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

Status
Not open for further replies.
Hi Andrea,
are measurements for the old Driscol boards done/posted somewhere? sorry if i overlooked that. I understand that the new Driscol is better, but is is also larger and more expensive.
Say for 22.5792/24.5760 SC-cut crystals that i still have from one of the previous group buys how much of a practical benefit would I get with TWTMC-DRIXO-SF vs the old design ?
This is to be used with Ian's FIFO.
Thanks.
 
Here you can find all the info and the measurements of all the new devices
The Well Tempered Master Clock - Group buy

I have not compared old vs new oscillators but the new ones perform much better so I expect sound improvement.

I know they are larger and more expensive, but keep in mind they are state of the art devices equal or maybe better than the best oscillators on the market.
 
Lite DAC question

Hi Andrea. I think I read your upcoming DAC Lite will use 5/6 clocks directly (no need for separate frequency doublers) and will have the squarer built in?

If yes to the above, I have the SC cut crystals from previous group buy. For your DAC Light I only need:
1) two new clock boards
2) ps boards

Will all the boards be available in the group buy for the DAC Lite? Or should we get the clock and related boards now since only the DAC itself (and any *new* boards) be available in the next group buy?

Sorry if I asked questions that have already been answered...

Frankie
 
Hi Frankie,

the new DAC Lite needs the new FIFO Lite to work because it uses a custom protocol and the FIFO Lite has the sine to square converters on board.
The system works at 5/6 MHz to get the best phase noise performance from the oscillators.

So, since you already own the crystals, you need the oscillator boards and the power supply.
I suggest the LiFePo4 battery supply because it's suitable to power the oscillators, the FIFO Lite and the DAC Lite (and also you have a 5V low noise regulator to supply other devices like USB to I2S board, I2S receiver and so on).

All the items of this GB will be available in the Lite GB, BTW I suggest to get oscillator boards and PS in this GB in order to avoid MOQ issue of the next GB (even finished and semi-finished boards have MOQ to start production).
 
Member
Joined 2007
Paid Member
Hi Andrea,

A bit late but this has been simmering in my mind for some days now ...

Inspired by MarcelvdG's comments about subharmonics e.g. in #2850 - and because I would like to be able to use these oscillators also with pure DS (DSD) converters - I have been thinking about what the subharmonics level is in the pure Driscoll oscillators (i.e. without the doublers)... Any chance you have measured this?

Cheers,

Jesper
 
Member
Joined 2007
Paid Member
Thanks Andrea for your feedback ... But would there be any amplifying circuits that may add harmonics? E.g. a differential amplifying stage or something else?

E.g. the very simple (BFR182 based) amplifier that was used in the previous version of the Driscoll oscillator in itself IME would have a quite high distortion level ... I cannot discern whether this somehow is counteracted in the way this amplifying stage is otherwise integrated into the oscillator design - but I reckon that this stage in isolation would add distortion ...

I hope it comes through what I am thinking of ... ?

Cheers,

Jesper
 
The oscillator itself and the buffers generate harmonics but not sub-harmonics.

The frequency doubler generates sub-harmonics because the base oscillator operates at lower frequency than the output carrier, while the pure oscillator works at crystal frequency without any multiplication.
So there is no way to generate sub-harmonics.
 
It is still possible that the MOQ for the 24 and/or 25 MHz crystals will be met.

There is now 5 pieces missing of each(if we assume that AOI and driscoll SC alternatives can agree on the same alternative in the end). There were more people that had interest in the beginning, so if you are still interested, don't forget to add yourself to the spreadsheet (and send order form), the sooner the better.
 
It is still possible that the MOQ for the 24 and/or 25 MHz crystals will be met.

There is now 5 pieces missing of each(if we assume that AOI and driscoll SC alternatives can agree on the same alternative in the end). There were more people that had interest in the beginning, so if you are still interested, don't forget to add yourself to the spreadsheet (and send order form), the sooner the better.

Hi Thom.
Where do you see that there is just missing 5 of each?

In the TWTMC_Summary of Orders I can see there is 3 orders at a SC 25MHz crystal and zero at a AT25MHz crystal. I don`t need 7 SC 25MHz or else I have ordered it ;o) maybe 2-3...
 
Member
Joined 2004
Paid Member
Is it correct?

If you want 24.000 or 25.000 MHz oscillator (for motherboards, switches or streamers) ...

I. 24.000 MHz sine output

1.) 6.0000 MHz AT-cut;
2.) TWTMC-PXO-AIO 12-24 MHz:
TWTMC-PXO-AIO-SF + TWTMC-PXO-AIO-XFMR + parts from Mouser

II. 25.000 MHz sine output

1.) 6.2500 MHz AT-cut;
2.) TWTMC-PXO-AIO 12.5-25 MHz:
TWTMC-PXO-AIO-SF + TWTMC-PXO-AIO-XFMR + parts from Mouser (different from the previous one - I.2. -??)

III. Sine to square converter
1.) TWTMC-STS-SX
or
2.) TWTMC-SFDO
 
Hi euro21. Who do you ask ;0)
The problem with SC6.25, AT6.25, AT6.00 is that the way to reach MOQ are longer, because theirs MOQ is 15. So in this moment I just see that SC25MHz with MOQ 10 is possible to reach, because there just need 7 more to order.
If there haven’t got a MOQ at the crystals, I have ordered the best solution a SC6.25MHz, the TWTMC-DRIXO-SF board, the TWTMC-DRIXO-XFMR 22-24 MHz and the TWTMC-SFDO.
 
Status
Not open for further replies.