of the audio spectrum is the most problematic these days.  Everyone wants fat punchy bass sounds.   In contradiction to my rant about not using effects during initial tracking, a good compressor with a brick wall limiter can fatten a bass quickly.  Dial in a bit of compression but not too much.  If the track is playing back and you hear the bass guitar wheezing, you have too much compression on it and it sounds like a valve opening and closing.  Compressors definitely color the sound and not always prettily.

It looks like I’m running out of space for this month.  Next installment we will walk through the recording session with some tips on how to make the work flow move faster as we move closer to the fun part of the project - mixing

Part 3