A special holiday edition of our hands-on troubleshooting column comes direct from sunny Turkey, where Paul White forsakes the beach to help Murat Yucel refine his band's recording setup.
This month's Studio SOS column goes international, coming as it does from the studio of Murat Yucel, which occupies a spare room in his apartment in Dalyan, a small riverside town in southwest Turkey. Imported recording equipment is fairly heavily taxed in Turkey, and the average income seems to be around a quarter that in the UK, so setting up any kind of studio is a bit of a luxury. Furthermore, because of flight luggage weight restrictions, I couldn't take along any rack gear or monitors, or a flightcased Hugh Robjohns for that matter!
The mainstay of Murat's studio is a Fostex VF16 16-track hard disk recorder/mixer. A hi-fi Minidisc machine is used for recording stereo mixes, connected to the VF16 via optical S/PDIF, and there is a small Behringer mixer at Murat's bar (the Blues Bar at the other end of town) which can be brought in when necessary. There's no outboard gear at all, so any processing is done using just the effects in the VF16, and there's not even a loudspeaker monitoring system — everything is done using a single pair of Fostex headphones. However, Murat has some good microphones, including a Rode NT2 with shockmount, an Oktava 219, an Oktava 012 (all capacitor models), a pair of Shure SM58s and an SM57. Other dynamic mics are available from the bar, where Murat plays most nights with his band, and because the bar operates seven days a week throughout the season and doesn't close until around three in the morning, finding time to record is also difficult. Anything involving 'real' drums must be recorded at the club before opening time, where space and noise is less of a problem than in the apartment.
Murat is a singer/songwriter who plays mainly acoustic guitar, though he can turn his hand to just about anything. Most of the time he records with friends and regular band members, where the instrumentation may be any combination of electric and acoustic guitars, electric bass and hand drum. He likes to record as many parts together as possible, so as to capture a good live performance feel, but in a small apartment room this clearly leads to separation problems. My challenge, therefore, was to try to come up with some suitable recording techniques for the instruments being used and to find a means to record electric and acoustic instruments together without incurring unacceptable amounts of spill.
I would estimate the studio room is approximately 2.5 x 3m, and 2.7m high. It is built of rendered brickwork as is common in this area and has a tiled concrete floor. At the advice of a friend, Murat had already improvised some soundproofing over the window, comprising three-inch thick plastic foam wedged between the window and a wooden board. It was quite effective as a barrier to sound, though the door was less so. This had been fitted with foam draft excluder around top and sides, but there was no sealing on the bottom and the glass panels in the door offered little isolation anyway. As it turned out, the techniques we finally arrived at necessitated recording with the door open, so this ceased to be an issue.
My first modification was to hang a thick rug (these are plentiful in Turkey!) over the window around three inches from the board holding up the foam. Leaving a space increases the effectiveness of the rug as an absorber as both the incident and reflected sound has to pass through it. The idea was that the main vocalist could record with his back to this absorber to help reduce reflected spill and to reduce coloration from the room.
There were a couple of foam cushions in the room already, as well as a block of foam stuck up in one corner on the ceiling. Additionally, there was a rail of clothes hangers with a few shirts on and a shelf full of bottled drinks to provide acoustic (and visual if consumed!) diffusion. These, combined with the rug over the window and another on the floor, tamed the room surprisingly well, though its low-end characteristics remained somewhat unpredictable. As no low-frequency sounds were being recorded in the room, this wasn't a problem, given that the monitoring was on headphones.
Initially we planned to make a test recording of vocals and guitar recorded together using the two Oktava mics, the 012 on the guitar and the 219 for the vocals. We set up the mics and playing position with Murat's back towards the window, so that the heavy rug would absorb most of the reflections. Murat had been told by a friend with some recording experience to record with the mic very close to a hard surface, presumably to get some boundary-effect low-end boost, but I felt the traditional way would work best, so we kept the mics away from walls but also not in the dead centre of the room so that we wouldn't emphasise any room modes.
There was no pop shield, so we improvised by using a plastic kitchen sieve with a nylon stocking stretched over it. A couple of elastic bands secured this to the mic leaving a couple of inches clearance between the stocking material and the mic — it worked perfectly with a singer-to-mic distance of around nine inches. However, we did have a couple of practical issues to sort out first.
Our first problem was that the capacitor mics refused to work with some new XLR cables Murat had bought. After pulling them apart, it turned out these were wired with unbalanced cable, so properly balanced versions were ordered, along with more mic stands. These arrived a couple of days later, but in the meantime we did a few tests using what cables I'd brought with me. Initially, we had the Oktava 012 taped to a music stand, as there was only one proper mic stand in the apartment (which we needed for the vocal mic), but as soon as the new stands arrived we gave it a more dignified mounting.
Problem number two was a mysterious rattle inside the Fostex recorder. Removing the hard drive access plate revealed two loose screws that should have been holding the circuit boards down. As it was impossible to replace these without dismantling the machine, we took them out. There were enough screws still in place to keep everything secure.
The resulting test recordings, made using no EQ or processing, confirmed that we were now getting a very decent basic sound. The acoustic guitar sounded a little more lively with the rug taken up, but as vocals and guitar are often recorded together, we decided the best compromise was to keep the rug down. When vocals and guitar were recorded together, a touch more separation could be obtained by angling the guitar mic downwards and the vocal mic upwards slightly. Although it would have been nice to use figure-of-eight mics to try to get more separation, this luxury was not afforded to us. The only variable pattern mic was the Oktava 012, which has interchangeable heads giving cardioid, hypercardioid and omni responses only. We used it in cardioid mode for this session, aimed at the neck/body junction around eight inches from the guitar (a nylon-strung classical guitar in this instance). This produced the most acceptable tonal balance, but, as always, I used the headphones while moving the mic to find the optimum position. The vocal/guitar separation was also quite adequate.
When we later tried to record two acoustic guitar players together, we had to move them as far apart as possible, each facing the centre of the room. This reduced crosstalk to an acceptable level, even in this relatively small room. One of the steel-strung guitars was fitted with a type of pickup I'd never seen before and which I was told was originally designed for percussion instruments. The resulting sound wasn't at all bad, so in situations where miking wasn't appropriate, or where there were more instruments than suitable mics, this was plugged directly into one of the line inputs of the Fostex. During our tests, a small amount of reverb was added to the DI'd acoustic guitar so that it better matched the ambience of the miked guitar. Even so, it still sounded better miked, so I suggested DI'ing only when the situation made miking impractical.
One major limitation of the Fostex recorder is that it only has two phantom-powered XLR mic inputs, which meant that only two capacitor mics could be used at any one time, unless the Behringer mixer from the bar was used to provide more mic amps. However, dynamic mics with jack leads could be used in the remaining inputs.
Recording a regular drum kit is not practical in Murat's apartment, but most of the time hand percussion (a Turkish darbuka drum) is used. Of the existing mics, I suggested the SM57 might be good for this purpose during a typical live session, though where a spare capacitor mic was available I felt this would give better results because of its superior transient response. For test purposes, we tried the Oktava 219 again, leaving the pop shield in place for convenience. There is generally no need to use a pop shield when recording instruments other than vocals, but leaving it on didn't seem to cause any problems and I wasn't sure about the elastic band situation!
The darbuka hand drum, which usually features a thin aluminium shell, lies somewhere between a bongo and conga in size and can be quite loud. It was quite impossible to record it in the same room as the acoustic guitar without incurring a huge amount of spill so, instead, I decided to try it outside the room in the hallway. During loud passages, the drum caused door panels in the hall to resonate, though very little of this picked up on the mic, which was only around six inches away from the drum.
Murat rightly suggested using more rugs to damp these when doing serious recording — recording gear might be scarce in Turkey, but rugs are not a problem! In fact a chair piled with rugs and cushions was placed in the open studio doorway as an acoustic barrier, which, combined with the few extra feet of distance, gave us far better separation than we'd expected. When soloing the acoustic guitar track, the drum was now barely audible. The guitarist was playing in the usual position with his back to the window, facing the open doorway, so there was good line of sight for communication between the two players.
Murat has a Carlsbro Electroacoustic amp in the studio room, so I asked him if we could try this for recording bass. As it turned out, it produced a really nice bass sound, but attempts to DI it from its slave output resulted in a lot of background noise, so we finally decided on miking it around six inches from the speaker. Miking produced a really warm, full bass sound with bags of definition and far less noise, but of course the bass spilled onto everything, so the amp had to be relocated to the adjacent lounge, where the amp was aimed at a soft sofa to try to minimise reflections. Because of the lack of headphone monitoring, the bass player had to stand between the amp and the studio door to hear both his own playing and that of the other musicians, but the solution was workable and provided excellent separation.
Electric guitar, and occasional bass, is usually furnished by Graeme Bickett, a Scottish friend of Murat's who plays guitar in the bar band during the summer season. Graeme uses a small Marshall Valvestate combo with a Fender Stratocaster, but it's still far too loud to run in the same room as any acoustic guitars, so we set this up just outside the studio door, around the corner and facing away from the studio door. It sounded fine miked with either the SM58s or SM57, but, as we had the Oktava 219 mic already set up in the hallway, I decided to see what kind of sound we could get from using it instead, as lead parts could be added as overdubs when this mic would be available for use.
The mic was set up six inches from the speaker, and initially pointed at the centre of the speaker cone. After making a test recording, we all felt that the result was rather bright, so I suggested moving the mic to one side of the speaker to warm up the tone, and also adding a little more low end on the amp EQ. Graeme said he'd been told that moving the mic to one side created a brighter sound, but we soon confirmed that the opposite was true (it's nice to know the same laws of physics apply in Turkey!), and with the mic aimed at the edge of the speaker cone we got very close to the way the amp sounded in the room. The main difference between using the Oktava and one of the dynamic mics was that the higher harmonics sounded more lively, helping to play down the usual close-miked sound.
All guitar players have to play at a certain volume to get a good tone, even with a master volume amp of this kind, so I raided one of the sofas for cushions to improvise an isolation box around the amp and mic. Rugs were then draped over the top for further isolation, but we left a small gap facing Graeme, so that he could hear what he was playing. Spill from either the bass or electric guitar to the miked acoustic guitar was minimal using this arrangement, and close to what I've heard in some professional recordings from big studios. Where the electric guitar was being recorded at the same time as the drum, the close miking of the drum avoided excessive spill from the guitar amp, provided that a minimum distance of four to five feet was maintained between the drum mic and the guitar amp, and that the drum mic wasn't directly in front of the guitar amp.
These measures dealt with the spill issues very effectively and, as long as the bass and guitar players chose positions where they could hear their amps and the other instruments, the solution was perfectly viable. However, I suggested that extra headphones and a headphone distribution amplifier should be placed near the top of the 'what to buy next' list, as this would enable all the players to hear a reasonable balance regardless of where their amplifiers were. This could be fed from the headphone out of the Fostex or from a pre-fade send.
Murat told me he'd had bad experiences with sound engineers applying excessive compression to his voice during live performance, so he was understandably suspicious of using it. He prefers to use mic technique to control his vocal levels, so I came up with the compromise of trying a combination of mic technique and mild post-recording compression, courtesy of the Fostex effects processing. I also felt that, while acoustic guitar can sound very good recorded 'as is' in a live environment, the vocals would work better with a suitably tweaked reverb setting, again using the internal effects of the Fostex.
We ended up using a compression ratio of between 4:1 and 6:1 for vocals, and adjusted the compressor threshold to produce sufficient gain levelling, but without causing audible side effects. The default attack and release settings were left as they were. Adjusting the amount of compression was quite difficult, as the VF16's compressors have no gain reduction read-out, so everything had to be done by ear. Also compression is only available on four of the VF16's channels, and can't be used at the same time as EQ on those channels.
For reverb, Murat wanted a fairly natural, warm sound so we ended up using the Old Plate preset and then adjusted the decay time to 1.8s and dialled in 57ms of pre-delay. Reverb was added mainly to the vocals, but was also used with the DI'd acoustic guitar to compensate for the lack of an acoustic environment. Enough was also added to the drum to give it more of a sense of space, something that close miking tends to lose, but we tried to avoid adding so much that it would make the mix seem obviously processed.
Murat was happy with setting the mix balance, so we didn't get as far as doing a complete mix. Indeed, being tourist season, it was impossible to get everyone together at the same time to make a complete recording, so we had to test the various setups in stages. Nevertheless, we were all happy that a viable recording method had been established and, despite the limited amount of equipment, the recorded results were well up to standard, with a surprisingly low level of spill.
To finish off, I made a few suggestions to Murat for his future work. Firstly, as the studio had no monitor speakers (and the apartment had no hi-fi system with aux inputs or any pretensions to accuracy), I recommended that headphone mixes be done using as little EQ as possible, especially at the bass end where you don't really know what is happening. As usual, I suggested that the mixes be checked on as many playback systems as possible, including car tape and CD players, but, because of the difficulty of evaluating low-frequency balance with headphones, I recommended that any material intended for commercial release be mastered at a studio with appropriate full-range monitoring.
As regards possible future purchases, I think a headphone distribution amp should be an essential acquisition, along with several sets of enclosed phones, for monitoring while recording. Some form of accurate nearfield monitoring system must also be a priority, but after that I suggest a voice channel with a built-in compressor for recording the main vocal parts — this would enable all three capacitor mics to be used at the same time, something not possible at the moment unless the club's Behringer mixer is brought in, and it would also provide a more controllable degree of compression, together with a visual read-out of gain reduction.
Because Murat's music relies heavily on voice and acoustic instruments, it would be nice to have more capacitor mics eventually, but at the moment the existing ones can be used for any critical parts when overdubbing, or for the main parts when playing live. For example, if the electric guitar and drum parts were being overdubbed, they'd probably sound best recorded using one of the capacitor mics, but where multiple players are recording together as a live performance, the lead vocal and acoustic guitars would benefit the most from the capacitor mics, with the dynamic mics being used for the remaining sources. Of the available mics, I'd use the SM57 for bass guitar, leaving the SM58s for miking the backing vocals and electric guitar.