Cybershaft rubidium clock...low priced option?

[zephyr24069]

It's two separate clocks. You get the OCXO Premium and Rubidum in one chassis. The one person I know who has this unit greatly prefers the OCXO Premium over the Rubidium and uses it the vast majority of the time.

 

Yes...that's me. The OCXO side of the dual clock from Cybershaft is superior in phase noise, etc...and more importantly causes a much more musical, not analytical and more natural sound-stage and musical experience in my system.

[zephyr24069]

That's correct...mine did to (I have a dual Rb + OCXO Premium Select) so there were 2 complete sets of test data on the specific unit. Nice touch,....I've never seen any manufacturer in audio give specific test results of the exact unit purchased AND give the buyer the right to review the test results before shipment and cancel the order if not satisfied (this was my experience with Kenji at Cybershaft....)

 

I plan shortly to install onboard the following OCXO:

 

http://www.pulsarclock.com/ds/Pulsar_Clock.pdf

 

The RMS jitter has been calculated at 20.174fs.

 

The part is about $400.

 

I'm actually very happy with the Crystek CCHD-950 which I recently installed, and is much cheaper.

 

Thank you for updating this table. The clock I'd like to see tested and added when it ships is the new Esoteric

GRANDIOSO G1. From what I can tell on the Japanese site translated, it appears to be a totally SOTA approach

to 10 Mhz master-clocking (and 22.xx Mhz master clocking if I read it correctly). No price yet but it would be

logical to assume it will come in within the normal Grandioso bands for price which would put it in league with

the Sforzato (price-wise).

[mozes]

I don't know much about external clocks, but I got interested as I have a Brooklyn DAC that has a wordclock in and out. So how do we know if this clock is compatible with the Brooklyn or with any other DAC? See my profile, I'm a newbie so bear with me

[Keith_W]

I don't know much about external clocks, but I got interested as I have a Brooklyn DAC that has a wordclock in and out. So how do we know if this clock is compatible with the Brooklyn or with any other DAC?

 

You find out by reading this: https://mytekdigital.com/download_library/manuals/Brooklyn_DAC_Manual.pdf

 

Seems as if your Brooklyn needs a digital input between 44.1kHz and 192kHz. This means you need a clock that can output these frequencies. Any clock that outputs a lab grade frequency (10MHz) is not suitable, unless you feed that output into a clock synthesizer, like this one: OCX HD 768 kHz HD Master Clock | Antelope Audio

[mozes]

Thanks Keith

[Elberoth]

I plan shortly to install onboard the following OCXO:

 

http://www.pulsarclock.com/ds/Pulsar_Clock.pdf

 

The RMS jitter has been calculated at 20.174fs.

 

The part is about $400.

 

I'm actually very happy with the Crystek CCHD-950 which I recently installed, and is much cheaper.

 

I'm not sure how you have calculated those jitter values.

 

Using the same calculator and range I used to calculate the jitter of the above clocks, I calculated rms phase jitter of this Pulsar XO at 446fs, not 20.174fs. 20x difference.

[zephyr24069]

Thank you for updating this table. The clock I'd like to see tested and added when it ships is the new Esoteric

GRANDIOSO G1. From what I can tell on the Japanese site translated, it appears to be a totally SOTA approach

to 10 Mhz master-clocking (and 22.xx Mhz master clocking if I read it correctly). No price yet but it would be

logical to assume it will come in within the normal Grandioso bands for price which would put it in league with

the Sforzato (price-wise).

 

Info from Esoteric's Japan site for the G1; looks like a Rubidium implementation. I wonder what the

phase noise characteristics, etc...will be. Interestingly enough, they allow another reference oscillator

10 Mhz or 1pps master clock device to input a baseline sync for this master clock. Wow,...master clocks

for master clocks, for reclocking master clocks. You could wind up having 3 clocking devices (external 10 Mhz

ref device into G1 then into G-01) in a high-end audio setup; kind of boggles the mind and kills the wallet

all at the same time. Price in Japan is stated as 1,600,000 Yen plus tax; no telling what the US price will be

but using the Grandioso line components as a reference may be indicative.

 

Best bet is to open this link with Google Chrome to get any translation possible;

http://www.esoteric.jp/products/esoteric/g1/index.html

[mac_and_dac]

I'm not sure how you have calculated those jitter values.

 

Using the same calculator and range I used to calculate the jitter of the above clocks, I calculated rms phase jitter of this Pulsar XO at 446fs, not 20.174fs. 20x difference.

 

See post #253 here:

 

LKS Audio MH-DA003 - Page 17

 

The tool focusses on jitter in the 10Hz-10Khz range.

 

Some calculated values appear in post #274.

LKS Audio MH-DA003 - Page 19

 

As stated, I currently use the Crystek CCHD-950, replacing the stock CCHD-575 in my machine. The 950 has a worse RMS jitter spec, but sounds better in my DAC. However, I have the 575 in another of my DACs and it sounds just fine. This suggests that experimentation is the best way to find the best XO for a particular DAC, rather than raw specs.

 

Installing the following DIP adapter on the board allows for fairly easy XO rolling:

1107741 Aries Electronics | Crystals, Oscillators, Resonators | DigiKey

 

This makes comparisons of sound a lot easier; easier, certainly than having to re-solder an XO every time.

 

I can't comment on external clocks as my DAC's don't support them (and I don't have the $$$ to play that ball game anyway).

[barrows]

See post #253 here:

 

LKS Audio MH-DA003 - Page 17

 

The tool focusses on jitter in the 10Hz-10Khz range.

 

Some calculated values appear in post #274.

LKS Audio MH-DA003 - Page 19

 

As stated, I currently use the Crystek CCHD-950, replacing the stock CCHD-575 in my machine. The 950 has a worse RMS jitter spec, but sounds better in my DAC. However, I have the 575 in another of my DACs and it sounds just fine. This suggests that experimentation is the best way to find the best XO for a particular DAC, rather than raw specs.

 

Installing the following DIP adapter on the board allows for fairly easy XO rolling:

1107741 Aries Electronics | Crystals, Oscillators, Resonators | DigiKey

 

This makes comparisons of sound a lot easier; easier, certainly than having to re-solder an XO every time.

 

I can't comment on external clocks as my DAC's don't support them (and I don't have the $$$ to play that ball game anyway).

 

Hahaha! Made me laugh. Trying to achieve low phase noise and then using a plug in XO... Also, does not the 575 achieve lower phase noise than the 950 at 100 MHz... At least that is what I see in published Crystek specs...

[Superdad]

Hahaha! Made me laugh. Trying to achieve low phase noise and then using a plug in XO... Also, does not the 575 achieve lower phase noise than the 950 at 100 MHz... At least that is what I see in published Crystek specs...

 

And actually, current production CCHD-575s, are well outperforming Crystal's own web published measurements. They sent us 3 samples of a mid-20s MHz rate, and they came with 3 individualized graphs--with 10Hz offset noise at -108, -110, and -112. I called and asked if those were hand picked and they said no, that they have just improved their process a great deal. Even Crystek's Sales Director could not give me a reason--other than ruggedness and installed clientele for the form factor--for the continuation of the 3 times the price CCHD-957 series.

(But I think Barrows knows all this already. )

[mac_and_dac]

Hahaha! Made me laugh. Trying to achieve low phase noise and then using a plug in XO... Also, does not the 575 achieve lower phase noise than the 950 at 100 MHz... At least that is what I see in published Crystek specs...

 

My 950 is actually soldered in, the board is there for those who want to roll. It's actually good to hear the difference that different XO's can make to SQ. And it doesn't always follow specs. I expect not many people have had the chance to hear how a different oscillator can impact the sound of the DAC.

 

Out of interest, how does the plug-in aspect impact the phase noise? If XO's are that sensitive, then surely having a clock offboard is also impactful? Not trying to be funny, just trying to learn more.

[julian.david]

You find out by reading this: https://mytekdigital.com/download_library/manuals/Brooklyn_DAC_Manual.pdf

 

Seems as if your Brooklyn needs a digital input between 44.1kHz and 192kHz. This means you need a clock that can output these frequencies. Any clock that outputs a lab grade frequency (10MHz) is not suitable, unless you feed that output into a clock synthesizer, like this one: OCX HD 768 kHz HD Master Clock | Antelope Audio

 

FYI, the MUTEC MC-3+ and MC-3+USB would also be very well suited for taking a 10 MHz reference signal and using it to derive a high precision Word Clock signals to send to DACs.

[mozes]

Based on the user manual of the Brooklyn, It seems that Mytek strongly recommends to use the Brooklyn internal clock for best performance.

[barrows]

My 950 is actually soldered in, the board is there for those who want to roll. It's actually good to hear the difference that different XO's can make to SQ. And it doesn't always follow specs. I expect not many people have had the chance to hear how a different oscillator can impact the sound of the DAC.

 

Out of interest, how does the plug-in aspect impact the phase noise? If XO's are that sensitive, then surely having a clock offboard is also impactful? Not trying to be funny, just trying to learn more.

 

Even more interesting is the variation from one sample XO to the next. A few years ago we had some XO samples measured for close in phase noise and hand selected the best ones for a project, if one has the equipment, time, and price margin available to do this one can achieve even higher performance.

Plugging any component, whether an IC OPA, XO, or even a voltage regulator into a socket is always going to result in lower performance, as the contact resistance will be higher than a well soldered connection to a well designed PCB. Indeed, using external oscillators subject to multiple plug in connections and a cable as well is not going to be ideal. Best approach is always going to be a dedicated, low phase noise fixed frequency XO right at the DAC.

Best to leave external clocks for the pro world, where the may be needed to sync together multiple components to one clock source. In the playback world, lets put a really good XO as close to the DAC as possible, and then send it out to the source for syncing.

[hurka]

In my dac,replaced the tcxo.Put ocxo,battery powered Lt3080,and dc choke filtered(10H).Sound get amazing.Total cost:50$

[Elberoth]

See post #253 here:

 

LKS Audio MH-DA003 - Page 17

 

The tool focusses on jitter in the 10Hz-10Khz range.

 

Some calculated values appear in post #274.

LKS Audio MH-DA003 - Page 19

 

As stated, I currently use the Crystek CCHD-950, replacing the stock CCHD-575 in my machine. The 950 has a worse RMS jitter spec, but sounds better in my DAC. However, I have the 575 in another of my DACs and it sounds just fine. This suggests that experimentation is the best way to find the best XO for a particular DAC, rather than raw specs.

 

Installing the following DIP adapter on the board allows for fairly easy XO rolling:

1107741 Aries Electronics | Crystals, Oscillators, Resonators | DigiKey

 

This makes comparisons of sound a lot easier; easier, certainly than having to re-solder an XO every time.

 

I can't comment on external clocks as my DAC's don't support them (and I don't have the $$$ to play that ball game anyway).

 

But those measurements are for diffrent ranges ! One was in the 1Hz-100Hz range and the other one - 10Hz-10Khz range.

 

You cannot compare XO jitter figures in different ranges.

 

Jitter is not unlike car acceleration. The measured car acceleration (measured in seconds) only gets any meaning after you apply a speed delta to it.

 

What you just did here, was comparing car A acceleration in 0-60 range, with car B acceleration in 60-120 range. Makes absolutely no sense at all !

[mac_and_dac]

Even more interesting is the variation from one sample XO to the next. A few years ago we had some XO samples measured for close in phase noise and hand selected the best ones for a project, if one has the equipment, time, and price margin available to do this one can achieve even higher performance.

Plugging any component, whether an IC OPA, XO, or even a voltage regulator into a socket is always going to result in lower performance, as the contact resistance will be higher than a well soldered connection to a well designed PCB. Indeed, using external oscillators subject to multiple plug in connections and a cable as well is not going to be ideal. Best approach is always going to be a dedicated, low phase noise fixed frequency XO right at the DAC.

Best to leave external clocks for the pro world, where the may be needed to sync together multiple components to one clock source. In the playback world, lets put a really good XO as close to the DAC as possible, and then send it out to the source for syncing.

 

Very interesting. With this in mind, I'm tempted to solder in the Ticha 994 op amps I use in the I/V section as these are also socketed (by the manufacturer).

 

When I got the DAC, the XO was sat on a separate 'clock' board with a custom voltage regulation circuit.

 

 

The XO's output ran a few cm along some coax - probably sub-optimal. I've soldered the new XO to the motherboard between the DAC's two ES9018 chips. It's as close as it can be, now. I still power the XO with the aftermarket board though, it just required some simple re-wiring.

 

I do wonder if there is any impact from an XO wired into two DACs rather than just one. I've never seen any other approach, but would be interested nevertheless in any views on this subject.

 

Great tip, by the way, on the XO batch variations. Maybe the 575 I had originally in my LKS was one of the runts of the bunch.

[mac_and_dac]

But those measurements are for diffrent ranges ! One was in the 1Hz-100Hz range and the other one - 10Hz-10Khz range.

 

You cannot compare XO jitter figures in different ranges.

 

Jitter is not unlike car acceleration. The measured car acceleration (measured in seconds) only gets any meaning after you apply a speed delta to it.

 

What you just did here, was comparing car A acceleration in 0-60 range, with car B acceleration in 60-120 range. Makes absolutely no sense at all !

 

Fair point on the comparison.

 

There is nevertheless a valid debate to be had on whether one XO type is 'better' than another for DAC usage based simply on a comparison of the published specs. The jitter ranges I have provided links to show (or at least attempt to show) a better context for audio DAC usage. My own listening tests have led me to conclude that following the published specs will not always give you better sound - there are other factors at play.

 

To conclude, I would not have believed that an XO (and possibly its location) could have such an influence on sound these days. My hearing is not the best, and I am usually pretty cynical when it comes to real vs perceived differences, but I have found that swapping the XO influences a DAC's sound perhaps more than any other component change I have tried.

[JohnSwenson]

Fair point on the comparison.

 

There is nevertheless a valid debate to be had on whether one XO type is 'better' than another for DAC usage based simply on a comparison of the published specs. The jitter ranges I have provided links to show (or at least attempt to show) a better context for audio DAC usage. My own listening tests have led me to conclude that following the published specs will not always give you better sound - there are other factors at play.

 

To conclude, I would not have believed that an XO (and possibly its location) could have such an influence on sound these days. My hearing is not the best, and I am usually pretty cynical when it comes to real vs perceived differences, but I have found that swapping the XO influences a DAC's sound perhaps more than any other component change I have tried.

 

The DAC does have to do something reasonable with the clock signal. I have seen many DACs that spend a lot of money on high performance oscillators, then do really stupid things with the signal, completely ruining any advantage that high performance oscillator could have provided.

 

I had an interesting experience with upgrading clocks years ago. I had built my own DAC with an I2S reclocking circuit fed from a low jitter oscillator. I got the I2S from a hacked up squeezebox, and fed the clock from the DAC back to the SB to sync it to the DAC. The signal integrity between the boxes was atrocious, but it didn't matter because the signal was reclocked

 

One day I decided to listen to the audio out of the squeezebox, WOW it sounded so much better, not as good as my homebuilt DAC, but WAY better than the original without the external clock, even though the clock from the DAC was highly distorted by the really bad digital interconnect I used. I expected it to sound worse given the SI of the clock I was feeding it, but the low jitterness of that clock managed to make it through.

 

John S.

[YashN]

In my dac,replaced the tcxo.Put ocxo,battery powered Lt3080,and dc choke filtered(10H).Sound get amazing.Total cost:50$

 

hurka, this is very cool, I like affordable enhancements like this.

 

Can I know which OCXO you used and how your battery packs are constructed? Thanks.

[hurka]

ohh,little old comment.Yes,ocxo in my dac12.000mhz,but powered now lt 3042 ultra low noise ldo with r core transf.without any chokes.

Better or same with ultracaps powered xo.I think tcxo or ocxo with this caps power should be very good!

[zephyr24069]

Just a quick note that those who purchase a Cybershaft clock will receive a certificate of calibration with measurements that show unit purchased meets or exceeds published specs. In my case (OCXO Premium), the phase noise was measured at -111.8 dBc/Hz @ 1Hz (vs spec of -110) and -132.98 dBc/Hz @ 10Hz (vs spec of -130). Phase noise was measured on a Symmetricom TSC5115A. Values other than phase noise that were also measured and certified are Frequency value, Allan Deviation, Output Level and Harmonic Distortion. Phase Noise and Allan Deviation graphs are also provided.

 

Definitely true and Cybershaft also as noted on the calibration certificates uses a BVA 8607 as a reference in their testing in addition to the Symmetricom TSC5115A and Agilent 53132A...

 

I've done some additional testing for them of a OCXO Ultra Premium prototype with a very high caliber OP14 module and a Schottky barrier-diode enabled power supply (used to be used only on the Limited when it was produced if memory serves) and the OP14+ level clock is even better than the outstanding performance of the OCXO side of my RBOCXO-01 'dual master clock'.

 

Good time to wake up this thread with the advent of the G1 Esoteric offering hitting the shores in the US, the new K1 coming out, etc....

[zephyr24069]

I'm told that later this year Cybershaft will be looking to release new offerings based upon various testing of new designs that will increase the performance of these ultra-low phase noise/ultra-low jitter master clocks demonstrably.  I'm using a custom-built version (with some of these specific design changes I had requested a while ago) at this point (see my system page on Audiogon or under SCAudiophile on WBF and AudioAficionado that has me very impressed and looking forward to playing music on my system even more than ever before!

[tboooe]

3 hours ago, zephyr24069 said:

I'm told that later this year Cybershaft will be looking to release new offerings based upon various testing of new designs that will increase the performance of these ultra-low phase noise/ultra-low jitter master clocks demonstrably.

Dang man I have been wanting to get one of these Cybershaft clocks since I started this thread!  However, I dont have any gear that accepts a clock input.  May have to get something like the SMS Ultra so I can try this out.

[zephyr24069]

That would definitely do it,..best of luck!