What causes DC offset problems?

  • Thread starter Thread starter NL5
  • Start date Start date
NL5

NL5

Unpossible!
I'm getting pretty bad dc offset on a track that someone else recorded. Why would this be?

Thanks,

NL5
 
it can be a variety of factors. a common one is bad sound card electronics
often. eg...consumer level sound cards.
before buying a sound card , record 10 seconds using it of blank audio then look at the dc offset figure. good sound cards have negligible dc offset.
peace.
 
manning1 said:
it can be a variety of factors. a common one is bad sound card electronics
often. eg...consumer level sound cards.
before buying a sound card , record 10 seconds using it of blank audio then look at the dc offset figure. good sound cards have negligible dc offset.
peace.


Well, it was originally recorded on one of the roland "all-in-one" hd recorders. Should be decent quality I would think. I ran it thru my analog mixer board, and then re-converted it thru my Motu 828mkII, which should be even better quality. Or, am I wrong on that?

Thanks,

NL5
 
I had the same problem trasfering stuff from a vs-2480. Those things are great sketch pads, but they really are kind of crappy.
 
Honestly, you are kidding yourself if you think that Roland or Korg offer anything professional regarding recording technology. Those companies bread and butter are synths... everything else they make is kind of crappy.

In fact, neither company has made a good synth in awhile... last good Roland was the JD800 or *maybe* JV series, last good Korg was the Z1.
 
Korg N1 isn't bad. Weighted keys and all that, but decent sounds. Anyway, to answer the question, there are a lot of possible causes for a DC offset. My guess is a poorly-grounded setup.

My favorite electrical problem was in a small home TV studio I used to run. The Cable TV line into the back of a VCR had a constant AC current going up the shield. Slight hum bars in the video, 60 Hz hum in the audio, and you'd get shocked when you touched anything... and that was after we tied shield from various parts of the system to a solid ground with about a 10 AWG solid copper wire....

Anyway, should be easy to fix. There's lots of software out there to remove a DC offset. It's nothing more than a simple subtraction from (or addition to) each sample.
 
Cloneboy Studio said:
Honestly, you are kidding yourself if you think that Roland or Korg offer anything professional regarding recording technology. Those companies bread and butter are synths... everything else they make is kind of crappy.

In fact, neither company has made a good synth in awhile... last good Roland was the JD800 or *maybe* JV series, last good Korg was the Z1.

Korg makes the best damn tuner on earth!!! And the Karma rocks (Especially with the MOSS board of Z1 fame).
 
...but now to get back on topic.

DC offset can also occur if you slice up audio in a digital editor, without thinking about the phase. Vibrations (audio) always have a positive and a negative part. The only good place to slice digital audio is at the exact point where the vibration is in the exact middle of the wave. Think of it as water, if you drop a stone into a lake, a wave is produced. That wave gets weaker and weaker, untill it dies out, but it can never be stopped completely at one single moment. With digital audio, it's the same. If you try to stop it without using a fade in/out, you are doing something that is not possible in the acoustical world. Therefor, you get an artifact, or a "click".

What I always do is use very fast fades (of a couple of milliseconds or less). That way, you can't really here the ending and beginning of an audio file, but you don't have any clicks either.
 
Halion, what you're talking about is a sudden square jump in a waveform, which is different from a DC offset. A DC offset is caused by DC (constant) voltage coming into the converters.

The thing about DC offset it that it is difficult to measure unless you have a section of the song which is ABSOLUTELY silent (i.e. no mics turned on, nothing). When you ask software to calculate DC offset from a varying waveform (music) it calculates the average value, which is usually zero or close to it. But a smallish reported DC offset might just be caused by the particular nature of the waveform you've recorded. Most songs will report some amount of offset when analyzed even though there isn't any actual DC voltage in there.

I should also note that the only side-effect of a DC offset is reduced headroom.
 
ive encountered dc offset due to poor grounding, usually cheap cables not enough insulation of the current. Power cables running parallel with audio cables will give a horrible hum. If they must be next to each other try to make them cross perpendicularly so that only in cross in one small section.
 
again...

DC offset is also a seperate issue from 60-cycle or other hum. Those are AC signals. DC is inaudible except when it is initiated or stopped. Like turning on a house mixer AFTER the main speakers will usually give a short burst of DC casuing a booming sound (though this is not really "true" DC due to its very finite time length). This is bad for speakers BTW.
 
The culprit is most likely the analog mixing board, but could be the motu. An analog mixing board should not pass DC, so if there is a DC offset in the original recording, it should have dissapeared at your board. Most analog gear filters DC at the input and output stages.
 
Back
Top