UPGRADING THE CRYSTAL (X0) ON ESSENCE STX

[thoppa]

Quote:

Originally Posted by b0dhi /img/forum/go_quote.gif PPM has nothing at all to do with jitter.

I've been told the exact opposite by several ppl, one of whom is an EE who is now a director of a PSU design company here in Asia. I've been told the ppm variation in the XOs accuracy is both long term and short term, the latter especially with temperature and voltage fluctuations, and is a major factor in jitter. I also did a lot of background reading about this so I'm quite skeptical about your assertion. Please explain in detail ! I don't care either way as I have no partisan opinions here. But I do like verifiable facts, not short little posts with no real info please. Thanks.

Quote:

Originally Posted by b0dhi /img/forum/go_quote.gif There are very high ppm oscillators with very low jitter and vice versa.

This is news to me - I'm very interested ! Got any examples ? Datasheets etc ?

Thanks,

Tom

 

[Pars]

Quote:

Originally Posted by thoppa /img/forum/go_quote.gif I've been told the exact opposite by several ppl, one of whom is an EE who is now a director of a PSU design company here in Asia. I've been told the ppm variation in the XOs accuracy is both long term and short term, the latter especially with temperature and voltage fluctuations, and is a major factor in jitter. I also did a lot of background reading about this so I'm quite skeptical about your assertion. Please explain in detail ! I don't care either way as I have no partisan opinions here. But I do like verifiable facts, not short little posts with no real info please. Thanks. This is news to me - I'm very interested ! Got any examples ? Datasheets etc ? Thanks, Tom

Guido Tent of Tentlabs would beg to differ
XO and VCXO

 

[b0dhi]

Quote:

Originally Posted by thoppa /img/forum/go_quote.gif This is news to me - I'm very interested ! Got any examples ? Datasheets etc ?

One of many examples: http://www.valpeyfisher.com/openlink...s/XO/VF140.pdf

Here's another: http://www.valpeyfisher.com/openlink...XO/VFXO321.pdf

50ppm, yet 0.09ps jitter.

There are also different ways of measuring jitter. You need to be careful about whether it's integrated or peak, what the integration band is, RMS, etc. Phase noise in dB at 10Hz is usually a good indicator but most manufacturers don't supply this value.

 

[thoppa]

He says low ppm error in an XO is no guarantee of low jitter, he doesn't say one doesn't affect the other. He also says there is no point in buying a TCXO, yet I have used one of his XOs in an SACD and a TCXO in a DAC and both proved excellent.

Thanks for the datasheets - interesting. It seems it is a must to check specs eh ?

Thanks.

 

[b0dhi]

Just ran across this by coincidence: DIYHiFi.org • View topic - PCM2707 - Crystal or Clock

Follow the thread and it talks more about it.

 

[thoppa]

It's interesting. One guy says some ppl call the ppm measurement a "dc" jitter but he thinks it is just poetry.

I thought I had bookmarked the teaching articles about this but I can't find the link now. Sorry.

The Vanguard TCXO has 1ppm and -125dB phase noise at 1Khz - low ppm and low jitter.

 

[thoppa]

My TCXO is working !

My thanks to Gworld ( Marco ? ) - your earth connection is what did it. THANKS !

You're the man !

I was using an earth from the expansion board pins on the ST and it should have worked but it didn't. I have no idea why. I had tried some different resistors, thinking it must be something to do with the circuit, but no joy.

Today, I removed the original crystal again, and this time the solder pad came with it. Oh dear.

The track was still there so I cleaned that and tested if I could get a good connection off it, and I could, so no problem.

I connected up my TCXO again this time using the earth on the nearby chip. I actually prayed earnestly too. And......success ! The power of prayer.....over poor soldering.....haha...so thanks to God too. We all need a little divine intervention every now and again....

The board looks a bit of a mess now but I couldn't care less because the sound quality change was obvious. First thing that took me by surprise was the rendition of detailed treble. Then I noticed the soundstage is improved. Then I noticed better bass definition. And that was all within about 30 seconds. It's an obvious change, and to me, quite a large one.

I certainly think this isn't one of this things that some people might notice and others might not. I honestly think everyone with even half-decent cans will hear this and, for those who haven't heard what low jitter can do, you will be surprised. Really.

For those of you who have had an STX and then an ST, you heard an improvement yes ? I'd say you'd hear the same improvement again but an order of magnitude higher. Say goodbye to mush. Say hello to much !

I have to say this is definitely a worthwhile upgrade (if you hadn't already worked that out....)

And here are pictures of my awful soldering. It needs tidying...but I don't want to pull the card out to do it because I'm enjoying the sound sooooo much.

 

[GWorlDofSPACE]

Nice to hear that. I am Marko. I told you the crystal upgrade is something major. You immediately notice the sound improvement. The card is sounding a new dimension.

I am thinking to open a new thread for discussing the sound quality of essence in comparison to all other DACs out there! I was thinking what would be the next step. This sound card is sounding really good and the question is how much more do I need to spend to have better sound.

 

[thoppa]

Hi Marko,

Yes, major is exactly right ! I couldn't have done it without your photos so thank you again.

Did you sort out a power supply ? I got some of my parts today to build one but the high-spec LM317AT are back-ordered so I can't do any more to improve the card at the moment. Do you really think you can get more from this card ? You already have so many changes, maybe only the power is left ? I'll be posting about this on the power supply thread :

http://www.head-fi.org/forums/f46/ma...ssence-426049/

Anyway, of the changes you made, what are your top three (in order...!). I'm sure people will be interested to know what they should upgrade first and how much it costs.

 

[ROBSCIX]

So the module was fine and it was a faulty connection? -I am with you on that point though, usually a ground is a ground is a ground. Weird that it wouldn't be a ground on the pin out.

Great that you got it working and are enjoying the results. I thought it might not make that much of a difference because of the clean up circuit but I guess I was mistaken as you said you notice immediate sonic gains. After more thought I guess it make sense that the cleaner the clock to begin with the cleaner the signal the clean up chip could produce. Anyway, great stuff. Enjoy.

 

[GWorlDofSPACE]

Quote:

Originally Posted by thoppa /img/forum/go_quote.gif Hi Marko, Yes, major is exactly right ! I couldn't have done it without your photos so thank you again. Did you sort out a power supply ? I got some of my parts today to build one but the high-spec LM317AT are back-ordered so I can't do any more to improve the card at the moment. Do you really think you can get more from this card ? You already have so many changes, maybe only the power is left ? I'll be posting about this on the power supply thread : http://www.head-fi.org/forums/f46/ma...ssence-426049/ Anyway, of the changes you made, what are your top three (in order...!). I'm sure people will be interested to know what they should upgrade first and how much it costs.

The PSU of Kingwa will not work. But I think to buy the board of q11, or two. I will discuss this on other thread.

I will not count which one is the first to do. You should do them all. The problem becomes how to rate the sound quality.

 

[thoppa]

Quote:

Originally Posted by ROBSCIX /img/forum/go_quote.gif So the module was fine and it was a faulty connection? -I am with you on that point though, usually a ground is a ground is a ground. Weird that it wouldn't be a ground on the pin out. Great that you got it working and are enjoying the results. I thought it might not make that much of a difference because of the clean up circuit but I guess I was mistaken as you said you notice immediate sonic gains. After more thought I guess it make sense that the cleaner the clock to begin with the cleaner the signal the clean up chip could produce. Anyway, great stuff. Enjoy.

Well, you see from my soldering that it was probably human error right ? And then I must have been checking the circuit wrong. Um. Still, it shows that if I can install an XO, anyone can !

I did go back to the shop and they looked at me with 'great patience' when I asked them to check the XO. It seems they think these things are very reliable.

 

[ROBSCIX]

I understand totally. It is great that it turned out to be such a good upgrade for you. Makes the time and effort worth it in the end.

 

[thoppa]

Quote:

Originally Posted by GWorlDofSPACE /img/forum/go_quote.gif The PSU of Kingwa will not work. But I think to buy the board of q11, or two. I will discuss this on other thread. I will not count which one is the first to do. You should do them all. The problem becomes how to rate the sound quality.

Yeah, lots of mods makes it harder to know which one was best right ? I tried replacing the power caps for the TPA6120 with SilmicII and the difference was very slight - easily passed off as imagination by some ppl - just slightly smoother vocal and better definition at the extremes of the frequency range. Changing caps is hard too. I might do the ones on the op amps if I find myself with nothing better to do one day.

 

[mojave]

The two XO's (Audio-GD and Vanguard) changed in this thread are both different. Is there a benefit to using one instead of the other? It looks like the Vanguard is the easiest to install. Would the Tent Labs XO also work for this application?