Hifidiy clock part value - Page 2 - diyAudio
Go Back   Home > Forums > Source & Line > Digital Source

Digital Source Digital Players and Recorders: CD , SACD , Tape, Memory Card, etc.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 13th March 2012, 02:57 PM   #11
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
Anything that effects the waveform can effect jitter. Changes in ringing effect transitions and therefore effect jitter.

Jitter can effect audio quality.

That said, listen to it and see if you like it.
  Reply With Quote
Old 13th March 2012, 03:12 PM   #12
diyAudio Member
 
schubert's Avatar
 
Join Date: Sep 2008
Location: Pennsylvania, USA
I liked it in the last CD player I put it into - this will be the second and third!
  Reply With Quote
Old 13th March 2012, 07:18 PM   #13
marce is online now marce  United Kingdom
diyAudio Member
 
Join Date: Jun 2007
Location: Blackburn, Lancs
Yes, clocks are the heart of a digital circuit, they control everything.
The reason why I asked is on some recent analogue/digital boards for communication we tested all the clocks and found quite severe overshoot, this was using the 'standard' 33R resistor. We used the signal Integrity Software to determine the value of resistor to use (and also did real life experiments) and found for this clock an 82R was the best match to the layout. This was not a suoper high speed clock, but 12MHz going to two codecs down less than 25mm of PCB trace, impedance controlled PCB 14 layers, every signal next to a contigous ground plane. I have some plots when I can find them, for different resistor vaklues from 22R to 120R. Ringing is a problem and should be reduced if possible, it also stresses the silicon so is best avoided. With a lowish frequency clock and a scope it can easily be seen, and matching the resistor to your interconnection will reduce the ringing.
For digital signals, especialy with todays high rise times impedance matching is critical, it is the rise time of the signal that determines the highest frequecies harmonics of the digital signal.
The very least use twisted pair cable as stated above, the signal needs to be in intimate contact with its return path, and the two wires should join the board as close together as possible, or use co-axial cable, as reccomended for the Tentlab clocks.
  Reply With Quote
Old 13th March 2012, 07:22 PM   #14
diyAudio Member
 
schubert's Avatar
 
Join Date: Sep 2008
Location: Pennsylvania, USA
Hmmm......I may try to scope this myself, as I can borrow an oscilloscope from the physics department......I hope I can figure out how to use it! I may be back for help with that, as I've never used one before. Have to get the parts ordered first, though - I'll include a few extra values for the 33R resistor.
  Reply With Quote
Old 13th March 2012, 08:19 PM   #15
marce is online now marce  United Kingdom
diyAudio Member
 
Join Date: Jun 2007
Location: Blackburn, Lancs
Trace was actualy only about 9mm long.
Notice how the rise time decreases as the resistor value increases, this reduces the high frequency harmonics. Going higher than 82R had little extra effect on the ringing but reduced the rise time further.
Attached Images
File Type: png 22R_Term_AP.png (41.6 KB, 117 views)
File Type: png 82R_Term_AP.png (36.5 KB, 114 views)
  Reply With Quote
Old 23rd April 2012, 12:34 PM   #16
diyAudio Member
 
schubert's Avatar
 
Join Date: Sep 2008
Location: Pennsylvania, USA
Quote:
Originally Posted by marce View Post
Yes, clocks are the heart of a digital circuit, they control everything.
The reason why I asked is on some recent analogue/digital boards for communication we tested all the clocks and found quite severe overshoot, this was using the 'standard' 33R resistor. We used the signal Integrity Software to determine the value of resistor to use (and also did real life experiments) and found for this clock an 82R was the best match to the layout. This was not a suoper high speed clock, but 12MHz going to two codecs down less than 25mm of PCB trace, impedance controlled PCB 14 layers, every signal next to a contigous ground plane. I have some plots when I can find them, for different resistor vaklues from 22R to 120R. Ringing is a problem and should be reduced if possible, it also stresses the silicon so is best avoided. With a lowish frequency clock and a scope it can easily be seen, and matching the resistor to your interconnection will reduce the ringing.
For digital signals, especialy with todays high rise times impedance matching is critical, it is the rise time of the signal that determines the highest frequecies harmonics of the digital signal.
The very least use twisted pair cable as stated above, the signal needs to be in intimate contact with its return path, and the two wires should join the board as close together as possible, or use co-axial cable, as reccomended for the Tentlab clocks.
Well, it's done - I followed marce's recommendation of an 82-ohm output resistor, and the connection was made with a twisted pair a little over an inch long. It sounds fantastic! The improvement in my RCD-971 was significant. For around $50 (mostly the crystal and shipping) you just can't beat this clock.
  Reply With Quote
Old 24th April 2012, 05:52 AM   #17
sesebe is offline sesebe  Romania
diyAudio Member
 
Join Date: Jan 2011
Location: Iasi
Default high jitter?

Hi,

I do not know for what you search, but the ignal is bad, realy bad. I do not understud exatly how influence the jitter the quality of audio signal but the clock that you show it is not good.
First, what type of probe you use? A standard 1/10 probe? This type of probe usualy generate a smal rinding in point of conection for high spead signal, the ~9pF can resonate with parasit elements of circuit. 9pF will change the rise/fall time of signal (for high spead you need to use an active probe low capacity <1pF or a low impedance adapted probe. I see that the scope is switched to 50 ohm imput but it is adapted?
Anyhow if you have a look to statistic measurement made by the scope you can see that STD DEV it is ~3 time biger comparing with the measured values. It must be, for a good measurement, at least 100 time smaller. That means that the measurement is made in a wrong way or the values have a very big dispersion, in this case for me equivalet with a very noissy signal, and probably a high jitter.
  Reply With Quote
Old 25th April 2012, 06:25 PM   #18
marce is online now marce  United Kingdom
diyAudio Member
 
Join Date: Jun 2007
Location: Blackburn, Lancs
The scope shots are the actual signal, these are from actual high speed 14 layer boards. The ringing is due to the current drive of the clock buffer IC.
Probes Tek active probe, cant remember as these are from 3 years ago Same board had DDR memory as well as other very high speed interfaces. All signal were checked with a 13GHz scope, and results confirmed with simulation software, including the modelling the scope probe in the simulations. The scope used in these shots was only a 2.4GHz. Also this is not from a DIY project. It is to illustrate how fast rise times can affect the signal and how the series resistor has to be matched to the line that is being driven. This is not an uncommon problem with digital design these days, the fast rise times of even the most basic logic is what gives the spectural content of the square wave.
when you look round digital signals these days that one is pretty good, have a look at DDR memory waveforms to see how un-square digital can be.
  Reply With Quote
Old 26th April 2012, 02:20 PM   #19
sesebe is offline sesebe  Romania
diyAudio Member
 
Join Date: Jan 2011
Location: Iasi
Hi Marce,

Not the shape of the signal is bad but the measurements made by the scope.

The Std Dev must be as low as posible for a good and stable measurement/signal. In you screenshot this it is very big. The shape of the signal looks OK.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
hifidiy.net hongrn Chip Amps 11 19th December 2011 03:23 PM
New to hifiDIY Tape Echo Introductions 2 6th December 2010 09:14 PM
HiFiDIY A20 mikecell009 Group Buys 0 18th May 2010 05:52 AM
Help for K.clock part Bassivus Digital Source 2 1st November 2006 09:59 AM


New To Site? Need Help?

All times are GMT. The time now is 03:28 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2