Audio Engineering Tips of the Day
by Ken Burke
Record a clean signal from a direct box off the guitar before it runs to the amp. This way you can re amp the guitar through a different amp or amp plug in later. You can also work quietly at home and monitor through a guitar processor, finish all your guitar parts and then go into the studio to mic your amps on “11”.
At every session, record a sample of every drum, one at a time, with and without room or overhead mics. Make sure to capture every cymbal as well. This way you have samples for triggering and material to help in editing. Soon you will have your own sample library. Often these sounds are triggered live to match the sound of the record.
Some singers and choirs prefer to work without headphones. This can make track leakage problematic. A simple solution is to record the singer with the music bleed. Then record just the music bleed with the singer standing in the same position in front of the mic. Now combine these two tracks with the bleed track out of phase and most of it will cancel out.
Choosing a vocal mic. I have a few favourite vocal mics that I rely on, however every time I record a new singer I will do a quick mic shootout. It only takes about 10 minutes to set up all the good mics you have and have the singer sing into each one. The difference between individual singers vocal sound on different mics is striking. What works for one person doesn’t always work for another. Let the singer choose the mic they like best. If you want a vocal mic for under $400 with some colour try the Shure SM7. If you are look for something closer to $500 start with the Audio Technica 4050. I consider this a better buy as it has multi-patterns and a more neutral sound.
Key ducking talkback microphone. Tired of always turning the talkback on and off and sometimes missing what the talent is saying? Tired of being blasted by drums or guitar when you leave the talkback mic open? Place a noise gate across the talkback mic and set it to key duck. Insert the drums or guitar mic signal (use a mult) as they key input to the gate. When the talent starts playing guitar the talkback automatically “ducks” shutting off the signal. When they stop playing the duck will release and the talkback automatically opens.
When mixing, make sure to reference to similar professionally produced records for perspective and comparison. This will get you closer to a good mix faster. Also take regular breaks, burn reference cds and listen to them on as many systems as possible. It is a good idea to check your mix in a car, through a TV, on your friend’s home stereo, on a ghetto blaster, iPod and even on a nightclub system.
Mono. If you are mixing a college band, keep mono in mind. Just over half of the college radio station in North America broadcast in mono. Wide stereo guitars and string pads will cancel out in mono leaving your mix sounding small. A good idea is to issue a separate mix just for mono radio. Befriend your local dj and ask them to play mixes for you on the air.
Having trouble finding a problem frequency in a mix? Boost the gain on a parametric equalizer by several dB and sweep the frequency control with a small bandwidth setting to find the problem frequency. You can also use a visual spectrum analyzer to pin point the frequency exactly. This is particularly helpful in finding the sibilant frequencies for a de-esser. When soloing instruments you will be able to see the fundamental frequency and all harmonics. Consider harmonic frequencies when equalizing. There are many free VST plug-ins for Macs and Windows such as Steinberg’s BS Spectrum.
Gating reverb send – To clean up my reverb return, I will duplicate the snare channel (mult) and place a tight gate one channel only. I do not use this gated channel in the mix, only route this through the auxiliary send to the reverb. This cleans up a lot of washy cymbal and high hat leakage from the reverb return making the high frequency area of the mix much clearer. I also do this with lead vocals that have excessive sibilance.
Key gating a sine wave. Run a continuous sine wave of around 40hz and place a noise gate on this channel. Send the kick drum signal to the key input of the noise gate and adjust the attack and release settings to match the kick. Every time the kick is hit a low rumble will be added. When doing live sound I tune the frequency to where the hall or glasses on the bar would vibrate. If you are using a drum loop sample you can send this to the key input and use the key filters to filter out the high end so the sine wave is only triggered on the kick drum. Listen to Ice T’s “Midnight” for a classic example.
Simple delay effects
Doubling effect with delay. Want to widen a mono part? Pan the original signal to the left and add a delayed signal (25-32 Milliseconds) with no feedback to the right and you will hear it as two people playing together.
“Elvis” style echo. A signal was sent to a tape machine, recorded and played back simultaneously. The time delay from the record head to the playback head is about 220 Milliseconds at 7.5 inches per second tape speed. The delayed signal was returned to the console and fed back into the tape machine again causing more repeats (the repeated signal was repeated). This is a very common effect in 50’s Rock and Roll, Country and Rockabilly. Adjust feedback to taste. This type of effect gives more clarity to vocals than reverb.
Simple delay effects part two
Flanging is created by mixing a signal with a slightly delayed (2-10 milliseconds) copy of itself, where the length of the delay is constantly changing (modulation). This was originally achieved by running two tape machines at the same time and varying the speed of one machine by placing your finger on the top part of the reel called the “flange”. A lot of feedback is used to create this effect.
Chorus differs from flanging in that the varying delay times (modulation) are much longer (20-40 milliseconds) and less feedback is used. Chorus simulates the effect produced when two people play together slightly out of time and out of tune.
This may seem to be an obvious tip but I consistently see students get caught on this one. It is important to leave time at the end of each session to back up your work. Keep two separate hard drives and copy the data from your work drive to a back up drive. After you have done so make sure to physically separate the two drives.
In editing an album you will make thousands of keystrokes. The Keyboard Focus application in Pro Tools will speed this up considerably by eliminating almost half of your keystrokes. In the upper right hand corner of the edit screen you will see a button labelled ‘a..z’. This turns on Keyboard Focus. All key commands are now one keystroke instead of two. For example, zoom is now ‘R’ and ‘T’ instead of ‘CTRL Bracket’, undo is ‘Z’ instead of “CTRL Z’ and break region is ‘B’ instead of ‘CTRL E’.
Zipper noise. When editing in any DAW, if you cut on a waveform at any point but the zero crossing, you will create a square wave that will make a loud pop or clicking sound. To avoid this you can cut right at the zero cross or use a crossfade of at least 14 milliseconds. It is fastest to finish all of your edits without regard to the zero cross point and the select all regions and crossfade them all with one command ‘CRTL F”. I keep my Pro Tools fade preferences set to 18 milliseconds and will “softedge” all regions with a single keystroke ‘F’.
Audio Engineering Tips of the Day
by Tony Marryatt
1) Starting a Mix. Once I start a mix I like to keep the flow. I don't want to stop and set up echo returns or buses or do re-patching etc. So, to avoid this I spend a sometime creating a start mix template. I'm speaking 'in the box' but this relates to your console mix too, maybe even more. Set up the usual track lay out, busses, reverbs and delays of different lengths in your largest typical production. It's easy to customize this for each new songs instrumentation by just deleting the tracks you won’t have.
2) Checking Phase 1. Any 2 mic's picking up the same sound source have potential for phase problems. Out of phase sounds can effect frequencies in your recordings adversely so it's a good idea to check the phase between them before you commit them to 'tape'. Sum the tracks in mono to listen for cancelations or visually compare the two tracks in your DAW.
3) Pan and Level Mix. Once my mix is loaded and I have everything set up in front of me the way I want, I'll get a pan and level mix. Set the levels quickly (temporary) and then spend some time experimenting with panning arrangements. Once things are sitting the way you like try flipping to mono and work on the levels from there. To a degree I like to think of these as separate mixes that fit together. The levels you set in mono should translate well back in stereo. Next do some EQing to make it fit together even better.
4) The 'Secret' To Great Drum Tracks. The secret to great tracks begins with the player. This tip is a reminder of the obvious but well worth repeating. The largest percentage of your success lies in the performer doing a great take. The point is, it presupposes everything to follow. The second largest factor would be the instrument they play. Thirdly the acoustic space in which it's preformed in. The mic placement would follow and then actual mic choices. After that you can talk about pre amps, compressors, and converters but it's important to keep it in perspective. Without the first, nothing that follows will matter much.
5) Vocal Send Track. Back in 'the day', mixers sent an aux signal to a recorder and took the signal off the playback head. At 15 Inches per second this delay is about 210ms. This clears things up and makes less reverb necessary. Although most reverb plugins have pre-delay parameters, I like to duplicate the vocal track, put a simple delay plug on it, (feedback set to 0) send it's 100% wet signal to the vocal reverb. This way I can EQ the send channel anyway I want before the verb and have the most control of the delay's timing.
6) Mixing By Sections. Depending on the song, I'll often start out the mix by treating the chorus (or any section for that matter) as if it's a separate song. In a DAW, such as Logic, I'll have the sections divided up by my markers and when in loop mode, I can 'command arrow' to toggle my loop selection between verse, chorus, bridge etc. After getting the chorus just right I'll think about what will be different in the verses and create separate verse tracks for those instruments by way of cutting and pasting. This is usually a lot easier than automating everything on the same channel. On a console you would do something similar by multing, for example, the vocal to 2 tracks and automate their mutes to get the same effect.
7) Make Your Own Multi-band Compressor. If you don't have a multi-band compressor in your DAW plug ins you can make one by using 2 or 3 separate channels, each with its own compressor and EQ. First copy the audio to 3 tracks and place a filter and a compressor in all of them. Second set up the first channel filter to pass only lows. Maybe somewhere around up to 250Hz. On channel 2, your mid band, use a band pass filter that picks up were your low pass left off and ends around 1kHz. Your last channel will have a high pass filter allowing 1k and up. Use as steep a slope as possible for all EQs. Now the combined 3 tracks are in effect a multi band compressor.
8) Dancing Delays. If your track isn't grooving as much as it could it's possible you can improve it substantially by creating delays with interesting note values. This also works with reverb decays and even compressor release times but with a more subtle effect. First ask yourself what division of the beat might add a little momentum. Try drumming along with the track and try to get a feel for what may help the groove. Maybe you have a very sparse 1/4 note pattern that could use some 1/8 th notes. Conversely if it’s on the busier side use slower subdivisions like 1/2 dotted quarters. Most plugins have note values and a sync option which can be convenient and fast but I recommend disabling it and just using your ear. Personally I have found the most interesting delay times by fishing around for the right groove. You can measure beats between drums with the time ruler in ms for setting delay times.
9) Overdriving Bass. Don't have a SVT in your 400 sqft condo to drive your bass track? You can get nice subtle (or not so subtle) distortion in the bass track without losing your bottom with your DAW. Create a duplicate track of your DI'ed bass and instantiate your favorite distortion plug or amp simulator in the second. Next follow the distortion plug with an EQ. Once you find a distortion sound that you like cut out most of the low end (experiment) with a high pass filter. Now when you combine this with your DI track it'll have the buzz and the bass. This comes from the much used practice of splitting the bass DI and sending it to a guitar amp such as a Marshall.
10) Mic with the mix in mind. Do you find yourself cutting bass and low mids out of almost everything when you mix? If so, next time you track think about what the depth of your mix might be like. Everything can't be in the foreground. Close mic'ing adds low mids and bass due to the proximity effect as well as increased high end so, if you want it back in the mix, record it back in the mix by taking an extra track of room ambience. It will sound more believable than EQing everything later. With unlimited tracks in DAWS now, I always record a room mic for distance.
11) Equalizing tracks. Consider what key the song is in before you reach for the EQ. For example, low E on a bass guitar is 41Hz. The second harmonic would be 82Hz. A4 on a piano would be 440Hz. There is a complete musical/frequency chart at http://www.phy.mtu.edu/~suits/notefreqs.html.
12) Defining The Back Wall. One of the best tricks for creating more depth in your mix is to pick an element of the mix that will be the 'back wall' of your mix and use this to gauge the depth of everything else you add. Your back wall could be different elements at different times but just thinking this way goes a long way to achieve more depth.
13) Kick' in Bass. As an alternative to Eqing the bass to the point of sounding like a ukulele to make it fit in with your massive kick sound, try a little side chaining. You can key a compressor on the bass track from the kick track and set it so that every time the kick hits the bass is attenuated as much as needed. Hint, have fun with the release times. By the same token you could also key an EQ or a gate the same way.
14) A.D.T. Artificial double tracking. I remember hearing that the Beatles had a machine that did automatic double tracking and I always wondered how it would work. I set out to get the effect by copying and delaying the 'to be doubled' track to a separate track and adding a chorus effect with 100% wet mix. When combined with the original signal it actually sounded like a tight double.
15) Parallel Bus. Compression can really make a drum track rock harder but if you overdo it your transients will be lost and you'll be left with a weird pumping effect. The trick to getting the cool part of 'over' compressing and retaining the transient is with a parallel bus compression scheme. The idea is you have your drum group sent to a stereo bus with a compressor, drive it as hard as you like and then blend it in with the uncompressed drums. Sometimes known as the New York compression trick.
16) Glyn Johns Drum Method. Get great drum sounds with only 3 mic's. Renowned engineer Glyn John (Led Zeppelin, Who, Rolling Stones) used this method to great effect and so can you. Set up and overhead mic, preferably a large diaphragm capacitor, over the snare looking down towards the space between the snare and kick beater. This position is how many Beatles mono drum recording were made using only a kick drum mic in addition. Next set up a second like mic just above the rim of the floor tom looking across the snare toward the hats. These should both be exactly the same distance (38-40") away from the centre of the snare. (use a tape measure) Next mic the kick as you normally do or perhaps a little farther out if your using another large capacitor. I often enhance this set up with a room mic, close tom and a snare mic but it can sound surprisingly good just as a trio.
17) Making Guides. If your getting guides tracks ready for a drummer to play along to, the way you prepare these can have a huge impact on the outcome of your drum tracks. It is extremely hard for drummers to play to a click when the guitars and vocals, for instance, are rushing ahead or are too loose to the click. It's also worth thinking about the impression your making on the drummer. Is he going to be inspired and able to imagine the song as the hit record you know it will be. In a nut shell 'it matters, big time' what your guides sound like, so take your time even thought there just guides. Maybe they aren't.
18) Defining Sections. One of the most common problems in songwriting is the lack of clarity between sections. When auditioning new material to be produced I listen for this and if there is any doubt about what section I'm listening to I try to come up with suggestions that can help contrast sections. There really shouldn't be any doubt about whether or not you’re on the chorus, verse, or bridge. Sometimes this means changing this first chord of the chorus if it’s the same as the verse. Other times it could mean chopping out or adding a few bars or beats. There are also many examples were the instrumentation and arrangement of the sections defines them clearly.
19) Which pad to use? When recording a loud signal, I am sometimes forced to use a pad (often-20dB) to lower the signal. Most condenser mics and consoles will have a pad, the question is, which sounds better? It only takes a second to compare the two and you will be surprised at the difference.
20) Small Room to Big Room. Are you recording drums in a small carpeted space such as a rehearsal studio? Set up a room mic anyway. When it comes time to mix you could send your small room mic track to a mono to stereo convolution reverb such as Altiverb or Logic's Space Designer and create the impression that it was recorded in a much larger room. You can also play around with delaying the room mic's region in your DAW by actually dragging it a bit later in your session.
21) Setting Compressors. I love the sound of compression on many things but setting it up wrong can really mess up your transients and introduce unwanted pumping . Try temporarily setting your compressors ratio high and the threshold, or input on some models, to be sensitive. With these settings you will really hear the effect of the attack and release. The attack can create an unwanted click sound when set too fast so make sure you’re letting enough through. 14ms should allow the attach through without creating a square wave. For the release time I try to get a musical rhythm going. You'll have to play around with this for each song. Once you get the attack and release sounding right it's time to set the ratio and threshold to a more subtle setting like 3:1 and maybe compressing 6db only on peaks.
22) Acoustic Guitar. Double tracking acoustic guitar and panning them left and right is a very popular and is a nice effect, although maybe a little unnatural. For an alternative approach, get a second guitar player to play the same part and record them together a distance apart using a stereo array such as an ORTF or Blumlein pair. The 2 guitars will bleed together naturally and create a nice stereo image. Additional guitars could be single mic'ed and placed across the stereo field. My mic choice for acoustic would be a small diaphragm capacitor like a KM84 or similar. You can also add a twelve string or high string guitar up the middle.
23) Turn a Pad Into a Rhythmic Element. You can turn a sustained pad into a new rhythmic element by using a keyed gate and a delay. First place a gate on the pad track. Pick a rhythmic track from your session that will trigger the gate open, like the kick drum or snare and send it to and Aux track. The output of the Aux will feed the key input of the gate. Set the attack and release for desired length and then instantiate a delay on the pad. Also try the reverse having the delay on the Aux track triggering the gate.
24) Ear Training For Mixing. Train yourself to recognize frequencies faster. Load a song you won't mind hearing over and over again, into your DAW. Preferably one that has a wide range of frequencies. Instantiate an EQ. Open up the automation lane for the track and automate a boost of 10db in every octave. 31Hz, 64Hz, 125Hz, 250Hz, 500Hz, 1kHz, 2kHz, 4kHz, 8kHz, 16khz. Pace them so that they're boosted for 2 bars and not boosted for 2 bars at 100 BPM. Also automate a mute in the music track for the not boosted 2 bars. Now do the same for cuts. When your done you will record yourself announcing every frequency after the boosts and cuts. Eg '500Hz cut' 16kHz boost etc. Now you will mix down those 2 tracks together so you have a recording of music boosted and cut at every frequency and an announcement following each one. Next, cut the region into 20 individual recordings of every frequency boosted and cut. Copy and paste these about 3 times and shuffle the order around randomly. You now have a 15 minute or so drill of random frequencies cut and boosted. Your job is to guess the frequency before the announcement. Do it every day for a month.
25) Drum Mic'ing. When setting up close mic's on a drum set, try to keep the mic's pointing in a parallel direction as much as possible to avoid phase problems. For example if the snare mic, high hat mic, tom mic's are all pointing the same direction the reflection will in better phase than if they're picking up reflection from different surfaces.
26) Classic Drum Dampening. Once I was playing a gig and noticed the drummer had an unusually tight funky snare sound. I looked over and saw he had his leather wallet folded over the side of the snare. In the studio you may need to use a bit of duct tape to keep it in place but it's a really cool effect. Another dampening method from the Beatles is the use of tea towels on drums for extreme dampening. It's a very cool sound and can be used to more or less a degree by having the towel taped 1/2 or 1/4 over the head of the snare or toms.
27) The Steed Effect. In the old days at Abby Road real chambers (rooms) were used for their reverbs. Delay was used as a pre delay for the chamber send but another use was the Steed effect, which stands for, send tape echo/echo delay. Meaning tape echo, to reverb chamber. The difference is for the steed effect they used controlled feedback to send multiple delays to the reverb chamber lengthening the decay time. Listen to Paperback Writer to hear a good example. Try this with your tape echo feeding your favourite reverb unit.
28) Recording Choir Sections. If you need to capture a large section and a soloist at the same time, for example a choir with a soloist. Place the soloist(s) in the back row so the main front mic's pick up everyone and give the soloist spot mic's (cardioid) behind the main group. This way the spot mic's won't pick up the choir directly.
29) The 3:1 Rule. When two mics are picking up a single sound, the time delay from the further mic combiner with closer microphone can cause phase cancellations. What sounds nice in stereo may cancel in mono. The 3:1 rules states that two mics should be placed apart from each other at least three times distance from the sound source to minimize phase cancellation.
30) Figure of Eight. Before the use of head phones in the studio engineers used the extreme null side of a figure eight pattern mic toward the studio speakers to record vocals without headphones. The null side of a figure eight mic has more rejection than the back of a cardioid but it is more narrow. You still had to keep the volume of the speaker at a minimum but this is how it was done and is a useful tip in reducing spill with a figure eight pattern.