Topic: Signal pumping at fs/4 using two ADI-2 Pro FS & two PC's

Testing some jitter related using two ADI-2 Pro FS with to PC's.

Issue: Signal pumping on the ADC input level display using signal exactly at fs/4.

. Using latest firmware and swapped the DAC ADI and ADC ADI gear and getting equal results. 

. The DAC / ADC is set at 19dBu. Int Clock source.

. The source signal is at -6dBFS

. SR on both gear with 44.1 kHz or 48kHz

. Connect both with a symmetric cable

. Feed out the signal and input of the ADC input level alters from -6.1 dB to -8.2dB (alters/pumping) within some seconds

. Recorded signal shows equal level pumping too

BUT: As soon I set the signal frequency different to fs/4 the pumping stops.

Is this know?

2 (edited by hpw 2020-03-23 12:04:51)

Re: Signal pumping at fs/4 using two ADI-2 Pro FS & two PC's

Also same pumping issue using BF out and ADI-2 Pro in with sine fs/4 and 44.1khz SR..

Wave file picture ...

https://1drv.ms/u/s!AnxOfA6VFTtxiwD7gF_ … Q?e=e345vO

3 (edited by KaiS 2020-03-23 15:15:04)

Re: Signal pumping at fs/4 using two ADI-2 Pro FS & two PC's

To me this looks like the typical aliasing you get with the sample rate of a lower resolution wave form display.
Comparable to the sometimes backwards turning wheels in Western films.

If I am right there is no audible effect, your signal is audibly stable.


I can reproduce this purely graphical effect by simply playing an 11025 Hz sinewave to the ADI-2 Pro from my iPhone via USB.
Display is pumping, but audibly the signal is stable.

This even means, display uses a fraction of the audio samplerate, but with a separate, non-syncronized clock generator — interesting.


To double check, if such a level variation is audible, I put out a mix of two sinewaves, 11025 Hz at -6dB and 11024 Hz at -30dB.
By using the "beat" effect this signal produces a real level variation comparable to the one displayed:
Clearly audible to me.

4

Re: Signal pumping at fs/4 using two ADI-2 Pro FS & two PC's

hpw wrote:

Testing some jitter related using two ADI-2 Pro FS with to PC's.

Issue: Signal pumping on the ADC input level display using signal exactly at fs/4.

. Using latest firmware and swapped the DAC ADI and ADC ADI gear and getting equal results. 

. The DAC / ADC is set at 19dBu. Int Clock source.

. The source signal is at -6dBFS

. SR on both gear with 44.1 kHz or 48kHz

. Connect both with a symmetric cable

. Feed out the signal and input of the ADC input level alters from -6.1 dB to -8.2dB (alters/pumping) within some seconds

. Recorded signal shows equal level pumping too

BUT: As soon I set the signal frequency different to fs/4 the pumping stops.

Is this know?

Yes, and it is no problem. The 'pumping' is a visual artefact, not the real signal and not something audible. Change the waveform display to samples and check their value and position.

Regards
Matthias Carstens
RME

5 (edited by hpw 2020-03-24 13:04:34)

Re: Signal pumping at fs/4 using two ADI-2 Pro FS & two PC's

While using DAC & ADC with equal SR using two independent (not synced) master clock & gear, what means, the master clock difference is may in xxx ppm regions.

IMHO, as pumping in 2.7 seconds, means within this period, a sample over or under run happen. So we could calculate the clock difference... big_smile

6 (edited by KaiS 2020-03-24 17:48:43)

Re: Signal pumping at fs/4 using two ADI-2 Pro FS & two PC's

hpw wrote:

While using DAC & ADC with equal SR using two independent (not synced) master clock & gear, what means, the master clock difference is may in xxx ppm regions.

IMHO, as pumping in 2.7 seconds, means within this period, a sample over or under run happen. So we could calculate the clock difference... big_smile

If you refer to the level display's pumping, it's one period of the 1/4fs sinewave used for testing (12kHz @ 48kHz samplerate).
This equals to 4 samples per 2.7s in your example.

But as this only happens in the display, and the audio itself is not pumping, it's of minor interest.


You don't even need two ADI-2s to see the pumping display, different internal audio clocks of the two ADI-2s is not the reason for this effect.

I want to stress: Audio is not pumping.