Making tracks

Discussing common issues the public can encounter with acoustic concepts, technical terminology, the interpretation and application of British Standards, and how to recognise good service in acoustic consultancy and professional competence in the industry.


If there is a point sound source, here shown marked A0, sound radiates from that location, here a nominal x=2m, y=2m, z=2m.

If a wall is located to the side, there is a reflection from that wall such that there is a virtual source or a mirror source located beyond that wall at a distance x=-2m so 2m beyond the wall.

If there is now an additional wall at right angles, we gain an additional reflection through that wall at y=-2 but also in the space behind the corner where x=-2 and y=-2 where each mirror reflects in the other mirror.  So adding 1 wall gives 1 virtual source but when there are 2 walls there are 3 virtual sources.

Adding a floor as shown in the final image gives an additional 4 reflections, now 8 sources are present, the original plus 7 virtual sources.

If the original source is at coordinates x=2, y=2, z=2 then the reflections are at locations:
1) x=-2, y=2, z=2
2) x=-2, y=-2, z=2
3) x=2, y=-2, z=2
4)x=-2, y=2, z=-2 
5) x=-2, y=-2, z=-2
6) x=2, y=-2, z=-2  

Areas of Acoustic work

'Design' is a term which means different things when used in different places.  We rarely if ever 'design'.  We do not have full control over a building or a concept.  We offer advice, based on testing, measurement, calculation and modelling.  We might find that a particular separating structure requires improvement if a given criterion is to be met.  Our sound insulation advice is perhaps what others would call sound insulation design - but ultimately the architect is the designer, we advise and propose changes and systems which architects may incorporate into their design.

Sound surveys or noise surveys are routinely undertaken across the north of England, and our work expands out into the rest of the UK too.  We routinely undertake surveys in Sheffield, Leeds, Hull, Manchester, Newcastle, Harrogate, and the areas around Yorkshire.

Public address systems and announcement systems can be tested and improved where necessary.  Intelligibility being critical.  Audio systems and sound system creation are of course in our area of expertise.  

We deal with performance spaces and theatres, as well as venues and nightclubs.  

Noise control is a primary area of work for us, allowing a radio studio to function adjacent to a major road network or railway line for example.

Planning applications and consideration of the whether a proposed development is acceptable in noise terms in a given UK location are commonly undertaken.  Whether the development is a dwelling or group of dwellings, a hospital or a care home, or on the other side of the coin, the development might be industrial or commercial, a restaurant or a power station.

Haas effect

If two clicks are generated simultaneously and played to a listener via headphones - one to the left ear and one to the right ear, the listener would perceive a click 'heard' in front of them.  This can be thought of as the average of the perceptions.  

If one of the clicks is delayed by 5ms so that it arrives at the ear later than the first, then the listener will still perceive the two clicks as being connected but will focus in on the perception information of the the first click and ignore the perception information of the second click.  The brain will consider the second to be a reflection.  

If the delay is increased to longer than 5ms then the listener will perceive two distinct separate clicks.  

For complicated sound source signals music as music and speech the limit would usually be higher and might be as high as 30ms or 40ms.  

This first arrival focus is called the Haas effect after Helmut Haas who was one of the first to write about it.   Hans Wallach did similar work around the same time.  Joseph Henry wrote about it in the 1850s and really the Haas effect should be the Henry effect.  Henry wrote “On the Limit of Perceptibility of A Direct and Reflected Sound” in 1851.

The effect is used in sound reinforcement where repeating distant loudspeakers are delayed so they are sent around 10ms after the physical sound wave from the stage pases them.  Even if the repeating signal is 10dB louder than the stage signal, the brain perceives the sound as coming from the stage as the first signal arriving came from the stage and so the perception data is tied to that.  The reinforcing loudspeakers increase the sound level but do not interfere with the source perception.     

Acceptable sound levels LAeq 

Different standards require different things and set different criteria.

Some room types are more sensitive to noise than others and as a rule of thumb some sources suggest the following criteria:
Broadcasting studio                                          15dB(A)
Concert hall                                                         20dB(A)
Classroom / conference room                         30dB(A)
Bedroom                                                              35dB(A)
Living Room / cinema / courtroom /library   40dB(A)
Private office                                                       50dB(A)
Restaurant                                                           55dB(A) 
Sports hall                                                           60dB(A)
Workshop                                                            65dB(A)

Perhaps the above might be adjusted for daytime, evening or night time considerations.  This might be 5dB(A) lower in the evening and 10dB(A) lower at night, depending on the issues.

Perhaps the above might be adjusted for location, say allowing the following adjustments:
Rural                +0dB(A)
Suburban        +5dB(A)
Urban              +10dB(A)
City                  +15dB(A)
Industiural      +20dB(A)

So bedroom in the day in a city might become 35dB(A) + 0dB(A) + 15dB(A) criterion 50dB(A)

Whereas bedroom in a suburban area at night might be 35dB(A) + 5dB(A) - 10dB(A) criterion 30dB(A)

These will generally not agree with BS8233 or WHO sound level criteria. 

0.5 seconds

Usually measurements that involve reverberation time correction are corrected relative to 0.5 seconds.  If a 0.88 second reverberation was measured at a particular frequency, it would be corrected by 10log(0.88/0/5) which in this case would be 2.5dB i.e. the sound level measured is 2.5dB louder than it would have been if the reverberation time had been 0.5 seconds. 0.5 seconds being the reference reverberation time.  ISO 140 states "The standardizing of the level difference to a reverberation time of 0,5 s takes into account that in dwellings with furniture the reverberation time has been found to be reasonably independent of the volume and of frequency and to be approximately equal to 0,5 s"  0,5 meaning 0.5 in UK terms.  'Reasonably independent' doing a fair bit of work here.  Larger rooms will have longer reverberation times, and it is usual for lower frequencies to have longer reverberation times - but this is not always the case.    


For acoustic engineers, octaves start with 1kHz and multiply and divide from there, so 2kHz, 4kHz, 8kHz, 16kHz, 32kHz etc, and 500Hz, 250Hz, 125Hz, 63Hz etc.

In music, octaves are also doubling and halving frequencies so if the A above middle C, A4 is 440Hz then the next A up would be 880Hz and the A below would be 220Hz.

Tuning of each note is much more complicated than you might expect and the 'correct' frequency of each note depends on the tuning system being used.  There is a choice and the frequencies used has varied through history and location.

Some tone combinations sound more or less desirable when played in combination and so different ages have solved tuning issues with different tuning system and temperament.

Octaves are absolute and doubling the frequency will give the octave up.

Tripling a frequency and then halving will give a perfect twelfth, then perfect fifth when brought down the octave.

Using this tripling and halving method the following notes can be found.  If the starting place is A4 = 440Hz then:

 | Action  | Note  | Frequency
 | Origin  | A4  | 440
 | Triple  | E6  | 1320
 | Halve  | E5  | 660
 | Triple  | B6  | 1980
 | Halve  | B5  | 990
 | Triple  | F#7  | 2970
 | Halve  | F#6  | 1485
 | Triple  | C#8  | 4455
 | Halve  | C#7  | 2227.5
 | Triple  | G#8  | 6682.5
 | Halve  | G#7  | 3341.25
 | Halve  | G#6  | 1670.625
 | Halve  | G#5  | 835.3125
 | Triple  | D#7  | 2505.938
 | Halve  | D#6  | 1252.969
 | Triple  | A#7  | 3758.906
 | Halve  | A#6  | 1879.453
 | Triple  | F8  | 5638.359
 | Halve  | F7  | 2819.18
 | Halve  | F6  | 1409.59
 | Halve  | F5  | 704.7949
 | Triple  | C7  | 2114.385
 | Halve  | C6  | 1057.192
 | Triple  | G7  | 3171.577
 | Halve  | G6  | 1585.789
 | Triple  | D8  | 4757.366
 | Halve  | D7  | 2378.683
 | Halve  | D6  | 1189.341
 | Halve  | D5  | 594.6707
 | Triple  | A6  | 1784.012
 | Halve  | A5  | 892.0061
 | Halve  | A4  | 446.003

As can be seen, the notes can all be found, and each octave up and down can be found for each note, but the A4 at the end is not 440Hz it is 446Hz, so there is an error in the system.  This is one of the reasons that other methods had to be explored.

It can be seen another way in the maths.  twelve 5ths would get around the circle of fifths. A back around to A.  we times by 3 and divide by 2 to get a fifth, so that is (3/2)^12 = 129.746.  This would be the same note 7 octaves higher so that is 2^7=128.  129.746 does not equal 128 and this is why there is a problem.

There are different tuning systems around the world and through time and each tuning system has its own strengths and weaknesses.  If 12 notes are used, as in most western music, then the above perfect fifth / circle of fifths idea almost words, but it doesn't quite work.

Equal temperament divides the octave into 12 equal parts which are 2^(1/12) apart.  Using this system the errors are evenly spread and each is very small and not noticeable.  All modern western music is a little bit of out tune as a result of this.  The error is small enough to not be noticable.

Another solution would be to have all of the perfect fifths as set out above, but to have one of them that is not correct.  Then most of the ways to use the notes would work, but there would be one note that is not right and all the error is loaded onto that note.


Soft floor covering f0

A soft floor covering on a separating concrete floor can make a significant improvement in terms of impact sound insulation.  Usually the improvement is at mid and high frequency and there will remain a thud at low frequency.  The impact sound insulation reference curves are more forgiving at low frequency in any case.  Improvement will be given above the f0 frequency.  

Mat thickness and elasticity play a role in performance and the thicker and/or softer the mat is the lower the f0 frequency.  f0 can be taken to be (k/m)^(1/2) where k is the stiffness and m is the tapping machine hammer mass of 0.5kg. k = ES/h   E is Young's modulus, S is area of the striking hammer which is around 0.0007m2 as it is 30mm diameter so pi*r2 is pi*0.015^2 is pi*0.000225 is 0.00070685835m2, h is the thickness of the material.

f0 is then

f0= (1/2pi)*[(S*E)/m*h)]^1/2

BS 5228-2:2009+A1:2014 Vibration Control on Construction Sites

When piles are to be driven and there is the risk of excessive vibrations emanating especially from the upper strata.  BS5228 suggests that the problem can sometimes be reduced by pre-boring.

Pre-boring involves removing some of the soil which would have been displaced by piling in in the initial part of the drive. 

Pre-boring might reduce the number of blows required to drive the pile.

Initially driving a open ended tube can allow the top soil to be removed in the form of a plug within the tube.  This can be repeated several times before the shoe is fitted and the close ended tube driven.


Diffuse soundwaves

A soundwave (or series of)  is perfectly diffuse if the sound pressure is of equal value in all places within the room and there is equal probability of the direction of soundwaves.  Perfect diffusion is impossible to achieve.  

Diffusion is increased by objects in the room that scatter sound and randomise the direction of the soundwave.

Irregularities in room surfaces such as walls also scatter incident soundwaves to a degree.

Sound in a furnished room will be more diffuse that in a unfurnished room.

Diffusion is easier to achieve at higher frequencies as the physical variation in dimensions of materials and objects is relatively large compared to the wavelength of the sound.  Diffusion at low frequencies it harder to achieve.

Absorbent material can also be used to increase diffusion, depending on the location of the material. 

Vibration dose value combining

Vibration dose value is a method to consider the cumulative impact of vibration episodes over time.  This might be measured in one long measurement.  However it might be measured as a series of measurements which might have to be combined.  The combination procedure is:

VDVtotal = (VDV1^4 + VDV2^4 + VDV3^4)^(1/4)

Acoustic precision and accuracy

Usually shots on a target is used to explain the difference between precision and accuracy.

Accuracy can be thought of as whether the average of the values hits the bullseye.

Precision can be though of as the spread of the values.

So 10 shots all on the bullseye is accurate and precise.

10 shots all clumped together but away from the bullseye is precise but inaccurate.

10 shots all spread out but in all directions from the bullseye is accurate but imprecise.

10 shots all spread out but not centred on the bullseye is inaccurate and imprecise.