Interesting...so you like this rubidium clock better than the scarlatti clock? What's the brand and about how much? I've found the scarlatti clock sensitive to clock cables and power...or maybe better said it responds to better cables nicely...
dCS Users
- Thread starter AudioExplorations
- Start date
Interesting...so you like this rubidium clock better than the scarlatti clock? What's the brand and about how much? I've found the scarlatti clock sensitive to clock cables and power...or maybe better said it responds to better cables nicely...
The rubidium clock provides a very accurate 10mHz signal which controls the Scarlatti clock.
Rubidium clocks are marketed by Esoteric, Antelope, SRS and others however the one I am testing is a prototype with still better short term accuracy. It is not currently a commercial product but that may change.
Rubidium clocks are marketed by Esoteric, Antelope, SRS and others however the one I am testing is a prototype with still better short term accuracy. It is not currently a commercial product but that may change.
The Scarlatti clock is not the ultimate in stability. If you have a chance, borrow the SRS Perf10 or Antelope 10M Rubdium clocks and try them with your Scarlatti clock.
That's certainly true, however you must define stability. The oscillator used by DCS has excellent short term stability. Most rubidium clocks will have worse short term stability but much better long term stability which is of no benefit to audio. Locking the two together can give you the better short term stability you need.
Stability is clock phase noise, and basicly, you can describe that as short term accuracy.
The clcok accuracy on the other hand, described by PPM figures that manufactures have been giving us for years, have little importance in world of audio.
Not quite true. Stability only has value if related to time. In simple terms both short term fluctuations or longer term drift could equally be defined as stability. The VCXO crystal oscillator has very good short term stability. The rubidium source has much better medium term stability however by locking the rubidium source to the crystal oscillator a major improvement in short term stability is possible.
Agreed, although the correct term that is used for long term stability is accuracy.
Short term = stability
Long term = accuracy
I would like to see some measurements that would back up the claim, that adding a second oscilator (be it a XO, VCXO, TCXO or Rubidium standard), with inferior phase noise performance (short term stability) to primary clock, can improve the performance of the primary clock.
The long term clock stability (which should be called accuracy) has no relation to sound quality.
Agreed on long term stability.
Hopefully if the experiments I'm currently doing with this rather special rubidium unit pan out, I will be able to give you some hard evidence. Currently reviewing possible patent filings on the underlying principle that this unit is using.
Well, I think what dSC components you want to include in your configuration depends on how you are going to use it. I think this checklist will help you decide.
1) Are you using a source with clock input?
2) What kind of digital outputs does your source have?
3) Does your source do any sample rate conversion?
If the source (transport/music server/PC) has a clock input you will benefit from the clock. If it doesn't have a clock input you will have no benefit from the clock even though you can slave both the upsampler and DAC from the clock.
If you intend to use an USB output from your source or your source doesn't do any SRC you will benefit from the upsampler. Otherwise you will not benefit from the upsampler from my experiance. The DAC does not have a USB input, so you need the upsampler if you want to use an USB output from your source. SRC in several components after each other will only mess things up from my experiance and is unwanted in my opinioon. At least it is not needed. So if your source has all the SRC you need and you don't want to use the USB output from your source, you will not benefit from the upsampler.
If the checklist indicates that you will benefit from upsampler or the clock and your choice is purely based on sound quality my experiance with the Paganini and the Scarlatti is that the dCS stacks sound best with all the boxes connected including the transport over a music server/PC.
1) Are you using a source with clock input?
2) What kind of digital outputs does your source have?
3) Does your source do any sample rate conversion?
If the source (transport/music server/PC) has a clock input you will benefit from the clock. If it doesn't have a clock input you will have no benefit from the clock even though you can slave both the upsampler and DAC from the clock.
If you intend to use an USB output from your source or your source doesn't do any SRC you will benefit from the upsampler. Otherwise you will not benefit from the upsampler from my experiance. The DAC does not have a USB input, so you need the upsampler if you want to use an USB output from your source. SRC in several components after each other will only mess things up from my experiance and is unwanted in my opinioon. At least it is not needed. So if your source has all the SRC you need and you don't want to use the USB output from your source, you will not benefit from the upsampler.
If the checklist indicates that you will benefit from upsampler or the clock and your choice is purely based on sound quality my experiance with the Paganini and the Scarlatti is that the dCS stacks sound best with all the boxes connected including the transport over a music server/PC.
Last edited:
It is a custom made clock for this purpose called Metronomo manufactured by Audeeva (the manufacturer of the Conbrio music server) and based on some original US army patents which have now been released if I understood correctly. It is made to be compatible with the 10MHz input on the dCS Vivaldi clock.
I am also going to experiment with slaving the Audeeva Metronomo Caesium oscillator to sattelite with a GPS antenna to find out what improvements really can be made with clocking. The advantage of a GPS-slaved clock is that it is constantly re-calibrated by multiple Caesium sources (4 in each of the 24+ GPS satellites), which are themselves synced to a Hydrogen Maser.
Last edited:
The Vivaldi DAC is fitted with a USB input. I had it on loan for a test as a stand alone DAC and the USB input is a major plus compared to the Scarlatti and the Paganini.
I had the Vivaldi on trial (since ordered) and I was amazed at the improvement achieved by using my external rubidium clock with both red book and hirez material. The improvement in imaging was clearly noticeable.
It seems to me that there are significant gains in performance to be had by adding a higher resolution clock, however be careful that any rubidium clock design is optimised for short term stability or you may be worse off.
It seems to me that there are significant gains in performance to be had by adding a higher resolution clock, however be careful that any rubidium clock design is optimised for short term stability or you may be worse off.
What Rubidium clock did you try ?
What did you use to divide the 10MHz signal ? Vivaldi Clock ? The 10MHz clock cannot be connected directly to the DAC.
I discussed the particular stability requirements with a Company that produces the highest quality rubidium clocks. They produced a prototype for me which I am very pleased with. There are no plans to produce an end user product at this time.
Of course the 10mHz signal is then used as the external reference for the DCS clock.
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |