The SOS team visit a London studio in a building directly under a busy road — and with another studio in the room next door!
As well as being a very talented singer-songwriter, Sunita Sagger plays keyboards, guitar, bass and drums. She has a small studio located inside a warren of other creative facilities, in an industrial building owned by a trust, not far from London's famous Portobello market.
So far so good, but the roof of the building is actually the A40 Westway flyover, and even though Sunita's studio is on the lower of the building's two levels, some road noise still manages to find its way in. Sunita called us in for advice on improving the sound isolation, as there was some bass creeping in from another studio next door (something she is hoping the landlord will see fit to attend to), in addition to the traffic noise. But mainly we came to see if we could improve the monitoring acoustics within the room itself, and Sunita also had a few small technical issues that she needed advice on.
When Hugh and I arrived, we were shown into the studio and offered tea and Hob Nobs, which has become something of a traditional greeting over the years. That gave us time to look around and evaluate the problems. The room itself was only a little over two metres wide and short of four metres long. There was a suspended ceiling, above which was an uninsulated void perhaps half a metre deep. The front wall was concrete, but the other three walls were all stud partitions with plasterboard, and the room had no acoustic treatment of any kind.
Rather than trying to do everything 'in the box', Sunita had just finished wiring in a second-hand Soundtracs 16:8:2 analogue console, which was connected to the rest of her studio via a patchbay. This seemed to work OK, apart from a couple of crackly pots, although its unbalanced line inputs made the rig susceptible to ground-loop hum. Hum and noise was very audible with all the faders and monitor levels fully turned up, but it was negligible at normal levels.
It's possible that some improvement in the noise levels could be achieved by making up a set of balanced-to-unbalanced cables. These are wired conventionally at the balanced end connecting to the patchbay and audio interface, but at the mixer end, a mono jack would be used, with the 'hot' wire connected as normal, the 'cold' soldered to where the screen would normally connect, and the cable screen itself cut off and left unconnected. This arrangement avoids making direct connections between the grounds of the interface and console, and can be surprisingly effective at removing annoying ground-loop buzzes. You can buy cables in this configuration from /shop.
We cleaned the input jacks on the Soundtracs mixer using Caig's Deoxit, a widely available contact cleaner. We then also tried it on the crackly gain pot. You can sometimes fix an upward-facing pot by removing the knob and then spraying Deoxit onto the shaft. Capillary action causes it to creep down the gap between the shaft and the housing, after which it spreads over the wiper assembly. We tried this on a particularly troublesome trim pot and it actually worked!
Sunita does her actual recording on an Intel iMac running both Logic Pro and Presonus' Studio One. She was using two audio interfaces (an M-Audio Profire 2626 and a MOTU 828), linked via Firewire and communicating via an 'aggregate' driver. Sunita seemed happy that this was working OK, but in my experience, aggregate drivers can cause an increase in latency, so I felt that a better solution might be for her to get rid of one of the two interfaces, and then add an eight-channel expander that could be connected to the remaining interface via its ADAT optical ports. As Sunita wanted the maximum number of analogue outputs possible, to allow mixing on the console, this would mean choosing an expander with both mic inputs and line outputs, rather than one offering only inputs. Suitable models are made by Focusrite, Presonus and Behringer, and these usually cost less than a second interface.
Next, we turned our attention to the monitor system. The easy part of the job would be to treat the reflective wall surfaces to improve Sunita's monitoring environment. She was using a pair of small ESI Near05 Classic speakers, which were sensibly chosen for the room size, in that they didn't generate more low end than was appropriate for the small space. These were initially perched on some scraps of foam, which we replaced with a set of Universal Acoustics Vibro-Pads (high-density foam speaker platforms). Hugh then adjusted the speakers to the correct angle using the included foam wedges.
While repositioning the speakers, Hugh noticed that their input controls were turned fully up, yet the monitor level control on the desk was barely off the backstop. Stereo level controls (such as the one in the Soundtracs desk) often have very poor level matching between the left and right channels in the lower part of their control range, so Hugh reset the monitoring chain's gain structure to allow normal listening levels to be achieved with the console's monitor control set to seven (about two o'clock). He first turned the speaker input controls right down, and then turned the monitor level up to seven. With a CD playing through two channels on the desk, and all relevant input and output faders set to their unity positions, Hugh adjusted the channel input gains to ensure that the CD track was peaking at a sensible level. After panning the two CD input channels to the centre (to create a dual-mono output), it was then a case of adjusting the input level on one speaker to achieve slightly less than the intended 'reference' listening level, followed by turning up the second speaker's input level control until the mono image moved to the centre line between the two speakers — thus ensuring that the two speakers were producing perfectly matched output levels. The CD-channel pan pots were then reset hard left and right, to restore normal stereo operation.
It's important to set up a known reference listening level that can be used for all mixing, to ensure consistent results, and seven on the dial is as good a setting as any, since it provides scope for cranking up the level to check for quiet anomalies when necessary.
Three of the studio walls were made of plasterboard (drywall) fixed to a studding frame, presumably with some sort of insulation in the middle, and the fourth wall behind the monitors was made of solid concrete, resulting in a lot of strong early reflections in the room. To address these, we used four Auralex four-inch Studiofoam Metro 24 panels in a fetching shade of purple, placing them in a pretty formulaic way to take care of the main early reflection points, sometimes known as mirror points. This required one foam panel on either side of the monitoring position at around head height, extending forward towards the corners of the room, and another piece on the front wall between the monitors. A fourth piece was deployed on the upper section of the rear wall. This would normally also have been at head height, but Sunita wanted to fit shelves to that wall to store a selection of musical toys that she uses in some of her recordings.
As I had my bag of tools with me, I fitted the two small shelves, which Sunita had already purchased, just below the foam. I also put up four guitar hangers that Sunita had bought to keep her Fender Strat and short-scale bass safe when not in use. Using threaded metal plasterboard fixings wherever there wasn't a wooden batten to screw into worked well for this. Once filled, the shelves helped break up reflections from the exposed part of the rear wall, while lifting the guitars off the floor gave Sunita a little more usable space.
Fixing the foam to the plasterboard walls was easy — we glued a couple of old CDs to the top edges on the back of the foam, then hung the holes in the CDs over a couple of screws driven into the plasterboard, making use of the spirit level on my iPhone to make sure they were level. We couldn't do the same with the concrete wall behind the speakers, however, and we didn't want to glue anything directly to the wall, because of the landlord's restrictions, so we simply propped the foam panel in place, then secured the top edge to the wall with some improvised double-sided adhesive made from a couple of loops of gaffer tape. As expected, the improvement in clarity and stereo imaging was immediately obvious, and while simple acoustic foam does virtually nothing for bass frequencies, the plasterboard walls contribute in that respect by either allowing the bass to pass through, or by absorbing bass energy as the board is forced into vibration.
Unfortunately, the 'advantage' of the plasterboard allowing low frequencies to pass through it was also a disadvantage, because the studio next door, which had much larger monitors, was sending some of its bass energy through the wall and into Sunita's studio. This could be heard and also felt as physical vibrations in the partition wall separating the studios. This kind of problem can only be addressed by adding a lot more mass and a layer of damping to the wall, which would require an empty studio, specialist tools, plenty of time, and a sensible budget — so it wasn't something we could undertake in an afternoon, but it is something the landlord could arrange. There are various acoustic products around that can either be applied as separate layers (such as alternating plasterboard and barrier mat) or that come as composite sheets where the damping layer is bonded to the plasterboard. While such an approach is unlikely to offer a complete solution, it should make a significant improvement — albeit at the penalty of making the small room even smaller!
The other weak area, as far as sound leakage was concerned, was the door, as it had no lower sill, so there was a gap under the door of around 10mm. Any air gap, however small, allows sound to pour through, so the only solution here would be to install a lower threshold and then fit a neoprene sealing gasket to all four sides of the frame. A compression-type door latch that pulls the door tight against the neoprene gasket is the best solution, as conventional latches usually have too much play to keep the door tightly closed. Again, this went on the list of landlord requests. In the meantime, however, we suggested wedging something (such as some scrap foam) in the gap beneath the door while recording, to help reduce speech noise from the communal area outside the studio.
That left the traffic noise from the busy flyover above. Given its proximity, the ingress of noise was surprisingly low, but tyre noise caused some wind-like sounds to emanate from above the suspended ceiling. These were high enough in frequency content that we suspected they could only be getting in via holes or ducts, not via structural conduction. If that was the case, these holes could be plugged using expanding foam filler, or even stuffed with other material, to improve the situation. This would entail bringing in a step ladder, removing some of the ceiling panels to make an inspection, and then filling the offending holes once located. Sunita said she'd already taken a look up there and had identified some holes, so going up again to fix them wouldn't be a problem. Expanding polyurethane foam in a can is the easiest way to seal small leaks, and a single can would be more than enough in this case. Filling the void above the suspended ceiling with loose rockwool or glass-fibre insulation would also help.
There was also a small air duct feeding the room, and it is possible that some noise was being transmitted along or within the air ducting. Ideally, the ducting should have proper noise traps but this didn't appear to be the case, and it isn't really something that Sunita could take on herself.
Our final task was to find the best way to record vocals in the small studio. Sunita already had a Vicoustic Flexiscreen Lite foam screen to place behind her microphone, but we explained that eliminating reflections from the wall behind the singer was also very important. The entrance door to the studio was set at an angle across one of the rear corners, so we thought the simplest solution would be to fix two self-adhesive hooks to the back of the door, onto which Sunita could hang a duvet or sleeping bag to act as an absorber. She just happened to have a couple of sleeping bags in the studio, so we did exactly that, which gave her a working vocal space that was reasonably dry from an acoustic perspective, and that was also as far from the equipment (and hence fan noise) as possible. That was about as much as we could do in one day, and Sunita now had a list of jobs for herself and the landlord to attend to, so we packed up our gear and headed for home.
Sunita: "Sitting in the mix position, I can really hear the difference — there is more definition and the sound is not bouncing around, so it's a lot clearer. I've taken on board the advice from Paul and will look into getting an ADAT expander so we have less latency, rather than using the aggregated device setup.
"The studio is a small room but with the guitars hanging up and the shelving on the back wall it feels a lot more spacious. I definitely feel this is a much more creative and inspiring space to work in. I can't wait to get stuck in and see what I come up with next!”
You can hear some of Sunita's music at www.phuzzbox.bandcamp.com.