Quick and dirty sound meter calibration.

Archived from groups: rec.audio.pro (More info?)

Hi,
acquired a sound level meter that had a broken sound level meter. The
moving coil meter (which was 50uA full scale) was broken. The only meter
I could get that was physically the same size was a 100uA unit. Clearly
this was going to under-read, so with no better alternative, I placed
the meter in an environment where the meter was half scale (i.e. in the
middle) and adjusted the calibration control until it was full scale,
since my replacement meter was only half the sensitivity.

I doubt there is, but are there any simple calibration techniques I
could use to get the meter about right? I've no idea if the calibration
control has ever been fiddled with, and although my meter needs double
the current, that does not mean it will actually read twice as much as
the original, as they will have different internal resistances.

I basically acquired this to make comparative measurements, but it would
be nice if it was right - or nearly so.

I did a couple of quick checks.

TV on in a room at a comfortable level - 50 to 60 dBA.

I've not had a hearing test, but I'm only 42, so I doubt it will be too
bad.

Does that sound way out, or semi-reasonable? Any better ideas how I can
get it calibrated, without the expense of buying calibration equipment?

Archived from groups: rec.audio.pro (More info?)

Scott Dorsey wrote:
> In article <424435b9@212.67.96.135>, Dave <nospam@nowhere.com> wrote:
>
>>Hi,
>>acquired a sound level meter that had a broken sound level meter. The
>>moving coil meter (which was 50uA full scale) was broken. The only meter
>>I could get that was physically the same size was a 100uA unit. Clearly
>>this was going to under-read, so with no better alternative, I placed
>>the meter in an environment where the meter was half scale (i.e. in the
>>middle) and adjusted the calibration control until it was full scale,
>>since my replacement meter was only half the sensitivity.
>
>
> What kind of meter is this?
>
> With some meters, the coil resistance may be important, and in most
> cases the meter dynamics will be important unless you're operating in
> SLOW mode.
>
>
>>Does that sound way out, or semi-reasonable? Any better ideas how I can
>>get it calibrated, without the expense of buying calibration equipment?
>
>
> What is it? Is it an old Gen Rad or something for which parts are
> readily available?
> --scott


It's a Castle Associates CS. 15C. The front panel is marked Scarborough
UK. It looks like them (or someone very similar) is in existance:

Castle Group Ltd
Salter Road
Eastfield Industrial Estate
Scarborough
YO11 3UZ(Road Map)
North Yorkshire
Tel: 01723 584250
Fax: 01723 583728
Internet: www.castlegroup.co.uk


There can't be too many companies with the name Castle, specilising in
sound meters in Scarborough.

I don't know what the resistance of the 100uA moving coil was, as in an
attempt to repair it, the coil got damaged.

There is slow/fast switch, an OFF/BAT CHECK/ON switch, a rotary switch
with positions of 40, 50, 60, 60, 80 and 90. The scale shows -5 dB to
+10 dB. It takes a PP3 9V battery.

I could photograph it an put it on a web site if it will be any help.

The electronics in this thing is really simple. It must be 20 years old.
Not an IC in site.

The battery check does not work properly now, as all readings are about
50% lower than they should be. But I could easily change a resitor to
fix that. (I am an electroincs engineer).

I don't fancy paying for a new moving coil meter. If it needs me buying
another moving coil, I'll probably just stick with it for qualitative
purposes. I was trying to find out what parts of my computer make the
most noise, as there are two external disks, and so having something
that is at least qualitative is useful. But quantitative, even if not
particulary accurate, would be nice.

I guess finding the resistance of the original meter would be useful. It
will almost certainly be higher than the 100uA unit I fitted.

I guess there is no harm in dropping Castle Group an email.

Archived from groups: rec.audio.pro (More info?)

>There can't be too many companies with the name Castle, specilising in
>sound meters in Scarborough.

Give them a call and ask if they have parts.

To calibrate the thing, you either need a sound source with a known level,
or a known-accurate meter. You can build a source with a known level that
is entirely mechanical; it used to be common to modify model airplane engines
to produce a calibrated sound pressure in a chamber. Or, you can buy the
Radio Shack sound level meter (in the US anyway) for twenty bucks and
assume it's accurately calibrated. Or borrow a meter from someone...
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Archived from groups: rec.audio.pro (More info?)

In article <42444b8a@212.67.96.135> nospam@nowhere.com writes:

> Here's a picture

That looks like a really cheezy meter (the thing with the pointer, not
necessarily the instrument) but unfortunately the sources for such
have pretty much dried up unless you're an OEM. But looking at the
range switch, this meter won't be of any use to you for measuring
computer noise. It's not sensitive enough.



--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

Archived from groups: rec.audio.pro (More info?)

Mike Rivers wrote:
> In article <424443d3@212.67.96.135> nospam@nowhere.com writes:
>
>
>>I don't fancy paying for a new moving coil meter. If it needs me buying
>>another moving coil, I'll probably just stick with it for qualitative
>>purposes.
>
>
> For less than a replacement meter, you can buy a Radio Shack sound
> level meter which, unless the one you have goes down to really low
> levels) will be find for any audio survey work. I think they've
> stopped selling the analog meters. Look for one on eBay. Should be $20
> or so.

I got the 100uA meter free. So far, the sound level meter and repair it
has cost me a grand total of zero.

> This is really difficult to measure. I know it's damn annoying, but
> it's really not all that loud.

It's showing 53dBA on the scale with the meter a couple of metres from
the computer. That is too much for the low end of the scale.

> You'll need a very silent room and a
> meter that's more sensitive than the Radio Shack. Does yours have a
> scale that goes down to 20 dB SPL? You also need to know the frequency
> spectrum of the noise in order to get a sense of what kind of "fixing"
> you're doing - assuming you're trying to improve your noise situation.


> You might do better with a microphone, a nice quiet preamp, and a
> computer equipped with a spectrum analyzer program.

Strange you say that. I found an electret mic and was going to put it
together with an op-amp and build my own sound meter, but I threw it in
the bin when I got this meter. I borrowed it from work, but found it was
damaged. They said at work to throw it out, so I thought I'd give
repairing it a go. So that is what I did. Took me half an hour or so.

Archived from groups: rec.audio.pro (More info?)

In article <4244caa0@212.67.96.135>, Dave <nospam@nowhere.com> wrote:
>
>The computer is a Sun Ultra 80 workstation, with 10,000 rpm disks, and
>two external 10,000 rpm disks. I think the machine is rated for use to
>about 50 deg C, and 10,000' above sea level, so there is a lot of
>cooling air. It has 4 CPUs and 4GB RAM - not a typical home PC.

I bet that you find most of the noise is drive noise. I have an Ultra 10
and a Blade 1000 here, and both are pretty quiet with the drives removed.
(REALLY quiet compared with the 1U Linux servers). I think this is just
because they have so much room for airflow inside. Most of the noise they
do produce is low frequency stuff... the whine from the drives is much more
annoying.

You _might_ be able to find some commercial external SCSI cabinets that
will cut the drive noise. You could take all four drives and put them on
the external buss if you aren't desperate for I/O bandwidth.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Archived from groups: rec.audio.pro (More info?)

On Sat, 26 Mar 2005 02:54:46 +0000, Dave <nospam@nowhere.com> wrote:

>Scott Dorsey wrote:
>>>There can't be too many companies with the name Castle, specilising in
>>>sound meters in Scarborough.
>>
>>
>> Give them a call and ask if they have parts.
>>
>> To calibrate the thing, you either need a sound source with a known level,
>> or a known-accurate meter. You can build a source with a known level that
>> is entirely mechanical; it used to be common to modify model airplane engines
>> to produce a calibrated sound pressure in a chamber.
>
>You are making me think there. I have several fans that are from
>professional companies, with data on noise level. If I know a fan
>produces x dBA in free air, and I did this in the garden, I wonder if
>that would work? It's not going to be to the accuracy of the National
>Physical Laboratory, but outside must be a pretty good representation of
>an anechoic chamber, as there will be little room for reflections.
>
>I have a few fans around that I could find out noise data on. I might be
>able to get some idea of the accuary if a few were compared.

Does the data say at what distance from the fan (and which
direction) the noise was measured?

>> Or, you can buy the
>> Radio Shack sound level meter (in the US anyway) for twenty bucks and
>> assume it's accurately calibrated. Or borrow a meter from someone...
>
>Radio Shack are not here in the UK any more. It used to be Tandy, but
>they don't exist any more either.

You could perhaps order online, I suppose they'll ship it over the
pond, thought the current price of USD $39.99 isn't quite as cheap as
the 'twenty bucks' it used to be:

http://www.radioshack.com/product. [...] id=33-4050

This looks like a different product from a few years ago which was
in a much more rectangular shaped case, but the operational features
appear to be the same - it's probably the same electronics, though
perhaps made with SMT.

>Maplin is an option, and I know they have a meter, but it is around $100
> if converted to US Dollars, so not exactly cheap.

-----
http://mindspring.com/~benbradley