Sound engineer with Sky chats about his work and gives advice about recording.
Andrew Jones — Sound Engineer with Sky
Andrew Jones is a prime example of the dedicated professional who goes quietly about his business and whose contributions are rarely acknowledged by the public. Having worked behind the scenes for the likes of David Essex, Sailor and Osibisa in the past, Andrew currently occupies the distinguished position of 'live' sound engineer with the rock group Sky. Here he chats about his own approach to sound mixing for Sky and imparts some helpful tips on such things as microphone placement, monitoring and mixing desks.
"I started off basically around 1968 working as a live musician, in fact, as a bass player with a small rock band in Birmingham. I wasn't very good so I looked towards the 'roadying' side of the business first of all. At that time PAs basically comprised of a mixer/amplifier and two 4 x 12" speaker columns. Then the band I was working for got signed up to a London agent. So I came down to London with them and I started mixing around with professional roadies, using larger equipment such as a WEM system and two Audiomasters.
I started getting more and more involved in the sound side of things and through this I started to meet people from small studios in London — demo studios, rehearsal studios — all with their own little mixing system. Because I didn't feel like carrying boxes around, week in, week out, I decided to get into the studio. I did this through a friend of mine, Alistair Crawford. He owned a studio called Hendene, which was basically a studio set up in his front room. We used to work with people like Colin Blunstone and Stomu Yamashta amongst others.
I never had a 'big break' in my career, it has been a gradual process of working my way up over 15 years or so. I worked with those people in the studio whilst doing live work at the same time.
Around this period (1973) I became very involved with a band called Sailor who I helped put on the road and whose sound equipment I helped build. We recorded their first album at Hendene studios on a couple of Revox A77 machines and a little Accuset four channel passive mixer.
The control room at Hendene was the hall of the house which was about 10 feet by 8 feet. We then moved on very quickly to a Soundcraft desk, as we had a couple of friends there when Soundcraft were first getting together. That was a 12 into 4 mixer which was a great improvement over the passive unit, as you could imagine.
That Sailor album was done very systematically by recording onto one Revox and bouncing down onto the second Revox whilst adding a new instrument live. So we had to be very careful with the mixing process as you only had one chance to play the part right. If you got it wrong you had to start all over again which led to a lot of hair pulling! It was a very successful album though.
Do you think that in those days, before multitrack techniques became widespread, that musicians generally were more competent players?
They had to be, I think. Competent and patient as well. You literally had to do a complete take; you couldn't stop and drop-in like you can now, if you make a mistake. Multitrack helps the musician a lot. You can play bar by bar if you want to. In those days the actual recording didn't seem to be as important as it is now. The quality was good enough, the ideas very simple and the overdubs were minimal. The songs on the first Sailor album were very simple; the complex arrangements came after that success when we found we could do a lot more with the studio than we were attempting to do before.
I did one tour with Sailor, working with Alistair as their technician on the sound side. Alistair and myself used to interchange our roles between sound mixing and being the general 'dog's body'.
I later went back onto the road for three years with Osibisa. Then I went out and actually used the money to buy my own sound system and formed a company, leaving the live recording thing to one side for quite a long time.
I then did some work with Wilko Johnson (from Dr. Feelgood) who wanted to record his album as it used to be done in the old days, straight onto tape with two microphones. This was in 1978. We did one tour with him then went down to a little farmhouse in Sussex and, using our live desk, turned it into a studio. We used a Teac A3340 four track, a couple of Roland Space Echo units for effects and monitored over Tannoy hi-fi speakers. For recording we used two AKG 451 condenser microphones.
So even though you were using down-market equipment, you still felt the need to use good quality microphones?
Yes. Even in the Hendene days we used to use Calrec — expensive, high quality mics. I've always worked on the basis that the source of the sound, the initial sound, has got to be of the highest quality, otherwise you're killing yourself before you've started. I approach everything that way, even on the road. You can't make a good sound out of a bad microphone.
Placement of microphones during that Wilko Johnson session was really trial and error. The room was basically a living room with big glass doors on one side, a fireplace, and we had a few mattresses to put around the drum kit. So we'd do a take, listen to it and reposition the microphones, if necessary, and do another take.
We tried to use a lot of distance between the pair of microphones, especially for the drums which we spread over two tracks, to get a wide feel. The positioning was quite critical and we ended up, in fact, facing the mics away from the instruments to pick up the room ambience. We actually got a better sound from the reflected rather than the direct sound. I felt we recreated that early rhythm and blues feel quite successfully.
We started off doing song demos to get a recording deal but ended up using three or four songs from that session on the final record. The record company didn't feel the quality was good enough on the rest of the tracks, and they wanted it to be re-recorded in a professional 24-track studio, but Wilko wouldn't have it.
During my travels I've come across many supposedly professional 8-track studios that are less well equipped than some 'home' studios I know of. In the middle seventies, 8-track studios seemed to spring up all over the place, but nobody seemed to put their heart into them. If you have your own home studio, you take personal pride in it. Those 8-tracks were only people putting in a bit of money and hiring out the facilities to bands who didn't know any better. They were being charged a fair amount of money for a pretty rough job generally.
After Wilko, I began to work with Sky on their first tour. This was during late 1978, early 1979. I got involved with them because I had also been working with David Essex on the road and Herbie Flowers was playing bass for David. Sky was really Herbie's idea since he'd worked with the founder members on studio sessions before.
Herbie invited me along to listen to the music they were putting down at Abbey Road Studio 3. So I went down there and experienced the band being born in the studio. There was this tremendous feeling of excitement amongst the band and the engineers. They came together with bits of paper in their hand with dots all over it, and suddenly they were producing this amazing music. At that point I just thought to myself that I had to work with this band. It was an engineer's dream — complex sound with varied instrumentation and fresh ideas about how to put the music across to the audience.
I didn't get involved with the recording of the first Sky album at all. My job was to put the live sound system together for the first tour. I worked very closely with the studio engineer on that tour and he came along for a few days to give me a helping hand. The tour initially was for 10 days but because it was so successful it was extended.
We'd had a few problems on the first few days with the monitoring. We wanted to take a studio approach and transfer that to a road situation. But when they came to do the second album the band wanted to take a live approach and put it into the studio.
The problem was that half the band wanted to be very quiet, the others very loud. Francis Monkman (keyboards) liked to have his monitor sound loud whilst Tristran Fry (drums) needed to be loud in order to hear the rest of the band above the sound of the drums.
How did you overcome the monitor problem then?
Well, we worked very closely with the sound company, Malcolm Hill, developing special monitors; basically Tannoy speakers in a wedge-shaped cabinet. We also screened off the guitar amps in a baffled box, so that Kevin just heard himself over the monitors, not from his amp. We were miking up the amp and just feeding him a low level signal. We contained the drum sound by placing perspex screens around the kit. From my point of view it allowed me the control of a studio mix, and I was able to take the drums out of the mix almost completely.
The feeling within the band was that they wanted to play to a family audience, and if they had young kids in the front rows they didn't want 6000 watts of sound blasting into their ears. It was a complete rethink of the normal stage situation.
The band basically rehearse in John Williams' front room, just rehearsing the musical parts, but we never do a live sound or lighting rehearsal until we get to the first show on the tour. Then on the day, we get to the venue very early and do half a day's rehearsal and a soundcheck.
How do you know where to set faders and EQ controls during the soundcheck when the hall is empty, so that when the hall is full the sound doesn't alter too dramatically?
It's something you pick up over the years anyway, but you can generally assume that the audience will mostly be wearing soft clothing which absorbs the sound and you tend to lose frequencies above 5kHz. You also tend to get a boost around 250Hz — I don't know why that is, it's probably due to skin reflection or something!
There's an awful lot of cues during a Sky concert, with different instruments coming in and out, so I tend to use the faders on the mixing desk as a gauge. I set all the faders at unity gain then bring up the master volume. Thus I tend to end up with the faders on the desk all in a straight line.
The keyboards have become more and more complex so I now spread those over 4 stereo sub-groups. Steve Gray, the keyboard player with Sky, is now using a Steinway Grand piano, William de Bley dual manual harpsichord, Yamaha GS 1 digital synth, an Oberheim OBXa, Roland Jupiter 8 and a little thing which we call the 'gob synth'; the Yamaha CS01 with breath controller.
On the piano we use two C-ducer contact strip microphones and an AKG 451 positioned over one of the bass sound-holes. This adds a bit of warmth to the sound as the C-ducer strips tend to be overly bright and require the bass to be rolled off quite a lot when used on the piano.
On the live stage the Steinway lid has to be closed to prevent pick up of ambient sound and feedback, and this makes the instrument sound 'squashed'. We tried experimenting with the positioning of the mics and especially the phasing between the three mics which is very critical. You find that you get an over-exaggerated peak in the midrange frequencies if everything is in phase. By simply switching one microphone out of phase with the rest, you can level that off quite easily without doing a lot of extra equalisation. The EQ on road desks tends to produce intermodulation between the frequencies and you get a ring modulation effect, especially if you're doing a lot of cutting or boosting. I try to work within the 9 o'clock and 5 o'clock parameters on any equaliser control. I feel that if I need to go further than that then there's something wrong with the source of sound or the microphone set up I am using.
All of the microphones go to Brooke Siren active stage boxes on stage, so that there is the smallest run of microphone cable before the signals get boosted up to line level and fed via multicores to the input channels of the mixing desk in the auditorium.
I am currently using a Malcolm Hill J2 Series 40/8/2 desk complete with a new equalisation section that uses eight frequency bands rather than four switchables. They are virtually phase correct which gives them a very natural, uncoloured sound.
The microphone picking up the sound source has to be of the highest quality and the next thing the microphone 'sees' is the desk. That again is a critical element in any system. I know a bad workman always blames his tools, but as an engineer, you can be limited by a fairly bad desk.
The harpsichord is treated just like the piano, miked up by means of another C-ducer contact strip. The piano has a stereo spread and I use a White 27 band graphic equaliser on the bass end of the piano pickup. It's very difficult to see what's happening with the White unit as it doesn't actually give you a graphic readout as it's supposed to because it uses potentiometers rather than sliders. I'm only using it because it was supplied with the system — I actually prefer Klark Teknik graphics.
The piano gets a separate stereo spread because it does come in and out during numbers quite a lot. Since I don't want to keep pulling three faders up and down, I just route the mics to a single stereo sub-group and fade that up and down when required.
The harpsichord is also fed to a stereo sub-group. There are eight stereo subgroups on the Malcolm Hill desk I use. The rest of the keyboards are then spread over two sub-groups, as I like to do treatments on the groups rather than the individual instruments.
Where do you position the instruments in the stereo image?
It really depends on the hall I'm working in. It's very unfortunate for people on the left or right not to hear, say, one set of instruments I might have had panned hard left. The only thing I do pan very wide is the drum kit but I like to give a little space to the keyboards as well.
John Williams only uses the acoustic guitar at the moment. He's bought a new Takamine acoustic which is about three hundred times better than the Ovation he used to use. That was a nylon strung instrument which gave us problems with tonal quality, especially with the bass end being restricted. So we used a little 10 band graphic EQ between John's guitar and the stage box to cut out the 250 Hz frequency which tended to be overemphasised on the A string. The feedback frequency was always around 240 Hz also. With the Takamine we don't get any trouble at all.
John monitors the sound on a single 10 inch Tannoy speaker at a very low level. For Kevin Peek we do the same for his classical guitar, although he still uses the Ovation and a 10 band graphic, simply because he only plays two numbers on it.
For Kevin's electric guitar, we place the amps in baffle boxes which we constructed ourselves and lined with four inches of rubber. He used to use two Roland Jazz Chorus amplifiers but he's now switched to RSD. It's a copy of the Mesa Boogie amp having a valve preamp section.
Do you mike that up or direct inject?
We mike it up, yes. We've tried several microphones over a long period starting off with Beyer 202s. We went over to 201s but found it still lacked in cut. It was a very smooth response but the cut had to be added at the desk EQ all the time, so we moved on to two Shure SM58 mics which give that presence peak that seems to work well on guitar. That was something we discovered from working in the studio, trying out various mics. We had the same problem with Neumanns as with the Beyer 202s — there was simply no cut there, so we ended up with SM58s.
Once again we used the microphone to obtain the sound source we wanted without having to do too much equalising, as equalisation tends to emphasise the system noise and hiss. Kevin's effects pedals are the worst offenders in terms of generating noise. At one stage, he used an AMS digital delay, but due to the finance situation reverted to his ordinary pedals. I use a Kepex noise gate on Kevin's channel to cut out background noise when he's not playing. Most of the outboard effects I use live are studio quality devices — they have to be really, just in terms of their quality, versatility and most important reliability.
Could you explain your approach to miking drums?
Well, we screen off the drums so that we can get a lot of separation between them and the other instruments. On all the tom-toms we use my AKG D190s which are very old microphones and one of my favourites. Once again this gives you the cut on the drums and the hard sound we like with Sky, without doing any EQ work. Basically all the drums are equalised flat which is most unusual in a concert situation.
For the bass drum I use a Sennheiser MD421, an MD421 above the snare, an AKG C451 underneath and another MD421 on the hi-hat. For the overhead microphones I've been using 441s because I find with Sky that the cymbal work is an important part of the music. Tristran doesn't play just crash and rides, he plays musical notes on his cymbals. The overheads add a lot to the overall drum sound so it's very important to use good quality microphones.
I normally use a stereo spread on the overheads, so you tend to get a tight tomtom sound with the width added by the overheads.
I put a noise gate, which is a small compressor as well, on the toms, so that they're not picking up anything too extreme. The noise gates operate on the two sub-groups, not on individual microphones. I also compress the bass drum a little bit and add a little expansion as well. Tristan is quite dynamic on the drums, it's not always the same beat. Some of it is at a low level which tends to disappear, so I use a compressor to even out the drum level slightly. It's an Audio Design Compex-Limiter. That's really the only effects device I use on the drum kit.
Snare, I try to keep as wide as I can and give it a lot of space by miking up quite a long way away. The top mic is usually five to six inches away from the snare head — the kit is quite wide so I get enough room to do that — and the bottom mic is maybe twelve inches away, directly below the head and positioned slightly forward. That's really only because the snare is too low on its stand to place the microphone far enough away.
Moving tom-tom mics an inch closer can give quite a significant modification to the sound. That's something I think a lot of engineers don't seem to notice. They tend to put a mic on a drum and either get the drummer to tune it to the way they want it to sound or try and EQ it. But mic positioning is very critical for what I want and I feel people don't pay enough attention to that aspect of recording.
I use four sub-groups on keyboards — all stereo and one on guitars. John Williams goes direct to the master faders, I don't put him through a sub-group at all.
That's basically to keep the noise down as the less electronics there are between his guitar and the PA speakers, the better the sound. Drums go across two groups and then one group for effects.
Herbie Flowers' bass guitar is direct injected using an active DI box. I always use active boxes simply to overcome the loading problems from output transformers. Herbie uses an amp on stage just so that he can 'feel' his sound behind him, but we also feed his part back through the monitors so that he can hear himself more easily. The amps are Roland Bass Cubes. Herbie is actually very unconcerned about his stage sound from the amplifier and as long as he can hear himself, he doesn't give us any trouble.
His real stickler though is double bass, which he plays quite a lot and that is what he really gets uptight about. He calls it his 'real' bass. It's a very old instrument so we have to be very careful with the varnish when sticking on contact mics, for example. Once again we're using C-ducer strips because they give an even response when placed above the soundpost, just in front of the bridge. We use the three-inch long strip for the bass rather than the eight-inch version which is used on the piano, simply because the double bass body is so vibrant that it tends to tear the strip off if used over a broad area.
We compress the electric bass slightly, but the dynamics within Sky's music tend to limit heavy use of compressors. From the quietest to the loudest parts of the music, there is about an 80 dB difference in signal level, which is a vast amount for any rock group. I tend to use the compressor as a limiter really, set on a 3:1 compression ratio with a very low threshold just to keep the loudest peaks in the sound under control. I'm not really happy in limiting anything but it's sometimes necessary to prevent damage to your PA system.
Apart from compressors, what other effects do you use to treat the live sound?
I use a Lexicon 224 digital reverb all the time and on just about every instrument. It has preset reverb 'programs' which you can modify. I have my own program variations but I choose a delay time that will work initially on drums. When you use reverb on a live concert mix you can't really hear it if the effect is subtle, so you are forced to go 'over the top' with the depth of effect. That's why I use the drums to set the reverb time, they are my guide. If the drums sound alright on a particular setting then everything else generally does too, I've found.
I use reverb more than I should but it does make the overall feel of the band that much better.
I set every instrument within a very narrow 'picture', in terms of the tonal quality of sound, to keep everything separate. But if I was to do a concert like that it wouldn't sound 'live', so I then try and blend the instruments together using the reverberation.
On the Lexicon I use a 60 milliseconds Pre-Delay setting and up to 3.5 seconds of actual reverb time, so that the sound lingers on for quite a while. The Lexicon is fed via an auxiliary send/return to the input channels. I'd really prefer to have it purely on the sub-group feed, leaving the input channels 'dry', as that way you don't still have the input instruments adding noise to the reverb when the sub-group fader is down. Unfortunately, on the current desk I'm using you can't do that, so I end up with a 'mushy' reverb sound.
I also use a Lexicon Prime Time digital delay on Kevin's guitar to give a 'double tracking' effect. There's also a few numbers where a repeat echo is required so I use a Roland Space Echo.
When you are preparing for a concert do you listen to the albums to find out what effect is required on each particular instrument, so that you can then duplicate it live?
What I tend to do is to go to the studio when the band are recording the album. I sit in on the session and advise the band on what effects and arrangements we can get away with live.
Working with a band that's already recorded the album limits you to a certain degree so I add little effects live that aren't on the album purely for my own creativity, such as panning. I quite often pan a synthesiser from left to right speakers as it fades away for example.
How much work is involved on the mix. Are you continually moving the faders or do they remain fairly static?
There's no way I could go through a set and leave the faders in fixed positions — purely from the point of view of noise. I'm constantly adjusting the levels of the instruments in the mix. The mics on the grand piano pick up so much ambient sound that they just create a cluttered sound mix. So I fade the piano down completely when it's not in use. Even if it's not being played for three bars, I'll still generally fade it out then back in — so I'm a busy lad when it's all happening.
You say you fade the piano down when it's not in use, is that done with the input fader or the sub-group fader?
It depends on how many microphones are being used for a particular instrument. I use 15 or so mics on the drums which would be impossible to fade down in one go so that's why I send them all to two subgroup channels, as I then only need to move their two faders to change the level of the complete drum kit.
A 40 channel desk is quite large so I lay out the instrument input channels in an order where instruments that need constant level adjustments are close at hand. The grand piano, for example, hardly ever needs adjusting so I control that not from the input faders but from the sub-group.
Do you introduce dynamics into the mix yourself, by increasing the levels, or is that left to the band?
I do create dynamics within the music by varying the fader level controls. I never touch the master left and right output faders except to build up the overall sound level gradually during the length of the concert. Sky normally work at a sound level of about 100dB on the loudest passages which is loud for such a band. On the quieter numbers I drop the master levels simply to reduce background noise and then increase the levels on the individual input faders.
Do you work from cue sheets during a concert?
I write out my own cue sheets for the beginning of each new tour. They contain detailed information on such things as instrument levels, timings, effects changes, settings, etc. Any new numbers I'll have a cue sheet for, but generally there's never time to look at them. After two days if I still have to look at a cue sheet then there's something wrong!
Sky recorded a live album in Australia last year and I was really responsible for the live sound, but also the miking situation. We generally used one set of mics and split the signal. One went to the mobile studio desk, the other to the concert hall mixing desk. The only mics we doubled up on were for the piano where we used a couple of Neumann's as well as the usual AKG and C-ducer contact mics. I couldn't take a feed from the Neumann's, basically because they were too sensitive. In fact the studio recording ended up using the piano sound picked up by the C-ducers. It was a very successful album in terms of sound quality.
I've found that the home recording situation has greatly helped bands on the road, to understand just what can and can't be achieved on a mixing desk. Working as engineers on their own recordings helps give them an awareness of their own sound and the limitations of the live set-up.
The most important factor to pay attention to is the sound at source. If you're not careful you end up with a situation where your initial sound isn't what you actually wanted — if its quality is lacking then you're lost basically. You've got to listen carefully to what you're playing and pick it up with the best quality mics. If you take enough time then you can record your sound straight down on to just two tracks.
With the complex music of today and the quality of sound required by record companies it's not always easy. That's the beauty of multitrack of course, since having the multitrack tape is like having a complete band on call all the time, and so you've got the versatility to mix and mix a track until it is absolutely perfectly balanced.
Working at Abbey Road Studios with Sky was quite a strange situation, as you've got Studio 2 downstairs which was where the Beatles did all of their recording, and that's like a shrine to the four-track machine. If the music is well rehearsed and simple you don't need any more than four tracks. The beauty of multitrack, though, is that it allows the individual a greater sense of freedom during the recording process, since he's able to play his music piece by piece on his own, which was never possible in the past.
However, I just feel that with multitrack recording you've got to learn to walk before you can run. If you can't use a two-track tape recorder and do it reasonably well, then you'll never do it right with a multitrack. You'll become completely overawed by the machinery as opposed to letting your creative talents flow!