• Registered: 2024-10-11
  • Posts: 3

Topic: Get DC square pulse signal into mic of sound card

Hello

I have a Fireface UFX II and acquire input from 4 microphones at 192 kHz. In addition, I use a 5th input channel in parallel to acquire a DC square pulse signal of 5V (it is a single square pulse, high state lasts 100 msec).
The square pulse is always out of sync with respect to the 4 audio channels, by roughly 30 msec. The offset changes every time (+/- 5 msec) and as a function of the buffer size of the sound card (smaller offset with higher buffer size).

I am aware that sound cards are not meant to deal with DC signals, but I would like to know if this out-of-sync issue is expected and if there is a way to fix it.

Thanks for any feedback!

2 Reply by unpluggged (edited by unpluggged 2024-10-12 01:41:51)

  • Registered: 2023-06-30
  • Posts: 161

Re: Get DC square pulse signal into mic of sound card

chiomatic wrote:

DC square pulse signal of 5V (it is a single square pulse, high state lasts 100 msec)

Excuse me, but this description makes no sense. So is it DC, or is it square waveform with T=200 ms, or is it a single pulse? What is the exact waveform of your signal, an what purpose does it serve in your recording setup?

Are you aware that square waveform signals (and so are single short pulses) are broadband and can exhibit phase distortions when digitized by an audio ADC?

Fireface UCX II + ARC USB > ADI-2 Pro FS R BE > Neumann KH 750 DSP + MA 1 > KH 120 A

3 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 25,599

Re: Get DC square pulse signal into mic of sound card

Adding a fifth channel from the UCX II can not cause such effecfts. All its ADCs are in sync. So I wonder if this 'description' talks about another unit added...

Regards
Matthias Carstens
RME

MC's Website

  • Registered: 2024-10-11
  • Posts: 3

Re: Get DC square pulse signal into mic of sound card

I apologize if something I wrote is not clear or wrong. And thanks for your feedback!

The signal on the 1st to 4th channels is audio signal recorded from 4 microphones.
The signal on the 5th channel is a single square pulse that goes from 0V to 5V, stays high for 100 msec, and goes back to 0V "forever".

I am aware that the square pulse can be distorted, and in fact it is, you can see the recorded pulse here https://pasteboard.co/xgA2XdRQyVbt.jpg. However, this is not a problem for me, because I am only care for when the onset of this pulse occurs. So I am fine as long as the pulse onset is reliable, i.e. the "true" onset does not change its timing with respect to the audio signal in channels 1-4 as it goes through the sound card, and I can easily detect the onset.

The purpose of this square pulse is to be able to synchronize (align in time) the audio signal of channels 1-4 with another stream of data (electrophysiological data) acquired by a separate device that has a different "clock" than the sound card. I do this time alignment after the recording is done, I do not sync the signals online.
My domain is not music... I am a neuroscientist working in a lab :-) The 4 microphones record sounds and vocalizations produced by a mouse, and the other stream of data is brain signal acquired in parallel that I need to align with the audio signal.

Thanks again for your help!

  • Registered: 2018-12-21
  • Posts: 322

Re: Get DC square pulse signal into mic of sound card

Interesting use case,
But what’s more helpful to know would be which other hardware you are using?
The UCX has only 2 mic inputs, so are any other microphone preamplifiers involved, and most importantly any other Analog-to-Digital converters? Are you using the AES, Adat or SPDIF inputs of the UCX?
If yes, is everything following the same clock?

  • Registered: 2024-10-11
  • Posts: 3

Re: Get DC square pulse signal into mic of sound card

I am using a UFX II https://www.rme-usa.com/fireface-ufx-ii.html.

The 4 microphones are plugged into the four AES input ports on the front panel via XLR connectors. These corresponds to input # 9 to 12. I am 100% sure that these work fine.

The 5th channel with the square pulse is plugged into input #7 on the rear panel. This signal is a 0V or 5V voltage fed directly into the sound card, without any other Analog-to-Digital converter.


If yes, is everything following the same clock?

I am not sure about your question. Can the 4 mic channels and input #7 follow different clocks?

7 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 25,599

Re: Get DC square pulse signal into mic of sound card

chiomatic wrote:

I am using a UFX II https://www.rme-usa.com/fireface-ufx-ii.html.

The 4 microphones are plugged into the four AES input ports on the front panel via XLR connectors.

These are Mic/Line inputs, not (digital) AES inputs.

chiomatic wrote:

The 5th channel with the square pulse is plugged into input #7 on the rear panel. This signal is a 0V or 5V voltage fed directly into the sound card, without any other Analog-to-Digital converter.

All analog input channels are in total sync. Unless you use FX, EQ etc on one channel only, or use insert FX in the record software.

Or you don't use ASIO but some unusual WDM based record software that does not adhere to industry standards when it comes to channel syncing. That would explain your varying offsets.

Regards
Matthias Carstens
RME

MC's Website