Influence of PS on clock modules

Status
Not open for further replies.
Jump to Latest
#23
Re: Crystal???

Elso Kwak said:


Fmak,
The SMT thingie is not a crystal but a complete crystal-oscillator ia a very small and thin package. If you use the screw at the right side of the laserunit on the audioboard you can place the oscillatorboard very close to the original SMT-oscillator and use a short < 5 cm wire to connect the clock.
😎
Click to expand...
--------------------------------
I know it's an XO; Jean-Paul is worried by stray and I am pointing out that 0 cm is better than 5 cm.

I don't believe that Jitter is just a function of the XO itself but an integral part of the player design. The XO contributes part.

If you don't know anything about the Sony XO, why did you say it was no good?
 
#24
Re: Clock

Elso Kwak said:
Hi, The gold coloured SMD clock module in the Sony SCD-1 is a big joke.
The LCaudioclock is much better but lacking bass...
Jitter specs tell you nothing about the sound...
I estimate powersupply is 20% responsible for the sound of the clock, roughly...
:devilr:
Click to expand...

dear Elso

Jitter specs do tell you about the sound, but not all. This is where the spectrum of the jitter comes in.

I think generalising the PS influence is slightly dangerous, as some soscillators are much more susceptible for the supply than others. I measured differences over 30 dB. In addition, the supply rejection depends onthe frequency as well.

best regards
-
 
#25
Re: What is more important,crystal or power suppy

Jesús Puerto said:

For example I bougth a Crystal of 2ppm in the Spain and the cost was
abaut 100$ and I have another with
less of 1ppm that is to more expensive
For this reason I´m agree with Elso
and I don´t belive tha quality of the
Sony put in your CD.

Jesús Puerto
Click to expand...

Hola Jesús

The ppm value often relates to the absolute frequency, but does not inform you about the jitter (only the very low frequent "day to day" jitter but we are not affected by that)

best regards
 
#26
Re: Re: Crystal???

fmak said:

--------------------------------
I know it's an XO; Jean-Paul is worried by stray and I am pointing out that 0 cm is better than 5 cm.

I don't believe that Jitter is just a function of the XO itself but an integral part of the player design. The XO contributes part.

If you don't know anything about the Sony XO, why did you say it was no good?
Click to expand...

Hi

I measured on various wiring and I can sleep safely with 5 cm twisted wiring.

You are very right about the "integral part" statement. Now some problems can be improved by (e.g.) improvng the supply

Other problems can be taken away b removing the crosstalk to that smae supply. For that reason it often helps to seperately supply all servo's (actuator, radial, focus and spindle) in the player, as they induce huge LF currents (Hz - kHz range) in the (often poorly designed) groundplanes.

Obviously you see that back in the jitter 🙂

all the best, to sleep now !
 
#27
Guido beat me to it.....

Yes, ppm is a measure of stabilty. It is a function of how clean the inside of the crystal can is. Impurities cause the frequency to drift over time.

Jitter is a whole other subject. And yes, the absolute value, i.e., pS, tells you little without knowing the spectra of the jitter.

I see where one well known oscillator company says that their XOs have only a few pS of jitter, but if you look closely at the data sheet, they say that it is for jitter frequencies >1 kHz. Which are probably easier to control than those around 60 Hz.

A crstal company that I have spoken to claims that jitter below 1 kHz is a function of the crystal, and the circuit; whereas jitter above that is a function of parts selection.

Jocko
 
#28
sony..

btw..i agree - the sonys do not sound good, can't help it, i think its that single/multi-bit thing....and i have actually owned several of these sony's xa7es and 777 sacd player which btw is identical to the scd-1 exept for the balanced out and the boxed trafos...
 
#29
Re: sony..

tbla said:
btw..i agree - the sonys do not sound good, can't help it, i think its that single/multi-bit thing....and i have actually owned several of these sony's xa7es and 777 sacd player which btw is identical to the scd-1 exept for the balanced out and the boxed trafos...
Click to expand...
------------------------------------------------

The SCD1 needs mods to sound best, such as removal of ground loops (that thick copper plate) and opamp changes. The XA7ES (are you sure?) has no relationship to the SCD 777. You need to do the changes step by step to see what is wrong. Just wrenching out caps and clocks is not the best way of doing it unless you have a lot of money to throw.

Many people go for Philips/Marantz whose sonics are controlled by the output caps and then talk about the role of the DACs and filters without isolating them. The quality of Philips boards is apalling-you lift a track when you 'touch' it. The drive is also cheap!!!
 
#30
Many people go for Philips/Marantz whose sonics are controlled by the output caps and then talk about the role of the DACs and filters without isolating them. The quality of Philips boards is apalling-you lift a track when you 'touch' it. The drive is also cheap!!!
Click to expand...

A lot of Sony players have output caps as well and some Marantz/Philips types don't have them ( CD 80 for instance ). And their role is not very dominating, but replacing them is a must when you want it to sound good. Best is to redesign the output stage and to build it discrete.

PCB tracks of the newer Sony types lift of too when you touch them. I am not saying Philips/Marantz oldies are the best you can find but they're good candidates for modding with very good results in the end for a price that can't be beat. I have a Marantz CD65 II that beats a lot of expensive Sony's with ease. Unmodded it sounded better than a 940 SACD cdplayer with normal cd's. I am not so sure which of the two brands makes the cheapest drives but my guess is that the difference is small.

There is no substitute for CDM 1/4/9 swingarm-mechanics however. 4 mm dropouts on CD's are accepted without skipping. CDR and CDRW playback is no problem, even with the 1985 types ! I agree on the quality of the newer CDM 12 and VAM mechanics. Although called Pro their quality is not the same as the older Philips ones.
 
#31
Re: What is more important, crytal module or popwer supply

fmak said:
Most people who tweak seem to spend large sums of money replacing clocks with £200 clock modules and power supplies. Often, the original clock modules seem top be high quality - the one in the Sony SCD1/777 actually measures about 3ppm. They all seem to be fed from 7805 type of regulators.
----------------------------------------------------------------
I have now partially answered my own question. I have just modified a 777ES for someone, putting in a Tent module powered by a VSL lo-noise supply. The mod DID not affect the sound out of the Sony's audio board signficantly, and sounds close to the unit before the mod. This is in what I believe to be a reasonably soundly designed CDP.

I then used the Sony's CO chip, connected it to a BC550/TL431 supply, with decoupling capacitor determined on a scope, and fitted it to a SCDXB 940. This unit has a defect in design in that there IS a lot of hf noise in the system. The unit has previously been modifified with changes in the I/V opamp to the AD825 and coupling cap to BGN, as well as shielding and digital section decoupling improvements.

Before the new clock, the sound was somewhat rough. With the 777's XO replacing a simple resonance type of crystal, the sound improved by a large margin. The roughness is gone and bass and timing much improved.

Conclusions:

777 XO not so bad all. When my Tent module and Euroquartz module arrive, I shall be able to judge the differences amongst all three!

Don't waste £200 on expensive clocks unless one knows where the problems are, and don't generalise.
Click to expand...
 
#32
Re: Re: What is more important, crytal module or popwer supply

fmak said:

Before the new clock, the sound was somewhat rough. With the 777's XO replacing a simple resonance type of crystal, the sound improved by a large margin. The roughness is gone and bass and timing much improved.

Conclusions:

777 XO not so bad all. When my Tent module and Euroquartz module arrive, I shall be able to judge the differences amongst all three!

Don't waste £200 on expensive clocks unless one knows where the problems are, and don't generalise.
Click to expand...

Hi,

Happy to see your confirmation. Low noise is the key, also in power supplies.....

enjoy
 
#33
Re: Re: Re: What is more important, crytal module or popwer supply

Happy to see your confirmation. Low noise is the key, also in power supplies.....

-------------------------------------------------------------

I now have further evidence. Using a TL431 series pass regulator, I have found it beneficial to optimise the ps for lowest noise in situ. I find that I am ending up with >150 uf OSCon on the output to the clock, 47 uf across the TL431, and 100 uf OSCon across the input inductor. Substituting OSCons for lytics result in quite large increases in noise voltages. This is expensive due to the cost of the OSCons but for a high end system may be worth it.

The result is a more relaxed presentation of music.

Anybody else have experience of this?
 
#36
Substituting OSCons for lytics result in quite large decreases in noise voltages.
Click to expand...

Maybe a stupid question but you substitute OSCONs for lytics. You mean that you remove the lytics and solder in some OSCON's. But the way I would write it would be like this:

Substituting lytics for OSCONs result in quite large decreases in noise voltages.

Because the OSCONs are a substitute for lytics. Am I right ?

Just want to know how to write this right in english, no criticism whatsoever.

Regards,

Jean-Paul
 
#37
jean-paul said:


Maybe a stupid question but you substitute OSCONs for lytics. You mean that you remove the lytics and solder in some OSCON's. But the way I would write it would be like this:

Substituting lytics for OSCONs result in quite large decreases in noise voltages.

Because the OSCONs are a substitute for lytics. Am I right ?

------------------------------------------
No, the other way round, sorry.
Click to expand...
 
#39
Re: Re: Re: Re: What is more important, crytal module or popwer supply

fmak said:

I now have further evidence. Using a TL431 series pass regulator, I have found it beneficial to optimise the ps for lowest noise in situ. I find that I am ending up with >150 uf OSCon on the output to the clock, 47 uf across the TL431, and 100 uf OSCon across the input inductor. Substituting OSCons for lytics result in quite large increases in noise voltages. This is expensive due to the cost of the OSCons but for a high end system may be worth it.

The result is a more relaxed presentation of music.

Anybody else have experience of this?
Click to expand...

Fmak,
Are you referring to this circuit by Cuno Snoeren?
http://www1.tip.nl/~t708955/schem/xoschema.jpg

😕
 
Status
Not open for further replies.
Top Bottom