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

Status
Not open for further replies.
See what I mean? This is a perfect example of what I am trying to convey.

You claimed for months now that the broadband noise of your oscillator is irrelevant (since you were shown better oscillators in this respect, and you also claimed the broadband noise degradation of the sine to square converter (9-12dB in your case) is also irrelevant), and it is only the 1Hz close in phase noise that matters (for audio) and now it appears the 1Hz close in phase noise is no longer that important, but the phase noise floor is critical, etc... A moving target, while you are trashing everything that comes in your way.



Otherwise said, "if you don't want to be insulted, you better shut up". Nice try, and BTW, the burden of proof is on you. Nobody could prove a negative.

No, I have claimed nothing you are pointing out.

I have merely reported facts:
"Firstly our DRIXO at 5.6448 MHz has the same phase noise of the Rakon at 10 Hz from the carrier: -155dBc.
It's 8-9 dB worse at 1Hz from the carrier (random walk of the crystal) and it has 12-13 db better noise floor."

Facts without any conclusions, it is you who draw the conclusions according to your thought.

Not a problem for me, if you keep questioning the measurements I will keep reporting the measurements of the oscillator you have built (for an ham friends at 6.144MHz).
No need to qualify your result, it speaks for itself.
 
No, but improper measurement conditions, according to the best practices, can lead to whatever (better or worse) results, depending on the time of day, temperature, season, and the moon phase.

There's more to measurements, in particular at these levels, that a nice plot presented as the absolute truth. There's error sources analysis, value certitude ranges, 3rd party validation (and no, the cheerleader crowd is not a valid 3rd party), independent labs etc... Not to mention the very basic requirement to correlate the measurement results with a final product performance metric, this is for the consumer market, not for a satellite tracking system.

Yeah but who on this earth is not influenced by the moon phase, i think the influence ist neglictable 😉

And If you dont believe the 3rd Party measurements (which seem to match the "original" measurements) then you seem to be indeed believing that the plots are faked (as i said, i dont believe that you can measure better phase noise by improper measurering techniques)

Greetings
Oli
 
Andrea, I have to re-iterate my few days old request: please stop insulting the experience and knowledge of everybody disagreeing with you. I know, you don't consider your stances as an insult, but according to everybody around that attempted a dialogue with you, after expressing any disagreements with your claims, they are.

If you can't stop re-iterating, then at least speak for yourself, as I've had a constructive discussion with Andrea about subharmonics somewhere around post #2850.
 
Last edited:
Let me ask the question a different way- Why not take a good oscillator that has a VCO input (they do exist) and modulate it with increasing levels of low frequency white or pink noise and see how that alters the sound? At some point its effects should be quite audible. I would hope its possible to quantify a lower threshold below which phase noise/jitter in any band is inaudible. And there may be a lower limit based on the resolution of the DAC/ADC.

Wow and flutter (gross phase modulations) have a threshold in the .1% range. Any crystal will be much better.

Its also a way to validate phase noise measurements.
 
Yes, I think we can stop now. All is said on the SYN08 topic.


For me it is very simple. Andrea designed a clock for audio use which should sound the best. That was his design goal.

He has a great tool to measure his design and compare values with other know clocks to see if he is closing in on his targets and being able to compare listening results and correlate with dB values…

The good thing is, that I and also many others are in absolute agreement, that the there is correlation between the dB values and perceived sound quality in many set ups…

So for me this is a success and nothing SYN08 and others are trying to post here will change that.

In fact there is no need for any response as I am totally happy with Andrea‘s clock and it is confirmed by many serious DIY, so let’s not waste any more post time - that would be nice.

I sincerely hope the next post I am invited to read are about the clock, it’s implementation and user experience.

Thanks to all to share my appeal
 
Last edited:
so many discoveries were made by chance, you see Cristoforo Colombo who discovered the America trying to go to india ... nobody disputes Columbus about how he sailed or in what conditions, we simply thank him for the discovery.
I don't think that's the case with Andrea, I think they worked hard to make an oscillator available for great music. for the best ever had to date.
and this is the fundamental POINT, we are not going to investigate the how and why.

It is useless to insist on contesting the measurement, when moreover, you are contesting only for theory as you do not have the same tools to ascertain any errors.

syn08 and C. your constant attacks turn out to be poerile and groundless. do something of YOURS! and then we will be happy to try them.

"the dogs bark but the caravan goes on"

Gavroche
 
Yes, I think we can stop now. All is said on the SYN08 topic.


For me it is very simple. Andrea designed a clock for audio use which should sound the best. That was his design goal.

He has a great tool to measure his design and compare values with other know clocks to see if he is closing in on his targets and being able to compare listening results and correlate with dB values…

The good thing is, that I and also many others are in absolute agreement, that the there is correlation between the dB values and perceived sound quality in many set ups…

So for me this is a success and nothing SYN08 and others are trying to post here will change that.

In fact there is no need for any response as I am totally happy with Andrea‘s clock and it is confirmed by many serious DIY, so let’s not waste any more post time - that would be nice.

I sincerely hope the next post I am invited to read are about the clock, it’s implementation and user experience.

Thanks to all to share my appeal
Well said and agree completely, for me I had to use the Ignore List function for the first time here and it works very well in removing these antagonistic posts.
 
Thanks for the reminder, it lead immediately to one of those never made claims we were talking about right above:
:rofl:

Yes, it's what I have said several times and in fact we have reached very good close in phase noise with our oscillators, maybe the best for DIGITAL AUDIO (I have never claimed for satellites or radars).
And other members (not me) wrote "in class with Wenzel and Oscilloquartz".
So nothing you were pointing out.
Tricks don't work.

Then the Rakon OCXO has better close in phase noise than our oscillator.
No doubts, 8-9 dBc better than our DRIXO in the range 0.1 to 1Hz from the carrier.
And so?
Is this a problem?

Not for us, we are a pair of poor hobbyists and not a NZD $101 million enterprise
Rakon - Wikipedia

And moreover you cannot use a 5MHz oscillator for digital audio unless you use your fantastic fractional-n PLL (no comment).
Maybe you can ask them to build a pair of similar OCXO at the right frequencies to be used in digital audio.
I'm not sure they'll do the job for you unless you order several hundreds or thousands of them.
Not to mention the price they would cost.

Then, since you still keep confuting the reliability of our measurements and you claim I have insulted your experience and knowledge pointing to the result you have reached with your oscillator ...
sorry ...
but who are you?

I don't think you are an authority in RF matter.
Of course we are not RF masters. RF masters are others guys such as Enrico Rubiola, Bruce Griffiths, John Miles, Arthur Everard, Richard K. Karlquist, maybe Gerhard and so on.
But not you.
I don't think you're a big contributor to time-nuts circle.

So, I repeat once again, a little humility wouldn't hurt.
 
- Perhaps because the many outlandish claims in this thread pegged the BS meter to the far right? Need examples/quotes?

- And perhaps because good engineering is perverted and distorted to accommodate these claims? Need examples/quotes?

A personal observation: in the majority of times (70-80%) when somebody jumps in to unconditionally support this thread main course and initiator, it turns out it's a long time member (over 5 years) with a handful of posts per year, mostly (over 60-70%) in this thread and the associated group buy thread. It's like an army of drones or bots comes to the rescue each time things get sour :rofl:. I would be very curious about the explanation of this obvious pattern.

Enough off topic stuff, everybody enjoy the benefits of the ultra low close in phase noise in the audio reproduction 😀. And, unless you are in the price-no-objection group, buy Andreas oscillator, the Rakon device is orders of magnitude more expensive 😀.
 
Last edited:
If you can't stop re-iterating, then at least speak for yourself, as I've had a constructive discussion with Andrea about subharmonics somewhere around post #2850.

We appreciate a lot the suggestions which make sense.

In fact, although it wouldn't be our priority since we use 5/6 MHz oscillators only for our DACs, we will investigate the opportunity of getting very low phase noise avoiding the frequency doublers.

I have already got some crystals samples at higher overtone in HC-47/U package, 11/12 MHz 5th overtone and 22/24/45/49 MHz 7th overtone.
Higher overtone means higher ESR which means greater in circuit loaded Q.
We hope to get similar performance we have reached with the doublers without using them.
Just we need to find the time to do the tests.


The same applies to just about any thread on diyaudio. It's almost all about overdesign and uncontrolled listening tests, so why make such a big deal of it this time?

I believe the reason is enough simple, I think one can understand quite well by reading the last 60-70 pages of this thread and ....
better if I stop here.
 
- Perhaps because the many outlandish claims in this thread pegged the BS meter to the far right? Need examples/quotes?

- And perhaps because good engineering is perverted and distorted to accommodate these claims? Need examples/quotes?

A personal observation: in the majority of times (70-80%) when somebody jumps in to unconditionally support this thread main course and initiator, it turns out it's a long time member (over 5 years) with a handful of posts per year, mostly (over 60-70%) in this thread and the associated group buy thread. It's like an army of drones or bots comes to the rescue each time things get sour :rofl:. I would be very curious about the explanation of this obvious pattern.

Enough off topic stuff, everybody enjoy the benefits of the ultra low close in phase noise in the audio reproduction 😀. And, unless you are in the price-no-objection group, buy Andreas oscillator, the Rakon device is orders of magnitude more expensive 😀.

Better if they got your fractional-n PLL oscillator since you “build and give out for free (including shipping) some of your stuff”.
 
Status
Not open for further replies.