Master Clock for your EtherREGEN

[AfterDark.]

Hi! 

The Mutec Nano is not quite up to performance when compared with Mutec Ref10. The OCXO inside the Mutec nano is not performing good at Short Term Stability (Allan Derivation) for some reasons.  The measurement of Allan Derivation is little way too high when compared below. This is not quite a same level when compared to Mutec Ref10 or others. 

The Typical measurement for AXTAL =  2.10 x E-12 in 1s interval as per above. 

Vs

The Typical measurement for OCXO = 5.93 x E-13 in 1s interval. 

This is one digit difference (E-12 vs E-13) Other OCXO has much lower measurement as below graph in BLUE. 

 

We want to share some of the OCXO with same Phase Noise -142dBC/Hz @10Hz offset. The Allan Derivation is always positve correlated with Phase Noise Performance.

 

The Allan Derivation  

This is another important measurement for Audio Application, the Allan Derivation is about the stability in frequency at very short time intervals e.g in 1s little second. The Allan variance is intended to estimate stability due to noise processes and not that of systematic errors or imperfections such as frequency drift or temperature effects. The Allan variance and Allan deviation describe frequency stability, i.e. the stability in frequency. We found this measure is positively correlated with Phase Noise measurement. E.g. OCXO with low Phase Noise figures, will have lower Allan Derivation. 

 

For audio application, the most important point is @ 1 sec = 5.93x10E13 (the lower the better) in BLUE.

 

We suggest when puchase the OCXO product, it is always look for indvidual measurement of Phase Noise and Allan Derivation. This can ensure the sound quality will not be compromised. 

 

Thanks and have a nice weekend.
 

Best Regards,

Adrian

AfterDark. 

[russellbobby]

Ok……Thanks for the info. Unusual that a competitor would come out and point out flaws in a product that has not even been distributed to the public. 
 

[Jakenz]

22 hours ago, Superdad said:

 

Well you are not breaking any rules but Thomann Music might be.  I am sure that Mutec’s exclusive USA importer will be none too thrilled to hear that a German dealer is trans-shipping this hot new clock for 28% off the official $2,499 list price. 

 

Expect removal of German VAT of 19% for export outside the EU accounts for much of this difference, so @russellbobbyis probably getting it cheaper than German locals can (assuming US customs duties and/or sales taxes are not material).

[Superdad]

24 minutes ago, russellbobby said:

Ok……Thanks for the info. Unusual that a competitor would come out and point out flaws in a product that has not even been distributed to the public. 

Besides, the Mutec REF10 Nano does not use that AXTAL OCXO…

[Superdad]

9 minutes ago, Jakenz said:

Expect removal of German VAT of 19% for export outside the EU accounts for much of this difference, so @russellbobbyis probably getting it cheaper than German locals can (assuming US customs duties and/or sales taxes are not material).

That misses my point. Dealer cost on the REF10 Nano is about $1,500–worldwide, convert that to whatever currency you wish—with various taxes being a separate matter. 
So you have one German dealer selling worldwide for about $300 over cost. That really sucks the incentive out of the product for Mutec’s entire dealer/distributor base. 

[Jakenz]

5 minutes ago, Superdad said:

That misses my point. Dealer cost on the REF10 Nano is about $1,500–worldwide, convert that to whatever currency you wish—with various taxes being a separate matter. 
So you have one German dealer selling worldwide for about $300 over cost. That really sucks the incentive out of the product for Mutec’s entire dealer/distributor base. 

Yes, I wasn't disagreeing with your point, indeed I agree the US distributor is unlikely to be at all happy with being undercut, just making an observation re a pricing factor in play.

[AfterDark.]

We noticed that the Mutec website changed the specification on Allan Derivation at 1 sec (Typically) for some reasons when we compared the figures on 24-Nov-2023 vs 25-Nov-2023. 

  

The Typical measurement for Mutec NANO =  2.0 x E-12 in 1s interval

Vs

The Typical measurement for OCXO = 5.93 x E-13 in 1s interval. 

 

This is one digit difference (E-12 vs E-13) Other OCXO has much lower measurement as below graph in BLUE. 

 

The Allan Derivation measurement figure is 330 times worse when compared, which may not an ideal grade for Audio Application.

 

Most Audiophiles and Studio Recording company will hear the difference when compared. The Allan Derivation in 1 second is extreme important figure, and it is describle frequency stability on OCXO itself. 

 

This is the specification on 24-Nov-2023:

 

This is the updated specification on 25-Nov-2023: 

 

 

 

[russellbobby]

Hey Adrian,

Put me something together with P.S for under 2k.  I will buy it with full understanding that you will accept a return with no restocking. I will pay shipping back to you. 
If it sounds better than the Mutec then we are all good. I plan on staying with a streaming Dac and can probably get away with one input/output. It will be mated with the Etherrrgen so need 75 ohm config.

please pm if you are interested. 

[Clockmeister]

I would like to suggest the Nano is NOT a replacement for the Ref10 it an affordable option for those that can't strech toa ref 10. With a quality LPS it may well exceed a Ref 10. However, as I suggested a few pages before you can have a low phase noise, HOWEVER is the spuri on the phase noise plot are many not just in quantity but amplitude as well it does negate a lot of the so-called figure show off's. This directly effects the performance of the phase noise, which is after all Jitter measured in the frequency domain,

In the lab we have many clocks that only just reach -100dBc @ 1Hz yet are superior in sound quality especially spatiality, overall coherence and linearity than one of two mentioned here but more than a small margin. And are not simply off the shelf evolution clock boards with a couple of LT3045 (yawn) in parallel on some average ali express LPS.

 

Far too much BS on this subject by many that really are shall we say 'looking for audiophile pinnacle' with the wrong information that’s been fed to them by third parties with an agenda.

 

designing a clock circuit is about many different aspects of electronics, impedance pathways, circuit board stack up, connectors, supply rails, various layers of regulation/ quality of components/ and the supporting cast. before we approach the subject of high-speed serial data alignment. this is purely about Signal integrity & how to preserve it not how may DIY knock's up I can sell in a month. Leaves a nasty taste in the mouth and does the whole industry zero favours.

 

[Markus8]

43 minutes ago, Clockmeister said:

However, as I suggested a few pages before you can have a low phase noise, HOWEVER is the spuri on the phase noise plot are many not just in quantity but amplitude as well it does negate a lot of the so-called figure show off's. This directly effects the performance of the phase noise,

Can you describe this simpler?

[Superdad]

38 minutes ago, Markus8 said:

Can you describe this simpler?

Just consider how harmonics (from poor power supply, poor layout, or wherever) that show up directly on a clock waveform can easily cause problems far worse than the base phase-noise performance of the OCXO used in the box:

If those harmonics appear on the waveform anywhere near the attached equipment’s clock trigger voltage, they will induce jitter far worse than some overall cleaner clock box, one that might be based on a more modest OCXO or  XO. 

 

Please also refer my recent prior post—and our paper—about how amplitude modulation turns into phase modulation and  vis versa. Those occurrences—happening all the time,  caused by clocks, power networks, layouts, and digital chips themselves—are what allow ground-plane noise to propagate throughout and between components. The impact of that ground-plane noise—on the DAC’s local master clock pin—is why we hear any of that upstream stuff. 

[Clockmeister]

On 11/25/2023 at 5:31 PM, Markus8 said:

Can you describe this simpler?

 

Very easy Markus

 

Like trying to reach the moon on a Vespa 125 with a flat tyre on bad luck Monday, in bad luck land,on a very bad luck day

[russellbobby]

The Mutec Ref10 Nano is in the rack. Mutec states a 14 day break in period. 
My room now is set up with Christmas tree so the seating configurations have changed, and not for the good. 
Good  news is that the Tubulus Concentus has a couple weeks breaking in and was really sounding good with the OCK 2

[Exocer]

10 minutes ago, russellbobby said:

The Mutec Ref10 Nano is in the rack. Mutec states a 14 day break in period. 
My room now is set up with Christmas tree so the seating configurations have changed, and not for the good. 
Good  news is that the Tubulus Concentus has a couple weeks breaking in and was really sounding good with the OCK 2

Excited to hear your take on these two. Other than the Etherregen, can you tell me where you’re using the clock? Cheers

[russellbobby]

‘Excited to hear your take on these two. Other than the Etherregen, can you tell me where you’re using the clock? Cheers”
 

The Bricasti M3 is a streaming Dac. So the EtherRegen is connected with the Mutec and the clean side into the Ethernet input of the Bricasti   Not sure if I will use any of the other inputs/sockets/connections. ( What is the proper nomenclature?)

[kiwibirch]

Whats your initial thoughts vs the OCK-2. I know the nano will get better as it runs in but sometimes the difference can be quite noticable from the outset...

[Exocer]

50 minutes ago, russellbobby said:

What is the proper nomenclature?)

On the Nano, those are outputs. Thanks!

[russellbobby]

Only been in for a few hours. Really do not want to comment. I think these clock/cables are much more sensitive to break in.

I do like the texture in the vocals that was apparent from the start. 
Like I mentioned with the Christmas tree up I have a couches on each side of of the room on the long wall and taking some high end energy away. 
 

The OCK 2 was sounding real good with the Tubulus Concentus cable when I pulled it out. 
 

[kiwibirch]

2 minutes ago, russellbobby said:

Only been in for a few hours. Really do not want to comment. I think these clock/cables are much more sensitive to break in.

I do like the texture in the vocals that was apparent from the start. 
Like I mentioned with the Christmas tree up I have a couches on each side of of the room on the long wall and taking some high end energy away. 
 

The OCK 2 was sounding real good with the Tubulus Concentus cable when I pulled it out. 
 

No worries, I’m considering the same upgrade. Will be interesting to see what you think once things settle

[russellbobby]

Can someone explain this to me like a 3rd grader?  So the Nano goes to the EtherRegen and into the Bricasti M3 ethernet input. No external clock input on the Bricasti.  Is it basically just giving it a cleaner signal to start with?

 

So what are the advantages of a dac that has a clock input other than the clock being  active on other inputs?  Does it bypass any internal

processing that way or basically the same with another way of getting there? 

 

Thanks

 

 

[MartinT]

On 11/17/2023 at 10:56 PM, russellbobby said:

Cabling was very important. The stock throw in and a 40 dollar BNC were very underwhelming. It wasn’t until I installed the Harmonic Technologies that I was starting to buy in. Martin on the forum here recommended the Tubulus Concentus . Breaking in very nicely. 
 

 

Clock cables are so important!  While the Canare and Belden clock cables were ok in my early experiments, a quality clock cable really makes the external clock sing.  First the Harmonic Technology DC-III but especially the Tubulus Concentus are amazing at pulling out extra detail while being less harsh - a real magic trick.

[MartinT]

On 12/7/2023 at 9:01 PM, russellbobby said:

Can someone explain this to me like a 3rd grader?  So the Nano goes to the EtherRegen and into the Bricasti M3 ethernet input. No external clock input on the Bricasti.  Is it basically just giving it a cleaner signal to start with?

 

So what are the advantages of a dac that has a clock input other than the clock being  active on other inputs? 

 

 

The whole objective is to minimise the noise superimposed on the data signal entering the DAC.  The lower the noise along the chain, the less gets into the all-important digital-to-analogue conversion, resulting in higher resolution and lower smearing of the resultant analogue signal.

 

Every stage of the digital chain matters, so using an EtherREGEN removes noise from the ethernet signal, as you said.  However, using a better clock into the DAC gives additional improvement over and above what the EtherREGEN has achieved.

 

I do exactly this using two AfterDark clocks, one for the ER and the other for my DDC/DAC combo.

[Superdad]

On 12/7/2023 at 1:01 PM, russellbobby said:

So the Nano goes to the EtherRegen and into the Bricasti M3 ethernet input. No external clock input on the Bricasti.  Is it basically just giving it a cleaner signal to start with?

 

It is very important to understand two things:

a) While low-phase-noise/low-jitter clocking is worthwhile for EtherREGEN, such is FAR more important for a DAC.  Phase-noise/jitter (essentially same thing, the former more usefully looked at in frequency domain, the latter a total in time domain) matters to packet-data interfaces (USB, Ethernet, etc.) almost entirely because of the ground-plane noise it induces in down stream equipment, ultimately infecting the DAC circuit board (and master clock pin of the DAC). We coined the term "clock threshold jitter." It is sort of a second tier affect.

 

b) The clocks inside a DAC--are at frequencies related to the audio sample rate (22.5792MHz and 24.576MHz for 44.1KHz and 48KHz sample rate families respectively; and sometimes twice those rates, or even something custom for the DAC's architecture)--and their phase-noise/jitter is DIRECTLY key to the performance of the DAC! These are the MOST important clocks--and everything we try to improve upstream (and most of the all the things we hear differences from in digital audio) are because of the very small effects they have on the DAC's master clock. Remember, any disturbance of the ground-plane--from which the DAC's clock references its voltage ticks--be it caused by leakage currents or phase-noise/jitter, will cause jitter of the DAC's clock. That is what you are hearing!

 

On 12/7/2023 at 1:01 PM, russellbobby said:

So what are the advantages of a dac that has a clock input other than the clock being  active on other inputs?  Does it bypass any internal

processing that way or basically the same with another way of getting there? 

 

Traditionally, very high-end DACs having an external clock input are generally accepting only a Word Clock input frequency. That is, an audio sample-rate related frequency. dCS DACs for example. And when they do, the external clock bypasses the internal clock of the DAC.  But really it would be better and easier for the DAC to have really good clocks built in--sitting just an inch or so from the DAC conversion stages.

 

Relatively few DACs accept an external 10MHz clock input. And just like the EtherREGEN, that 10MHz must then be converted (synthesized) into some other frequencies. For a DAC those are the above mentioned audio-rate-related non-whole numbers. For the EtherREGEN we synthesize multiple differential clock lines--two at 25.0MHz and two at 250MHz--for the switch chip and PHYs, and for the ultra-high speed, ultra-low-jitter reclocking flip-flops.

The quality and performance of clock synthesizer chips and supporting circuits (clock buffers, single-ended versus differential distribution, etc.) matters a great deal. If not done right (and some of those parts are expensive!) there will be no benefit to an external clock--and potentially it could be much worse.  It is all in the details. As always... 

[Oggo]

8 hours ago, Superdad said:

But really it would be better and easier for the DAC to have really good clocks built in--sitting just an inch or so from the DAC conversion stages.

 

Alex, could you tell us, what phase noise ratios - in your opinion - a good built-in clock should have at least?

[Superdad]

On 12/13/2023 at 3:00 AM, Oggo said:

 

Alex, could you tell us, what phase noise ratios - in your opinion - a good built-in clock should have at least?

 

Well it is a really complex subject and ultimately I am not an expert to answer.

A few of the factors involved include:

a) What kind of device are you thinking of? A DAC, a network streamer, a computer, a switch?  

 

b) Depending upon device, both the requirements and the frequencies of the clocks vary greatly.

 

c) The implementation of the clock--its type (HCMOS or sine), the power networks surrounding it, the routing and conversion of its output (single-ended>differential?)--are all VERY important.

 

d) And even if just comparing phase-noise specifications (of the XO or OCXO itself), one has to be mindful of the carrier frequency in order to be able to come close to a fair comparison.  That's because even with equivalent design/performance, the lower the clock rate, the higher the phase-noise spec can be at a given offset (say 10Hz). It is about a 6dBc/Hz per octave difference.  In other words, other factors aside, you can consider a 20MHz clock that measures -110dBc/Hz (@10Hz) to perform equally to a 10MHz clock that measures -116dBc/Hz (@10Hz offset).

 

 

By the way, either late today or sometime tomorrow (Friday) I will make an exciting announcement regarding UpTone Audio and clocks... 

Will start a new thread about it.  But I promise nobody here (except for my popular industry friend expert in the UK) is likely to guess right...

Cheers,

--Alex C.