Thank you Andrea!
Excellent.
Yes, I saw the measurements. It's hard to believe that these PN measurements are from a simple Pierce + AT xtals.
This level of performance was only available in very costly custom-made OCXO many years ago.
Fantastic work.
What is the difference in the new Pierce osc. design?
Excellent.
Yes, I saw the measurements. It's hard to believe that these PN measurements are from a simple Pierce + AT xtals.
This level of performance was only available in very costly custom-made OCXO many years ago.
Fantastic work.
What is the difference in the new Pierce osc. design?
Last edited:
The new Pierce oscillator is BJT based and followed by a buffer/amplifier.
Its output is sine wave instead of CMOS.
BTW, the 5/6 MHz AT-Cut crystals have exceptional specs, ESR below 5 ohm and Q around 700-800K, so I expected good phase noise results.
Unfortunately the specs of the AT-Cut crystals at higher frequencies are not so good, so I don't expect very low phase noise.
Its output is sine wave instead of CMOS.
BTW, the 5/6 MHz AT-Cut crystals have exceptional specs, ESR below 5 ohm and Q around 700-800K, so I expected good phase noise results.
Unfortunately the specs of the AT-Cut crystals at higher frequencies are not so good, so I don't expect very low phase noise.
Hi Andrea.
Sorry, I am a little bit lost in these long threads and my only needs is a 25mhz crystal and a PCB, there should be used in a switch. Is it correct that there is just one 25mhz to order= the 25mhz AT-cut Crystal?
But what PCB board to order, does this crystal only fit at the old designs?
Can you please give me a advice?
Regards Basil.
Sorry, I am a little bit lost in these long threads and my only needs is a 25mhz crystal and a PCB, there should be used in a switch. Is it correct that there is just one 25mhz to order= the 25mhz AT-cut Crystal?
But what PCB board to order, does this crystal only fit at the old designs?
Can you please give me a advice?
Regards Basil.
I follow this project from the side and I'm especially interested in Pierce oscillator with CMOS output.I see here published mostly phase noise plots.Would love to see the shape of the real output signal
of the Pierce oscillator in the time domain and frequency range of the 16.9344 MHz oscillator.
So if anyone has such recordings,I would kindly ask him to put them on the forum.Thank you in advance.
of the Pierce oscillator in the time domain and frequency range of the 16.9344 MHz oscillator.
So if anyone has such recordings,I would kindly ask him to put them on the forum.Thank you in advance.
The 25 MHz AT-Cut crystal is suitable for the old oscillator versions TWTMC-C and TWTMC-P and also for the new Pierce TWTMC-PXO.
The real issue is that there is a MOQ of 10 pcs from the manufacturer, so you could join the GB foer the crystal and then we will see if there is enough interest for this frequency.
Not too commonly known is that the BVA crystals are AT cut (not ST) with the special suspension. And they still have lower close in phase noise. Just not as low further out.
I am also interested in 25.000 MHz for a music server.
I'm also interested in 24.000 MHz as I have those on my server motherboard, on my usb card as well as in my DAC usb interface.
In the state of the art (boutique, or modded industrial) switches (mostly with separated optical network possibility) this one of the commonly used built-in clock:
Clock
If Andrea provides a better (cheaper) alternative, can expect a wide range of interest.
I also interested 2 pcs. 25Mhz good crystal.
In principle I have to agree with TNT, using a well designed and well isolated FIFO buffer the quality of the source does no longer matter.
Such devices use PLL, so the high performance of the oscillator could be vanished.
Anyway, taking a look at the linked oscillators I believe you can achieve the same or better performance with some of the new solutions at lower price.
In relation with the pink faun "OCXO" I believe you can get better phase noise performance with the new Driscoll oscillator (TWTMC-DRIXO) with crystals at 24-25MHz.
While if you look at the performance of the pink faun "ultra OCXO", the better solution could be the same TWTMC-DRIXO with crystals at 6-6.25 MHz followed by a couple of doubler.
We are also thinking to offer a "All in one" solution, the new Pierce oscillator with crystals at 5/6 MHz followed by 1/2 doublers in the same PCB, to get 11/12 or 22/24 MHz.
In this case you can get 24/25 MHz starting from crystals at 6/6.25 MHz.
As I have already said the problem is the MOQ for each frequency, so I think you should reach an agreement on a single solution, oscllator type and crystals, to understand if there is enough interest to meet the MOQ (10 or 15 pcs).
Such devices use PLL, so the high performance of the oscillator could be vanished.
Anyway, taking a look at the linked oscillators I believe you can achieve the same or better performance with some of the new solutions at lower price.
In relation with the pink faun "OCXO" I believe you can get better phase noise performance with the new Driscoll oscillator (TWTMC-DRIXO) with crystals at 24-25MHz.
While if you look at the performance of the pink faun "ultra OCXO", the better solution could be the same TWTMC-DRIXO with crystals at 6-6.25 MHz followed by a couple of doubler.
We are also thinking to offer a "All in one" solution, the new Pierce oscillator with crystals at 5/6 MHz followed by 1/2 doublers in the same PCB, to get 11/12 or 22/24 MHz.
In this case you can get 24/25 MHz starting from crystals at 6/6.25 MHz.
As I have already said the problem is the MOQ for each frequency, so I think you should reach an agreement on a single solution, oscllator type and crystals, to understand if there is enough interest to meet the MOQ (10 or 15 pcs).
Last edited:
I'd be in on that from the get go, Andrea. (that is exactly how I put together my wish list in the group buy list)
D
Deleted member 537459
This sounds like a very good idea!
That would be my favourite configuration if available.
So far the theory, but practice shows that it is not that easy after all.
So i'm also interested in 24.000/25.000 MHz.
Very interesting. Why do you think this is?
Oscilloquartz used SC-cut in their BVA OCXO8607
The 'York' 10MHz SC-cut OCXO white paper I linked to is as good as the 8607
close to the carrier (-123dBc at 1Hz) and quite a bit better away from the carrier.
TCD
- Status
- Not open for further replies.