[Tony[_3_]]

Brian Marston wrote:
Background noise generators can work BUT can take a week or more to
acclimatize to, with the volume increased each night in very small
increments over 2-3 weeks.

I wonder whether people are really referring to white noise, or just using
the term to mean a continuous unidentifiable noise. The point of continuous
noise is to mask intermittent, rhythmic or tonal noise which is more
disturbing to sleep. (That is the theory, and it works for me anyway.) To
do this at all effectively, the masking noise must have energy in the
frequency region close to the frequencies contained in the interfering
noise. If it also has energy in other frequency ranges, it will be louder
than is necessary for effective masking and so more likely to disturb sleep
on its own account. White noise has a lot of high frequency energy, while
most common interfering sounds do not (car alarms and burglar alarms
excepted).
I would start off by measuring the maximum level of the interfering noise,
preferably in 1/3 octave bands but at least in octave bands, and then
tailoring the noise spectrum and level to that.
I've never heard that any sort of noise, as such, is conducive to sleep.
Maybe adverts try to imply it but if they actually say it, it would be
misleading. In all my experience, the quieter the better, down to a very
low level indeed. But then of course minor noises might wake you up. That
might even be useful - I was once woken by an extremely quiet burglar. I
would never have heard him, but it was a very still night in a place far
from traffic and industry.
--
Tony W
My e-mail address has no hyphen
- but please don't use it, reply to the group.

[Brian Marston]

Tony wrote:
Brian Marston wrote:

Background noise generators can work BUT can take a week or more to
acclimatize to, with the volume increased each night in very small
increments over 2-3 weeks.

I wonder whether people are really referring to white noise, or just using
the term to mean a continuous unidentifiable noise. The point of continuous
noise is to mask intermittent, rhythmic or tonal noise which is more
disturbing to sleep. (That is the theory, and it works for me anyway.)
SNIP
I would start off by measuring the maximum level of the interfering noise,
preferably in 1/3 octave bands but at least in octave bands, and then
tailoring the noise spectrum and level to that.

Agreed - I would only recommend masking noise where alternate solutions
would vastly exceed the client's budget and an external sources is
beyond control - a solution of 'last resort'.

I've never heard that any sort of noise, as such, is conducive to sleep.
Maybe adverts try to imply it but if they actually say it, it would be
misleading. In all my experience, the quieter the better, down to a very
low level indeed. But then of course minor noises might wake you up. That
might even be useful - I was once woken by an extremely quiet burglar. I
would never have heard him, but it was a very still night in a place far
from traffic and industry.

I would bottom out noise control at 20-25dB(A) with masking noise only
up to 35dB(A) maximum. Masking can either be a neutral bland noise or
recordings of a collection of familiar 'safe' sounds. It's the client
who wants almost anechoic conditions that worry me.