• T0N
  • New member
  • Offline
  • Registered: 2023-05-11
  • Posts: 4

Topic: Strange DC on ADI 2/4

Hello,

I finally decided to go with the ADI 2/4, which has just arrived. So far I’ve only used it as a basic DAC and headphone amp. Which it does extremely well. But it caught my eye, that the display’s meter for the analog inputs shows a ~10 dB difference between L/R. Also when nothing is connected but power.

So I hooked it to my computer and took a look in DigiCheck NG, the bitstatistics and noise module. I’m posting the result below taken at each sensitivity step.

When I switch on the analog input’s dc filter (rme) all bit states are green. DC is reduced, yet still jumping quickly in levels between -108 dB to -144dB/UFL.

I have tried cutting the power to everything else in my apartment. Tried 3 different outlets as well, but it hasn’t made a difference.

Could this be sign of a faulty unit?
If the source is outside the unit, it would seem strange that one channel is more affected than the other?
Also that they don’t follow same pattern as they jump in levels?

Please let me know what you think.


Taken at +24 +19 +13
https://abload.de/img/241cfnz.png
https://abload.de/img/19wmdzu.png
https://abload.de/img/13toc76.png

  • T0N
  • New member
  • Offline
  • Registered: 2023-05-11
  • Posts: 4

Re: Strange DC on ADI 2/4

Last two screendumps at +7 and +1
https://abload.de/img/7bzewu.png
https://abload.de/img/1wfeiz.png

3 Reply by KaiS (edited by KaiS 2023-05-14 01:00:58)

  • KaiS
  • DAW Mastermind
  • Offline
  • From: Germany
  • Registered: 2020-02-05
  • Posts: 3,525

Re: Strange DC on ADI 2/4

Measuring an open, unterminated ¡nput is invalid.
Repeat your measurements with 200 Ohms termination, or shorted input.

We just had this here (in German language):
https://forum.rme-audio.de/viewtopic.ph … 97#p202097


Even a resitor has a thermic self noise, this would be the lower physical limit of any amp.
ADI-2/4 Pro SE open input = 90,000 input impedance

Resistor self noise 90,000 Ohm -103 dBu.
Resistor self noise 200 Ohm -130 dBu.
https://www.beis.de/Elektronik/Nomogram … Noise.html


The following is ADI-2/4 Pre FX Meter readings.
The meter shows Peak Unweighted measurements.
Noise usually is measured RMS A-Weighted, RMS A / dBA RMS, giving ca. 12 dB lower results.

  • FS=44.1 kHz
    open input
    Ref        Meter
    Level     PreFX, ca.
    +1dB    -79dB
    +7dB    -85dB
    +13dB  -91dB
    +19dB  -97dB
    +24dB -102dB

  • FS=44.1 kHz
    Termination 200 Ohm
    Ref        Meter
    Level     PreFX, ca.
    +1dB    -108dB
    +7dB    -111dB
    +13dB  -111dB
    +19dB  -111dB
    +24dB  -111dB

KaiS's Website

4 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 25,586

Re: Strange DC on ADI 2/4

I think this post is mainly about DC, which is unavoidable and no issue at these low numbers. And there is zero point in switching the DC filter off (default on), as done here, unless you are doing scientific experiments or dedicated  examinations of certain things.

Regards
Matthias Carstens
RME

MC's Website

5 Reply by T0N (edited by T0N 2023-05-14 17:12:26)

  • T0N
  • New member
  • Offline
  • Registered: 2023-05-11
  • Posts: 4

Re: Strange DC on ADI 2/4

Hello guys,

Thanks a lot for the response and for calming my nerves! I'm glad to hear none of this is sign of a faulty item.

So the DC filter should basically always be on for my (musical) use, check.

The device's peak meter readings for noise with open input (DC filter on) matches the ones listed. If not slightly better.

KaiS wrote:

Measuring an open, unterminated ¡nput is invalid.
Repeat your measurements with 200 Ohms termination, or shorted input.

Just for future reference, this means shorting (or placing a resistor between) pin 2 and pin 3 of a XLR plug?
Would this eliminate any measurable DC?

6 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 25,586

Re: Strange DC on ADI 2/4

No. The DC is a function of the analog input stage (the inherent error of the opamps, and partly even the ADC itself), and can therefore not be completely zero even with adjustment (which the ADI has, via trimpots) - unless you filter it in the digital domain.

Regards
Matthias Carstens
RME

MC's Website

  • T0N
  • New member
  • Offline
  • Registered: 2023-05-11
  • Posts: 4

Re: Strange DC on ADI 2/4

I see, cool.

Thanks again for the detailed response.