Input voltage for high gain amplifiers and the dream of XLR cables.

[antcollinet]

dualazmak is talking about induced noise, which gets worse the longer your cables are.
I was talking about transmitted noise, which does not meaningfully change with cable length.
If you have a GND loop in your system, then it doesn't matter if the cable is 1m or 10m long, the GND loop will stay.

Yes, it will.

But the ground voltage difference from one end of the analogue interconnect to the other (which is what is picked up as noise in an unbalanced interconnect) will be equal to the ground loop current multiplied by the ground impedance between the two devices. This will be related to cable length.

 

[dualazmak]

A passive attenuator should be purely resistive, and so have no negative impact on the frequency response (or other sound quality aspects) of the system.

Edit: By my mistake and misunderstanding, I was discussing about "passive attenuators in SP high-level signal" in this post. I hided the contents, therefore, under the spoiler cover. @tonycollinet, thank you for your kind reminder.

Spoiler

You are "theoretically" almost right, I assume. I too, for long time have been thinking about the same as you pointed. But,,,

As you know, passive attenuators have metal-to-metal physical sliding contacts which have inevitable tendency of getting "worse" over long period. When I have disassemble and fully cleaned-up (washed and cleaned all the components) of relatively old but very nicely built rather big and heavy attenuators of YAMAHA NS-1000 (same attenuators for NS-1000M), the total sound improvement was clearly audible with my and my wife's ears. (I cleaned-up with reference to https://shorturl.at/qsGKO.)

Furthermore, when I very carefully investigated the elimination of my passive attenuators from the SP high-level signal paths, I found "your point" would not be always true relating to the slight impedance changes and zero-cross distortions looking from the amplifier. You would please carefully read through these my posts, especially #248;

- In depth insights on SP attenuators and their elimination in multichannel system: #248, #251, #99(remote thread), #100(remote thread), #101(remote thread)

In my system, after these careful investigations, I fully eliminated the passive attenuators, but only inserted parallel (not series) 22 ohm tuning resistors for midranges, tweeters and super-tweeters.

As for the fine tuning parallel (not series) 20 - 22 ohm resistors giving small excessive power load to amplifiers, the discussion here would be of your reference; there @DualTriode kindly cited "page 14 https://rephase.org/projects/JBL_M2_crossover.pdf"; for some of the sophisticatedly designed JBL SP units, they dare use "parallel" tuning resistors (R2 in page 14 of the document) to give a slight overload (above zero cross) to amplifier. @DualTriode also kindly wrote,"What this (parallel tuning resistor) does is operates the amplifier at a higher power output where the amplifier operates at a higher SNR. Also the amplifier operates into a flatter, more resistive load."

 

[antcollinet]

You are "theoretically" almost right, I assume. I too, for long time have been thinking about the same as you pointed. But,,,

As you know, passive attenuators have metal-to-metal physical sliding contacts which have inevitable tendency of getting "worse" over long period. When I have disassemble and fully cleaned-up (washed and cleaned all the components) of relatively old but very nicely built rather big and heavy attenuators of YAMAHA NS-1000 (same attenuators for NS-1000M), the total sound improvement was clearly audible with my and my wife's ears. (I cleaned-up with reference to https://shorturl.at/qsGKO.)

Furthermore, when I very carefully tried the elimination of my attenuators from the SP high-level signal paths, I found "your point" would not be always true relating to the slight impedance changes and zero-cross distortions looking from the amplifier. You would please carefully read through these my posts, especially #248;

- In depth insights on SP attenuators and their elimination in multichannel system: #248, #251, #99(remote thread), #100(remote thread), #101(remote thread)

In my system, after these careful investigations, I fully eliminated the passive attenuators, but only inserted parallel (not series) 22 ohm tuning resistors for midranges, tweeters and super-tweeters.

As for the fine tuning parallel (not series) 20 - 22 ohm resistors giving small excessive power load to amplifiers, the discussion here would be of your reference; there @DualTriode kindly cited "page 14 https://rephase.org/projects/JBL_M2_crossover.pdf"; for some of the sophisticatedly designed JBL SP units, they dare use "parallel" tuning resistors (R2 in page 14 of the document) to give a slight overload (above zero cross) to amplifier. @DualTriode also kindly wrote,"What this (parallel tuning resistor) does is operates the amplifier at a higher power output where the amplifier operates at a higher SNR. Also the amplifier operates into a flatter, more resistive load."

Woah - hang on:

SP high level attenuators. No way! Then you are just messing up the output charcterisitics of the amplifier - especially output impedance, and the speaker impedance curve will pull the system frequency response all over the place. Also - big, expensive.

Aside

As far as I can tell from your articles, you were putting attenuation into the crossover for individual drivers - that does not apply here: we have a source with 2V (unbalanced) or 5V (balanced) output into an amp with two alternate gain settings, and (presumably) existing speakers already with their crossover - OP is all about managing system gain staging

I'm talking about input attenuation between source (dac) and amp : fixed value resistive U pads or H pads. Allowing the DAC to run at close to full output, while still having listenable volume output from the amp. Essentially applying negative gain to the input - restively rather than via digital volume control - to offset a higher than needed gain of the amp.

And if you are worried about an additional set of connections (but why - you have to have connectors - if they don't work then your system doesn't work anyway) and are prepared to DIY, then attenuation can be built into a cable (edited to add: Or inside the amp end connector) with soldered connections.

 

[dualazmak]

Woah - hang on:

SP high level attenuators. No way!

Understood; in my post #22 above, I was talking about SP high-level attenuators by mistake. Please ignore it as you are focusing on passive attenuators in line level between DAC and amplifier.

In my setup, I use "HiFi integrated amplifier with excellent gain/volume controller" as active attenuator...