• Hello and Welcome to HiFiVision.com - an online community for the home entertainment and tech enthusiasts!

    If you would like to ask a question, participate in a discussion and view attachments please Register yourself.

D C Coupled amplifier

Wharfedale EVO 4.1 4.2 Speakers

soundbuff

Active Member
Joined
Dec 26, 2015
Messages
287
Points
43
Location
Bangalore
Hi Folks,
I am curious to know something about Direct Current or DC coupled transistor amplifiers in layman lingo. From surfing around I gather that they are good at reproducing low frequencies, in fact starting at 0 hz. I also found it mentioned that they pass DC current from upstream components to the speaker and can potentially damage them. Why do some designers choose this topology. Please do keep in mind that I have no background in electronics except words I have picked up here and there.

Thanks
 

Robert Bose

Member
Joined
Oct 24, 2008
Messages
65
Points
18
Location
Bangalore India
Most amps are not DC Coupled

Music Freq starts from 20 HZ on wards

Most amps block DC with Coupling capacitors

from Input, Voltage Gain Stage, to Power Amp power stages

the Caps cause Freq related Phase Shifts

But if the amp is DC coupled, it needs Servo circuit to maintain DC

offset close to zero, so more complicated

also though the amp may be DC coupled the gain at Freq less than 10 HZ
can be reduced when compared to the output at 1 Khz

only thing is the quality of coupling capacitor has an influence on the sound

as well as the freq related phase shift issue, in listening tests comparing

DC coupled amp, and well built non DC Amp, with good quality caps, not much difference can be heard, more important is low global negative feedback in the circuit, amps with low global negative feedback tend to sound harmonically natural, but maybe less exciting

Robert Bose
 

square_wave

Well-Known Member
Joined
Aug 16, 2006
Messages
3,248
Points
113
Location
Bangalore
The dc coupled designs ussually have coupling capaciors at the amp output to protect loudspeakers
 

soundbuff

Active Member
Joined
Dec 26, 2015
Messages
287
Points
43
Location
Bangalore
Thanks a lot Folks,
Even though what you have written is extremely lucid, there is only so much I have understood which I set out below:

1. Music frequency starts from about 20 hz so 0 hz is not much point. Does this also include sounds that we feel rather than hear? May be some of the body in music comes from the sub-20 hz frequencies
2. The "sonic advantages" if any, of DC coupled may have something to do with the absence of coupling capacitors which cause frequency related 'phase shifts' which get blocked in conventional amps. (I think I know what 'out of phase' sounds like, not sure if that is relevant here)
3. It is useful to enquire how much global feedback is provided by a DC coupled amp. The sound characteristics of the amp is connected with this.
4.One should make sure that a design has coupling capaciiors at the output to protect speakers

I did not quite get the point about reduced gain below 10 hz, is this is important from the point of operation or design? I am afraid "Servo circuit to maintain offset close to zero" was Greek to me.

Muchas Gracias
 

Fantastic

Well-Known Member
Joined
Jun 1, 2010
Messages
866
Points
93
Location
Third Rock
Thanks a lot Folks,
Even though what you have written is extremely lucid, there is only so much I have understood which I set out below:

1. Music frequency starts from about 20 hz so 0 hz is not much point. Does this also include sounds that we feel rather than hear? May be some of the body in music comes from the sub-20 hz frequencies
2. The "sonic advantages" if any, of DC coupled may have something to do with the absence of coupling capacitors which cause frequency related 'phase shifts' which get blocked in conventional amps. (I think I know what 'out of phase' sounds like, not sure if that is relevant here)
3. It is useful to enquire how much global feedback is provided by a DC coupled amp. The sound characteristics of the amp is connected with this.
4.One should make sure that a design has coupling capaciiors at the output to protect speakers

I did not quite get the point about reduced gain below 10 hz, is this is important from the point of operation or design? I am afraid "Servo circuit to maintain offset close to zero" was Greek to me.

Muchas Gracias


I wrote a long explanation but it got lost ! So here is a quick short one.
Most music lies above about 30 Hz . The disco type bass , kick drum etc. lies around 50 to 100 Hz ! Lower than 20 Hz is usually only from an organ or LFE effects in movies like thunder , explosions, .....running Dinosaurs etc ! :)
The sonic advantage is due to eliminating the input capacitor. But this is not due to the phase shift it introduces with frequency. It due to the behaviour of the capacitor , which is highly debated. Like polypropylene sounds better than polyester etc. Some agree and others do not. But they sometimes do sound different to me.
Feedback can be dc or ac coupled. However dc coupling only affects the bandwidth of the amp. Response goes down to dc rather than rolling off at say 10 Hz or so, if it is dc coupled!

Capacitors at the output of an amp is not good and not used in most modern amps. It used to be done in old generation amps or amps that have a single supply . This means that it will have dc at the output ( about 1/2 supply Voltage ) and so you need a capacitor to block the dc. These are large electrolytic capacitors and are not as good as small film caps used typically at the input.So if you want to block dc , do it at the input.
The amp should have a fast acting dc sensing circuit at the output which will operate a protection circuit , often a relay , to disconnect the speaker if a large dc voltage is found at the speaker terminal.
DC servos are only to correct a small dc offset at the output that all amps have. That works only in the milli Volt range.

When you use a capacitor at the input,the frequencies below the roll off frequency ( say 20 Hz) will get reduced ( lower gain !) . It ( the signal level) keeps dropping as the frequency falls tending to zero at dc ( blocking it !).
 
Top