OCXO 16.9344MHz LOW JITTER CLOCK

[Moddang]

Originally Posted by canvas

Mine is a custom made 33.8688M OCXO. Usually, the manufacturer will not just make one for you. They need to make a batch of 10 or 20, so I waited for two months. OCXO is a bit expensive (to me), but still cheaper than those hi-end cables.

How many do you have in your stock?
I have some of MV14V1CD it's 16.9344 VCXO. Can we exchange? Probably you can send the private message to me if you are interested in about this.

[canvas]

Moddang,

I only keep one. Yet, I've seen someone is selling OCXOs in this forum. Crystal is extremely sensitive to power, so you need a very clean power for OCXO. This is more important than the OCXO itself.

[davidsrsb]

OCXOs are potentially noisier than a room temperature XO as they regulate to a constant high temperature.
Cheap module oscillators have horrible internal circuits. Variations on cmos gate circuits have terrible phase noise.

[Moddang]

Originally Posted by canvas

Moddang,

I only keep one. Yet, I've seen someone is selling OCXOs in this forum. Crystal is extremely sensitive to power, so you need a very clean power for OCXO. This is more important than the OCXO itself.

Yeb, i have my own version of ultra(may be) low noise regulator. It just simple discrete circuit (no new technology) but components are not simple . For conventional XO, TCXO, VCXO or what ever if the gaurantee specification of maximum jitter (period, cycle-cycle, or whatever) from vendor is 1pS RMS i think it's OK and everthing is upto the performance of PSU and bypass capacitor.

Moreover than that the real performance is upto PCB layout and circuit designed. I see many people try to reclock by using Flip-Flop and it's not my cup of tea in the real work it's not operate as ideal, may be added more jitter or skew? Do we concern about clock uncertianties?. I don't know.

For me if jitter is less than 50-60pS it's OK because i'm a very bad sound perception guy