On 20/11/2009, at 9:21 AM, Eric Benjamin wrote:
> Uh oh! Someone used my name
>
> The basis for my original statement of 8 dB self noise for an AT3032 was that
> I had measured one as having 8 dB self noise. I have very good facilities
> for measuring microphone self noise and I can generally do a pretty accurate
> job down to about 5 dB or so. Below that, the impact of environmental noise
> may limit my ability to measure the true self noise of the device. Note that
> the usual sort of errors work in the direction of causing the measurement
> result to be high, not low.
>
> Could I have been in error in my measurement? Of course I could! But I
> don't think so. I was so surprised by the result that I did the measurements
> over again, and then I did a listening test to confirm. As it happens, the
> Schoeps Mk2 microphones are very consistent in terms of their self noise. I
> have measured three of them and they were all within a fraction of a dB of
> the specified 11 dBA. The reason for this is that a particular resistor in
> the circuit limits the self noise, not the FET. And the AT3032 was
> noticeably quieter than the Schoeps Mk2.
>
> Could other samples of the AT3032 have different self noise? Of course. I
> haven't analyzed the AT3032 circuit so I don't know what limits the
> performance. It could be pretty variable. That could explain why AT chose
> to rate the microphone as having 16 dBA. I dunno. I only measured the one
> sample. I do have access to about a half dozen of these microphones so I
> could sit down and measure the whole crop.
>
> But that doesn't answer the question about the AT4022. Lack of funds
> prohibits my purchasing one just out of curiosity. But if there is someone
> in the SF bay area who has one and would like to have it measured, I would be
> happy to make the comparison.
>
> Eric
>
Hi Eric,
Sorry to use your name in vain, and thanks for taking the time to respond and
clarify your tests.
I have three of the AT3032 two purchased from a US online retailer and one old
stock from an Australian music retailer. I'm currently using a the "old
stock" mike and one of the newer production mics as a pair because they are a
fairly close match. The second newer production mike was substantially
mis-matched to the other two and I was having to adjust gain to L-9 R-9.25 on a
scale of 10 get a reasonable match on levels. The "odd" mic also seemed to
have a tonal balance that was tilted in respect to the other two. I think this
points to a reasonable level of sample to sample variability between the
AT3032's so I wouldn't discount the possibility that your 8dB sample was an
"outlier". If you had the time and interest I'd be more than happy to send my
3 AT3032 to you to test.
cheers
Paul
|