Church of St. Bartholomew, St. Albans

The Problem

It was outlined by the Parish Council that the system that had been installed approximately four years earlier had not been a success and, whilst it was believed that the acoustics of the building were not conducive to good speech intelligibility and therefore played a part in the poor performance, it was also suspected that there may be some potential for improvement.

The Building


The building which was completed in 1962 is an irregular hexagonal shape and from a height of approximately 6 metres all sides slope inward to form an irregular pyramid rising to a point centrally above the congregation.

A local affectionate name for the building of ‘The Lemon Squeezer’ describes the building quite well!

The Investigation

We went to the church on an evening and there met Fr. Tim Edgar the parish priest and some parishioners who make up part of the Parish Council.  After the system was demonstrated using radio microphones and agreeing that it was largely unintelligible, suffering from a very ‘boomy’ characteristic, we suggested that as a point of interest that we carry out some tests to measure intelligibility using the speech transmission index.

The speech transmission index is a recognised standard of test and measurement which sets out to ensure that an electro-acoustic system achieves adequate levels of intelligibility and enables us to avoid dependence upon subjective opinion.

Measurement techniques use synthesized test signals that do not sound like speech but which have common characteristics to real speech signals the resulting sounds are then received to a meter through a calibrated microphone.

Our Findings

The reading measured in the congregation seating at St. Batholomew’s church with the old system was 0.29 which is the worst that we have ever measured and, as a point of reference, is significantly worse than that required of voice evacuation systems in public spaces. The tests confirmed that there was something seriously wrong with the existing installation and that there was, indeed, much potential for improvement. Whilst the ‘front end’ components of the system were good units that we could work with particularly an Allen & Heath iDR8 digital mixer, albeit following reprogramming, the root of the problem was believed to be the earlier choice of loudspeakers more resembling near field monitors and offering very little dispersion control in a fiercely reverberant space. As there are no supporting columns or other obstructions within the interior upon which to mount loudspeakers there are no opportunities to reduce the distance from loudspeakers to listeners and so the existing loudspeakers had been situated almost equidistant from the nearest microphone and the nearest listener with all three being within the reverberant field resulting in a total disaster for intelligibility and gain before feedback. In addition the distance from the altar to the furthest congregation seat is 14 metres and during busy services it is not unusual for latecomers to stand at a greater distance from the altar under the Choir Gallery.

As a part of the earlier system installation small loudspeakers had been installed under the Gallery with a view to providing reinforcement during busy services but these were operating into an even more reflective space and created a greater sound level than the main congregation speakers further degrading intelligibility throughout the open space. It became clear that a complete re-design would be necessary.

Conclusions and Our Proposal

When designing for speech intelligibility it is crucial to take into account the prevailing acoustics of the building and to design a system that uses loudspeakers that are correct for the application. It is also essential to remember that no amount of front end electronics, no matter how sophisticated, can correct an acoustic problem such as long reverb decay time or dominant reverberant field. Taking into account all of our findings we decided that the only possible solution was to propose the use of one loudspeaker of ‘steerable array’ design and, because the reverberant ‘noise’ was so severe, it would need to be the tallest loudspeaker that could be both afforded and accommodated. The reason for the need for a tall loudspeaker was that our tests had indicated that most of the problems were of a low to mid frequency nature and that it is a fact that a tall loudspeaker system would be more likely to offer the sort of dispersion control that was necessary. The single loudspeaker approach would also mean that there would be less chance of multiple signals arriving at listener’s ears at different times to potentially cause a degradation of intelligibility including those listeners standing under the gallery.

The Demonstration

Whilst the science of test and measurement is essential it is also desirable to have the confidence of hearing a possible solution before placing an order, particularly for a client who has suffered the unpleasant experience of having previously purchased an unsuccessful installation. We therefore borrowed a Tannoy Q Flex 24 loudspeaker, tried it in a number of locations and then installed it on a temporary basis as close as possible to the planned final ‘resting place’ and fed it with a re-engineered signal from the existing digital mixer. Once Fr. Tim, the Parish Council and ourselves were assured that the Q Flex was a sufficient improvement it was decided to leave it in place for a weekend so that it could be tried during a number of services.

The Result


Happily, the demonstration test proved to be entirely successful with parishioners attending the services commenting to us very positively in addition to our measuring speech intelligibility in the region of better than 0.6; not a fantastic figure but certainly quite an achievement in such a difficult space. Following receipt of an order installation soon began and was completed within five days including a smaller equipment rack built inside some existing furniture now that amplifiers were not required and freeing up work surface space in the Vestry.

As the loudspeaker is powered with it’s own amplifiers and on board processors, it was necessary to arrange for a 240v spur to be installed adjacent to it so that it can be switched and protected from the rack using a sequential power switcher and conditioner. All installation and engineering work was carried out directly by Sound Systems UK qualified electrical engineers.

Following one month of use Fr. Tim was pleased to offer the following testimonial reference:

“I don’t think there has been one negative comment about the new speaker. The Archbishop of Westminster was here for the Golden Jubilee of the Parish and he was very pleased with it, as was Bishop John Sherrington who was here at the beginning of October for his Visitation of the Parish. From the celebrant’s perspective, one has to work less hard generally and it is certainly easier to project above the ambient noise of children and toddlers. The installation is very discreet and one hardly notices the speaker at all. I think, from the Parish perspective, this was a very successful undertaking and has brought an end to years of complaints about audibility”.