kwak clock issues

Status
Not open for further replies.
Jump to Latest
#43
The difference is (how I look at it)

1 - what you like
2 - what is most accurate

There is a bunch of people that like the typical tube sound of badly designed tube amps, because it gives a big cloud of spacious sound. I don't like that.

For the same reason some people don't like the sound of my filament modules, because it takes away colorations that they are used to.

I can make oscillators (by adding specific amounts of jitter) that will please everyone, but won't help in the search for accuracy.

It is a matter of taste, and what you're after. The ear is enough in case 1, but one needs more than that in case 2.
Click to expand...

I must say dead right!

The trick is to talk and understand what individual wants and likes, but to be able to produce accurate digital reproduction that only low noise / jitter will give. We all have different tastes and the main thing when it comes to modifications is to understand what one likes. Specific amount of jitter (different amount at different spectrum) can definitely colour (tailor) the sound to individual needs.

Regards,
Extreme_Boky
 
#44
I listen to various music, on various recordings

and not to one particular song or recording. So therefor I want a set that re-produces the live like sound for all those recordings. From that standpoint Guido's statement that he is trying to realise the most accurate (neutral) solution. As SY mentions the brain can be easily fooled and I therefor am convinced that listening to music is a learning process (like learning a language) and listening to a set that reprodices the music in a coloured way (non-neutral) can't be corrected by the listener (i.e Elso) without
1) someone else (that has larned listening to music, through un-coloured way)
2) measurements of aspects that reduce/add coloration to the sound.

my 2 cents

Henk
 
#48
Well, finally a gif of my kwak layout with divider.
(have been solving motorboating problem in pre-amp with succes in the meantime)

Two screenshots combined, that why there is an edge off.
 

Attachments

  • kwak_printje.gif
    kwak_printje.gif
    58.4 KB · Views: 577
#49
Re: completely useless remark?

Elso Kwak said:


Geeh Glassman did design that board. 🙄

Well I was hoping you could comment on it with ur general guide lines in mind....no?.......:bigeyes:
Click to expand...



Hello Elso

I would just like to add to Guido's comments , firstly that the placement of the decoupling capacitor at the gnd pin versus the supply pin is not just the opinon of Guido, and it is not a minority opinion, if you care to look elsewhere on the net you will also find people who support the idea that a decoupling capacitor should be connected to the ground pin of the device first with a track going to the VCC pin. This approach gets rid of ground bounce, doing it your prefered method creates ground bounce which is a far more serious problem than the other way round.


Regards
Arthur
 
#50
Re: Re: completely useless remark?

PHEONIX said:




Hello Elso

I would just like to add to Guido's comments , firstly that the placement of the decoupling capacitor at the gnd pin versus the supply pin is not just the opinon of Guido, and it is not a minority opinion, if you care to look elsewhere on the net you will also find people who support the idea that a decoupling capacitor should be connected to the ground pin of the device first with a track going to the VCC pin. This approach gets rid of ground bounce, doing it your prefered method creates ground bounce which is a far more serious problem than the other way round.


Regards
Arthur
Click to expand...
Really Arthur?, and the inductance of the long narrow meandering ground track does not give ground bounce?

🙄 🙄
 
#51
tubee said:
Well, finally a gif of my kwak layout with divider.
(have been solving motorboating problem in pre-amp with succes in the meantime)

Two screenshots combined, that why there is an edge off.
Click to expand...


Hi Tubee, I kind of like it!
You could elongate the ground plane from the decoupling cap at pin 1 of the IC to fill the empty space in the center.
I also like the symmetrical way you layed out the powersupply.
But the rrack from +5V to the pot is dangerously close to the crystal, might pick up the oscillation.l
What's the size of ur PCB?
:angel:
 
#54
You could elongate the ground plane from the decoupling cap at pin 1 of the IC to fill the empty space in the center.
I also like the symmetrical way you layed out the powersupply.
But the rrack from +5V to the pot is dangerously close to the crystal, might pick up the oscillation.l
What's the size of ur PCB?
Click to expand...

Thanks Elso for comments, and to speak as with Guido: we all learn from this.

What ic you are you referring to, the 16 pin divider? Or the comparator? Can i connect the MAX913 to the ground-plane above it, or do I get a ground loop then?

Matter of fact: At the divider chip: pin 1 is pin 16, i had to combine a SMD chip with a DIL chip, did it on the PC with TCI for the first time, after adding the divider realising its the component-side you see when drawing in TCI !

I see the track to the pot is indeed near cristall. The center tap of pot is decoupled, is that not enough?

I apply usual (with p2p) a central ground meandering through all parts about the middle: short lines to earth. With this PC design there is some shielding also without using a doublesided pcb. The divider will get a copper foil shield above it also.
Divider gets 5V from small 78L05 & bead. The PCB is 10 by 6.5 cm.
 
#57
About the supply resistor of pot along cristal: Maybe its beneficial to ground the housing of cristal in this case, or put a small copper screen between resistor and cristal. This last solution seems a somewhat messy way to me but if it helps its fine.
 
#58
Re: completely useless remark?

Elso Kwak said:


Geeh Glassman did design that board. 🙄

Well I was hoping you could comment on it with ur general guide lines in mind....no?.......:bigeyes:
Click to expand...


I could, but why would I ? I wrote down my guidelines, so everyone can check and comment accordingly, unless they don't agree with my philosophy. Their pick.

And Elso, honestly, it would be nice if you react on content and arguments rather than avoiding the fundamental discussion.
 
#59
Re: Re: completely useless remark?/ Ground Bounce

PHEONIX said:
Hello Elso
I would just like to add to Guido's comments , firstly that the placement of the decoupling capacitor at the gnd pin versus the supply pin is not just the opinon of Guido, and it is not a minority opinion, if you care to look elsewhere on the net you will also find people who support the idea that a decoupling capacitor should be connected to the ground pin of the device first with a track going to the VCC pin. This approach gets rid of ground bounce, doing it your prefered method creates ground bounce which is a far more serious problem than the other way round.
Regards
Arthur
Click to expand...

http://en.wikipedia.org/wiki/Ground_bounce

and references given there
and:
http://www.home.agilent.com/cgi-bin..._KEY=199279&LANGUAGE_CODE=eng&COUNTRY_CODE=US

http://www.eetchina.com/ARTICLES/2003OCT/PDF/2003OCT27_ST_HBM_AN.PDF
😎
 
#60
Status
Not open for further replies.
Top Bottom