The BPT Array

Reverberation radius by mic polar pattern: a shielded fig‑8 mic can be placed 2.4 times further from the source than an omni mic, and still pick up the same ratio of direct to ambient sound.

The creator of the Blumlein‑Pfanzagl‑Triple mic array describes its development, as well as its use at a recent Salzburg Festival event.

The Salzburg Festival of Classical Music is one of the biggest summer festivals in Europe, with more than 200 performances delivered over a period of roughly six weeks during July and August, to a total audience of around 250,000. It involves some unusual venues, one of which recently hosted a competition that required me to adopt a rather different approach to recording than engineers would typically choose to capture orchestral performances in a concert hall.

In this article I’ll describe the technique I used at this event, along with a little background on its development and why it worked so well here.

Objective Recordings In An Unusual Venue

The event in question was the Herbert von Karajan Young Conductor’s Award 2021. The use of spot mics was out of the question, since the aim wasn’t to deliver the most subjectively pleasing balance but, rather, an accurate rendition of the sound in the hall at the ears of the listener — and, more specifically, the ears of the competition’s ‘jury’, who were judging the performances conducted by these aspiring talents.

The particular acoustic situation at the Felsenreitschule venue complicated my decision‑making in two respects. First, having been carved out of the Mönchsberg rock, one of the two natural ‘hills’ which determine the characteristic look of downtown Salzburg, the stage area is surrounded by stone. While this makes it visually very appealing, it also means there’s quite a long natural reverberation — the last time I measured it suggests this is about 1.9 seconds at 1kHz, when unoccupied. Second, this venue has less in terms of reflective ceiling structures above the stage than is typical. In conventional venues these are very helpful in creating a balanced orchestral sound, because the reflections can compensate for or interact with the highly complex and frequency‑dependent radiation characteristics of acoustic instruments.

That second factor meant that to achieve a good balance of all the instrument groups within the orchestra, I would need to place the main microphone/array further back than usual — about 10 metres (11 yards) from the orchestra. And that posed a question: how could I achieve a healthy balance between the direct sound and the naturally long reverb when miking at such a distance? My answer was to use a unique mic, the Nevaton BPT, which has three capsules, stacked vertically, and each with switchable polar patterns. For this event, I ended up using all three in fig‑8 mode, effectively augmenting a Blumlein Pair with a fig‑8 centre mic, and I placed an acoustic shield behind it.

Nevaton BPT-1 three-capsule microphone.

This may seem rather a radical decision for an orchestral recording, but it was based on a setup that my team have developed over several years for use in opera and concert recordings at our three main halls. Before the Felsenreitschule event, we tested the technique for the archival documentary recording of opera dress‑rehearsals at the Grosse Festspielhaus, a venue with 2200 seats and a stage‑front width of around 20m. Here, we positioned the BPT in the floor section, in the eighth row back, roughly 6.5m (20 feet) behind the conductor, and at this ‘best seat in the house’ position, the balance between the singers on stage and the orchestra in the pit was just perfect (assuming the conductor was doing a good job!). So we already knew it could work well, and just needed to establish the best placement in this different venue.

Developing The Technique

It’s perhaps helpful if I first explain how and why we arrived at this technique, and the advantages it can offer. For several years, we’d been making archival recordings (which occasionally resulted in a CD release) by mixing spot mics from the orchestra pit with vocal spot mics splayed out along the rim of the stage.

This process required an additional sound engineer to do the balancing but not only had I begun to think that it must be possible to devise a more efficient approach — I wasn’t entirely content with the results we’d been getting. Usually, the sound was only half‑satisfying for the sensitive ear, primarily because the combination of close placement of the instrumental spot mics and more distant (many metres) placement of the vocal spot mics created a sort of ‘acoustic inversion’ of what listeners heard out in the hall, at floor level. The guest sitting there hears a strong ‘direct’ signal from the singers on stage, while the sound from the orchestra in the pit mainly arrives in the form of a first reflection off the stage ceiling, or at least ‘bending around obstacles’ in the case of low‑frequencies. The singers, therefore, sound ‘upfront’ and reasonably close, while the orchestra will sound somewhat ‘indirect’. This is generally desirable, since the singers should always have the ‘leading role’ in an opera — but it wasn’t the case for the sound coming from the desk!

Therefore, we began to explore the potential of other mic setups. Especially in live‑sound reinforcement but also recording, most engineers seem to limit themselves to the use of omni and unidirectional microphones — cardioids, subcardioids, supercardioids, hypercardioids and so forth — and I think it’s fair to say that figure‑8 microphones get less attention than they deserve, and often that’s because their rear‑facing lobe can pick up sound from unwanted sources. But fig‑8’s can offer some significant advantages too. For example, the classic Blumlein Pair of fig‑8 mics provides very spacious, natural‑sounding results, with a good perceived sense of front‑to‑back stage depth.

The Nevaton BPT mic augments a Blumlein pair with a switchable‑pattern centre capsule. (The second pattern, fig‑8, was used for this recording.)While the Blumlein pair showed some promise for our application, it wasn’t perfect. A notable characteristic is that the inter‑channel level differences increase at higher frequencies (since, in the real world, fig‑8 mics don’t achieve theoretical perfection; they become increasingly directional as you move higher up the frequency spectrum), and this tends to result in a ‘hollowness’ in the centre of the stereo image. In my experiments, which ultimately led to the development of the three‑capsule BPT design, I soon came to appreciate the addition of a central capsule to the L and R capsules of the traditional Blumlein Pair. The central mic provided the flexibility to emphasise the middle of the sound stage when that was needed — for example, in the case of a soloist performing with an orchestra — simply by riding the centre‑capsule’s fader.

Usually, I’ve found that the level of the centre capsule sounds best when set somewhere between 6 and 3 dB lower than the signals of the other two capsules; this tends to provide a nice‑sounding, stereophonic image. But if greater focus on the centre is required, the level can be raised above 0dB. There are no problems with sound‑coloration due to comb‑filtering effects and the like.

Young Conductor’s Award Recordings

With all of that in mind, let’s return to the initial challenge: recording the Young Conductor’s Award performances at Felsenreitschule. In this venue, we found that, for the desired balance, the BPT mic needed to be placed further back — about 10m (nearly 11 yards) behind the conductor. This put the BPT outside the critical distance or ‘reverberation radius’ of this hall (see the 'Reverberation Radius' box). Thus, the centre capsule would need to have a directional pattern: one usually wishes to pick up mainly ‘direct’ sound from an instrument, and then have the option to mix in the ‘spatial feeling’ of more distant ‘room mics’ to taste.

As the main graphic shows, the ‘effective reverberation radius’ for different directional polar patterns (cardioid to fig‑8) can be multiplied by a corresponding factor: thus, mics with different patterns can be placed at different distances from the sound source and still provide the same direct‑/diffuse‑sound ratio as an omnidirectional capsule would provide when positioned at the reverberation radius of the hall.

The technique not only sounds good — it makes good acoustic sense for documentary recording purposes.

In a concert hall like the Salzburg Festival, the reverb radius is about 5.5m for midrange frequencies. At the much greater distance of 10m, though, the string sections around the conductor will still be within the reverb‑radius of a fig‑8 capsule when an absorptive panel is placed behind it. While this is obviously not effective at all frequencies, it does shield off the noticeable diffuse sound components coming from the rear of the hall, and since these components account for 50 percent of diffuse sound on the capsule, the effective reverb radius doubles to 11m. So the technique not only sounds good — it makes good acoustic sense for documentary recording purposes to have a BPT microphone in this position, with the acoustic panel behind it. Each fig‑8 capsule effectively becomes a ‘half figure‑8’ one, with its rear lobe having been ‘deactivated’.

It’s possible to set up the shielded BPT array with separate microphones, but the Nevaton BPT makes it rather easier and less obtrusive to rig!

But what about the instruments in the rear of the orchestra, which are even further away than the strings in the front section? Well, as well as most being higher in level (eg. brass, timpani and, to a certain extent, woodwinds) they also have highly directive sound radiation characteristics and, therefore, naturally have a larger reverberation radius than the average for the orchestra as a whole. Thus, the signals of these instruments will exhibit a ‘healthy’ direct/diffuse sound ratio at the mic capsules.

Easy As BPT!

I suspect some readers may be put off by all the detail I’ve provided, but working this way with the Nevaton BPT is really easy. There’s no need for any processing, and it just comes down to looking for an appropriate location to erect/hang the mic and setting the desired level for the centre capsule relative to the L and R ones. So, really, a good set of ears and decent monitors are all you need to complete the job. For large sound sources, though, I like to add a large A‑B pair of omni outriggers, which can be used to add a nice amount of spatial impression.

You can hear the results I achieved with this technique in the Facebook promotional video for the Salzburg Festival: https://sosm.ag/salzburg-festival. You can try the same technique with individual mics, but it takes more work to rig and refine the position and, visually, it will be rather less discreet, which can be an important consideration.