Right now I’m gambling with idea to build ultimate CD transport build on Philips CD Pro2 transport. One of the first things is change original 8.4672 MHz crystal oscillator.
Rubidium master clock is used in ultra stable application such as GPS satelite modules, labolatory clock or GSM cells.
Operational temperature is 50C. input about 50W first 5 minutes than cca 15W.
SA.22c Precision Rubidium Oscillator
– Phase noise (@10 MHz):
1 Hz <-72 dBc/Hz
10 Hz <-90 dBc/Hz
100 Hz <-128 dBc/Hz
1 kHz <-140 dBc/Hz
10 kHz <-148 dBc/Hz
• Jitter: <10 ps RMS
An Rb clock may have amazing long term stability, but for CD playback you are looking for shorter term jitter, and in this they are VERY poor. These chips only address higher frequency offset phase noise
These are 10MHz Rubidium standards with frequency adjusted or ‘pulled’ to 11.2896MHz.
At their native 10MHz frequency, the phase noise specified is pretty good. By pulling the frequency to 11.2896 there would
be some degradation in phase noise but I don’t know how much.
They spec phase noise of -100dB at 10Hz from carrier at 10MHz and -125dB at 100Hz from carrier. This would be better than most
Oscillator (OCXO) that is fixed 11.2896MHz and specs -120dB at 10Hz from carrier which is 20dB better (10x) than
these Rubidium oscillators at their native 10MHz frequency. An OXCO standard can have very low phase noise such as -115 dBc/Hz @ 1 Hz offset with a -174 dBc/Hz floor noise.
The best so far I find is Tent labs XO http://www.tentlabs.com/Components/XO/index.html
- One of the most cons which i find is that the aboslute frequency isn’t so neccesary.
- It is not necessary to have long term stable freqency (low excitation = high LF jitter)
- It is not necessary to have temperature stabilisation for TCXO (outer case), more important is vibration dumping.