That begs the question what you are after: knowing the facts or just any answer you like.
He's spot on with his analysis - those diodes are a waste of money and space. That's why no serious designer uses them.
You yourself said you have limited electronics knowledge.
This is a free learning opportunity!
Jan
Thanks guys.
First attempt: I will reduce the DAC output which we currently run at 3V. 2V output and 1V output are also user-selectable options.
If this is still not enough to deliver usable range to the volume knob: I neglected to mention that we currently run a passive preamp that also offers (additional) -10dB / -20dB attenuation before the stepped attenuator, so I will be giving this a try.
Only as a last resort, I will mess with the QUAD 306 NFB.
My reluctance to messing with factory design NFB levels stems from my own (many) tests with our tube amp and GNFB. The impact on the tube amp were massive and easily measurable - but I am learning from others here that this must be very circuit-dependent.
With circuit context in mind, what I am hearing from more experienced minds, is that the QUAD 306 circuit doesn't mind a bit of additional NFB.
Thanks for all the helpful feedback. Appreciated.
First attempt: I will reduce the DAC output which we currently run at 3V. 2V output and 1V output are also user-selectable options.
If this is still not enough to deliver usable range to the volume knob: I neglected to mention that we currently run a passive preamp that also offers (additional) -10dB / -20dB attenuation before the stepped attenuator, so I will be giving this a try.
Only as a last resort, I will mess with the QUAD 306 NFB.
My reluctance to messing with factory design NFB levels stems from my own (many) tests with our tube amp and GNFB. The impact on the tube amp were massive and easily measurable - but I am learning from others here that this must be very circuit-dependent.
With circuit context in mind, what I am hearing from more experienced minds, is that the QUAD 306 circuit doesn't mind a bit of additional NFB.
Thanks for all the helpful feedback. Appreciated.
Last edited:
As the '306 is a current dumping amp, there is more than one fb path.
Make sure you modify the right one so as not to upset the bridge balance that gives it its great sound and low distortion.
Jan
Make sure you modify the right one so as not to upset the bridge balance that gives it its great sound and low distortion.
Jan
The level you experience is not in the DAC selection. it is already embedded in the media you play which is accordingly .9.49 , 6.42 and 3.16 times less. than the levels you state. The only exception is if you are very strangely playing a Test Disc with cautions that 0DBFs is on it. So no need to make any changes to the Quad 306, ( I have used one for over 12 years ) Easiest is to use it with a passive volume control (attenuator), ideally that caters for adjustable shunt resistance and has no adverse loading to the source component impedance, but you will have to look around for that, which is well worth it and where your time should be.Thanks guys.
First attempt: I will reduce the DAC output which we currently run at 3V. 2V output and 1V output are also user-selectable options.
If this is still not enough to deliver usable range to the volume knob: I neglected to mention that we currently run a passive preamp that also offers (additional) -10dB / -20dB attenuation before the stepped attenuator, so I will be giving this a try.
Only as a last resort, I will mess with the QUAD 306 NFB.
My reluctance to messing with factory design NFB levels stems from my own (many) tests with our tube amp and GNFB. The impact on the tube amp were massive and easily measurable - but I am learning from others here that this must be very circuit-dependent.
With circuit context in mind, what I am hearing from more experienced minds, is that the QUAD 306 circuit doesn't mind a bit of additional NFB.
Thanks for all the helpful feedback. Appreciated.
Attachments
Thanks for these thoughts, Chris. I can see you are big fan of the QUAD 306!
I hope you are correct - and hope that what has worked for you in your system, will work similarly as well in our system.
I take your point that it is the music content embedded in the media that ultimately determines the playback volume level. Taking this precept to the extreme: a brief silence between verses in the music program gives no playback volume at all - no matter how sensitive your amplifier is. I understand.
I don't really understand this statement. Perhaps I have missed a deeper point you are making above, but it is undeniable to my ears that the playback volume - when music is playing - is louder with the 3V DAC output, than when that same music is playing, with only 1V DAC output. Have I somehow got this wrong and my expectation bias has tricked me?
If I am correct about this, it makes sense to me to start by reducing the 3V DAC output, in the event that I have insufficient volume knob range using our passive preamp partnered with the QUAD 306.
Perhaps I have not explained this so well, but I hope this makes sense.
Our passive is a TISBURY Mini Passive Pre II. It's a nice, affordable little English product that is superbly backed by its maker. As you will see below, it also offers two levels of fixed attenuation in addition to a stepped attenuator.
I really don't know if this unit "...caters for adjustable shunt resistance and has no adverse loading to the source component impedance..." That said, it does sound very good - and the additional output jacks allow us to drive our preferred headphone amplifier also!
In any event, I will report, once the QUAD 306 arrives and reveals all... in our system.
I hope you are correct - and hope that what has worked for you in your system, will work similarly as well in our system.
I take your point that it is the music content embedded in the media that ultimately determines the playback volume level. Taking this precept to the extreme: a brief silence between verses in the music program gives no playback volume at all - no matter how sensitive your amplifier is. I understand.
I don't really understand this statement. Perhaps I have missed a deeper point you are making above, but it is undeniable to my ears that the playback volume - when music is playing - is louder with the 3V DAC output, than when that same music is playing, with only 1V DAC output. Have I somehow got this wrong and my expectation bias has tricked me?
If I am correct about this, it makes sense to me to start by reducing the 3V DAC output, in the event that I have insufficient volume knob range using our passive preamp partnered with the QUAD 306.
Perhaps I have not explained this so well, but I hope this makes sense.
Our passive is a TISBURY Mini Passive Pre II. It's a nice, affordable little English product that is superbly backed by its maker. As you will see below, it also offers two levels of fixed attenuation in addition to a stepped attenuator.
I really don't know if this unit "...caters for adjustable shunt resistance and has no adverse loading to the source component impedance..." That said, it does sound very good - and the additional output jacks allow us to drive our preferred headphone amplifier also!
In any event, I will report, once the QUAD 306 arrives and reveals all... in our system.
Yes I am a big fan of the 306. Could we know your DAC make and model and if it has a schematic we can discern what it is doing. It should be all that occurs is you are providing the possibility of higher available headroom - such as to be allowing the playing of a Test CD, but all normal CD's stick to embedded consumer line level, so despite having available headroom, your passive will perform well with the 306 It might be your DAC is adding some current drive, so the source impedance is lowered further.
Adjustable shunt resistance can be beneficial, where the traditional volume control is reversed in purpose, namely that the series resistance is made extremely high at very lowest volume ie 25 megaohms or more affording silence, The shunt resistance that is no longer just boring grounding, can then be made adjustable, so there is no adverse loading at any time to the source impedance, as volume is increased.
Adjustable shunt resistance can be beneficial, where the traditional volume control is reversed in purpose, namely that the series resistance is made extremely high at very lowest volume ie 25 megaohms or more affording silence, The shunt resistance that is no longer just boring grounding, can then be made adjustable, so there is no adverse loading at any time to the source impedance, as volume is increased.
Our DAC is a CHORD Qutest, from England.
I have had a few exchanges with the designer - Rob Watts - and he has confirmed that the unattenuated output of the Qutest design is 3V.
The user-selectable options of 2V and 1V are both achieved via attenuation within the Qutest itself - in the digital domain - which in Rob's viewpoint, is sonically invisible.
I understand that the three different Voltage Output options are there to provide different strengths of signal, to help better match a variety of preamp or power amp sensitivities.
These three different Voltage Output options have proven invaluable with the various different amps we run at home.
We run the Qutest with an advanced bespoke SMPS that also powers our RUPERT NEVE RNHP Headphone amp, courtesy of a mate who is very much at the leading edge of audio design. I can take no technical credit, aside from having made the effort to try his SMPS design against other PSUs... Needless to say, it's killer! 😊
I have had a few exchanges with the designer - Rob Watts - and he has confirmed that the unattenuated output of the Qutest design is 3V.
The user-selectable options of 2V and 1V are both achieved via attenuation within the Qutest itself - in the digital domain - which in Rob's viewpoint, is sonically invisible.
I understand that the three different Voltage Output options are there to provide different strengths of signal, to help better match a variety of preamp or power amp sensitivities.
These three different Voltage Output options have proven invaluable with the various different amps we run at home.
We run the Qutest with an advanced bespoke SMPS that also powers our RUPERT NEVE RNHP Headphone amp, courtesy of a mate who is very much at the leading edge of audio design. I can take no technical credit, aside from having made the effort to try his SMPS design against other PSUs... Needless to say, it's killer! 😊
Last edited:
Surely digital attenuation is the worst possible method as it represents in reduction in resolution?
Hasn't this threaad been a bit of a storm in a tea cup? The dac output is switchable and you have a decent passive attenuator/volume control so you never had any reason to alter the 306's NFB.
Hasn't this threaad been a bit of a storm in a tea cup? The dac output is switchable and you have a decent passive attenuator/volume control so you never had any reason to alter the 306's NFB.
I would agree, which is why we run the Qutest at 3V.
But that said, with a DAC that sounds as good as the Qutest does - for the money - I rather think Rob Watts has already forgotten a load more than I will ever know on the subject. 😉
But that said, with a DAC that sounds as good as the Qutest does - for the money - I rather think Rob Watts has already forgotten a load more than I will ever know on the subject. 😉
I am surprised Chord think a 3v output option is necessary. 2V rms is the usual max.
They would have been better to fit a switchable gain function in the analogue / IV stage.
They would have been better to fit a switchable gain function in the analogue / IV stage.
Perhaps...
Can we all agree that a 3V output will sound louder than a 1V output - given the same playback program, same playback chain, with no other changes?
Or am I on my own here?
Can we all agree that a 3V output will sound louder than a 1V output - given the same playback program, same playback chain, with no other changes?
Or am I on my own here?
Which the media never provides as level, unless you habit and ignore the warnings of 0DbFS on your Test CD, over breakfast !! , i think its more so a manufacturer sadly pandering for no proper reason, to a certain other forum where consumer line level is never understood.
Not necessarily. If the following stage cannot handle a 3v signal, it may clip. (Louder isn not the correct term to use)
Also depends on how Cjhord have done the digital attenuation. If 3v is the native dac output, this means full dynamic range is available and attenuating in the analogue domain retains this.
Also depends on how Cjhord have done the digital attenuation. If 3v is the native dac output, this means full dynamic range is available and attenuating in the analogue domain retains this.
Assuming none of the following stages clip - and that a water main doesn't burst and fry the whole system...
Can we all agree that a 3V output will sound louder to the human ear, than a 1V output - given the same playback program, same playback chain, with no other changes?
Can we all agree that a 3V output will sound louder to the human ear, than a 1V output - given the same playback program, same playback chain, with no other changes?
Of course it will if the amp and speakers can handle the extra ~10db of signal.
However as you won't be listening at these levels, you will have reduced the 3v to some lower level anyway, hence why I say 'louder' is not the correct term.
Perhaps you should ask Chord to clarify what the dynamic range is at 1v, 2v and 3v output? (and hence the resolution of a 16 & 24bit data stream at these output levels)
However as you won't be listening at these levels, you will have reduced the 3v to some lower level anyway, hence why I say 'louder' is not the correct term.
Perhaps you should ask Chord to clarify what the dynamic range is at 1v, 2v and 3v output? (and hence the resolution of a 16 & 24bit data stream at these output levels)
No it won't because the media contains the level, not the equipment, All you are doing is providing headroom, like a bigger road for your mini that never goes faster to drive on. If the equipment is raising voltage it just risks distortion from doing so. I think you will find the manufacturer is selecting slightly different current capability, giving you the impression of change. Test each so called level with a distortion analyzer, or at the least Audacity, where you can see Peak and RMS level, to confirm, what it claims to be doing. Compare what you see in Audacity to my earlier graph, which is Mike Oldfields Amarok, ie play the same CD via your DACPerhaps...
Can we all agree that a 3V output will sound louder than a 1V output - given the same playback program, same playback chain, with no other changes?
Or am I on my own here?
Last edited:
A Quad 306 needs .375v for 50watts.
So it matters not what the dac output is, you still have to attenuate it with your passive preamp.
the only question is whether you are happy with the reduced travel of the pot needed. But your preamp also has switchable attenuators, so -10db will require about 1.2v for full output.
Chris is correct when he referred to the media level because you cannot be certain that any recording uses the full dynamic range, so in practice the 3v output may never reach 3v rms.
Most recording engineers will keep well below the max possible level to ensure no peaks can stary into digital clipping.
Otherwise your argument is like saying a car with 200bhp will be faster than one with 150 - it will but only if you use it. So the 200bhp car has 50bhp of extra headroom.
So it matters not what the dac output is, you still have to attenuate it with your passive preamp.
the only question is whether you are happy with the reduced travel of the pot needed. But your preamp also has switchable attenuators, so -10db will require about 1.2v for full output.
Chris is correct when he referred to the media level because you cannot be certain that any recording uses the full dynamic range, so in practice the 3v output may never reach 3v rms.
Most recording engineers will keep well below the max possible level to ensure no peaks can stary into digital clipping.
Otherwise your argument is like saying a car with 200bhp will be faster than one with 150 - it will but only if you use it. So the 200bhp car has 50bhp of extra headroom.
No they are not opposing views, as maximum level embedded in the CD as consumer line level remains the same. You are just allowing for the very unlikely playing of a Test CD 0DbFS CD, which no one should ever play with audio equipment - fine with a scope though You would see a warning like this : " This CD contains tones at frequencies and levels that can easily destroy expensive loudspeakers/tweeters and/or amplifiers "
There may be though instead of voltage some current buffering enabling lowering of the source impedance, You might be ascertaining that as level change. All fine up until distortion is added by the device, which needs to be checked. Your passive will be sensitive to source component output impedance changes. But much less so where the shunt resistance is adjustable, allowed by the series resistance being extremely high at lowest volume.
https://en.wikipedia.org/wiki/Line_level
Attachments
Last edited: