Just curious.
Context: I don't really need a preamp for my system. Everything has high enough output (and drive current) that a passive attenuator is all I need. I can even bring the gain down a bit on the output of my sources. Would it make the most sense to let the line-level signals pass unattenuated to the amp and attenuate with transformers between the amp and speakers?
I've read about vol pots before, in and after a preamp circuit. I have read about tvcs in the same location. I have mocked up different scenarios and can't hear a difference (that I can't rationalize as psycho-perceptual). Even with quality transformers vs quality preamp/pot (or even resistor based stepped, series, pass shunt, etc), I think most would agree (and has been well documented) that given the perfect storm of drive current and voltage amplification, no attenuation has the most merit in terms of maintaining electrical headroom, matching impedance between devices, inserting as little noise as possible, etc.
Is attenuation after the amp a better option? Could be a transformer or resistor based attenuator. Could be talking about ones used for whole-house distribution. It could also be high end esoteric. Consider merits of circuit physics only.
I have used light-dependent resistors in preamp circuits, but couldn't get the devices to match enough (always ate at me as I listened) and don't want to buy 5oo of them to test and sort. If I could easily get some that are matched, I might consider using them at the preamp level. However, they still present as resistors when providing any attenuation. Super easy to implement in any number of schemes with an Arduino controller, however. Active volume control with digital or analog opamps isn't really an option as I'm looking for less, not more.
System:
Rod Elliot zen pre-amp as the output on my cd player (among others);Pass Labs aleph clone used in balanced config, 22k input impedance; Theil cs2 speakers.
The amp might run a bit cooler with the input at max as more dissipation through the speakers and less through the heatsinks. If I recall from when I built it many years ago, it is a constant current sink. Any reason this would stress out the amp any more with the transformer on the output?
Context: I don't really need a preamp for my system. Everything has high enough output (and drive current) that a passive attenuator is all I need. I can even bring the gain down a bit on the output of my sources. Would it make the most sense to let the line-level signals pass unattenuated to the amp and attenuate with transformers between the amp and speakers?
I've read about vol pots before, in and after a preamp circuit. I have read about tvcs in the same location. I have mocked up different scenarios and can't hear a difference (that I can't rationalize as psycho-perceptual). Even with quality transformers vs quality preamp/pot (or even resistor based stepped, series, pass shunt, etc), I think most would agree (and has been well documented) that given the perfect storm of drive current and voltage amplification, no attenuation has the most merit in terms of maintaining electrical headroom, matching impedance between devices, inserting as little noise as possible, etc.
Is attenuation after the amp a better option? Could be a transformer or resistor based attenuator. Could be talking about ones used for whole-house distribution. It could also be high end esoteric. Consider merits of circuit physics only.
I have used light-dependent resistors in preamp circuits, but couldn't get the devices to match enough (always ate at me as I listened) and don't want to buy 5oo of them to test and sort. If I could easily get some that are matched, I might consider using them at the preamp level. However, they still present as resistors when providing any attenuation. Super easy to implement in any number of schemes with an Arduino controller, however. Active volume control with digital or analog opamps isn't really an option as I'm looking for less, not more.
System:
Rod Elliot zen pre-amp as the output on my cd player (among others);Pass Labs aleph clone used in balanced config, 22k input impedance; Theil cs2 speakers.
The amp might run a bit cooler with the input at max as more dissipation through the speakers and less through the heatsinks. If I recall from when I built it many years ago, it is a constant current sink. Any reason this would stress out the amp any more with the transformer on the output?
A class A amp will dissipate constant power all the time.
A class B amp will dissipate more power the louder the source if its music.
Only with a single tone does the amp dissipation fall as the input is really large - with music you get gross clipping well before this point.
A transformer on the output is going to change the impedance seen by the amplifier. This can have various consequences. Its not usual.
It is pretty common on distributed audio, where individual rooms might have their own wall vol control. Not talking about 70v matching transformers.
And yes, it is a constant current amp.
And yes, it is a constant current amp.
When hearing 70 V (or, on this side of the pond, 100 V), I'm thinking the likes of in-store audio, where fidelity isn't really of primary concern.
In hi-fi, people were utterly glad that they could get rid of output transformers in the mid-late '60s. Pesky buggers, those - expensive, nonlinear and bandwidth-limiting. Heavy, too.
You should consider yourself lucky that the Thiel CS2s seem to have a rather well-behaved impedance response for a 3-way - a lot of speakers would devolve into a peaky mess in current drive. (EQ / room correction is still very likely to be required.) It is an approach arguably best employed in an active speaker configuration, and then preferably in a "hybrid" configuration that transitions to voltage driving around fs.
If you were to ask me what the best configuration for the volume control would be, before or after the preamp, I'd say both.
If it's before only, you have superb input level handling but may have difficulty getting noise down to an acceptable degree, especially with high-sensitivity speakers, with output dynamic range having to be pushed way beyond what you actually need at any one time (like 125+ dB).
If it's after only, the preamp circuitry is potentially burdened with super high levels all the time, not helping distortion, and you may even have difficulty reaching as much gain as you want while still providing a decent input overload margin. In return, these can be very low in noise. (Some transformerless vacuum tube circuits are employing this approach. It is clever since at line level, their high input noise matters little, and with a high enough B+ you've got plenty of voltage headroom - you just need to sort out the linearity.)
In a nutshell, it helps getting by with way less circuit dynamic range - and that potentially means better linearity and/or reduced power consumption.
Unsurprisingly, essentially every truly high-end preamp from the late '70s and '80s used 4-gang volume pots to employ a two-stage volume control, and even a few integrated amps eventually did. If the Zen pre-amp you are referring to is the Balanced Zen line stage, then that has a two-stage volume control, too. Any hi-fi worthy PGA is a two-stage affair as well, usually combining input attenuation with variable gain. (It would not have been possible to build tiny MP3 players with decent performance without these.)
Some of the more clever constructions actually included some range-limiting resistors in one stage, given that nobody really needs the insane volume control range of two pots in series in practice. (120 dB or something?)
In hi-fi, people were utterly glad that they could get rid of output transformers in the mid-late '60s. Pesky buggers, those - expensive, nonlinear and bandwidth-limiting. Heavy, too.
You should consider yourself lucky that the Thiel CS2s seem to have a rather well-behaved impedance response for a 3-way - a lot of speakers would devolve into a peaky mess in current drive. (EQ / room correction is still very likely to be required.) It is an approach arguably best employed in an active speaker configuration, and then preferably in a "hybrid" configuration that transitions to voltage driving around fs.
If you were to ask me what the best configuration for the volume control would be, before or after the preamp, I'd say both.
If it's before only, you have superb input level handling but may have difficulty getting noise down to an acceptable degree, especially with high-sensitivity speakers, with output dynamic range having to be pushed way beyond what you actually need at any one time (like 125+ dB).
If it's after only, the preamp circuitry is potentially burdened with super high levels all the time, not helping distortion, and you may even have difficulty reaching as much gain as you want while still providing a decent input overload margin. In return, these can be very low in noise. (Some transformerless vacuum tube circuits are employing this approach. It is clever since at line level, their high input noise matters little, and with a high enough B+ you've got plenty of voltage headroom - you just need to sort out the linearity.)
In a nutshell, it helps getting by with way less circuit dynamic range - and that potentially means better linearity and/or reduced power consumption.
Unsurprisingly, essentially every truly high-end preamp from the late '70s and '80s used 4-gang volume pots to employ a two-stage volume control, and even a few integrated amps eventually did. If the Zen pre-amp you are referring to is the Balanced Zen line stage, then that has a two-stage volume control, too. Any hi-fi worthy PGA is a two-stage affair as well, usually combining input attenuation with variable gain. (It would not have been possible to build tiny MP3 players with decent performance without these.)
Some of the more clever constructions actually included some range-limiting resistors in one stage, given that nobody really needs the insane volume control range of two pots in series in practice. (120 dB or something?)
Last edited: