Page 1 of 1

Question regarding noise floor

PostPosted: Sun Sep 09, 2012 6:40 am
by xaumpl
I just built my first four modules (T15, D11, A12 and N72), and let me just say
Wow. It was so much fun, and so educational. Let me add my voice to the chorus
of adoration for SCA.

My only prior experience with mic preamps has been in audio interfaces and cheap
mixers. Which is to say I know nothing about how good preamps are supposed to
behave or sound.

So, here's my question: Is the noise floor on these pres supposed to be
significantly lower than on pres found in cheap mixers and audio interfaces?
With the N72 set at 55 dB of gain there is a very audible hiss of about -61 dBu.
With one of my Fireface's pres set at the same 55 dB of gain, the noise floor
sits at about -58 dBu. So the N72 is only about 3 dB lower. And I'm seeing
similar numbers from the T15 and the A12.

I've eliminated variables to the best of my ability. The room is dead silent.
I've swapped out all the mic cables; tried different mics; plugged the chassis
into an outlet on a separate circuit; plugged each module into the power supply
individually -- all with identical results. Each module was assembled very
carefully and cleanly, and passed all the tests.

Maybe my expectations are unrealistic, and this is a totally normal amount of
noise. Like I said, I know almost nothing about good pres. Any info would be
appreciated.

PostPosted: Sun Sep 09, 2012 10:39 pm
by xaumpl
Maybe a better way to put it is: Can someone just confirm that at 55 dB of gain
a hissing noise around -63 dBu / -73 dBu RMS (+-5 dB) is to be expected, so I
know there's nothing wrong with this unit?

PostPosted: Mon Sep 10, 2012 11:12 am
by tpryan

PostPosted: Mon Sep 10, 2012 11:49 am
by xaumpl
Wow, ok, that makes perfect sense. Looks like these units are operating totally
to spec. Thanks!

One small question, though: I'm probably missing something obvious, but how do I
take the EIN numbers on a module's spec sheet and arrive at something like "-130
dBu"?

PostPosted: Mon Sep 10, 2012 12:05 pm
by xaumpl
Heh, all the sudden I'm really impressed that (if I understand this correctly)
these modules are operating close to the thermal noise limit of a 150 ohm input
resistor. As in, it doesn't get any quieter.