Home

Recording Secrets for the Small Studio

1. 393 APPENDIX 3 Quote References 423 APPENDIX 4 Picture Credits 439 INDEX OF NAMES 441 SUBJECT INDEX 443 Contents This page intentionally left blank ix This book hasn t been the easiest of births and would not have been possi ble at all without the many people who have helped it on its way First of all I d like to thank everyone who has emailed me about my writing and those who have peppered me with questions at the various seminars workshops and conferences I ve been involved with thereby sowing the seeds for this book I m also very thankful for the forbearance of all the musicians who have helped me carry out the numerous practical tests which were vital in road testing the advice offered here Mike Zufall has also been instrumental in mak ing this project possible by developing a content management system for the Cambridge Music Technology web resource pages which has saved me an inor dinate amount of time ddling with HTML I must also express my gratitude to Bruce Lee Dr Elisabeth Wadge and Rovio s Angry Birds franchise for much needed motivation during the lengthy writing process In addition I d like to applaud all the
2. Trevor De Clercq Karl Coryat Tim Crich Joe Dochtermann Wes Dooley John Eargle F Alton Everest Carlos Lellis Ferreira Lynn Fuston Nikolay Georgiev Mike Gray David Greeves Christian Hugonnet Sam Inglis J rg Jecklin Mike Major J rgen Meyer Mallory Nichols Alec Nisbett Harry F Olsen Bobby Owsinski Acknowledgments Acknowledgments x Adrian Revill Hugh Robjohns Mike Ross Trevor Francis Rumsey Eberhard Sengpiel G nther Theile Pierre Walder Paul White Michael Williams Helmut Wittek Chris Woolf Wieslaw Woszczyk and J rg Wuttke The unenviable task of reading through early drafts of this text fell to Timo Carlier Daniel Plappert and Sam Inglis and I am indebted to them all for their in depth feedback insightful suggestions and continual encouragement I would also like to express my appreciation to the Editorial Department of Sound On Sound magazine in general who have been extremely supportive during this book s rather extended gestation In preparing suitable images I have been assisted enormously by Kim Campbell of Linn Records Stefan Gienger at Munich s Mastermix studios Matt Houghton at Sound On Sound magazine Blake Lewis of Stewis Media Neil Rogers of Cambridge s Half Ton Studios Matthias Schaaff Rainer Schwarz and all at SAE Munich and Chris Woolf of Microphone Data I also extend my sincere thanks to all the staff at Focal Press whose patience and expertise have played such a large role i
3. ing called howlround or howlback Howlround isn t easy to ignore so you ll know pretty quickly if you ve got a positive gain feedback loop in your recording chain but if the gain around the feedback loop is less than unity the DIGITAL CLOCKING AND JITTER If all you wish to do is digitally connect a stereo playback machine to your recording system then you shouldn t have to tangle with anything more complicated than a single S PDIF or AES EBU cable However some larger scale DAW systems may incorporate several independent digital converters at which point things get a bit more complicated Either you have to fi nd some way of synchronizing all their sample rates because the destination device can t slave itself to several sources at once or you have to have additional circuitry at the inputs to continuously convert the incoming sample rates to match the destination sample rate The former solution is cheaper to implement so it s the one you re most likely to meet in small studios but if you re not careful it may lead to digital glitches and can also introduce subtle sample rate fl uctuations called jitter which make analog to digital conversion less accurate Because such problems may affl ict any of a recording system s inputs even this chapter s simple stereo recording task may require you to tangle with them so here are some tips that should hopefully keep you out of trouble The most common schem
4. so you ll usually get a cleaner recording if you use balanced connections wherever possible Most playback devices have stereo outputs which means you ll need to make con nections for both the left channel and right channel audio signals There are a few extra things to be aware of with XLRs Because male and female XLR plugs interlock it s possible to chain several cables into a single longer cable run Many female XLR sockets plugs have a locking design so you won t be able to unplug them until you press the special lock release buttons shown in Figure 1 1 Little alignment dimples on XLR plugs and sockets mean that the plugs will only insert fully when the internal pins are correctly oriented This means you don t actually need to see an XLR socket to plug into it correctly you can just twiddle the XLR plug between your ngers while holding it gently against the socket until you feel the dimple engage whereupon the plug can be inserted fully Some XLR inputs are able to feed a 48V phantom power supply back along the signal cable which may damage line level outputs so make sure the phantom power is switched off Occasionally you may come across a so called combi jack XLR socket which looks a lot like a normal XLR socket but which can also accept a jack plug instead if you wish There s nothing extra you have to know about using these inputs though just insert the plug as y
5. No Mics PART 1 PART 1 This page intentionally left blank 3 Recording a Machine CHAPTER 1 CHAPTER 1 Although much of the magic in many productions stems from a marriage of live performers and microphones I d like to start this primer by eliminating both these variables from the proceedings so that we can rst concentrate on the bedrock studio techniques you ll need for pretty much every recording job you attempt As such our rst goal will be learning to record samples from playback machines that directly output an electrical signal things like radios TVs CD players and mobile devices On the face of things this might seem a rather unedifying prospect but bear with me because it s actually the quick est way to fast track your basic technical skills If you can learn to do this task right every time then you ll avoid the embarrassment of elementary goofs and the frustration of unnecessary delays once musicians are in the room In addi tion this activity should iron out the most common small studio con gura tion problems thereby preparing your core recording system for the rigors of serious music sessions 1 1 HOOKING THINGS UP So what do you need to know to connect the outputs of such equipment to your recording system First off you should realize that plugging anything in can generate powerful signal spikes which could easily damage your monitors and or ears Seeing as deafening yourself probably isn
6. The way most of these algorithms squeeze the data bandwidth is by reducing the recording resolution whenever the human hearing system wouldn t normally notice contrary to what you might expect we don t actually perceive everything in a musical signal all the time and background details in particular are often rendered inaudible by strong foreground sounds This drop in resolution would be fi ne if all you were planning to do with your recording was play it back through one side of a pair of earbuds natch but the moment you apply any mixdown processing to the audio you ll fi nd yourself unveiling previously inaudible and unmusical distortions introduced during the data compression process 1 3 3 Adjust Gain Through the System Now make sure your monitors are muted and restart your playback device it ll save time if you can shuttle to the loudest section for level setting pur poses Adjust your rst gain stage while watching the meter for the following level checkpoint The reading you re aiming for will vary depending on the type of meter you re using so here are some guidelines Digital Peak Meter This shows the instantaneous level of a digital signal sample by sample At the top of a digital meter is 0 dBFS which is the high est level the system can capture before overloading The basic level setting tactic here is to make sure the signal is as hot as possible but without ever hitting the top of the scale Dig
7. a simple step which con rms that what you re metering is really the playback device and not some other unwanted signal Now restart the player pull down the recorder track s monitor channel fader unmute your studio monitoring system and carefully fade up the recorder track to con rm that what you re hearing tallies with what you re seeing All of which should take no more than 20 seconds if all s well after which you re ready to set recording levels If nothing comes through though then it s time to troubleshoot 1 2 2 Divide and Conquer The secret to line check troubleshooting is to make like Daft Punk Keep a cool head Unless you work methodically you ll waste masses of time going down blind alleys When a signal isn t reaching the recorder track your rst call should be the recording system s input metering in other words the very rst meters the input signal hits after exiting the connection cable These might be single LEDs next to the input sockets or something much more sophisti cated within a computer soundcard s software control utility but unfortunately they re not always that easy to nd the input channel metering in most DAW software rarely ts the bill for line checking for example because a signal can so easily be misdirected by the audio drivers prior to that Whatever form the system s real input meters take it s paramount that you nd out where they are because they al
8. a soldering Jedi then you can save yourself a packet by constructing your own cables in the rst place Plastic bodied con nectors are also a false economy in my opinion as the casings have a tendency to crack even at the best of times and only get more brittle with age As long as a line level audio cable has the appropriate plugs it ll usually do the job but there are a few exceptions that you need to be careful about if you re rooting through an unknown box of leads Firstly some TS jack cables are designed for connecting the high level output of an ampli er to a speaker and don t offer as much protection against external electrical interference The best way to identify these speaker level cables is to unscrew one of the TS jack plugs and have a look inside see Figure 1 6 a line level cable will usually have a single plastic insulated core for the audio signal surrounded by an earthed wire mesh screen whereas a speaker cable typically has two insulated cores of a thicker gauge without any mesh screening Cables designed for digital audio signals are also constructed differently from normal analog leads but because they re usually clearly labeled by FIGURE 1 4 Where an unbalanced stereo output is provided on a TRS jack or TRS minijack connection cheap adaptors are widely available to convert between these plug sizes bottom right or indeed to make the left channel and right channel signals available
9. and ensembles understanding how different recording methodologies suit different artists and genres and being able to adapt your approach to get the best out of low budget gear and untreated acoustics While I make every effort to fast track the learning process within these pages the art of recording is by nature tremendously complicated and full of nuanced decision making so this book is ultimately intended to re ect that reality rather than brushing all the subtleties under the carpet I m also not going to show you how to operate any speci c brand of studio gear that s what equipment manuals are for In essence this book is primar ily about recording technique rather than recording equipment The infor mation here is deliberately platform neutral so that you can use it with whatever studio hardware or software you happen to have access to BEFORE YOU START Although I ve done my best to make this book accessible to studio newbies there is nonetheless some basic background knowledge that you ll need in order to get the best out of what I ll be writing about In particular I m assum ing that the reader already understands something about the fundamental physics measure ment and perception of sound has some idea of the main stages involved in the multitrack production process Introduction Introduction xii can identify the main functional components of hardware and software recor
10. can do to x a damaged CD or Minidisc but at least you can save yourself some troubleshooting time by spotting the nature of the fault swiftly Cable Issues When audible glitches are combined with sporadic error mes sages from your digital recording device along the lines of Digital Clock Unlocked or Digital Synchronization Error this usually means you ve got a problem with a digital cable or your digital clocking sychronization scheme isn t set up properly It s most likely to occur if you re using nor mal XLR or RCA phono cables instead of correctly speci ed AES EBU or S PDIF cables but can also happen if an optical S PDIF cable s internal lament has been damaged Even bending an optical cable too sharply en route between input and output may cause the data to become unreliable if it causes too much light to refract out of the optical ber rather than being re ected successfully along it However if you use purpose designed digital cables and treat them carefully this kind of data corruption shouldn t ever make it onto your radar MULTI CHANNEL DIGITAL CONNECTIONS Although we re only focusing on recording stereo in this chapter it s not unusual for even quite unassuming studio setups to have a multi channel digital connection somewhere in the recording chain most frequently between the output of a multi channel analog to digital converter box and the corresponding digital input on a di
11. for your recording rig Like VU meters they re primarily intended for use with analog equipment but they respond much better to transients For this reason you can set a recording level only about 10 dB below the analog overload point on a PPM without any real danger of distorting unmetered signal peaks Irritatingly though there are several different labeling standards commonly used for PPMs some of which use decibel markings while others use an arbitrary numeric scale so you ll need to work out roughly where the 10 dB below overload point appears in your speci c instance if you re going to use PPMs to judge your gain settings Uncalibrated Meters Metering is one area where manufacturers of budget equipment tend to cut corners most commonly by replacing fully featured metering with a single LED The simplest design just lights up when it senses any signal signi cantly stronger than the unit s own noise oor and is there fore mostly just a line checking tool More useful for level setting is the overload clip LED which warns you of impending distortion on signal peaks so your primary concern while setting recording levels is to turn up the gain as far as you can without triggering that These two single LED meters complement each other quite well so they re often twinned in practice or else have their functions combined via a single variable color LED When you ve achieved the meter reading you re afte
12. is distributed in this form or in a data compressed fi le format directly derived from it you should at the very least record at 44 1 kHz 16 bit if you re planning on releasing anything to the general public However to make best use of the CD noise fl oor it actually makes sense to record at a higher bit depth so that the digital noise fl oor of your recording won t rise above that of a CD even if you increase the levels of your recordings during mixing and mastering For this reason I suggest working at 24 bit resolution instead which drops the digital noise fl oor well below the noise fl oor of any other equipment in a typical small studio at which point you can stop worrying about it The downside of 24 bit audio is that it takes 50 more storage space than 16 bit but nowadays this really isn t a big deal given the ridiculously low cost of digital storage Some DAW platforms give you the option of recording at 32 bit fl oating point as well but I wouldn t waste further disk space on that frankly it s overkill for practical recording purposes unless you re the kind of person who stores their CDs in the fridge to keep them fresh The choice of sample rate for recording is a more contentious issue In addition to the CD quality rate of 44 1 kHz a 48 kHz rate has long been standard in the broadcast and fi lm industries on account of its ease of synchronization with video equipment To be honest it matters very little whi
13. of the recording system and using that as an earthing extension gaffer taping the free end so that its earth conductor touches some earthed metalwork Alternatively if you have any earthed audio gear you re not using for your recording a redundant connection between that and your recording system might also do the job FIGURE 1 13 Audio equipment powered via a wall wart mains adaptor left or a gure eight mains plug right typically makes no connection to the mains earth PART 1 One Source No Mics 28 1 4 3 Avoiding Digital Errors Errors in the data stream of a digital audio signal are usually identi able as little click pop sounds that don t bear any obvious relation to the rest of the audio signal they don t get worse for louder signals and you can t usually predict exactly when they ll occur There are several common causes for them in the small studio each of which calls for different remedial action Damaged Optical Media If the surface of a CD or Minidisc is damaged and that suf ciently interferes with the player s laser optics to defeat the medium s built in error correction mechanism then you may get audio glitches and drop outs Unlike most types of digital error these often occur in a regular rhythm directly related to the disc s revolution speed a useful troubleshooting clue especially if you can actually see the disc spinning in the player There s not a tremendous amount you
14. preferable to the silence of the grave FIGURE 1 12 Two cables which can reduce earth loop hum when connecting the outputs of unbalanced gear The upper schematic shows a lead for connecting an unbalanced TS jack input to an unbalanced TS jack output the lower schematic and the photograph show a pseudo balanced lead for connecting an unbalanced TS jack output to a balanced XLR input Recording a Machine CHAPTER 1 27 Transformer isolation will also work with balanced connections but in a fully balanced scenario there s also a cheaper option You can actually break the loop by snipping the earth conductor away from the connector pin at one end of the audio cable without compromising its operation Cutting the earth conductor on unbalanced connectors is inadvisable because it makes the signal more vulnerable to unwanted interference However in my experience this dodge is hardly ever necessary in smaller systems so I wouldn t recommend it unless you ve already tried all other possible solutions because it makes your customized cable ineffective for interfacing with unbalanced gear and it ll need soldering to repair Whatever you do though don t ever try to break an earth loop by disconnecting the earth conductor of a mains cable unlike the earth conductor in an audio cable a mains power cable s earth conductor is there to stop people getting barbecued by electrical faults In the grand scheme of things a little hum s pr
15. site for achieving this is that you set appropriate signal levels throughout your recording chain For one thing low level noise will inevitably be added to your recording by any recording equipment you use so you want to keep your audio signal level higher than this noise oor the higher the bet ter in fact in order to maximize the recording s signal to noise ratio On the other hand overloading or clipping your recording gear by feeding it levels that are too high will produce unwanted distortion and the further you push the level beyond any unit s capabilities the more audible this distortion will become So in the normal run of things your aim is to keep signal levels high enough to minimize noise but not so high that you trigger undesirable clip ping distortion Recording a Machine CHAPTER 1 15 This seems quite simple on the face of it but there are a couple of complica tions in practice Firstly different stages in your recording chain may require different optimum signal levels and secondly there are often so many meters and gain controls on hand that it s easy to get confused about which ones to SAMPLE RATES AND BIT DEPTHS The limit of CD quality sound is largely determined by its standardized 44 1 kHz sample rate which restricts the upper frequency response limit to around 20 kHz and 16 bit sampling resolution which results in a noise fl oor at roughly 96 dBFS Given that most commercial music
16. t the best way to begin a recording course do ensure that you mute your speakers or headphones at the outset The simplest way to do this is by silencing your system s monitor ing outputs in some way you could mute your mixer s master output you could hit your monitor controller s Mute button or disengage its output selec tor switches or you could just turn down the volume control on the ampli er feeding your speakers or headphones 1 1 1 Connector Basics Most modern playback devices and sound modules output analog signals at line level which is an alternating voltage roughly in the 1V ballpark Getting PART 1 One Source No Mics 4 this signal to the inputs of your recording system usually involves one of the hardware connectors illustrated in Figure 1 1 Each of these provides one earth or ground conductor and at least one con ductor for carrying audio signals Where two audio conductors are provided for example in XLRs and TRS jacks both still typically carry only one audio signal but in a so called balanced con guration that better protects it against FIGURE 1 1 The most common plugs and sockets for making line level analog audio connections Recording a Machine CHAPTER 1 5 electrical interference en route to its destination Connectors such as RCA pho nos and TS jacks don t offer this additional defense and their unbalanced signals are often transmitted at slightly weaker levels too
17. Recording Secrets for the Small Studio This page intentionally left blank Recording Secrets for the Small Studio Mike Senior First published 2015 by Focal Press 70 Blanchard Road Suite 402 Burlington MA 01803 and by Focal Press 2 Park Square Milton Park Abingdon Oxon OX14 4RN Focal Press is an imprint of the Taylor amp Francis Group an informa business 2015 Mike Senior The right of Mike Senior to be identi ed as author of this work has been asserted by him in accordance with sections 77 and 78 of the Copyright Designs and Patents Act 1988 All rights reserved No part of this book may be reprinted or reproduced or utilised in any form or by any electronic mechanical or other means now known or hereafter invented including photocopying and recording or in any information storage or retrieval system without permission in writing from the publishers Notices Knowledge and best practice in this eld are constantly changing As new research and experience broaden our understanding changes in research methods professional practices or medical treatment may become necessary Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information methods compounds or experiments described herein In using such information or methods they should be mindful of their own safety and the safety of others including parties for whom they have a professional res
18. any signal conductors in the process and the gizmo s outputs will then safely accept TS jack and or RCA phono plugs to feed your unbalanced recording inputs Although transformers will inevitably alter the recorded tone a fraction I wouldn t lose any sleep over this if you nd your rig needs them unbalanced inputs usually only appear on the cheapest record ing hardware nowadays in which case any minuscule transformer related side effects should be a long way down your list of quality control concerns While normal TRS jacks mostly carry balanced signals in the studio a half size minijack version of the same connector is frequently employed on con sumer computer hardware for unbalanced stereo line outputs In this scenario the two audio signal conductors carry the left and right channels respectively while the earth conductor is shared Headphone outputs use TRS jack mini jack connections in a very similar way so can be treated as line level outputs for recording purposes Functionally speaking TRS jacks and TRS minijacks FIGURE 1 3 Not all balanced TRS jack outputs will operate correctly with a TS jack plug inserted If in doubt it s safest to unbalance the signal via a transformer isolator box such as ART s T8 or Ebtech s Hum Eliminator Recording a Machine CHAPTER 1 7 are identical so it s easy to convert between them with cheap adaptor plugs and there is also a wide variety of adaptors that will take a ste
19. are musicians waiting to play The biggest obstacle small studios face with earth loop problems is that they usually only become worryingly audible once there s a whole tangle of earth loops interacting with one another at which point you simply can t trou bleshoot them in the same straightforward way you can a duff cable or a bit of distortion there are too many variables As such these two preventative mea sures are a better recourse Use Star Mains Wiring One way to reduce the impact of any earth loops within your system is to elimi nate as much of the building s mains wiring from them The way to do this is to connect your whole recording and monitoring system to a single power socket Whether you do this via a dedi cated multi output power distribution unit or just by chaining several multi output plug boards together is a question of budget as much as anything but either approach should substan tially reduce the audible effects of any earth loops Use Balanced Connections Wherever Possible While this won t break earth loops it can decrease the level of the unwanted noises enough to sidestep the problem If these measures don t give you a workable system or are impossible to implement in your speci c studio scenario then the only reliable way to deal with earth loops is by rst unplugging all the audio cables in your system with the exception of the cables connecting your studio monitors to t
20. as you turn down the gain If it does then clipping is happening before that gain It may seem obvious but you ll feel like a prize chump if you dismantle your whole recording rig in search of setup gremlins only to discover that the signal degradation you re hearing is actually part of the playback medium itself Recording a Machine CHAPTER 1 23 control in the signal path if the degree of distortion reduces then clipping is happening after that gain control Where noise is the concern ask your self whether the signal to noise ratio remains consistent as you turn down the gain in other words whether the noise level reduces roughly in pro portion with that of the desired signal If it does then unwanted noise is being added before that gain control in the signal path if the signal to noise ratio gets worse i e the gain change doesn t affect the noise as much as the desired signal then the unwanted noise is being added after that gain con trol These questions answered you can now reach for another gain control within the portion of the recording chain still under suspicion and keep repeating the process to home in on the primary culprit SOME USEFUL ALARM BELLS One of the things engineers gain with experience is a kind of sixth sense which provides advanced warning that something somewhere is wrong with their setup Since this instinct is usually hard won a reward for the litany of embarrassing mistakes ev
21. ch you choose for music work although I marginally favor 44 1 kHz so that I don t have to convert the sample rate for CD mastering a process that can have audible side effects However in recent years manufacturers have begun offering higher rates as well based on doubling and redoubling the 44 1 kHz and 48 kHz standards to 88 2 kHz 96 kHz 176 4 kHz 192 kHz and beyond extending the upper limit of the captured frequency range well beyond the 20 Hz 20 kHz zone commonly regarded as the range of human hearing The extent to which frequencies above 20 kHz do actually infl uence our listening experience is very much a moot point but there are also well understood technical reasons why elevated sample rates can actually sound better even below 20 kHz and many professionals have already voted with their ears and wallets by moving to 88 2 kHz 96 kHz in particular On the face of it this should be a strong incentive for ambitious small studio owners to follow suit but there are two big downsides to factor in The fi rst is that working at a doubled sample rate not only doubles the storage space you need but it also doubles the strain on every data buss and digital signal processor in your entire studio which frequently translates into fewer simultaneous record playback tracks fewer effects plug ins at mixdown and more complicated digital cabling The second thing is that the difference in resolution brought about by the increase in sample rate deman
22. chain In many cases the clock input s circuitry will already provide this in which case all you need to do is connect the cable but there are some units that can switch to a higher input impedance as well The higher impedance termination is provided as a means of avoiding the processing delays incurred by each unit s word clock circuitry should these begin to compromise synchronization between the fi rst and last units in a clocking daisy chain The way this works is that the fi nal unit in the chain is connected to the master with a 75 Ohm impedance as usual but all the intervening units maintain a higher input impedance which allows them to tap into this signal using special BNC T piece splitters without rendering it unreadable See the third setup in Figure 1 15 FIGURE 1 14 A BNC cable and a T Piece splitter as used for digital word clock connections Recording a Machine CHAPTER 1 31 problem becomes more insidious Where the feedback level is reasonably high or the feedback loop s gain is near unity you may hear a pronounced pitched ringing or a strangely metallic in a drainpipe timbral quality but if the feed back is slight there may only be a subtle though usually undesirable tonal change to show for it You also need to be on guard against situations where a portion of your input signal is inadvertently split off delayed by some means and then mixed back in with the undelayed signal further d
23. computer recording system s audio interface Again the dashed lines indicate which gain control and which meter would be used for setting each checkpoint s level in the rst instance all other gain stages would remain at their unity setting unless the rst choice control s gain range was insuf cient Recording a Machine CHAPTER 1 19 master output meters the cables from the studio mixer s group outputs mea sured from its group output meters and the analog to digital conversion stage measured from the digital meters most closely following it i e those within the standalone converter unit itself Note that the cable between the standalone con verter unit and the audio interface wouldn t qualify as a level checkpoint because it s not an analog connection I d have to keep my eyes on the meters for all three checkpoints when setting levels to be sure of capturing the cleanest signal DATA COMPRESSED FILE FORMATS Most affordable digital studio equipment now offers at least 44 1 kHz 24 bit recording resolution but older gear and some more modern mobile devices may offer you the option of maximizing the recording time available by using a data compressed fi le format such as MP3 or AAC Despite the fact that the sound quality captured by such compression schemes can sometimes be almost indistinguishable from uncompressed linear or PCM digital audio I would personally advise against them and here s why
24. ction or by storing your cables in loose folds or on a cable drum For more information on these methods check out this chapter s web resources FIGURE 1 6 Here are the TS jack plugs from three different audio cables with their casings unscrewed Examining the internal conductors can help to identify which are suitable for line level signals left and center and which for speaker level signals right Spare a thought for wear and tear because although studio cables are pretty hardy there s no sense in throwing money away on unnecessary repairs replacements Obviously do your best not to tread on them but also try to anticipate anything that might roll over them wheeled ightcases of ce chair casters trolleys crush them doors table legs high heels or melt them PART 1 One Source No Mics 10 lamps radiators and take evasive action Special mention needs to be made of optical digital cables however because these are quite fragile and won t even tolerate being bent beyond a certain angle without damaging the optic ber so keep the kid gloves on when handling those 1 2 THE LINE CHECK Returning to our stated task namely recording the output of a stereo play back device in an ideal world you should now be able to start playback make any necessary internal signal routing assignments within your record ing system and see the appropriate meters on your recorder lighting up Back on Planet Ear
25. der a recording unusable earth loops digital errors and feedback loops You never quite know when any of these will crop up but they re a lot more common on low budget sessions so I think it s important for small studio users to learn to recognize them and respond appropriately right from the outset An earth loop is an unbroken ring of electrical earth conductors which can arise whenever you cable audio devices together If two pieces of studio gear have power cables that access the shared earth conductor of the studio s mains electrical supply then the earth conductor in a signal cable running between them can complete the circuit as shown in Figure 1 11 The earth loop then effectively operates as a kind of aerial typically picking up a continuous pitched humming or buzzing from the alternating current of the mains supply itself sometimes accompanied by various clicks chirps and swizzles induced by other electrical equipment in the vicinity particularly mobile phones and refrigerators I ve even been treated to bursts of police radio on a couple of sessions FIGURE 1 11 A simple earth loop between two pieces of audio equipment in this case a playback machine and a mixer Recording a Machine CHAPTER 1 25 One of the reasons to address any earth loops in your system at the earliest opportunity is that they can be a pain in the posterior to remedy so you really don t want to be faf ng around with them while there
26. ding studios knows how to set up a small stereo monitoring system Many modern musicians will already have absorbed this stuff without realizing it just by coming into contact with other like minded people and following the activities of their favorite commercial artists However if you feel you might bene t from a quick refresher on any of that or you d like to clarify my usage of some of the essential technical terms involved then check out Appendix 1 where I ve provided a super condensed overview of this material Most people reading this book will I imagine have some recording system available to them but if you don t have any studio gear at all then the web resources page at the end of Appendix 1 suggests a few cost effective entry level systems to get you off the starting blocks HOW TO USE THIS BOOK Because this book has been speci cally designed as a step by step primer you ll get best results if you work through it from beginning to end Later sec tions rely on material covered in earlier chapters so some aspects of the discus sion may not make the greatest sense if you just dip in and out The complexity of the subject matter also increases progressively throughout the book so later chapters may suddenly feel rather daunting if you skip ahead At the end of each chapter there is a Cut to the Chase box which allows you to review a summary of each chapter s main secrets before proceeding fu
27. ds equal resolution of your recording and monitoring hardware and many lower cost devices simply haven t been designed with frequencies above 20 kHz in mind As a result although it s diffi cult to dispute that elevated sample rates make an audible difference to the audio quality I wouldn t recommend for anyone on a budget to bother with them In my opinion the 44 1 kHz 48 kHz rates are more than a match for practically every small studio I ve ever been into so upgrading your entire studio rig to handle elevated sample rates will rarely ever be an effi cient use of resources especially when seen within the wider context of a music market increasingly reliant on MP3 AAC fi les and media streaming technology which fall well short of even CD fi delity PART 1 One Source No Mics 16 use Neither does it make things any easier that there s no standard recording rig these days and that many small studio users are frequently working with shared or borrowed gear that s unfamiliar So in response to all this I d like to explain a step by step procedure that I ve found to be pretty foolproof for set ting decent levels no matter what studio setup you happen to be faced with 1 3 1 Find the First Gain Stage Firstly try to locate the very rst gain stage in your signal chain To give a sim ple example imagine that I m using an unbalanced splitter cable to record the TRS minijack headphone output of a portable MP3 player in
28. e No Mics 1 CHAPTER 1 Recording a Machine 3 CHAPTER 2 The Human Element 41 CHAPTER 3 Instruments with Pickups 81 PART 2 One Source One Mic 115 CHAPTER 4 Basic Vocal Recording 117 CHAPTER 5 Beyond Vanilla Vocals 155 CHAPTER 6 Single Mic Instrument Recording 171 PART 3 One Source Multiple Mics 215 CHAPTER 7 Multimiking in Mono 217 CHAPTER 8 Multimiking in Stereo 243 PART 4 Multiple Sources Multiple Mics 269 CHAPTER 9 Ensemble Recording with a Dominant Array 271 CHAPTER 10 Ensemble Recording with Peer Arrays 321 CHAPTER 11 Going Freestyle 365 CHAPTER 12 Conclusion 381 APPENDIX 1 Music Studios and the Recording Process An Overview 383 APPENDIX 2 Who s Who Selected Discography
29. e for synchronizing digital equipment together is to distribute a dedicated digital word clock signal from a master unit to which a number of slave units are synchronized In small systems the usual approach is for one of the units to daisy chain its word clock output to each of the slaves in turn using special digital clock cables whereas more complex multimedia rigs may use a dedicated multi output master clock unit to feed each slave directly instead See the fi rst two setups in Figure 1 15 In either scenario every slave unit needs to be switched into its designated word clock synchronization mode When using the daisy chaining approach it s best for the master unit to be something that s doing analog to digital conversion because slaving any digital converter to an external word clock signal will usually increase its jitter If there s more than one ADC in the system then choose the higher quality unit to act as the slave because it ll normally have superior clock reading mechanisms to minimize the added jitter Word clock cables are very similar to digital audio cables in design but use BNC connectors featuring a bayonet style locking system to hold them in place a quarter turn of the plug s rotating collar secures it in place One complication with BNC word clock cabling however is that every chain of BNC cables needs a 75 Ohm impedance at the end to stop the clock signal refl ecting back and forth along the
30. eferable to the silence of the grave EARTHLESS SYSTEMS Having too many earth connections can cause earth loops but you can also encounter similar sounding problems if your studio has no mains earth connection at all Fortunately an earthless fl oating system is pretty rare because there s usually at least one piece of equipment in any setup that connects to the mains electrical earth via its power cable However low budget home recordists often rely heavily on consumer grade electronic equipment powered via external wall wart power adaptors or fi gure eight mains connectors Figure 1 13 neither of which make an earth connection So for example if you were recording a battery powered MP3 player via a USB powered audio interface to a laptop with a wall wart power adaptor and you were monitoring on headphones there d be nothing electrically connecting the whole system to earth In practice for line level signals this probably won t impinge drastically on the signal quality as long as you get your gain settings right If it does though you can usually scotch the unwanted noises by improvising an earth connection Try making contact between a bit of the recording system s exposed metalwork and some other earthed item such as an unpainted plumbing pipe or the metal casing of an earthed domestic appliance If that s awkward to do in practice you can make it easier by plugging a spare cable into an unused audio socket
31. enough help here then fi nd someone who can talk you through it As well as generally minimizing undesirable signal loss or tonal coloration for the recorded sound this also does two things that are absolutely essential for the level setting process It matches the unit s input and output gain so that you get roughly the same level coming off the tape as you put onto it Therefore the tape machine s input metering can be used to optimize the signal level in its input and output circuitry simultaneously It determines how hard the tape is driven magnetically for a given input level You d need a dedicated fl uxmeter to measure this directly something which would be uneconomical on studio tape machines Instead the idea is to establish a known relationship between electrical input level and magnetic fl ux level using a special test tape during alignment so that the tape machine s input metering can guide you in minimizing noise and distortion incurred by the tape itself What this all means is that you can treat an analog tape machine as a third type of level checkpoint in Section 1 3 let s call it a tape checkpoint and you should use the machine s own input metering to measure the signal level Although the noise distortion response of the tape will depend on the specifi cs of the alignment process the suggested meter readings in Section 1 3 3 still provide a reasonable starting point However it s important to real
32. eryone makes as they learn hopefully I can spare you a few blushes by sharing a few little things that always set off my own mental alarm bells If any of your left channel gain settings don t roughly match the corresponding right channel gain settings There s probably a stereo balance control set wrong somewhere If the left and right meters of a stereo recording appear to be responding identically This usually means you re inadvertently recording the same mono signal twice If you have to really crank any gain control when recording a line level signal Most of the time line level signals shouldn t need much gain adjustment If turning up a gain control doesn t change your meter readings as much as you d expect There may be an undetected overload or some unnoticed signal processing going on between the gain control and the meter If you spend several minutes trying to decide on the merits of any ne adjustment It s not that it s wrong to do this it s just that it s really easy to imagine differences that aren t actually there something that has made every engineer in the history of the universe feel like a total idiot at least once in their life So do yourself a favor First give the control a good wrench to check it s actually doing what you think it is before you focus on minutiae To demonstrate these concepts in practice here s a quick case study Imagine I m hearing a disto
33. gear you re using so try to get into the habit of thinking this way right from the outset It might seem a bit pedan tic when you re just recording your radio but unless it becomes second nature you ll quickly ounder on more complicated sessions especially when you ve got frustrated musicians clients breathing down your neck Another way to apply the divide and conquer principle is if you nd that only one side of a stereo signal is coming through properly as in Figure 1 8 for FIGURE 1 8 If one side of a stereo signal s coming through but not the other swapping over the cables can help diagnose the source of the problem PART 1 One Source No Mics 14 the sake of discussion let s say the right channel s arriving safely at your sys tem s input meters but the left channel s missing in action First con rm that the right channel cable is correctly connected to the player by brie y remov ing the plug in the player s right channel output socket if all s well the right channel meter reading should disappear as you do this Then plug the right channel cable into the player s left channel output socket instead look ing at your meters to see whether the signal reappears If it does you know that both of the player s outputs are working so you should probably regard the left channel cabling with suspicion if it doesn t then either the player s left channel output is bust or there s a stere
34. gital recorder or computer audio interface There s no need to panic about multi channel digital connections though because they re usually no more complicated to deal with than stereo connections The eight channel ADAT standard uses the same Toslink connections as optical S PDIF so budget gear frequently has Toslink sockets which are software switchable to use either data format Eight channel AES EBU works electrically using thick multi core cables with D Sub connectors 9 pin or 25 pin equipped with locking screws that can be fastened by hand Unbalanced audio lines are best kept away from any D Sub cable if possible to avoid collecting modem like interference on your recordings It s important to treat D Sub plugs with care because the pins are thin enough to be bent if handled roughly and you should be aware that the allocation of signals to pins varies between different manufacturers so be prepared to consult your equipment manuals when hooking things up for the fi rst time Recording a Machine CHAPTER 1 29 Sample Rate Mismatch Whenever you send a digital signal down a cable it s important that the sampling frequencies of the source and destination units are properly synchronized even if they re nominally operating at the same sample rate otherwise the receiving unit will periodically misread the incoming data causing a series of digital pops and clicks Most budget friendly equipment will usually prevent you getting
35. hain as well Listen to the Unwanted Signal One of the quickest ways to track spe ci c signal degradation problems is if you ve learnt to recognize their tell tale sound character For example analog clipping is generally smoother sounding than digital clipping and both are usually more related to changing signal levels than the kinds of distortion generated by faulty connections gear Likewise the generic hiss of broadband circuit noise is usually dealt with quite differently to things like hums buzzes or digital clicks pops If your ear can distinguish between these sonic signa tures you ve immediately got a head start Divide and Conquer Again If none of the above provides a de ni tive answer then you ll need to adopt a divide and conquer approach similar to that described in Section 1 2 2 Occasionally you may discover a mismatch between the noise distortion levels of the left and right chan nels of your stereo signal in which case repatching cables as before can help tracing the signal path position of a problem In most situations though both channels will be equally af icted so rather than repatching cables the trick is to grab a gain control somewhere in the middle of the signal chain preferably one of those you used while setting the metered levels and try gradually backing it off while you listen Where distortion is the concern ask yourself whether the distortion remains consistent
36. heir ampli er In this state the monitors shouldn t hum or buzz unless they re broken or a wasp s own into the vent You can then start rebuilding your studio setup one unit at a time listening after each addition for earth loop problems The moment you hear something untoward see if you can nd a way to break the earth connection in the audio cables you most recently connected Where an unbalanced cable is responsible for creating an earth loop then the easiest way to tackle it is with the same kind of transformer isolator box I ve already mentioned in Section 1 1 1 since this interrupts the earth connection As we ll see in Section 3 2 1 a DI box can be pressed into service to do the same job but a simpler transformer isolator is usually more cost effective with line level signals Alternatively if you re a whizz with a soldering iron you can reduce the level of the hum by inserting a 100 Ohm resistor and a 100 nano Farad capacitor in parallel between the earth connector and the earth conduc tor at one end of the cable and a similar scheme can also be applied to the unbalanced to balanced adaptor cable previously shown in Figure 1 2 both these con gurations are illustrated in Figure 1 12 Don t ever try to break an earth loop by disconnecting the earth conductor of a mains cable that s there to stop people getting barbecued by electrical faults In the grand scheme of things a little hum s
37. im down the variables further as illus trated in Figure 1 7 For example assuming a no show on your recording sys tem s input meters try to nd any readout on the player itself or try plugging headphones directly into its headphone socket If something s actually com ing out of the player then the nger of blame points at your cabling whereas if the player s silent then it s time to dust off its instruction manual or ring the repair shop Alternatively let s say audio is indeed arriving at your com puter system s recording inputs but not showing up at the recorder track Try inspecting your DAW s input channel metering if you get a reading then you can concentrate on tinkering with the routing and channel assignment options FIGURE 1 7 The divide and conquer troubleshooting approach in action Step 1 divides the eld of inquiry in half by using the recording system s very rst input meters those on the computer audio interface itself Whichever section of the recording chain Step 1 implicates is then divided again in Step 2 narrowing the options further Recording a Machine CHAPTER 1 13 in your software mixer if not then the input signal has probably been misdi rected somehow by your audio interface s con guration settings or perhaps the DAW is just using the wrong driver This kind of divide and conquer approach is the cornerstone of effective troubleshooting irrespective of the
38. interviewers who have done such an immense service to us all by shedding light on top level studio practice John Baccigaluppi Matt Bell Howard Bilerman Bruce Borgerson Richard Buskin Bill Bruce Rick Clark Larry Crane Eli Crews Mark Cunningham Dan Daley Tom Doyle Maureen Droney Chris Eckman Hans Dietrich Faulhaber Eddi Fiegel Tom Flint Jimmy Foot Matt Frost David Greeves Keith Hatschek Jason Hiller Nigel Humberstone Sam Inglis Blair Jackson Michael Jackson Mark Lewisohn Dave Lockwood Howard Massey Chris Mayes Wright Alex McKenzie Stephen Murray Henry Owings Bobby Owsinski Joey Ramone J Robbins Andrea Robinson Anthony Savona Jonathan Saxon Sue Sillitoe Dave Simons Craig Smith Roman Sokal Philip Stevenson Paul Tingen Marsha Vdovin Pete Weiss and Peter Wetherbee The following engineers and producers have also generously taken it upon themselves to share their insight into the recording process and I also offer them my sincere thanks Chuck Ainlay Steve Albini Tom Dowd Tony Faulkner Oz Fritz Jimmy Jam Kevin Killen Gerry Kitchingham Daniel Lanois Roger Nichols John Merchant Shep Pettibone Jack Joseph Puig Phil Ramone LA Reid Eric Mixerman Sara n Al Schmitt Mike Stavrou Ron Streicher Bruce Swedien Butch Vig and Michael Wagener Furthermore many thanks are due to the numerous music technology authors who have helped inform my own opinions on the subject of recording Bruce Bartlett
39. ital distortion is one of the nastier sound ing varieties There s no need to be too nicky about things though if the loudest signal from the playback device registers within the top 6 dB of the meter s scale you re ne Volume Unit VU Meter This is normally used for analog signals and doesn t respond nearly as quickly to fast moving waveforms as a digital peak meter does As such it tends to favor lower frequencies in general as well as seriously under reading short term level spikes often called tran sients that are responsible for both the percussive attack of drums and the note onset de nition of many other instruments such as acoustic guitar piano and tuned percussion For this reason the 0 dBVU reference level marking on such meters is usually designed to correspond to an electrical PART 1 One Source No Mics 20 level about 20 dB below the actual overload point of the surrounding ana log circuitry Therefore you can set your gain control to give a reading at the meter s reference level and you ll still have around 20 dB of headroom to accommodate the unmetered transient peaks without distortion If your VU meter eschews the dBVU scale in favor of dBu markings then you can usually treat the 4 dBu level as 0 dBVU Peak Program Meters PPMs These meters are pretty uncommon in small studios so you re only likely to come across them if you ve scav enged some broadcast gear
40. ize that refi ning your level settings in Section 1 5 2 may involve not only revising your target meter reading but also adjusting your initial alignment parameters PART 1 One Source No Mics 22 1 4 TROUBLESHOOTING SIGNAL QUALITY 1 4 1 Pinpointing the Problem In order to solve any audio quality problem you need to nd out what s caus ing it and whereabouts in the signal chain Here are some sleuthing tips Check the Source It may seem obvious but you ll feel like a prize chump if you dismantle your whole recording rig in search of setup gremlins only to discover that the signal degradation you re hearing is actually part of the playback medium itself Listen to that LP tape disc le on another play back system if you ve not already done so or plug headphones directly into the player as a con dence check Check Your Monitoring In a similar vein brie y mute the destination recorder track to con rm that any unwanted noise you re hearing through the monitors is actually coming through your recording chain and not from some other unnoticed source Another simple thing to check is that the signal isn t inadvertently clip ping on its way from the recorder track through to your monitoring hardware the big giveaway is when reducing the track s monitor channel fader level cleans up the distortion in which case you may need to apply the gain management techniques of Section 1 3 to your monitoring c
41. low you to speedily eliminate a whole section of the stu dio setup from your inquiries if those input meters light up but you re still not hearing anything then something s awry in your recording system if they don t then you need to scrutinize your sound source or its connections RECORDING FROM TURNTABLES Despite being superseded in the mass market by digital playback formats the vinyl record shows little sign of dying out especially now that scratch DJs have elevated it to the status of musical instrument Capturing the output of a turntable isn t any trickier than dealing with regular line level sources as long as you realize you can t just record the cartridge s raw signal it ll require both amplifi cation and heavy RIAA standard equalization fi rst Fortunately most DJ mixers and turntable equipped hi fi s have built in RIAA preamplifi cation to handle all that so the trick is to take your recording feed from the DJ mixer s master outputs or from the hi fi amplifi er s dedicated line level recording outputs rather than using the RCA phono sockets on the turntable itself Also be sure that the turntable itself is correctly earthed otherwise the cartridge won t be properly shielded from unwanted electromagnetic interference This may require a separate earth wire to be connected between dedicated binding posts on the turntable and preamplifi er PART 1 One Source No Mics 12 Armed with this information try to sl
42. ly common sense but I ve seen enough small studio catastrophes over the years that I m still inclined to proffer a few basic tips The most frequent mistake is using leads that are too short thereby suspending the cables in the air as rst class tripwires and stressing the internal solder joints and signal con tacts of your sockets not a good idea with budget studio equipment in par ticular where the socketry may be pretty fragile to begin with It s also preferable to keep audio cables particularly unbalanced ones as far away as possible from your mains electrical wiring or any mains transformers in order to minimize the amount of unwanted interference these may add to the signal you re trying to record Mains transformers can be pernicious in this regard because they re so often tucked away out of sight inside everyday elec trical appliances and domestic lighting systems Networking devices such as modems routers and hubs are also best given a wide berth if possible TWISTED CABLES A classic newbie mistake is coiling up cables for storage in such a way that they get twisted A twisted cable is a complete pain in the ass because the twist s slight torque causes the cable to wrap around itself so that it refuses to lie fl at on your studio fl oor and generally creates an almighty tangle Fortunately it s easy to avoid this either by using one of several special coiling techniques which neutralize the normal twisting a
43. mmediately be visible For simi lar reasons remove any unnecessary plug ins from your DAW s recording project If you re using an external audio interface connected via a data connection such as USB or FireWire try to connect it directly rather than via any kind of multi port hub and try different sockets on the computer in case some are handled by its motherboard in different ways Finally nd the audio interface driver s con guration settings and try increasing the buffer size parameter which reduces CPU drain albeit at the expense of increasing the processing delay incurred between the audio inputs and outputs which shouldn t cause any practical problems with this chapter s recording task If none of that helps then try the soundcard manufactur er s technical support channels for ideas and search the Internet for simi lar reports from any other users of your particular interface there s usually someone out there who s already got the solution 1 4 4 Feedback Loops and Comb Filtering While setting up your recording rig it s not hard to unwittingly send a signal from some point in your recording chain back to an early point creating a cir cular signal path called a feedback loop If the overall gain around the loop is above unity then the signal level in the loop will rapidly accumulate and PART 1 One Source No Mics 30 overload typically resulting in an unpleasantly distorted high pitched squeal
44. n bringing this project to fruition in particular Carlin Bowers Emma Elder Mary LaMacchia and Ana s Wheeler Above all however I d like to thank my wonderful wife Ute for her unwaver ing love and support as well as for taking on the worst of the bibliography and referencing tasks so graciously I d be lost without you my love And nally I d like to thank my trainee limb models Lotte and Lara I can scarcely believe what pros you re turning out to be xi WHAT YOU LL LEARN FROM THIS BOOK How to achieve release quality recordings on a budget within a typical small studio environment by applying power user techniques from the world s most successful producers In my work for Sound On Sound magazine I notice the same recording mistakes cropping up time and again and while it s sometimes pos sible to salvage a usable mix from poor recordings that way of working is incred ibly long winded and laborious This book is about helping you make better use of your time and energy by leading you step by step through a series of in depth practical exercises that demonstrate how to record sensibly in the rst place WHAT YOU WON T LEARN If you want a How To Record My Band In Three Easy Steps quick x guide you ll have to look elsewhere This book has a much broader remit To teach you how to be a con dent small studio recording engineer That means having the skills to record a full range of instruments
45. nnels back to the recording inputs creating a feedback loop through the recording system or the input channels to the mixer s monitor outputs mixing the original undelayed signal with the delayed signal returning from a digital recorder Likewise many computer audio interfaces now incorporate complex DSP powered mixers that operate independently of the DAW software and they seem to take a perverse glee in encouraging very similar routing gaffes It may only be your monitoring signal that s suffering not the recording itself but it s still important to banish this malaise before hitting Record because it can disguise various sonic subtleties you may want to assess especially when we start working with microphones later on 1 5 BEYOND REASONABLE REFINING YOUR RESULTS By this point you should be able to hit Record on pretty much any compar atively modern studio system and capture a passably clean sounding version of whatever your playback machine s putting out But if you re serious about the art of recording then that simply isn t good enough You want to get the PART 1 One Source No Mics 32 FIGURE 1 15 Three different digital clocking arrangements for interfacing several external analog to digital converters with a single computer audio interface 1 a simple daisy chain system 2 a system synchronized from an independent digital master clock unit and 3 a daisy chain system using BNC T piece adaptor
46. nterface drivers and DAW soft ware mixers in particular often have little gain controls tucked away in nooks and crannies all over the place and if just one of those has been pulled down inadvertently it can throw a spanner in the works Some gain controls such as 4 dBu 10 dBV sensitivity switches may offer no unity gain position at all while others such as headphone volume knobs may have no calibration in either case start off with those at their lowest gain settings Don t forget to check that any Pan or Balance controls are centered too because both of those are essentially stereo ganged gain controls Also con rm that there are no Mute Solo buttons engaged and that any signal processing facilities have been reset or removed from the circuit particularly the channel EQ if you re recording via an analog mixer Once you ve set your whole record ing chain to unity gain do the same for the monitoring chain Recording a Machine CHAPTER 1 11 Now you can start the line check Level meters should be adequate for the pur pose so your monitors can remain muted for now Set the playback machine going turn up any output level or headphone volume control it has and examine your recorder track s level meter Don t worry too much about what the level actually is for the moment just check whether there s something there If you can see something on the readout then stop the player and check that this reading also dies
47. o balance control set funny some where under the hood Where both of the player s outputs are actually func tioning correctly your next move might be to restore the player s output socket plugs to their original positions and then swap the plugs feeding the relevant pair of recording system inputs instead If you still only get the right channel input meter lighting up in that instance then you ve removed the cables from the equation in other words there s probably something wrong with the left channel input itself Exactly the same troubleshooting techniques can be applied if you encoun ter problems getting signals out of the recording system for monitoring pur poses Again you should take pains to identify which metering stage most reliably re ects what s appearing at the physical output sockets of your record ing system because this gives you concrete information about whether or not your monitoring hardware is at fault The channel swapping technique is also very handy if you re only hearing one side of the stereo picture but with one important safety caveat You should switch off all ampli ers including headphone amps and powered speakers before replugging any monitoring connections otherwise you risk damage to your listening equipment and or hearing 1 3 SETTING LEVELS You don t want just any signal reaching your recorder track though you want something that actually sounds good The fundamental prerequi
48. og cable I call this a cable checkpoint is converted between the analog and digital domains a conversion checkpoint Optimizing the signal level at a given checkpoint should pretty much guaran tee that you ll steer clear of noise and distortion problems between that check point and the previous one provided that you were conscientious in setting unity gain throughout the signal path back in Section 1 2 1 The most suitable meter for assessing the level in each case will be the closest meter before a cable checkpoint the digital meter closest to a conversion checkpoint If you can t nd a meter for a given checkpoint that satis es these conditions before reaching another checkpoint or the rst gain stage then it means that it can t reliably be measured so strike it from your list and continue working through the remaining checkpoints as if it didn t exist Theoretically speaking disregarding the signal level at any checkpoint raises the possibility of noise dis tortion concerns but in practice problems very rarely arise because manufacturers of studio equipment tend to provide metering where it s required Bear in mind that some meters may be able to measure signals in a variety of signal chain posi tions for example the main meters of a small analog mixing console will often display the control room monitoring signal allowing you to measure the level of different channels gr
49. on separate mono connectors such as RCA phonos bottom left and top PART 1 One Source No Mics 8 the manufacturer and often noticeably less exible when handled you re unlikely to confuse them in practice As it happens you can sometimes get away with using analog RCA phono or XLR leads to transmit coaxial S PDIF or AES EBU digital signals over short distances without problems but if you hate the thought of sacri cing session time to data glitches as much as I do then spending 45 30 on a dedicated digital cable is a no brainer Another thing to realize when recording a digital source is that your recorder s sample rate must be synchronized to that of the incoming data so you ll only get the audio coming through correctly once that s been done The simplest means of FIGURE 1 5 The most common plugs and sockets for stereo digital audio connections Recording a Machine CHAPTER 1 9 doing this is to slave the recorder s sample rate to that embedded in the incoming digi tal data stream but more complex systems are also possible see Digital Clocking amp Jitter in Section 1 4 4 for more details On computer systems the synchroniza tion options are usually located somewhere in the audio interface s driver settings although somehow they always seem to be playing hide and seek whenever I need to nd them 1 1 3 Cable Layout Physically arranging cables in the studio should be most
50. ou d normally expect and away you go Recording an unbalanced signal through a balanced TRS jack input should be ne as long as you use an unbalanced TS jack cable for the connection This effectively earths one of the two signal conductors inside the TRS jack socket so that it can accept the unbalanced signal Balanced XLR inputs can also record unbalanced signals directly if you use a special cable such as the one shown in Figure 1 2 FIGURE 1 2 Schematic diagram for creating a lead to connect an unbalanced TS jack output to a balanced XLR input PART 1 One Source No Mics 6 Connecting balanced outputs to unbalanced inputs is a bit of a mine eld because the outcome depends on the design speci cs of the output cir cuitry Although modern equipment frequently has TRS jack output sockets that ll happily permit unbalanced operation if you insert a TS jack plug you shouldn t assume you can do this unless it s clearly condoned on the unit s rear panel or in its user manual some output circuits might go up in smoke Rather than take any risks in this respect I prefer to use a dedicated trans former isolation box to unbalance the signal if in any doubt especially now that companies such as ART and Ebtech offer respectable quality transformer isolation for only around 30 20 per channel A balanced output can be cabled directly to the transformer isolator s TRS XLR input without disconnect ing or short circuiting
51. oups and returns just by hitting their Solo buttons Solo buttons may sometimes be labeled PFL Pre Fader Listen or AFL After Fader Listen to indicate where in the channel path the signal is being measured Let s look at how these principles apply to our two setup examples In Figure 1 9 there d be two level checkpoints the headphone splitter cable and the sampler s analog to digital converters However assuming there was no meter ing within the MP3 player which is quite common then there d be no way of measuring the level prior to the rst checkpoint so I d disregard that one and focus all my attention on the second checkpoint instead relying on the sampling workstation s digital input meters In theory by ignoring the rst checkpoint I d risk overloading the MP3 player s analog output circuitry if I turned the headphone level up too high In practice though this would be very unlikely given that most MP3 players are designed to play back even the loudest digital les at maximum headphone volume without signi cant distortion which is why they seldom require output metering Turning to the more complicated setup in Figure 1 10 the checkpoints would be the cables from the DJ mixer s main outputs measured from the DJ mixer s PART 1 One Source No Mics 18 FIGURE 1 10 A more complicated recording chain from a turntable through a DJ mixer a studio mixer and a standalone ADC and then into a
52. own the signal chain The time difference between the undelayed and delayed signals causes a tonal effect called comb ltering or phase cancellation which often sounds quite similar to a subtle feedback loop You might think this kind of comb ltering would be an unlikely eventuality but it s actually quite easy to fall foul of given that every piece of digital audio equipment inevitably causes a small but signi cant processing delay between its inputs and outputs The most challenging thing about both these problems is recognizing them in the rst place so it pays to tune your ear to how they sound under controlled conditions because you ll then spot them more quickly in the wild I ve pro vided some pertinent audio examples in this chapter s web resources Even then you may sometimes question whether you re hearing subtle feedback loop or comb ltering artifacts or whether your ears are just playing tricks on you in which case the best sanity saver is to use headphones to compare the sound of the player s raw output with your main monitoring signal Once you re aware what s wrong it s not usually too dif cult to diagnose the causes In my experience both these issues are most likely to arise where you re recording through the same small mixer that s also routing the record er s monitoring signal to your loudspeakers headphones One wrong button press can easily route the monitor cha
53. ponsibility Product or corporate names may be trademarks or registered trademarks and are used only for identi cation and explanation without intent to infringe Library of Congress Cataloging in Publication Data Application submitted ISBN 978 0 415 71670 3 pbk ISBN 978 1 135 87950 5 ebk Typeset in ITC Giovanni Std By MPS Limited Chennai India www adi mps com To my parents Bound to Create You are a creator animation games audio media communication web design or by nothing except your own creativity and determination Focal Press can help For over 75 years Focal has published books that support your creative goals Our founder Andor Kraszna Krausz established Focal in 1938 so you could have access to leading edge expert knowledge techniques and tools that allow you to create without constraint We strive to create exceptional engaging and practical content that helps you master your passion Focal Press and you Bound to create We d love to hear how we ve helped you create Share your experience www focalpress com boundtocreate Whatever your form of expression photography lmmaking theatre you simply want to create without limitation Bound vii ACKNOWLEDGMENTS IX INTRODUCTION XI PART 1 One Sourc
54. r ther Underneath it are one or two Assignment boxes which suggest prac tical activities to help consolidate your understanding of each chapter and these could also serve as coursework tasks within a more formal educational framework Each of these assignments is deliberately limited in scope so that it doesn t call on topics before they ve been properly discussed but these limita tions naturally reduce as the chapters progress and are removed completely in Chapter 11 Finally the Web Resources box leads to a separate website con taining an extensive selection of related links and audio les all of which may be freely used for educational purposes This book is based on my own wide ranging research into the studio practices of more than 200 world famous engineers drawing on more than 5 million words of rst hand interviews The text therefore includes hundreds of quotes from these high iers If you don t recognize someone s name then look it up in Appendix 2 to get an idea of the most high pro le records they ve worked on you ll almost certainly have heard a few of those If you d like to read any quote in its original context which I d heartily recommend then follow the little superscript number alongside it to Appendix 3 where there s full refer ence information for each one Finally if you have any further questions or feedback feel free to email me on ms cambridge mt com One Source
55. r for your rst checkpoint grab whichever gain control is immediately after the rst checkpoint and adjust it if necessary while looking at the next checkpoint s meter Continue in a similar manner until you ve got appropriate meter readings for all the checkpoints Returning to our examples in Figure 1 9 there s only one measurable check point so I d just turn up the headphone volume control until the loudest play back signal registered just under 0 dBFS on the sampling workstation s digital input meters In Figure 1 10 I d rst grab the Input Trim knob on the DJ mixer s turntable channel and set that for a 0 dBVU reading on the unit s master output VU meters Then I d shift my focus to the studio mixer s group output VU meters and check whether those were reading around 0 dBVU if not I d adjust the Input Gain knob on the mixer channel s receiving the DJ mixer s stereo output Finally I d check the digital meter on the standalone converter unit and if necessary adjust the input sensitivity controls on the converter to achieve the most sensible reading Recording a Machine CHAPTER 1 21 One potential y in the ointment though is if a speci c gain control doesn t have enough juice to achieve your target meter reading Here s what to do in that scenario If you need more boost use the next gain control in line as long it still precedes the metering point If you need more cu
56. reo signal on a TRS jack minijack connector and split it out onto a pair of TS jacks or RCA phonos Occasionally it s possible to make a direct digital connection from a playback unit to a Digital Audio Workstation DAW system bypassing the analog domain entirely In that event you ll probably have to deal with one of the three digital transmission formats shown in Figure 1 5 coaxial S PDIF usu ally on unbalanced RCA phonos optical S PDIF usually on ber optic Toslink connectors or AES EBU usually on balanced XLRs All of these formats transfer stereo audio but you ll need a ded icated format conversion gadget to convert between them if you re faced with incompatible socketry 1 1 2 Choosing Cables Once you ve muted the monitors decided which audio plugs sockets you re going to use and placed your sound source within easy reach of your record ing system you re ready to hook things up Even cheap as chips consumer hi cables will carry line level signals OK but I d de nitely recommend using the more robust professional style cables designed speci cally for onstage studio use You shouldn t have to spend more than about 20 15 each for workmanlike specimens of up to 10 m in length but do try to avoid anything with molded on connectors because they re almost impossible to repair if one of the internal solder joints fails separate connectors can be resoldered Speaking of which if you re
57. rted signal coming through the rig in Figure 1 10 First I plug headphones into the DJ mixer and solo the turntable channel con rm ing to my satisfaction that the source signal s clean Then I brie y mute the DAW s recording track to check that a distorted signal isn t inadvertently arriv ing via another route and try waggling the recording track s software fader to see whether the degree of distortion changes this doesn t remedy the prob lem so I know I m not being misled by a fault in my listening system The smoother sound of the distortion suggests analog clipping to me so I try fading down the studio mixer s channel input gain control momentarily then its group output fader The former reduces the distortion the latter doesn t so I m now pretty sure there s something amiss in the mixer itself Perhaps the channel fader s been knocked accidentally some processor has been left PART 1 One Source No Mics 24 in circuit by mistake or there s some aky electronic component under the hood whatever species of technical gremlin it is I m a whole lot closer to exterminating it now that I know where it s skulking 1 4 2 Managing Earth Loops So far I ve mostly focused on tactics for minimizing broadband hiss and clip ping distortion because this is something that s fundamental for any session However there are a few other signal quality concerns that arise from time to time which can easily ren
58. s maximum quality out of a given system especially if you re at the mercy of out dated and or low budget gear which may struggle to deliver truly usable results even at the best of times So let s look at how we can re ne the basic approach so far in order to boost quality Recording a Machine CHAPTER 1 33 1 5 1 Improving the Source One of the essential truths about any recording job is this The better your source sounds the better your recording will sound Furthermore the sonic bene ts of adjusting a sound source directly will usually dwarf any improve ments attained via your studio gear and often cost less time and money into the bargain Admittedly there may not be anything you can do to improve the sound of the simplest playback devices but if you re recording from a cassette deck say you ll almost certainly get a bigger step up in quality by cleaning and aligning its playback heads than you d get by doubling the cost of your record ing system or switching to a 96 kHz sample rate Given the impact that the source quality has on the recording one of the big learning curves of studio engineering is discovering how to get the best out of each speci c sound source you re presented with so if you re not happy with the recorded sound of any playback device DAT LP cassette wax cylinder either do a bit of background reading to harvest general purpose tips or nd someone who can clue you in on the bes
59. t use the previous gain control in line So if for example the DJ mixer s Input Trim in Figure 1 10 didn t give me an adequate level on its master output meters then I might apply further gain by pushing the turntable channel s fader above its 0 dB mark and even the DJ mixer s master fader too Alternatively if setting the optimum level on the stu dio mixer s group output meters resulted in overloading the standalone con verter unit s digital meters even at the lowest input sensitivity setting I might turn down the studio mixer s group output fader to reduce the level 1 3 4 Listen With a bit of luck you should now be able to fade up your studio monitors and hear what you re about to record clean and clear With the best will in the world however there will still be times when the signal is unacceptably degraded in some way either by a quirk of the equipment or by good old fashioned user error So what can be done to x such problems WHAT ABOUT ANALOG TAPE Because analog tape machines feature in very few small studios these days I ve not complicated my main level setting discussion by including them However if you re faced with recording to tape then here s how to incorporate it into the scheme Firstly unless you have the luxury of a studio technician you ll need to acquaint yourself with the overbiasing and alignment procedure of your specifi c tape machine if the tape deck s manual isn t
60. t specialist tweaks 1 5 2 The Aesthetics of Noise and Distortion When I talked about setting initial levels using your meters back in Section 1 3 I deliberately erred on the side of caution with my suggested meter readings on the basis that excess distortion is usually trickier to remedy at mixdown than excess noise In doing this however I ve allowed you to dodge an important judg ment call that every engineer must make at each level checkpoint in their own unique rig What do you think is the best balance between noise and distortion Answering this question is partly a question of learning the nuances of your own equipment so if you re starting out I d suggest fast tracking your devel opment in that respect by which I mean experimenting with alternative gain settings at this point while there aren t any musicians hanging around getting bored Work through each of the level checkpoints you identi ed in Section 1 3 2 and try hitting them with a higher signal level bearing in mind that you may need to turn down a gain control later in the recording monitoring chain to avoid unintentionally overloading later checkpoints Discovering how each of your pieces of gear responds to overloads is an important factor in getting the best from your studio not least because it tells you in practical terms how much headroom your speci c VU meters and PPMs provide and thus what target meter readings are sensible And once you accept the ine
61. th however it s actually very common for source signals to go AWOL which is why where s the signal is probably the most common setup problem in any studio As such one of the recording engineer s primary tasks before every session is to line check the rig ensuring that all signals are reaching their intended destinations There s nothing worse on a session and nothing worse on an engineer than when things are just not working right comments Al Schmitt 1 echoing similar comments from Simon Climie 2 and Stephen Hague 3 It s like they say in the boy scouts continues Schmitt Be prepared Be on top of your game Make sure everything s working 1 2 1 Unity Gain To make the line checking process as straightforward as possible you should rst try to make sure nothing in the recording path is changing the level of your audio signal at all which means setting every gain control that precedes the recorder track to its 0 dB or unity gain position There can be hundreds of gain controls even in entry level recording systems manifesting themselves as knobs switches or faders labeled Gain Volume Level Sensitivity Pad or Trim or sometimes just calibrated in decibels Digital gain controls usually default to their 0 dB setting while analog controls often have their unity gain position marked or detented so the trickiest part of zeroing them is usually just tracking them all down Computer audio i
62. to the TS jack inputs of one of those little all in one sampling workstations perhaps one of Akai s iconic MPC range as illustrated in Figure 1 9 My rst gain stage there will probably be the MP3 player s headphone level control A more complicated setup is shown in Figure 1 10 a vinyl turntable being recorded via a DJ mixer and a studio mixing console to a stand alone stereo analog to digital converter which in turn feeds the digital inputs of a software DAW system In this case the rst gain stage will likely be the Input Gain knobs on the DJ mixer s turntable channels FIGURE 1 9 A very simple recording chain from a portable MP3 player to a sampling workstation showing the position of the gain controls level checkpoints and meters The rst checkpoint is disregarded because it has no meter to measure the signal level there Setting a preliminary recording level using this setup would therefore involve adjusting the MP3 player s headphone level control to achieve a sensible reading on the sampling workstation s digital input meter see dashed lines Recording a Machine CHAPTER 1 17 1 3 2 Identify the Important Meters Next narrow down which meters you need to concentrate on while setting your levels The way to do this is to work your way through the signal chain all the way from the rst gain stage to the destination recorder track looking for checkpoints where the signal passes through an anal
63. vitability of a trade off between noise and distortion it becomes perfectly defensible on tech nical grounds to allow some overloads if their audible consequences turn out to be a reasonable price to pay for an improved signal to noise ratio While you can often afford to leave lots of headroom for signal peaks when recording at 24 bit resolution with decent project studio equipment there s also nothing to say you shouldn t light up a few red LEDs with your signal peaks on a cheap and noisy mixer or atten the odd transient against the 0 dBFS digital ceiling of an old 12 bit sampler if your goal is substantially lower noise
64. yourself into dif culties here by refusing to unmute a digital input unless the unit s sample rate has been set to synchronize with the incoming audio data stream s embedded clocking information Equipment aimed more directly at the professional market may not be as forgiving however and the remedy will usually involve a bit of menu manual sur ng in order to track down the necessary digital clocking and synchronization settings Software Driver Problems If your recording system is based around a com puter by far the most common cause of unwanted digital detritus on the audio signal is the computer s software con guration The permutations and pitfalls here are innumerable so rather than getting too bogged down in speci cs let me suggest some general strategies that usually lead me to a suitable x First invest in a dedicated computer recording interface instead of trying to use the built in audio hardware you can get something for under 100 70 that ll work more reliably and record at a much higher quality Then check that you have the latest version of your audio interface s software driver and that you ve correctly selected it for recording purposes in your DAW s user preferences Close down any unnecessary applications running alongside your DAW software to minimize the load on your com puter s CPU resources bear in mind that some programs may be running as background processes so they may not i

Recording Secrets for the Small Studio

Contents

Download Pdf Manuals

Related Search

Related Contents