The Well Tempered Master Clock - Building a low phase noise/jitter crystal oscillator

Status
Not open for further replies.
Hi Andrea,

Good news. A lot of things were made in less than one year... If I remember we started the "TDA1541 make it yourself" project in february or march !

The simultaneous delay mode managed by the XMOS is a good news if it can be done easliy !

Ah, an isolator is missing and the trade off is between now : a Wave I/O + I2Sto USB or an Amareno ! Money is the judge of peace for each of us 🙂.

But for the shortest path Amareno stacked with your clock and the Distinction-1541 is a great improvement

Amareno is at 70 euros bucks (soon at 1:1 with the USD !) at Audiophonics, if it can be updated by ourselves and the crystal removed by ourselves, the TCO (less the time we spent... it's a hobby !) is lowered as well?

You got a PM as asked...

Bravo,

Eldam

You dont even have to do anything to the XOs on Amanero. Amanero has a firmware version to slave to external master clock.
 
PCB

PCB preseries arrived, ready to build and test (sorry for the bad quality of the pictures).
Crystals shipment scheduled on March 23.
 

Attachments

  • TWTMC-C_pcb.jpg
    TWTMC-C_pcb.jpg
    252.7 KB · Views: 753
  • TWTMC-D&D_pcb.jpg
    TWTMC-D&D_pcb.jpg
    657.8 KB · Views: 733
  • TWTMC-DIL_pcb.jpg
    TWTMC-DIL_pcb.jpg
    61.2 KB · Views: 690
Laptech Sc-cut crystal quote

While I'm waiting for crystals delivery (shipment from Laptech scheduled on March 23), I got a quote from Laptech for some SC-cut crystals (attached).

With the 11.2896 MHz crystal (Q around 850K or greater) and a suitable Driscoll oscillator, that assure high loaded Q, one can expect a phase noise around -140dBc at 10 Hz from the carrier.
To be experimented in the next step (obviously with my very long times).
 

Attachments

Those look really interesting. The close in performance will probably be limited by the following:
1) Active device noise
2) Temperature stability
3) Vibration
Ensuring that the oscillator will be optimized will requre some very specific measurements.
Can I buy in on the sample run?
Time to explore some really low noise circuits. 2X and some varicaps to phase lock and you have the core of a phase noise measurement system capable of what the HP system can do.
 
😕
Quote:
11.2896 MHz sc-cut 3rd overtone
Frequency: ( 11.2896MHz ),
Cut mode: SC3
Package: HC43 Cold weld.
Resonant mode: fundamental with polish finish15 pcs m.o.q.

Now is it a fundamental mode crystal or third overtone??

😕😕😕

I'm enough sure it's 3rd overtone, I ask them for 3rd overtone mode at 11.2896 MHz (maybe they were wrong copying data from another quote). I ask Laptech to confirm.
 
Those look really interesting. The close in performance will probably be limited by the following:
1) Active device noise
2) Temperature stability
3) Vibration
Ensuring that the oscillator will be optimized will requre some very specific measurements.
Can I buy in on the sample run?
Time to explore some really low noise circuits. 2X and some varicaps to phase lock and you have the core of a phase noise measurement system capable of what the HP system can do.

Hi Demian,

I will ask Laptech for some samples (I'm interested also), BTW I believe the MOQ will be at least 5 pcs for each frequency also for sample. When I did order samples about two year ago, I ordered 5 x 11.2896 MHz fundamental and 5 x 33.8688 MHz 3rd overtone.

Andrea
 
A little more expensive than a Wenzel streamline. About 1/2 of a Wenzel ULN Bluetop. And around 10% of an Oscilloquartz BVA. Really high performance oscillators are very expensive. And 95% of the cost is in solving issues that probably do not matter for audio.
 
Can I buy in on the sample run?

There is no chance to get a few sample for SC-cut crystals.
Laptech revisited the quote, but MOQ is almost the same of the previous quote:
15 pcs for 11.2896 MHz and 10 pcs for 22/24/45/49 MHz crystals (see attachment).

I'm interested to get a few samples, such as 1-2 pcs for each frequency.
If there will be other members interested on these SC-cut crystal, to reach the MOQ, I can arrange the group buy.
 

Attachments

11.2896MHz SC cut

Hi Andrea,
I would buy one or two of the 11.2896MHz. It much cheaper than buying a OCXO in that frequency anyway. And the only one I could find readily available is the is the extremely expensive Pulsar which needs extra circuitry to crank up to 5V logic output.
 
SC-cut crystals interest list

I start a list for SC-cut crystal, please add yourself to the list if you are interested on.

As quoted by Laptech, MOQ are 15 pcs for 11.2896 MHz and 10 pcs for the other frequencies.

- andrea_mori: 2 x 11.2896 MHz + 2 x 22.5792 MHz + 2 x 24.576 MHz + 1 x 45.1584 Mhz + 1 x 49.152 MHz
 
Status
Not open for further replies.