E1DA Cosmos ADC and ATOM AMP 2 as a buffer for DAC measurements?

[salmo]

Hi!

I found that the day has finally come to learn a little more about taking measurements. I managed to acquire a Cosmos ADC grade B. While I don't have a Cosmos Scaler, I thought that maybe it wouldn't be such a bad option to use the Atom Amp 2 amplifier as a buffer - its performance is very good, input impedance is 10kΩ , and the output impedance ~0.7Ω.

Thus, for the moment I came up with such a measurement chain:
(USB 1) laptop on battery -> ADC
(USB 2) laptop on battery -> USB isolator -> DAC (L) -> ATOM AMP 2 (L) -> ADC (L)

I connect only the L channel from the DAC to the AMP, and then only the L signal to L input on the ADC. The ADC itself works in stereo mode. I set DIP switches at the bottom to non standard setting: between 1,7 and 2,7V. I've made myself an RCA-XLR cable according to the guidelines, because I'm also only measuring unbalanced DACs. I generate a 0dBFS signal in REW, and then set the gain on the ATOM AMP for the final of -0.50dBFS for measurements. I'd like to ask if you could take a look at these screenshots, as it might be better to adjust the FFT settings a little bit. I also know that I shouldn't compare 1:1 results with those from AP by amirm

First, the results for the Topping DX1 and the SMSL SU-1. I guess the DX1 come out nicely, the SU-1 a little worse than I would have expected, but there are differences between the copies too so it could be the case here:

DX1

SU-1

I also measured the ifi ZEN DAC (first version). Here - on the other hand - the results seem too good to be true?

This DAC was modified by me - I replaced the capacitors in the power section with better ones (kemet), and in addition also in the audio section with Elna Silmic. It is possible that such treatment lowered the harmonics a bit? SINAD of -95.4dB looks suspiciously good here. Although, on the other hand, looking at the specification [<0.0015% @ 0dBFS (UnBAL/BAL)] maybe it came out quite reasonable. So here I am asking for tips for a very beginner in the world of measurement. I feel this is going to be a fun adventure!

 

[Pow1]

Random question:

If the DAC has a volume control, like a few that you've measured, do you control the volume of both the Atom amp and the DUT to get the final result?

Also, how do you know how much of the Atom's volume to use and how much of the DUT's volume to use? For example, I could set my DUT volume to max and then start adjusting Atom amp from there to reach the gain level I want. Or I can start my DUT volume at zero and the Atom's volume close to max and then slowly increase the DUT's volume. I'm not very experienced with measurements but I have an inkling that both of these scenarios may give different results. Or do you just play around with settings until you get the maximum SINAD possible?

If the end goal is just to use as a buffer for the input and output impedance, maybe there might be a way to set the atom at unity gain and then just adjust volume using the DAC and the sensitivity settings on the ADCiso. An alternative would be to use the Cosmos Scaler which has an autoranging function I believe.