What are the implications of turning off DC filtering on the inputs?

[Nickerz]

I saw that in the ADC measurement tests, there is a significant increase when one disables DC blocking either through modification or through firmware changes. Now I was under the impression this was to prevent accidentally sending CV or some other weird situation where you accidentally plug in the wrong thing into your interface. But appears this could be introduced with bad mastering or sound files that are too loud?

A user on another forum got a "DC voltage detected" on their RME interface when playing a music file from Apple Music.

I was planning on having an RME FS BE handling my ADC duties and having it set off. I run real analog gear into my interface, but this is all legit studio stuff. e.g. Brand name mic pres (not knockoff stuff), DIs, compressors etc.

What am I getting myself into running this without DC blocking? And how is this relevant in day to day operation?

 

[Scrappy]

Ha when I was young, I built a lil amp from a cash box and piece of a HT in a box receiver circuit board, with a red stripe cap for a knob (oh the college days..) Anyhow, i bypassed the DC caps for some ungodly reason. Not good. Huge popping sounds. FWIW amp board was isolated from chassis...

 

[Ken Tajalli]

I saw that in the ADC measurement tests, there is a significant increase when one disables DC blocking either through modification or through firmware changes. Now I was under the impression this was to prevent accidentally sending CV or some other weird situation where you accidentally plug in the wrong thing into your interface. But appears this could be introduced with bad mastering or sound files that are too loud?

A user on another forum got a "DC voltage detected" on their RME interface when playing a music file from Apple Music.

I was planning on having an RME FS BE handling my ADC duties and having it set off. I run real analog gear into my interface, but this is all legit studio stuff. e.g. Brand name mic pres (not knockoff stuff), DIs, compressors etc.

What am I getting myself into running this without DC blocking? And how is this relevant in day to day operation?

I am no expert, just talking out of my head.
By DC, you mean direct current , right?
A lot of amps or buffers, may have tiny amounts of DC offset on their outputs. Since ADC's work on miniscule levels of input (-100db), then any DC present on the signal gets picked up by ADC and should switch some lower bits constantly on.
I don't think ADC's like that.
Wait for an expert's reply.

 

[antcollinet]

I am no expert, just talking out of my head.
By DC, you mean direct current , right?
A lot of amps or buffers, may have tiny amounts of DC offset on their outputs. Since ADC's work on miniscule levels of input (-100db), then any DC present on the signal gets picked up by ADC and should switch some lower bits constantly on.
I don't think ADC's like that.
Wait for an expert's reply.

ADC's (as far as I know) don't give a rats ass about DC. Where it all goes tits up is when you send DC out to your amp, which amplifies it and sends it to your speakers.

Speakers really really don't like DC. They get so uppity about it they are wont to inform you of their distress using smoke signals.

 

[RayDunzl]

Wait for an expert's reply.

That ain't me, but:

A DC bias in the signal chain could offset and clip one side of a waveform that would otherwise be contained within the limits of the recording medium.

 

[Ken Tajalli]

ADC's (as far as I know) don't give a rats ass about DC. Where it all goes tits up is when you send DC out to your amp, which amplifies it and sends it to your speakers.

Speakers really really don't like DC. They get so uppity about it they are wont to inform you of their distress using smoke signals.

Good one, but do DACs have DC output?
I think, no DC comes out of any DAC.
@RayDunzl explanation seems more plausible.

 

[restorer-john]

For linear voltage measurements, DC coupling is required, but your offset better be zero.

For audio recording, there is no good reason to have an ADC that attempts to go to DC.

 

[LTig]

AFAIK all DAC chips emit DC if the digital signal contains it. The designer of the following analog stage decides whether DC will be blocked or not.

 

[Ken Tajalli]

AFAIK all DAC chips emit DC if the digital signal contains it. The designer of the following analog stage decides whether DC will be blocked or not.

Do you know of any DACs (not chips) that is not DC blocked?
Isn't it common practice to have a servo, or atleast a cap. in the way?
I am seriously asking, because I don't know, but think so.

 

[RayDunzl]

I wonder how much "DC" a musical instrument that operates via blowing air creates.

 

[restorer-john]

As far as I am concerned, any Audio DAC that can output DC is faulty by design. If they are direct coupling the outputs, fine, but there must be a DC servo and/or DC protection.

There are many, many pure DC coupled preamplifiers, power amplifiers and integrated amplifiers that will simply trip their DC protection in that situation.

I once had one of those silly little $10 DACs for TV sound. When plugged into a proper pre/power combination, I noticed a physical woofer offset and measured the DAC. It was never used again, as they were too cheap and too stupid to employ a blocking capacitor.

 

[mansr]

Do you know of any DACs (not chips) that is not DC blocked?

About half of the models on the market, if not more.

 

[mansr]

As far as I am concerned, any Audio DAC that can output DC is faulty by design. If they are direct coupling the outputs, fine, but there must be a DC servo and/or DC protection.

 

[antcollinet]

Do you know of any DACs (not chips) that is not DC blocked?
Isn't it common practice to have a servo, or atleast a cap. in the way?
I am seriously asking, because I don't know, but think so.

I really don't know - (happy to go with RJ's opinion though). I just think it would be a really good idea not to have digitised audio with DC in it.

 

[Ken Tajalli]

About half of the models on the market, if not more.

I take your word for it.
Can you explain why is that?
Is there a good reason for any source device to have any DC output?

 

[mansr]

I take your word for it.
Can you explain why is that?
Is there a good reason for any source device to have any DC output?

A source should never output DC unless present in its input. Where opinions differ is whether a DAC should filter out DC from the input signal. Some DACs do, others don't.

 

[Ken Tajalli]

A source should never output DC unless present in its input. Where opinions differ is whether a DAC should filter out DC from the input signal. Some DACs do, others don't.

Are you a politician by profession? have you ever been?
I like your answer.

 

[levimax]

I take your word for it.
Can you explain why is that?
Is there a good reason for any source device to have any DC output?

No device should output DC and most don't or at least not very much but adding any DC filter adds some costs and some amout of compromise i. e. phase shift, group delay, etc.

 

[Ken Tajalli]

No device should output DC and most don't or at least not very much but adding any DC filter adds some costs and some amout of compromise i. e. phase shift, group delay, etc.

Thanx for your reply.
It seems you are in agreement with @restorer-john , in that DC output from a DAC is wrong, period. Those that do, is because of costs and side effects. He called it irresponsible, cheap and a flaw, that I pretty much agree with.
At any rate , we have been talking among ourselves so far!
What about an answer to OP's original question?

 

[solderdude]

Most amplifiers are AC coupled anyway and the better ones that are full DC usually have either DC servo (which, within limits) removes DC at the output or have DC protection that disconnects the speaker/headphone.

There may well be designs that do not have any protection which would be a severe design fault... owners might not be aware.
Usually a loud 'pop' when switching inputs/devices can be a dead give away.
After such a 'pop' you really won't know until the woofer/headphone starts to smoke or stop working. Maybe, when the DC is not very high you may never know it is there.

For ADC's the biggest issue is asymmetric clipping (@RayDunzl) and the fact that a DAC could potentially have DC components on its output as well (not al DACs go down to DC)

In any case.... a constant DC should NOT be in a recording. There can be DC 'components' in recordings where there basically is a higher 'average' signal on the + or - side of the 'null volt' line. A reproduction chain that is full DC is technically more 'accurate'. So an audio system can be full DC to xkHz but there should be NO constant DC voltage on the input NOR on the output of the chain (ADC to speaker/headphone out).

Mainly for reasons already addressed by @tonycollinet.