Examination of Witnesses (Questions 160
- 179)
160. CHAIRMAN: I remember you from before.
161. MR ELVIN: Mr Thornely-Taylor, you
have prepared a set of Powerpoint slides which you are going to
speak to, and I am simply going to let you do your presentation.
I will only interrupt from time to time, if it is necessary to
do so. So, if you would please commence.
162. CHAIRMAN: I hope that if any of
my colleagues want to ask questions about this—because it
is not easy stuff—they do so as they see fit.
163. MR ELVIN: My Lord, I was going to
suggest—and, of course, it is entirely a matter for you
and the Committee—that I suspect Mr Thornely-Taylor would
welcome questions as he went along.
164. CHAIRMAN: I think it is going to
be easier that way than trying to go back to some particular point
in the presentation.
165. MR ELVIN: Indeed.
(Mr Thornely-Taylor) My Lords, I will do my best to make
what I know is generally thought of as quite a difficult subject
as clear as possible. I will begin, if we can see slide 4, by
talking about the basics of noise and vibration, and do my best
to make clear the nature of the decibel scale and all the indices
that are used for measurement and assessment for noise.[42]
Then I will talk about standards for the assessment of noise and
vibration. I will go on to talk about mitigation—using that
in the environmental assessment sense ways of minimising noise
and vibration—and I will say something lastly about the
way in which levels of noise and vibration are predicted.
166. MR ELVIN: Mr Thornely-Taylor, just
so this is absolutely clear, a draft of this presentation was
sent to Messrs Sharpe Pritchard, who are the Agents for the two
London local authorities—London Borough of Havering and
London Borough of Camden—which have outstanding general
noise issues on behalf of the London local authorities, and that
you will make appropriate comments as you go through to reflect
qualifications which they referred to in subsequent correspondence.
(Mr Thornely-Taylor) Yes, I will. If we can move to slide
5, clearly there are several ways in which noise and vibration
are of interest in this project.[43]
There are construction issues, both in terms of surface construction
of the relatively conventional kind, worksites at ground level
with construction plant operating in them, and then of course
there is tunnel construction, which is the driving of tunnels
underground, the excavation of stations and other works underground.
From the operational point of view, there is surface operation
of the railway, the passage of rail vehicles along the track;
there is also the operation of a very substantial amount of what
I have referred to as "fixed plant", meaning everything
from tunnel ventilation systems to station air-conditioning, transformers—all
items of that kind—and there are other surface operation
areas such as depots and sidings, which also come under that heading.
Then, last but by no means least, there is the operation of Crossrail
underground—that is the passage of trains in the tunnels.
As I think we all know, there are tunnels in London that you can
hear, and I will be talking quite a lot about that topic. May
we move to slide 6.[44]
I will begin with the basics. Noise is traditionally described
as unwanted sound. Sound is vibration of the air as a result either
of a surface vibrating and causing radiation of air vibration
waves away from it, or by aerodynamic or thermodynamic vibration
of air itself. I think most people now are familiar with the use
of the word "decibel", although it tends to be abbreviated
into dB—a little "d" and a capital "B".
The reason for the capital "B" is it is named after
Alexander Graham Bell who actually devised a system called "a
bell", which was rather large and it has been divided into
ten to call it the "decibel". The most interesting thing
about this slide is the capital letter A, and we will hear quite
often reference to either dB(A), or later on I will be talking
about other indices which also include the capital letter A. That
is an important feature of the scale that is used, because the
human ear is nothing like a physical instrument. Its response
to sounds of low pitch and high pitch is not at all the same as
its response to sounds around the middle of the pitch scale, and
we have to take that into account. If you took a basic sound level
meter that did nothing but measure the amplitude of those air
vibrations I was referring to, the number it would give you would
not correspond to human perception of loudness because the human
ear is very bad at picking up low-frequency sounds—rumbles,
roars—and bad at picking up very high-frequency sounds.
Instruments have to be fitted with electronic weighting networks
to make them behave slightly more like the human ear so that the
index that is used to measure loudness and other forms of human
response to sound comes up with a better relationship between
the measured number and people's actual judgment of what it sounds
like. I will say quite a bit more about this effect later on.
167. Just to stop you there, of course the dB(A)
is sometimes referred to as the `A-weighting'.
(Mr Thornely-Taylor) That is quite correct.
The `A' is A-weighting, and I will be saying a little bit more
about some other forms of weighting on a later slide. For most
everyday noises, the basic feature of the decibel scale is that
you use it to express changes in proportions. It is a scale for
saying when things have doubled and when they have halved.[45]
It is not like a temperature scale, it is not like a measure of
concentration of atmospheric pollutants where you are simply stating
in old physical terms how much of something there is out there.
The decibel scale is uniquely about changes and that happens all
the way through and this is the basic one to bear in mind, that
for everyday noises, about a 10 dB increase is perceived as about
double the loudness, and that happens all the way up the scale,
30 to 40, 40 to 50, 50 to 60 and so on because the decibel scale
measures proportions and it is telling you each time it goes up
by ten that there is about a doubling of the loudness. That becomes
quite important when we look at ways in which varying noises are
expressed in terms of a composite index. The opposite of course
is true if you have a 10 dB decrease and the loudness is about
halved. There are some exceptions to that which I will come to,
but that is a very useful rule of thumb to start with. If we could
move to slide 8, we now go to much smaller changes.[46]
If the sound which we can hear from the projector in the room
were suddenly to change by one decibel, 1 dB, or if any other
continuous, unvarying sound suddenly dropped by 1 dB, it would
only just be possible to detect a change in loudness. If we all
went out of the room and came back in and that noise source had
changed by 3 dB, which is the next slide, slide 9, we would just
about perceive that its loudness had changed.[47]
If we were sitting in the room and there was a sudden change of
3 dB, that would be quite clearly noticeable, but, if there is
a lapse of time between the noise changing from, say, 60 to 63,
that is about the smallest change we would notice unless it happened
instantaneously. Moving to slide 10, it is quite common to show
in textbooks and talks of this kind either a thermometer or some
scale explaining what decibel levels correspond to for sounds
in everyday life, and it is always difficult to do that.[48]
We have this amazingly quiet level at the bottom of 10 dB(A),
falling leaves, and the difficulty is that I do not know anywhere
on the planet which is quiet enough to be able to measure 10 dB(A).
I think the quietest place ever measured was the rim of the Grand
Canyon which was 14 dB(A), and I have measured 19 on Foulness
Island, but it is effectively silence, and indeed the whisper
at 20 dB(A) is so quiet that most people would consider it no
noise at all.
168. So what we commonly think of as a silent,
very quiet background in the country is likely to be somewhere
in the region of 19 or 20 dB?
(Mr Thornely-Taylor) Yes. My experience on Foulness Island
was at the time of my first experience, not in this room but another
one nearby, and I am embarrassed to say it was 37 years ago for
the Maplin Development Bill. I went to Foulness Island and on
a totally still day with no vehicles for many miles, no wind,
no sources of any kind that one could notice, 19 was the lowest
I could measure and even that probably had a contribution from
the instrument's own internal circuitry. You have to get up to
about 30 which is there as a bedroom, and that is regarded as
an ideal noise level for a bedroom. Again, it would be difficult
to hear very much. There is a step up to 43 which is identified
as a wind turbine. It depends obviously where you are as to what
the level is. This would be a figure at a substantial distance
from a wind turbine at a residential location and we move gradually
up to the living room, office, in the car 77, stereo music, well,
maybe 87—
169. LORD YOUNG OF NORWOOD GREEN: I think
there is a special index for stereo music for teenagers! That
could be up to the pneumatic drill level, I would think!
170. MR ELVIN: My Lord, you took the
words out of my mouth!
171. LORD YOUNG OF NORWOOD GREEN: I was
thinking about iPods in a train with that horrible tinny sound
that is audible from one end of the carriage to the other.
(Mr Thornely-Taylor) And indeed very high levels can be produced
by those and hearing damage is a real issue with people who wear
iPods. Indeed, once we get up into the 90s, we start talking hearing
damage. With industrial noise at 95, you would have triggered
action levels according to the Noise of Work Regulations because
continual exposure for a lifetime at 95 would expose a proportion
of the population to permanent hearing loss, and 115 is a very
unpleasant noise indeed. One of the first loud noises I was ever
exposed to when I first came into this field was a sudden burst
at 124 at a factory that was testing compressed aircraft engine
starters and, as a callow youth, I immediately clapped my hands
to my ears and crumpled to the ground; it was extremely unpleasant.
The 141 would be immediately damaging, and it says "aeroplane",
but you would be standing within metres of an aero-engine and
it is most unlikely that anyone would really do that. The range
is enormous, as you can see, and the human ear is capable of hearing
a vast range of sound levels. If we move on to slide 11, here
we move into the realms of sound that is much more complex than
the examples I gave earlier of a continuous sound suddenly changing
in level.[49]
In circumstances such as this, we are frequently dealing with
fluctuating noise levels, noise from individual events like passing
vehicles, aircraft or trains, and, if you stood by a railway or
under a flight path with a simple sound level meter, all you would
get is a lot of different levels. It would fall to low levels
when nothing was passing and it would reach quite high levels
if an aircraft flew overhead or a train went by and then it would
fall again. If you were charged with quantifying that noise environment
in a way which could be used to say, "The noise environment
for this community is this level and there is a significant effect",
matters of that kind, you need an index which will deal with rising
and falling sound, peaks of sound and get it into one number.
Clearly, you can measure the maximum noise level when aircraft
go over or when trains go by and we do that by Lmax, as it is
called, and we may put the `A' in front of it to signify the A-weighting
I was describing earlier. LAmax is a useful measure of the passage
of vehicles, for example, but, if there is one vehicle a day or
one aircraft a night, that is not as significant as 10, 20 or
100 vehicles in the period, so some means is required to take
account both of sound level and of the number of times the event
occurs or how long the event goes on for, and we use an index
which, in full, is called `equivalent continuous sound level',
which is abbreviated as `Leq' (for equivalent) and nearly always
it is A-weighted, like the dB(A), and then it becomes LAeq. It
is an index which crops up in almost every case when environmental
noise and indeed even occupational noise is being assessed and,
consequently, it comes into the firing line from people who consider
that it is in some way inadequate. It helps to understand exactly
what it does because some of the criticism of it is misplaced,
and I will do my best now to explain what Leq does and show it
can usefully be used for these time-varying environments. Slide
12 is a hypothetical example of noise varying with time.[50]
This is actually artificial, I made it up, but it is sound levels
bobbing along around 45 dB(A) until something happens, it could
be the passage of a vehicle, it could be the switching on of a
piece of plant for a time and then switching it off, and then
we go back to ambient noise, if I could use that term, which frequently
applies to the general goings-on acoustically in the environment,
low levels interspersed by minor noise events. It is not an untypical
case and the scale on the left is the dB(A) scale that I have
been describing and it is going up in tens, so to the extent that
the ambient gets up to around 50 and my hypothetical event gets
roughly to 70 while that event is on, if it went up to 60, it
would be double the loudness, but it is actually going up to nearly
70, so that moment is about four times as loud as these moments.
Now, that is a classic case of needing to put a single number
to a very variable noise environment, and this is how it is done.
The next slide, slide 13, looks very similar, but there is one
essential difference, that it is not dB(A) anymore, but the plot
is identical and we have what is known as a `logarithmic scale
of energy' which is what the dB(A) scale or at least the pure
decibel scale is really measuring.[51]
One of the reasons why the decibel scale confuses everyone is
that, when I was saying that from there to there was a doubling
of loudness, in energy terms you will see that these energy units
have gone up tenfold. The human ear does not respond to an increase
in energy at anything like the rate you would expect it to if
you looked purely at energy units. If a sound level meter gave
you a reading in watts instead of in decibels, you would see the
number of watts per square metre go up by a factor of ten and
you would say, "That's odd. It only sounded like twice",
and that is one of the psychophysical effects of human hearing,
that it is not as sensitive as the increase in energy would imply.
The corollary of that is that where this peak went up to about
four times the loudness, the energy has shot up and, if we go
to the next line up, it has gone up tenfold and going to the top
of the scale, it has gone up a hundredfold. Then, as in slide
14, if you put exactly the same environment on to the much more
familiar linear scale, which would be the same if it was particles
of pollutant or anything like that, each line is the same, so
we are not talking about tenfold increases, we have a linear scale,
that makes that event in the middle really show up for what it
is, a huge reservoir of energy, so to speak.[52]
What LAeq does is, first of all, to get into this state and average
the energy and, then if we move to slide 15, going back into decibels
we find that, having averaged the energy, it is a much higher
result than had we just averaged the decibel levels.[53]
You hear lay people who are making a case about noise environments
of one kind or another talking about LAeq being an average. They
sometimes think it is this kind of average and, if it were, it
would be a pretty poor way of expressing this entire sequence
of events because it would not give anything like the weight that
this noisy moment merited, whereas the LAeq scale, because it
is driven by the hidden store of energy in the high levels, comes
out much higher than an actual arithmetical average. It is nothing
like as bad an index as you often hear people say as dealing with
environments with elements of high noise level superimposed over
periods of relative quiet.
172. CHAIRMAN: Is the green line the
figure without the peak?
Mr Thornely-Taylor: No, the green line includes
the peak. The green line is the arithmetical average of all the
decibel levels.
173. CHAIRMAN: Including the peak?
(Mr Thornely-Taylor) Including the peak, yes, my Lord. That
is one of the concerns people have about the LAeq scale. The other
concern—and it is a very important one—relates to
its faithfulness in dealing with the frequency content of the
noise. I began by explaining a bit about the A-weighting scale
and the fact that the human ear is very insensitive to low frequency
noise, particularly insensitive to very low frequency noise. I
describe the A-weighting scale as an electronic network that goes
into sound level meters to make the meter a bit more like an ear.
To make things more difficult, as sounds get louder, that inequality
that the ear has becomes less and once you get up to those horrendously
high noise levels, like my unfortunate exposure to 124, the ear
is pretty much linear in its frequency response at low frequencies
and at high frequencies it responds with about the same displeasure,
in that case, for the same sound pressure level, no matter what
the frequency. When sound level meters were first manufactured,
they came out with a minimum of three different weighting networks,
A that I have been talking about, also B and C which were meant
to address loudness at higher levels. You were supposed to use
the B curve if you were talking about something around 70 decibels
taking 1,000 hertz as the reference point and 1,000 hertz means
1,000 cycles per second and is a whistle. You were meant to use
the C weighting curve when you got right up to about 120. For
reasons more of convenience and practice, the A curve has come
to be used at all levels of sound, so if you are measuring the
noise in a recording studio at very low levels, the noise at the
roadside in the 70s or 80s or the noise in a factory in the 80s
or 90s the practice now is to use the dB(A) even though it is
the wrong weighting network for the higher levels. It gives rise
to a justifiable criticism that the A weighting curve unfairly
discriminates against low frequency noise as it does at medium
to high sound levels. We are going to be concerned in these proceedings
with some fairly low sound levels. There are issues in many petitions
about noise from the operation of underground trains and I understand,
my Lords, that you will be visiting a location, I think, at the
beginning of next month to listen to some underground train noise
and we will see what the sound levels are and they will be around
the upper 30s up to about 40. Therefore, I need to spend just
a moment looking at this frequency weighting issue because it
is very important. The next slide—
174. MR ELVIN: Just before we go there,
Mr Thornely-Taylor, while we have this up, this is your hypothetical
example that you referred to earlier. If we were going to have
a look—because you referred to it earlier—the LAmax
the max would be here (Indicating)? Is that right?
(Mr Thornely-Taylor) It would indeed. That might be a moment
to explain two types of LAmax, called LAmax, S for "slow"
and LAmax,F for "fast". What those two subscripts mean
is that one is much more sensitive to variations than the other.
Early sound level meters used to have moving needle indicators
on them and if you measured a sound that was not totally steady
with the meter on the fast setting, the meter needle would move
fairly rapidly. It would be quite hard to note down a repeatable
level for the maximum sound level so they also were fitted with
something called the "slow time constant" which more
or less dampened the movement of the needle. That, in fact, has
an eight times longer time constant, there is dampening to the
needle, a little bit of averaging goes on and the difference between
fast and slow is just a smoothing of the curve. We will be looking
at both fast and slow maximum sound levels very shortly. If I
could go on to slide 16, this is to look a little bit closer at
the important business of low frequency noise and the way we hear
it and the way we measure it.[54]
I was talking a short while ago about the existence of several
different weighting curves and referred to the fact there is an
A, a B and a C. You will even find in the literature D and E,
but I will not go into those. The A-weighting curve is plotted
on this scale here—it is difficult to see the colours—it
is this line (Indicating). It is this sort of shape. If we use
the whistle, 1,000 cycles per second, 1,000 hertz, as our reference
point, the A-weighting curve down, for example, at 63 hertz and
that is a rumble, quite a low rumble, the curve has actually gone
up by 26 decibels. What that means is if you listened to a sound,
look at the sound level on a simple sound level meter with no
weighting in it and have alongside a more sophisticated meter
with A-weighting in it and you measured a sound of a whistle at
30 dB, and then if you turned that off and switched on a low rumble
which a jury of people would consider to be the same loudness
as the whistle, the meter that was fitted with the A-weighting
scale would still show 30, but the simple meter with no weighting
in it, which is merely measuring the physical amplitude of the
sound, would measure 56 because that 26 is the ear's inability
to hear low frequency sounds with the same sensitivity that it
does whistles and other high frequency sounds. There are several
curves on this slide and there is an important reason for that.
Perhaps the most interesting one is not the curve but the group
of triangles. If you go into loudness in great detail and you
study the response of numbers of people, juries of people, who
answer whether they consider one noise to be the same loudness
as another, you end up with a family of curves called "equal
loudness curves" which has now been standardised, there is
an ISO standard 226, and they actually use a unit called the "phon",
which is one of the very earliest acoustical measurement units.
The only point I want to make, I do not want to confuse the issue
by talking about phons—we will never refer to the term again
except for the next slide—the important thing is what the
phon is showing you is that the whistle at 1,000 hertz if you
go down to the rumble at 63, the phon is telling you that the
A-weighting curve is underweighting at this level and that the
true loudness is actually less than the A-weighting meter would
show you. There are also other more sophisticated ways of rating
noise which we may have to deal with later in the proceedings
because they are used for rating noise in places like auditoria,
recording studios, special places where the sound environment
is critical. They approach things in a slightly different way,
there are two rather similar indices, one called "NC"
which stands for noise criterion, and that appears in an ISO standard
for the recording industry, and there is a European-based, very
similar, system "NR" called noise rating. The fundamental
point is that they too adversely weight low frequencies to very
much the same degree as the A-weighting curve. That is for sounds
around about 30 dB(A) level. For simplicity's sake we are talking
about single frequencies: whistles, hums, rumbles. If we move
on to slide 17, we have gone up to double the loudness, we have
gone up to 40, but the same picture emerges.[55]
The only thing which is apparent is there is rather more uncertainty
at the low frequency end, the different systems agree slightly
less about what is of equal loudness. Again, there is no fundamental
problem with the A-weighting network overdoing things at low frequency.
I am not going to prolong things by showing you curves going on
higher—50, 60, 70—but if I did we would then start
to see the A-weighting curve falling down progressively more and
the higher levels, dB(A), are justifiably criticised for not measuring
low frequencies very well. Down around the 30 and 40 mark all
the international systems and the basic loudness assessment show
that it is about the right shape. The only thing which is important
to bear in mind with dB(A) levels at low frequencies at these
low sound levels—and I think this will become apparent on
the site visit when we go to listen to underground train noise—is
that the rule of thumb I almost began with, that 10 dB(A) is a
doubling of loudness, does not hold good all the way down to very
low frequencies. If we are dealing in rumbles, the change in loudness
is rather more than that and we will be probably be lucky enough
to hear some trains go underneath the room, we will be in at around
40 dB(A); we will hear some others go by at about 30. You can
judge for yourselves whether you believe that to be a halving
of loudness or a bit more, but you may consider that it is a bit
more than a halving and that is one of the consequences of the
A-weighting scale and one of the reasons why we would be in difficulty
if we were using it at much higher levels than this.
175. LORD BROOKE OF ALVERTHORPE: You
referred to all the international systems. Are there many? Is
there a common one in Europe?
(Mr Thornely-Taylor) There are many systems for detailed
evaluation of noise, my Lord, like the noise rating, the noise
criterion. There is a system for certificating aircraft noise
which uses yet another weighting method. I think if I were to
write down all the different ways used for measuring noise I would
fill an A4 sheet of paper quite quickly. When we are dealing with
environmental noise and assessment of it everybody uses the LAeq
scale to a greater or lesser extent. Additionally they use maximum
noise levels LAmaxS and LAmaxF, and I will show you some international
guidance shortly. There are also sub-indices of the LAeq that
enable you to calculate it using something called SEL, which either
means sound exposure level or single event level according to
which standard you are using. I do not need to go there at present
but I think, largely to keep my colleagues in work, we have invented
more indices than any other field of endeavour. Unfortunately
it remains so complex that consultants are needed to sort things
out.
176. MR ELVIN: Mr Thornely-Taylor, just
to pick up Lord Brooke's question, in terms of the World Health
Organisation and the advice which it gives on an international
basis, what is its favoured set of indices?
(Mr Thornely-Taylor) I will shortly show a slide in which
most of the guidance is LAeq, but there is also important advice
about LAmax, the maximum noise level measurement, as well.
177. MR ELVIN: Which is one of the reasons
why we are focussing on it.
178. LORD YOUNG OF NORWOOD GREEN: Before
you move on, just to get a fix on this, in terms of what is detectable
by the average human ear, and that is a weighted phrase in itself,
is it not, as you say with frequency we are not very good at low
and we are not very good at high and these are age-related as
well, but what is the sort of range we are talking about? Are
we talking about an average human ear at its peak of hearing?
(Mr Thornely-Taylor) You may have noticed a story in the
press recently about a device called a Mosquito which is intended
to discourage young people from congregating in shop doorways.
It produces a very high frequency sound. As a very young man playing
around with sound oscillators I can remember what it was like
to hear 20kHz. Now most of us hear 20kHz in the form of tinnitus
internally generated within our heads. As you get older something
called presbiacusis comes in which is a natural reduction in the
sensitivity of the ear to very high frequencies. Birdsong is a
bit lower than that and fortunately it is usual to be able to
continue to hear birdsong until a very advanced age, that would
typically be around the 2-4kHz area—1kHz is a whistle. As
you go down in frequency there is not normally an age-related
effect except that disorders of the hearing can reduce low frequency
sensitivity due to mechanical malfunction of the ear. Most importantly
noise-induced hearing loss—damage to hearing caused by high
noise levels at work, or excessive exposure to loud music—has
a very specific effect. When cases come before the courts claiming
noise-induced hearing loss what you look for is a dip at 4kHz;
it is localised around that area. When claimants come forward
alleging loss of hearing due to their employment if there is not
that characteristic dip you have to look very much more closely
at their hearing and the history of their exposure to noise. I
do not know whether I have answered your question or not.
179. LORD YOUNG OF NORWOOD GREEN: Just
at the low frequency end, what are we talking about in terms of
the average ear? What can you detect down to? What sort of frequency
are we talking about?
(Mr Thornely-Taylor) The textbooks will say 20Hz is the start
of the audible range; but what happens down there is you start
to confuse audible sound with vibration. If you live close to
a road and heavy lorries go by they may set your windows going,
they will set the room vibrating and you will perceive it more
as a whole body effect than just as an audible effect. As the
frequency goes down there is just a transition from perceiving
a heard sound to a felt sound, and you would possibly even describe
it as vibration even if it was still coming through the air and
not necessarily through the ground.
42 Crossrail Ref: P4, Outline of Presentation (LINEWD-RTT01-004) Back
43
Crossrail Ref: P4, Scope of Noise and Vibration Issues (LINEWD-RTT01-005) Back
44
Crossrail Ref: P4, Basics-Noise (LINEWD-RTT01-006) Back
45
Crossrail Ref: P4, Basics-Noise (LINEWD-RTT01-007) Back
46
Crossrail Ref: P4, Basics-Noise (LINEWD-RTT01-008) Back
47
Crossrail Ref: P4, Basics-Noise (LINEWD-RTT01-009) Back
48
Crossrail Ref: P4, Scale of Decibel Levels (LINEWD-RTT01-010) Back
49
Crossrail Ref: P4, Basics-Noise (LINEWD-RTT01-011) Back
50
Crossrail Ref: P4, LAeq is no ordinary average ... (LINEWD-RTT01-012) Back
51
Crossrail Ref: P4, LAeq is no ordinary average ... (LINEWD-RTT01-013) Back
52
Crossrail Ref: P4, LAeq is no ordinary average ... (LINEWD-RTT01-014) Back
53
Crossrail Ref: P4, LAeq is no ordinary average ... (LINEWD-RTT01-015) Back
54
Crossrail Ref: P4, Low Frequency Noise-A-weighting curve 30 dB
(A) Level (LINEWD-RTT01-016) Back
55
Crossrail Ref: P4, Low Frequency Noise-A-weighting curve 40 dB
(A) Level (LINEWD-RTT01-017) Back
|
