A Hall Choir, a Reverberant Nave and One Afternoon of Microphone Placement

In mid autumn Marcus rang us from a regional arts centre. He produces a short televised concert series, and the next episode featured a community choir in a converted stone chapel. His rough recordings were thick and distant: every consonant blurred, and the quieter verses vanished under the room.

He had already tried more microphones, assuming that more channels would mean more clarity. What he had instead was a tangle of overlapping pickups and a mixing desk with every fader nudged up a little. He asked whether we could look at the set-up before the real recording day, and I drove over the following week with a small case of my own microphones and a pair of closed headphones.

I should say that Marcus is not a sound engineer by training. He is a producer who had inherited a technical problem, and he asked good questions. That made the visit easier than many.

Before touching a microphone I sat in the nave for ten minutes while the choir warmed up. Stone walls, a high timber roof and a bare floor gave a long decay, noticeable even on a clap. The reverberation was not a fault to be removed; it was the character of the place, and the job was to decide how much of it the viewer should hear.

I walked the room clapping softly, noting where the sound smeared and where it stayed tight. Near the side walls there was a harsh flutter between parallel surfaces. Close to the chancel steps the sound was clearer. That told me where the choir should stand and where the microphones should not.

Marcus had assumed the room was simply too live to use. By the end of the first hour he agreed it was usable, provided we chose positions deliberately instead of covering everything.

We removed six of his eight microphones. A coincident pair on a stand above and slightly behind the conductor's position captured the balance of the choir as a listener would hear it. Two spaced microphones further back picked up the room, kept low in the mix to add depth without washing out the words.

Distance mattered more than model. Moving the main pair half a metre closer made the diction clearer, and raising it by roughly the same amount reduced the weight of the floor reflections. We changed one thing at a time and listened to each change on headphones, then in the chapel's quiet side room where the sound was more neutral.

Marcus noticed something I often see: with fewer channels there was less to manage, and the choir itself began to sound like a choir instead of a collection of voices.

• One main pair for balance, placed above head height
• Two room microphones kept low in the mix
• No close microphones on individual singers
• Cables run along walls, away from the singers' feet

The second problem was how Marcus had been judging the sound. He was listening on the small speakers of the portable recorder and on consumer earbuds, both of which flatter muddy recordings and hide low-frequency build-up. Decisions made that way tend to fail the moment the programme reaches a proper television chain.

We switched to closed-back studio headphones and set a fixed listening level, so that his ears were not constantly recalibrating. I asked him to check the same passage at the start and end of each session, which sounds trivial but catches drift in fatigue as much as in equipment.

Once he heard the low-mid congestion on the headphones, he could name it himself. That is the real value of good monitoring: it lets a producer say what is wrong in words an engineer can act on.

On the recording day the choir arrived with a small audience in the pews, which changed the room again. Bodies absorb sound, and the chapel became noticeably drier. We nudged the room microphones up slightly to compensate and kept the main pair where it had been during rehearsal.

Marcus ran the desk himself with a simple plan written on one sheet: levels, the order of pieces and a note of where to stop and reset. Between takes we played back short sections and listened for the same few things: clarity of consonants, the tail of each ending and any rustle from the audience.

We recorded the programme in three sessions with short breaks, because the singers tired and so did our ears. The final takes were not perfect performances, but they were clean, balanced and consistent, which is what a broadcast chain needs to work with.

• Reset levels after the audience arrived
• Played back each take on headphones before moving on
• Kept a written running order at the desk

The job was small, but it showed again how often sound problems begin with listening conditions rather than equipment. Marcus did not need a larger kit; he needed a quieter way of deciding what to capture and an honest way of hearing it.

A few weeks later he sent a short note saying the finished episode had gone out and that the choir sounded like themselves. He has since kept the same two-plus-pair layout for other venues, adjusting only the distances.

For anyone about to record in a reverberant space during these colder months, my advice is to visit the room early, sit through a rehearsal and resist the urge to add channels until the basic pair is working.