Help with me setting up my Focusrite to measure a headamp please

[Nkam]

I have a question about measuring a headphone amp with my Focusrite Scarlet 2i2

When I plug into one of the channels from the out put of the headamp do I set it on ‘ Line’ or ‘ Inst’?
I’m assuming ‘ Line’

Also where do I put the Direct Monitor volume dial?
And the large Monitor dial ?

I use the line outs on the back to connect into the headphone amp and use the Scarlet to generate the 1khz tone from REW.

I’ve used both Line and Inst , but get different results.


Thanks in advance !

 

[AnalogSteph]

I have a question about measuring a headphone amp with my Focusrite Scarlet 2i2

I hope you realize that plenty of headphone amps can outperform a 2i2 (which generation?).

When I plug into one of the channels from the out put of the headamp do I set it on ‘ Line’ or ‘ Inst’?
I’m assuming ‘ Line’

Yep. The instrument buffer is not generally doing the distortion performance any favors.

Also where do I put the Direct Monitor volume dial?

You don't want to do any direct monitoring, so... all the way to Playback, I guess?

And the large Monitor dial ?

That's what controls your output level.

You'll probably need to activate the input pad to get a decent handle on distortion performance, and actually further external attenuation may be required.

Your best bet at determining output residual noise level is adapting to XLR (3.5 mm or 1/4" stereo to 2x XLRm cables are available off-the-shelf, but 1/4" jack to XLRm cables should also exist). The main hurdle is establishing an absolute level reference. I'd suggest asking the multimeter with a sine at -12 to -6 dBFS and then turning the REW generator level way down to like -60 dBFS or something, low enough that you can crank input gain almost all the way up. Note down your measured output and generator level, how far you turned down the generator level and at which input level the sine ends up.
Random made-up example:
Generator -6 dBFS = 2.00 Vrms out.
Generator -60 dBFS = 2.00 Vrms + 6 dB - 60 dB = +6 dBV - 54 dB = -48 dBV
Input sine level at -60 dBFS generator level = -7 dBFS
--> 0 dBFS = -48 dBV + 7 dB = -41 dBV ~= -38.8 dBu.
At this point you can convert amplifier output noise floor as seen in dBFS into absolute dBV or dBu (and thus µV). Take note of measurement bandwidth.
I am making the assumption that headphone amp output impedance is negligible vs. both the ~3 kOhm mic input impedance and the multimeter's megohms worth. Which seems reasonable enough.

You may encounter ground loop issues when doing unbalanced loopback, in which case custom adapter cables would be required to eliminate those. Basically, connect unbalanced signal to XLR pin 2 and ground to XLR pin 3, leave 1 open. (Same for TRS input: Signal = tip, ground = ring, sleeve left open.)

 

[Nkam]

I hope you realize that plenty of headphone amps can outperform a 2i2 (which generation?).

Yep. The instrument buffer is not generally doing the distortion performance any favors.

You don't want to do any direct monitoring, so... all the way to Playback, I guess?

That's what controls your output level.

You'll probably need to activate the input pad to get a decent handle on distortion performance, and actually further external attenuation may be required.

Your best bet at determining output residual noise level is adapting to XLR (3.5 mm or 1/4" stereo to 2x XLRm cables are available off-the-shelf, but 1/4" jack to XLRm cables should also exist). The main hurdle is establishing an absolute level reference. I'd suggest asking the multimeter with a sine at -12 to -6 dBFS and then turning the REW generator level way down to like -60 dBFS or something, low enough that you can crank input gain almost all the way up. Note down your measured output and generator level, how far you turned down the generator level and at which input level the sine ends up.
Random made-up example:
Generator -6 dBFS = 2.00 Vrms out.
Generator -60 dBFS = 2.00 Vrms + 6 dB - 60 dB = +6 dBV - 54 dB = -48 dBV
Input sine level at -60 dBFS generator level = -7 dBFS
--> 0 dBFS = -48 dBV + 7 dB = -41 dBV ~= -38.8 dBu.
At this point you can convert amplifier output noise floor as seen in dBFS into absolute dBV or dBu (and thus µV). Take note of measurement bandwidth.
I am making the assumption that headphone amp output impedance is negligible vs. both the ~3 kOhm mic input impedance and the multimeter's megohms worth. Which seems reasonable enough.

You may encounter ground loop issues when doing unbalanced loopback, in which case custom adapter cables would be required to eliminate those. Basically, connect unbalanced signal to XLR pin 2 and ground to XLR pin 3, leave 1 open. (Same for TRS input: Signal = tip, ground = ring, sleeve left open.)

Yeah I figured out the ‘solution’ is not to use the Focusrite for both sending the 1khz signal and measuring the incoming signal.
That get the rest of the monitoring knobs out of the equation.

So I used my other DAC. a Benchmark DAC3 to send the signal and the Focusrite to do the measuring. The AD part.
It’s way better.
But still not hardly the best since the Focusrite is pretty bad specs wise as you mentioned.

 

[Nkam]

I hope you realize that plenty of headphone amps can outperform a 2i2 (which generation?).

Yep. The instrument buffer is not generally doing the distortion performance any favors.

You don't want to do any direct monitoring, so... all the way to Playback, I guess?

That's what controls your output level.

You'll probably need to activate the input pad to get a decent handle on distortion performance, and actually further external attenuation may be required.

Your best bet at determining output residual noise level is adapting to XLR (3.5 mm or 1/4" stereo to 2x XLRm cables are available off-the-shelf, but 1/4" jack to XLRm cables should also exist). The main hurdle is establishing an absolute level reference. I'd suggest asking the multimeter with a sine at -12 to -6 dBFS and then turning the REW generator level way down to like -60 dBFS or something, low enough that you can crank input gain almost all the way up. Note down your measured output and generator level, how far you turned down the generator level and at which input level the sine ends up.
Random made-up example:
Generator -6 dBFS = 2.00 Vrms out.
Generator -60 dBFS = 2.00 Vrms + 6 dB - 60 dB = +6 dBV - 54 dB = -48 dBV
Input sine level at -60 dBFS generator level = -7 dBFS
--> 0 dBFS = -48 dBV + 7 dB = -41 dBV ~= -38.8 dBu.
At this point you can convert amplifier output noise floor as seen in dBFS into absolute dBV or dBu (and thus µV). Take note of measurement bandwidth.
I am making the assumption that headphone amp output impedance is negligible vs. both the ~3 kOhm mic input impedance and the multimeter's megohms worth. Which seems reasonable enough.

You may encounter ground loop issues when doing unbalanced loopback, in which case custom adapter cables would be required to eliminate those. Basically, connect unbalanced signal to XLR pin 2 and ground to XLR pin 3, leave 1 open. (Same for TRS input: Signal = tip, ground = ring, sleeve left open.)

I did get this measurement from my Zen Can with the refined method mentioned.

I’m sure using something like an RME ADI Black Edition would yield much better results.


Thanks so so much !

Oh another thing which is odd. I made a cable with two 250ohm resistors to provide a load and used another one with no load and there were no differences.
That I don’t understand. If you know what’s going on with that

I does kinda match with the official Zen Can specs below though