SCA N72 Questions (I just built one)

[armanbohn]

Ok - I'm not trying to be a paranoid noise freak, but I just built my first N72 and testing it with a 150 ohm resistor from pin 2 to 3 I get a noise floor of about -85db. This seems a little higher than my onboard pres on my Black Lion MOTU.

This is not a problem for me, I just want to know if I built it right.

Can someone give me a reference on the noise floor of their N72?

[b3groover]

Did you post about this on the Seventh Circle board?

[armanbohn]

yeah - it's been on there for a few days with no responses. I've fished around and can't seem to come up with any concrete answers.

I think I'll take the pre down to the lab and throw it on the scope.

[SeventhCircle]

I get a noise floor of about -85db.

This is a meaningless measurement without specifying the reference level. If it's -85 dBFS on your converter, you first need to find out what actual input voltage produces 0 dBFS. If, say, +4 dBu equals -22 dBFS (like an Apogee), then you may have a noisy N72. If -10 dBV equals -10 dBFS, then you may not have a noisy N72. At any rate, make sure the converter is set for +4 dBu levels, not -10 dBv, and make sure you're plugging into a line level input, not a mic input. Also make sure that if you're going from XLR to 1/4" TRS that pin 3 is actually connected to something.

[armanbohn]

Hey Tim,

Thanks for responding here!

I'm going into a line level input on a MOTU 828mkii that has been modded by Black Lion. It is set at +4 not -10.

The input was terminated with 150 ohm res. across pins 2 and 3. The XLR out from the pre is going to TRS 1-sleeve, 3-Ring, 2-Tip.

I;m not sure how to figure out what input voltage produces 0 db. I will probably have to go use a scope for this. I want to got do the fine biasing anyway so let me know if there is a reference test for this.

[Jim Williams]

The standard test is the equivalent input noise test, or EIN. It's a sum of the maximum gain + the S/N at maximum gain. To be honest, A weighting should not be used, just the open loop response. 20hz to 80k bandwidth is the measurement range. An analyzer may be needed to do this accurately, compensating for input load resistance as I use a 150 ohm metal film resistor across pins' 2+3 for the test. Shorting the inputs will remove the 150 ohm source input impdance so the true self noise of the preamp can be measured. Good designs will spec around -135 db EIN with a shorted input, the best a transformer based input design can do is -127 db EIN due to transformer impedances and insertion loss, typically 2.5 db.