naturerecordists
[Top] [All Lists]

Re: Sony MD Microphone preamp... Design and settings...

Subject: Re: Sony MD Microphone preamp... Design and settings...
From: Walter Knapp <>
Date: Sun, 09 Feb 2003 13:25:58 -0500
evertveldhuis  wrote:
> Hi guys,
> 
> After some searching I found this very nice document: 
> http://www.geocities.com/ferocious_1999/md/sony_preamp.htm
> 
> It deals with the designs, and in short this is what it says:
> - The preamp setting low gives a gain of 5.4 times
> - The high setting gives a gain of 50 times.
>   [own remark: In both cases this is done by letting the same signal
>   go thru the same opamp, but with 1 resistor more or less in the
>   circuit. So overdriving the opamp will happen at the same SPL no
>   matter the switch position.]
> - Models without a switch have a gain of roughly 25 times.
> 
> I haven't seriously tested the difference between the hi and low 
> setting, I will do it next week. But what is the reccommended setting?
> 
> I always thought it was best to record to about -6dB to -3dB. I try 
> having as little analog gain as possible, and having almost no 
> digital cutting (attenuation). 
> Because when you have a microphone preamp in a mixer (large studio 
> device type) it is common - recommended, I should say - to have the 
> fader at zero dB (no attenuation at the end of the chain), and have 
> the preamp setting with as little as pre-gain as possible (as little 
> gain as nessecary in the beginning of the chain)
> This instead of a high gain at the beginning together with a high 
> attenuation at the end of the chain because this way the hiss and 
> noise is way higher then the first mentioned approach...
> 
> You might know that I always use the low gain setting of my MZR55 and 
> have the digital recording level almost always at 75% to 100%.
> 
> Is this approach wrong? (judging from what I have read last week I 
> tend to think it is wrong, but why?)
> 
> Should I use more analog gain and then digitally bring down the 
> recording level? Has anybody already tested this with results that 
> could be repeated by performing the same test based onm the test 
> procedures?

One should note that the link above is from a rock concert recordist. 
And has a lot of application in that sort of recording. It's less 
relevant to nature recording.

It's been some time since I recorded regularly with a MZ-R30. But, I did 
spend the fall going through all my recordings including the MZ-R30 ones.

I found the difference between the high and low setting on the MZ-R30 in 
actual use to be about 14-15 dB. As I've noted, over 90% of my recording 
was done on the High setting. Even once I got the Telinga this continued.

It may be time for a bit of a reality check. From the link above, there 
is a table giving the maximum "safe" sound levels vs preamp headroom and 
mic sensitivity. Taking the most pessimistic view on that table (most 
sensitive mic given) we find that we are "safe" using a high setting up 
to recording a sound level of 93 dB. Or with the low setting 107 dB, a 
calculated difference of 14 dB, which agrees with what I found. For a 
sony md without the switch it looks like about 100 dB.

Now just what are these sound levels, what, out in nature recording will 
we be encountering? From this link:
http://www.abelard.org/hear/hear.htm

90 dB - Heavy Truck at 15m, Busy city street, passing motorcycle, Lawn 
mower, Loud shout, Screaming child. Very annoying. Can damage hearing 
after 8 hours exposure per day

100 dB - Jet takeoff at 500m, Train horn at 30m, Diesel truck, Pneumatic 
drill/jackhammer. Can damage hearing after 2 hours exposure per day.

105 dB - Helicopter, Motorized/power mower. Can damage hearing after 1 
hour exposure per day

110 dB - Sand-blasting, Squealing pigs, Inboard motorboat, Typical night 
club, Unsilenced motorcycle from 7m away. Maximum vocal effort. Can 
damage hearing after 30 minutes.

Now, I don't know about other nature recordists, but I tend to avoid 
such situations for recording nature. Where I record is quieter than that.

I also have a moderately good sound meter, and have often measured the 
sound levels at my recording locations. At the mic, most locations are 
well below 80 dB, the highest I've measured at the mic was 94 dB, which 
was in the middle of a very large and dense frog chorus. I've heard 
chorus's that were louder before I got the meter. The more typical 
recording site reads 50 - 60 dB or less. And that during the calls. I 
did, one night, stalk calling Squirrel Treefrogs with the meter. At 1' 
they put out 84 dB. That's a lot closer then I would record them. Most 
frogs won't let me get that close to measure. Eventually I plan on 
measuring as many species as I can.

Note, what these measurements say is that most of the time you will be 
well below the limiting safe level for the preamp. Which is why my 
experience is that the high setting is just fine most of the time.

There is one additional parameter to take into account. This is for a 
parabolic microphone or a pzm mic. For a parabolic, the reflector 
provides extra gain before the mic. This is effectively a higher sound 
level for the mic and pre. Reports vary as to how much to use, between 5 
to 20 dB increase is probably reasonable. You have to subtract that from 
the 93 or 107 dB to determine a safe level for the center axis of the 
parabolic. This is why the Telinga can require more use of the low 
switch setting. And, if close to your subject, even this may not be 
enough. In such situations, for my survey work, I'd simply aim very 
slightly to one side, dropping the sound level to avoid problems.

A pzm or boundary mic also has gain before the mic of at least 4 dB to 
account for.

My approach is to have the gain on a preamp somewhere in the middle 
third of it's range if possible. Usually the cleanest, best signal will 
come from a preamp somewhere in that middle range. There is a "sweet 
spot" which you have to find for each setup. So, when using the MZ-R30, 
I'd choose the high/low setting based on that. And as I noted, that 
resulted in almost always running on the high setting.

It's not a set in cement rule as to which setting to use. You use the 
appropriate one for the conditions you have.

I did find that when I accidentally bumped the switch to low when it 
should have been high, and then recorded, that such recordings would 
have a higher level of unwanted noise when normalized.

As far as setting a level vs the OdB (clipping) point, it is ideal to 
have that just a few dB lower. And if you are god and can predict 
exactly the loudest point you are recording you can set it exactly. The 
rest of us usually allow a bit more pad for our own lack of judgment. In 
most situations I set at -15 to -20. And that after monitoring the 
levels before recording. Digital recorders have a huge amount of dynamic 
range compared to the environment we are recording.

Once I bring my recordings into the computer, I digitally normalize them 
to -2 dB for archiving. But I will adjust that if, for instance I'm 
making a audio CD. I'll try to get it so that the listener does not have 
to continually adjust the volume control between tracks. The highest 
point for all tracks will be very close to the 0dB level, but some 
tracks may be much lower.

Walt




________________________________________________________________________
________________________________________________________________________

>From   Tue Mar  8 18:23:18 2005
Message: 5
Date: Sun, 09 Feb 2003 10:45:38 -0700
From: Doug Von Gausig <>
Subject: Re: Re:Where did I go wrong?

At 12:02 AM 2/9/2003, John Moore wrote:
>Coincidentally I was listening yesterday to some recordings I made
>in 1968 in the Sahara desert on my Uher and I couldn't find any hiss
>problems on them.....

"Ambient noise" is often nearly impossible to perceive - it's there, but we 
"tune it out" - our brains toss it out as insignificant. The mic and 
medium, though, pick it up faithfully, and play it back to us in a 
completely new setting - one in which our brains allow us to hear the 
"truth". This is very often perceived as a noisy recording system, when it 
is actually a noisy venue. There are two ways to test this. One is to 
record the sound of a very, very quiet room (as Walter suggested) - and 
those are really hard to find! The other is to make simultaneous recordings 
with the quietest mic around - (like a MKH-20) and the mic you are testing 
on the same recorder, then transfer the recording digitally, "equalize" the 
two recordings, so you aren't just hearing a difference in gain, and compare.

This "noise" problem is very common when you first start recording, and it 
gets better with experience. You can and should train yourself to listen 
for the true sound of an area. I like to monitor the mic with one earphone, 
then vary the recording level until the sound level is the same from the 
earphone as it is in your bare ear. This will give you a setting that is 
usually far quieter than where you would normally set the level using just 
earphones, but will be a more accurate representation of the actual sound 
level in the environment. You may learn that you are setting the level for 
your recordings unnaturally high, thereby getting abnormally high ambient 
noise. Try it.

Doug
Doug Von Gausig
Clarkdale, Arizona, USA
Moderator
Nature Recordists e-mail group
http://groups.yahoo.com/group/naturerecordists



[Non-text portions of this message have been removed]



________________________________________________________________________
________________________________________________________________________

<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU