I have always been fascinated by the sound of church bells. In many countries around the world, bells are part of the soundscape. In France, bells give rhythm to the lives of the inhabitants. They ring out the hours of the day and are also part of religious festivals, weddings, masses, etc. It is an important sound heritage. Bell are always made of bronze, an alloy of copper and tin.
During the 15 years that I have been working in the sound industry, I have always taken time to record the bells I encountered while traveling, and I noticed that there is a big diversity of pitches and timbres. I also noticed that it is often difficult to have a truly clean recording because of noise pollution (traffic, humans, birds).
The Project:
A year ago, an idea emerged: to build a sound library with bell sounds that would meet 3 important criteria:
– Quality: The sounds must be free of noise (traffic, birds, voices), have a very good signal/noise ratio, and be precise. Indeed, it is easy to move away or degrade a sound in post if necessary, but the opposite is impossible.
– Diversity: It should contain enough different sounds to represent the variety of sounds found in France and Europe and cover almost any situation, from the small chapel bell to the giant bourdons of the largest cathedrals.
– Accessibility: The sounds must be easy to find, and well sorted, with the note of the bells and complete metadata like year of fabrication, weight, etc.
How to meet the quality criterion?
Over the years, I have perfected my sound recording skills and acquired some very good equipment, but that is not enough. Bell towers are often placed in the center of villages or towns and near roads. In the countryside, they are often inhabited by birds. There is always noise pollution.
To ensure a good signal/noise ratio, I began by recording the bell towers at quiet times: winter and night. I also chose quiet places. I quickly realised that this was too restrictive. In fact, many bells do not ring at night and many bell towers are located in noisy places – in cities or near roads. So I tried to reduce the sound recording distance. It is well known that when we get closer to a sound source, we increase the signal and therefore improve the signal-to-noise ratio.
During my first attempts, I was in a quiet village with beautiful ringing bells, but a flock of starlings completely ruined my take. The church priest suggested that I record the bells from inside the bell tower. I accepted with joy. However, I had a preconception: I feared that the sound would be too loud, distorted, and not natural compared to the sound we are used to hearing. I was also worried that we would hear the bell mechanism too much.
This sound recording removed all my fears. The Sennheiser MKH 8040 easily cope with the sound volume of nearby bells and even if, to the ear, the volume of the sound is absolutely unbearable without hearing protection (around 120 dB), the recorded result is excellent. It’s a bit like a drum kit – it’s harsh to listen to nearby but close-miking gives the best results. However, the minimum gain on my Sound Devices 633 was a little too high to handle the levels delivered by the MKH 8040 mics. I didn’t want to use pads so I switched to a Sound Devices MixPre 10 II with great results.
I thought it would be great to record all the sounds in bell towers… but it’s not that easy; accessing the bell towers is often tricky. Some bell towers are not passable, too dangerous, or the stairs or ladders are broken. Sometimes access is refused.
When I couldn’t access bell towers, I used the following techniques:
– Choosing the most silent places.
– Placing mics outside as close as possible to the bells with giant poles or mic stands.
– Finding the most isolated places from background noise (behind a wall; between two houses).
Recording several takes to be able to edit out car pass-bys or birds.
– Slightly cleaning the sounds in post (especially to remove birds).
Keeping only the best recordings, and trashing many others !
How to meet the diversity criterion?
The discipline that focuses on bell ringing is called “campanology.” There are two main methods for ringing a bell: hit it with a hammer or swing it.
Hitting it with a hammer emits a very bright, very stable sound, with a regular decay of each harmonic.
If you swing it, a clapper hits the inside of the bell with each swing. In this case, it makes a sound that modulates in frequency, thanks to the Doppler effect, but also in timbre, because the orientation of the bell causes the amplitude of the different harmonics to modulate. This results in a less stable sound yet one that’s softer and more lively.
Struck bells are used in carillons and clock towers because the number of strikes can be easily controlled. Swinging bells are used for religious ceremonies and festivals.
Since the ’30s, bell founders have had the tools to precisely tune their bells after casting. It means that modern bells tend to have a standardized sound. For the library, I recorded modern and older bells. With old bells, I’ve never found two bells with the same sound. The tuning, the formants, the balance… every aspect of the sound changes.
Before that experience, I thought all bells had more or less the same sound, and that it was just a matter of pitch. In fact, it’s not the case at all. The sound of a bell is very complex with a lot of harmonics. Each bell has its specific sound. I did an experiment of tuning different bells to the same fundamental note and none of them sound the same.
There are more than 80,000 bells in France. If I wanted to record them all, at a rate of 2 per day (which is completely unrealistic) I would have to work for more than 100 years!! At first, I set a goal of 50 different bells, but this was such an exciting quest that I travelled around 12,000 kilometers in 6 months to record more than 200 sounds in 63 locations. I managed to get 80% of the sounds recorded inside bell towers.
Realization:
I started by doing an inventory of the sounds I had recorded in the past. I noticed how difficult it is to get really clean sounds. Among the 40 bells that I already had in my personal sound library, only 3 or 4 met my quality criteria due to background noise.
I undertook a major documentation work around bells in France. Which bell towers are accessible? What notes do the bells make? Where are the biggest bells found?
The easiest bells to find are those with a pitch between G3 and C4. These are a bit like the “standard bell,” and weigh a few hundred kilograms. Higher-pitched bells are found in some small chapels or carillons and are sometimes still operated by hand.
Bells that emit notes lower than C3 are called “bourdons.” They weigh several tons and are found in cathedrals, basilicas, and some churches. Recording a bourdon is a unique experience.
It is the only musical instrument capable of emitting such a deep tonal sound with so much power. Your whole body vibrates as if you were next to a huge subwoofer. Each time I felt very strong emotions. The sounds of a bourdon can often be heard several kilometers away.
Recording bells requires a lot of work upfront. You have to find the bells that have interesting sounds, find the people responsible for them, make many phone calls, send emails, ask for permissions, make appointments, and travel. Fortunately, there are hundreds of videos filmed by bell lovers online. The sound is distorted most of the time, but that gives you an idea of the tones of the bells of the most famous buildings without having to travel.
Recording is not the hardest part. However, once you’re in the bell tower, you have to find the right spot to record.
It requires finding the right balance between a few parameters :
– Keeping the mics away from motors to reduce mechanical noises.
– Placing the mics at the right distance from the bell to keep a natural stereo image and avoid huge variations of dynamics, especially for bourdons that travel several meters when they swing.
– Recording above the bell lacks basses for acoustic reasons.
– Placing the microphone lower than the bell reduces background noise because walls act as sound barriers.
While most swinging bells are swung by an electric motor, there are still a few hand-operated bells. I had the opportunity to record with two different bell ringers. Hand-swung bells have a pace that is a bit less regular. There are more dynamics and no motor sound at all, which makes them a very interesting choice for sound recording.
I also had incredible opportunities to record two carillons.
Carillons are quite hard to play, even for a pianist, because you play with your fists and you need a lot of force. The carillons sounded so great that I decided to make virtual instruments out of them. As a composer, I have experience in sampling and programming virtual instruments and I’m not sure that there are carillon libraries available. The entire process – recording, editing, cleaning, and programming – took about one week, but I think it was really worth it !
The Closing Bell
It’s been a fantastic journey, I learned a lot of things about bells, and I met great people. It’s crazy to find musical instruments that have stayed outdoors for centuries and that still sound great.
I would like to warmly thank everyone who helped me bring this project to fruition: priests, sacristans, guides, and campanologists, in particular père Luc Caraguel, père Grégoire Corneloup, Philippe and Christian Wathelet.
A big thanks to Nicolas Titeux for giving us a behind-the-scenes look at his journey of recording the sounds of bells!
