How much parts selection can impact the sound ?

[ginetto61]

Hello,
sorry for the trivial question.
But i wonder how much a very good selection of parts keeping the same circuit can increase the unit performance.
I was reading about a dac where a different choice of caps and resistors was able to give 3D depth to the sound that was missing in the original ummoded unit. I have a soft spot for soundfield depth.
I found this as interesting as minimalist approach for instance.
It could be the case that a very basic design with a good parts selection could give excellent sound, even more than a complex design with generic parts ...
Thanks and regards,
gino

 

[DF96]

It could be the case that a very basic design with a good parts selection could give excellent sound, even more than a complex design with generic parts ...

It depends on whether by 'very basic design' you mean 'competent but simple' or 'incompetent'. Also, does 'complex' mean 'important details attended to' or 'all hobby horses ridden simultaneously' or 'all audiophile buttons pushed'.

Circuits come first. Build layout comes second. Component choice (in proper engineering terms e.g. capacitor dielectric) comes third. Component brand comes a distant fourth.

Actually, there is no need to start a new thread on this. A little searching and reading would give you whatever answer you desire.

 

[Fsatsil]

Is there a particular DAC you had in mind or do you just mean in general? Id be skeptical that a different cap/resistor choice would be "able to give 3D depth to the sound that was missing in the original unmodded unit". The circuit design is much more important then parts selection.

 

[KatieandDad]

Generally only the parts that are in the signal path will seriously affect the sound quality.

There are obvious gains to be made by carefully designing your PSU and more importantly how you implement the PSU, especially in a mixed digital and audio environment.

If you look at the Pass F5 or JLH 10W or the Hiraga for example, these are some examples of simple designs that have been very well implemented. There are loads of threads already on these designs where guys have experimented with various components throughout the designs.

 

[ginetto61]

It depends on whether by 'very basic design' you mean 'competent but simple' or 'incompetent'. Also, does 'complex' mean 'important details attended to' or 'all hobby horses ridden simultaneously' or 'all audiophile buttons pushed'.
Circuits come first. Build layout comes second. Component choice (in proper engineering terms e.g. capacitor dielectric) comes third.
Component brand comes a distant fourth
Actually, there is no need to start a new thread on this. A little searching and reading would give you whatever answer you desire.

Thanks a lot for your valuable advice.
Now i got a hierarchy ... i asked because me and a friend were amazed by the changes in sound after replacing a feedback resistor in a preamp.
And I was amazed after changing a electrolytic with a plastic cap in series with a tweeter. Even bad connections or thin wires on a high current path can influence sound.
Maybe there are point in a circuit particularly delicate. In general i know that i have to look at the signal path first.
Thanks a lot,
gino

 

[DF96]

And I was amazed after changing a electrolytic with a plastic cap in series with a tweeter.

Don't make the common mistake of extrapolating from a high power filter component to a low power coupling component. The former are sensitive to component parameters for two reasons (power, filter); the latter are not.

Even bad connections or thin wires on a high current path can influence sound.

I would drop the "even".

 

[ginetto61]

Is there a particular DAC you had in mind or do you just mean in general?

Hi the 3D is this one ...

http://www.diyaudio.com/forums/digital-source/239441-xmos-ak4399-single-board.html

Id be skeptical that a different cap/resistor choice would be "able to give 3D depth to the sound that was missing in the original unmodded unit".
The circuit design is much more important then parts selection

I do not know. I think that high grade parts in strategic positions can make a real difference, the problem is how much.
For instance, resistors are usually very linear and low noise.
And still there are some much better than others.
I am referring to Caddock or Vishay best ones.
If i had to choose an approach i would prefer simpler topologies with more accurate parts selection instead of complex topologies with generic parts.
Yes i would do this way.
Thanks alot
gino

 

[bear]

Well the power supply is "in the signal path".

There are certainly some components that have now well established differences in parameters and frequency response that have been *measured* and compared.

Caps, resistors are in this list.

So, in theory it is possible that the use of "better" quality components may yield an audible difference in perceived "sound". Better or worse is still highly subjective.

Yes, electrolytics are prime offenders in crossovers.

_-_-bear

 

[analog_sa]

Parts quality, especially in simpler, low/zero loop nfb circuits can be crucial. It can completely make or break the final result, although measurements won't show much.

Parts, which are perhaps more audible than others are coupling and PS caps, cart or dac loading resistors, anode, drain, collector load resistors, gain setting resistors in nfb circuits, chokes, transformers.

Signal path? As long as it includes devices which are in series and shunt to the signal. And of course all PS parts... which includes mains cabling too.

 

[ginetto61]

Generally only the parts that are in the signal path will seriously affect the sound quality.
There are obvious gains to be made by carefully designing your PSU and more importantly how you implement the PSU, especially in a mixed digital and audio environment.
If you look at the Pass F5 or JLH 10W or the Hiraga for example, these are some examples of simple designs that have been very well implemented. There are loads of threads already on these designs where guys have experimented with various components throughout the designs

Thanks a lot for the very valuable advice
I have seen a lot of comparison between capacitors, for PS and signal coupling.
Some like the famous Black Gate are object of cult.
I like the Pass approach very much.
Thanks again,
gino

 

[DF96]

If you understand circuits then you will know which components are critical to sound quality, and which components cannot affect sound quality. If not, you will have to decide who to believe.

 

[AndrewT]

It depends on ........................

Circuits come first. Build layout comes second. Component choice (in proper engineering terms e.g. capacitor dielectric) comes third. Component brand comes a distant fourth.....................

Picking up on point three: Component choice, means selecting the correct TYPE of component to suit the duty required of it.

There is no point in using a high tolerance, high tempco, resistor where precision is required.
Similarly a leaky electrolytic with high tolerance is no good where leaking DC ruins the circuit performance and the (filter) component requires precision.

 

[Fsatsil]

There's also that common misconception that more expensive always means better when choosing parts. Just because a resistor cost $10 doesn't mean it will sound any better then the a 10cent Dale RN55. I learned this the hard/expensive way when I first got into diy audio years ago

 

[nezbleu]

A little searching and reading would give you whatever answer you desire.

Amen.

 

[nezbleu]

Parts quality, especially in simpler, low/zero loop nfb circuits can be crucial. It can completely make or break the final result, although measurements won't show much.

If only I had my magicometer to measure the things other than frequency response, phase response, noise, and distortion which so drastically affect sound (and only sound, not fighter jet controls or satellite communications or medical equipment).

 

[Fsatsil]

If only I had my magicometer to measure the things other than frequency response, phase response, noise, and distortion which so drastically affect sound (and only sound, not fighter jet controls or satellite communications or medical equipment).

Your magicometer could probably only measure those things if it was made with special magic components and connected with cables made of 99.9999% pure admantium

 

[DF96]

Fighter jet controls use secret military technology. It is a myth that they use standard (analogue or digital) servo loop theory to design them. To gain audio access to this technology you have to give a lot of money to certain people. In order to protect the military secrets these people are deliberately chosen so they don't understand the technology themselves, so they can't accidentally give away any technical information in their marketing.

 

[7n7is]

Some good articles about capacitors

http://www.scribd.com/doc/2610442/capacitor-sound

Picking Capacitors

Humble Homemade Hifi


SPEAKER CABLES: Science or Snake Oil
https://passlabs.com/articles/speaker-cables-science-or-snake-oil

 

[gmphadte]

If you want to modify with most benefit, start with the power supply. Everything else is also to be taken up but this will surprise you.
And there is no regulator that can be said...YES, this one. You have to assemble and listen to each and decide. Long process.

Gajanan Phadte

 

[DavidL]

Really? Regulators all sound different? Let's see that hard objective data to prove that. Please don't tell some one to start down that slippery slope of just "trying" one part after another thinking they might hear differences when in fact they aren't. it just wastes time and money and sanity.