How to build a 4.2336 MHz TCXO clock?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
B

bob91343

Member
Joined 2010
You don't need a TCXO. The temperature variation in the room isn't enough to make an audible difference in the operation of the unit.

So any oscillator will work. Yes you can divide a higher frequency but it's a lot of trouble. Why do you want to do all this? If the original oscillator is working, leave it alone.

And you can special order any frequency you like if you are willing to pay the setup charge.
 
N

noyan

Member
Joined 2004
bob91343 said:
You don't need a TCXO. The temperature variation in the room isn't enough to make an audible difference in the operation of the unit.

So any oscillator will work. Yes you can divide a higher frequency but it's a lot of trouble. Why do you want to do all this? If the original oscillator is working, leave it alone.

And you can special order any frequency you like if you are willing to pay the setup charge.
Click to expand...

Thanks for the answer!
I changed the 11.2896mhz crystal with TCXO clock unit in my Marantz CD80 and i have very pleasured the difference. That's why i intended to do the same way in CD104.
Noyan
 
amc184

amc184

Member
Joined 2007
Why do you want to do all this? If the original oscillator is working, leave it alone.
Click to expand...

To reduce jitter and power supply contamination. If you use one of the common TCXO kits on eBay, which is likely what noyan has done, you'll get a substantial improvement in sound quality. They're not the best way to go for a low jitter clock, but they are effective, and they're good value for money.

Noyan, there are two solutions to your problem:

- Buy a TCXO kit off eBay that includes a divider. This one here will do the job, use its MCK/4 output.
- Build something like a Kwak clock with the original crystal out of the player. It might cost more, and it requires more effort, but the results will be better than the first option.
 
N

noyan

Member
Joined 2004
amc184 said:
To reduce jitter and power supply contamination. If you use one of the common TCXO kits on eBay, which is likely what noyan has done, you'll get a substantial improvement in sound quality. They're not the best way to go for a low jitter clock, but they are effective, and they're good value for money.

Noyan, there are two solutions to your problem:

- Buy a TCXO kit off eBay that includes a divider. This one here will do the job, use its MCK/4 output.
- Build something like a Kwak clock with the original crystal out of the player. It might cost more, and it requires more effort, but the results will be better than the first option.
Click to expand...

Thank you!
My CD80 clock mod is here: http://www.diyaudio.com/forums/digital-source/171384-marantz-cd80-clock-installation.html

Yes, i saw this clock unit in ebay :) Actually i want to build my clock myself like in Marantz CD80. Therefore, i need a schematic of simple frequency divider. If i can not, i will buy.
Noyan
 
amc184

amc184

Member
Joined 2007
This site here shows some example divider circuits. You'll want to use two divide-by-two configured flip-flops in series.

That said, if you're keen to build it yourself (and you seem to be), build a Kwak clock instead. You'll get a far better result than with a TCXO with a divider.
 
amc184

amc184

Member
Joined 2007
Actually, that's just a conservative guess. I use a clock closely related to the Kwak clock, and I've used it all the way up to 33.8688MHz. I've only previously used it as low as 11.2896MHz, but tonight I plugged in a 4.2336MHz crystal into it (one of my clocks has a crystal socket), and it oscillated just fine, a nice, clean TTL square wave.

As for the actual limits of the design, I don't know what they are. When I had the clock running at 33.8688MHz the rise time was starting to slow, though this could have partly been my 'scope.
 
N

noyan

Member
Joined 2004
amc184 said:
Actually, that's just a conservative guess. I use a clock closely related to the Kwak clock, and I've used it all the way up to 33.8688MHz. I've only previously used it as low as 11.2896MHz, but tonight I plugged in a 4.2336MHz crystal into it (one of my clocks has a crystal socket), and it oscillated just fine, a nice, clean TTL square wave.

As for the actual limits of the design, I don't know what they are. When I had the clock running at 33.8688MHz the rise time was starting to slow, though this could have partly been my 'scope.
Click to expand...

Very interesting! Thank you very much for this good news :) . I will build it.
Noyan
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom