Master Clock for your EtherREGEN

[LowMidHigh]

5 hours ago, PYP said:

I'm asking my question in this thread since it didn't belong in the listening impressions thread.  Did you DIY your Canare cable?  I'm asking since I like the idea of a short cable, but haven't seen a pre-terminated 0.5m 75 ohm Canare cable.  The pre-terminated ones tend to be longer that 0.5m.  Thanks.   

 

 

I use five of my Mutec REF10 outputs, all 75ohm, with DYI cables. 

 

I use Furutech Fc-61, very well constructed, at $6.75 per meter (red or blue) :

http://www.bellatoneaudio.com/index.php?route=product/product&product_id=199

 

For terminators, I use Audioquesst Bnc ITC. You need to expose sufficient cable and push it on. Super simple. There's a YouTube clip. 

 

Later on, when I mastered soldering, I started doing just that. 

 

Your other option is crimping. 

 

Good luck! 

[LowMidHigh]

4 hours ago, Ricardo007 said:

Hello Meister

I read from Alex Crespi from UpTone Audio that thé etherregen likes to see square waves and not sinewaves that's thé reason I asked thé question, thé Sotm master clock being sinewaves... I also have other SoTM products to "discipline" (but not in à dungeon) so I thought better for them to have an SoTM master... 

Actually hesitating to buy between Sotm Mutec and Cybershaft

Any help ? 🐸

Isn't Mutec the only one with a square wave? If so, and given that according to John that' shape is optimal the ER, then you have your answer. 

[LowMidHigh]

7 hours ago, audiotunesx said:

Lot's of great insight in this thread, many thanks I learned a lot.

 

The BG7TBL/DAPU clock is a very interesting solution.  We'd have some control over the sound because we have an assortment of power supplies that we can use to match the needs of our systems. I messaged the fellow HamRadioshop on the original aliexpress link days ago.  He got back to me today and said that he can make a 75 ohm version!  So cost would be $125 unit + $475 for say an sbooster mkii, so $600 CAD.  Not bad.  I wish the LPS 1.2 had the power, but the JS-2 would be the ideal PSU I'd imagine (power both the eR and clock with it).  I've had the $99 ifi 12v plug providing 18watts without issue to my NAS, so one could potentially get away with that for experimentation.

 

Teac 10m.  This has only 50 ohm outputs.  Inside, the transformer is quite small.  Beautiful casing, and I'm sure it will last.  A simple comparison is here:  https://dime.jp/genre/604669/  (use Chrome for auto-translate).  There's a ton of foreign reviews.  It seems that it can improve imaging and soundstaging as expected by a clock but only to an extent.  Teac Direct on eBay has a few open box clocks.  They are taking offers.  To get this to work Cybershaft has a 75 ohm to 50 ohm adapter that would connect to the eR.  They also have 2 50 ohm cable options, both very reasonable cost.  The review above reviews the basic cable, and they seem to have liked it.  Almost all reviews are with the associated Teac dac, or a Esoteric.  There's a youtube review with before/after audio footage.  I don't hear that big of a change.  Same reviewer has before/after with the Bonn N8, much bigger change.  Cost (CAD) $1400 open box + $150 for cybershaft accessories.  I think this would be worth it if my dac could use the external clock as well.  From what I gather, cascading a second eR would likely have a greater effect or a Farad power supply with a reasonable cable at much less cost.

 

Cybershaft Op-14.  Prices have gone up a lot over the last two years.  Now about the same as Teac retail.  The review says it provides more than the Teac but sounds sharp.  Inside looks to be an R core transformer.  The clock itself is shielded but looks to be a Chinese 4G clock as well.  Big capacitors near the clock must help, but I do wonder if we give the BG7TBL/DAPU clock a mid or top range power supply how close it would get to the low range Cybershaft solution.

 

Mutec - sounds like it has the right sound qualities for my taste.  With that price though, I'd wait for the Uptone ClockRegen.

 

I might just hop on MartinT's bandwagon and go with the BG7TBL/DAPU clock with 75 ohm terminals for now.  Entry level 75 ohm BNC cables actually seem reasonably priced.  1m Wireworld's Chroma is only $30 USD, Starlight 8 $100, Cardas Parsec $140.

 

I'd really like to know:

- How hot does the BG7TBL get?  Does it need a lot of ventilation?

- Does it increase sibilance, harshness, or make the sound forward?  Any change to warmth, or the treble/bass balance of the sound?

 

I'm a lowly layman, who is guided by John"s views on such matters. 

 

As I mentioned above, according to John, the ER, as other audio equipment, is designed around a square wave signal. You may want  to take that into account. 

 

In addition, as he explains in a different thread, the 75/50 ohm adaptors and toggles, as Cybershat employs, are ineffective and don't perform as intended. That topic was discussed when a proposal of equipping the ER with said toggle was floated up and   summarily rejected. I suppose Mutec offers two batteries of BNC ports, 50ohn and 75ohm respectively, for that reason. 

[LowMidHigh]

4 hours ago, Clockmeister said:

Hello AbAndgate

 

Old skool boolean logic been a while since collage for that one!

 

A great many DIY electronics enthusiasts tend to use  these impedance converting devices, its not really a complete solution if you are looking for a true impedance match.  Ideally all things being equal keeping the pathway the same Z all the wat through from Tx to Rx.

 

I appreciate however that most of the after-market clocks are specific 50 Ohms.

When we measure a 75 ohm impedance path ways and cables we use a set of true 75 Ohm calibration tools and have our VNA or TDR devices set to 75Ohm. Totally understand 99.7% of audio users are not going to have this facility.

 

What I can do is set up an example of a 75 ohm cable being used with 50 Ohm measurement parameters, this will give you an understanding of how the reflections can be seen on the waveform.

 

Think of it like a converter from SPDIF<>AES or single end to balanced always something lost in translation I feel.

 

Will it be better than a total mis match 50<>75? possibly yes, possibly no, sorry for not being more definitive.

 

 

 

Logic suggests that converting impedences would outperform a 50/75 mismatch. But only people with your type of equipment and knowledge can test test that hypothesis. 

 

On a different note of yours:

"Think of it like a converter from SPDIF<>AES or single end to balanced always something lost in translation I feel." 

You feel, or that's a fact? Again, logic would poise said loss, but can you validated with measurements? 

 

 

[LowMidHigh]

1 hour ago, Clockmeister said:

As promised some results from using impedance mis matched cables and adapters.

 

I used the lab's Mutec Ref10 SE-120 master clock as the signal generator, scope has been calibrated, all test cables 18Ghz reference ultra low loss 50 ohm, audio cables were 75 ohms matched. 50 Ohm adapter is an R&S reference 8Ghz model.

 

First images are a base line set, 50 ohm output, 50 ohm feed 50 ohm receiver, also I enhanced the waveform to highlight the very slight overshoot to give you an idea of  of a close up of the wave form top edge definition.

 

 

 

The photos are explanation are imminently clear. Thank you!

 

So in a case of mismatch, what would be most detrimental to the operation of the enslaved device?

 

1) Would the signal occasionally not trigger gate? It doesn’t seem the case.

2) Would the timing be off? The raise and fall in the mismatch cases seem slanted; not as plumb as when impedances are matched.

3) Would the ripple/reflections introduce fresh jitter?

3) Something else?

[LowMidHigh]

37 minutes ago, AnAndGate said:

Hi @MartinT,

I am trying to understand Mutec mc-3 clock distribution capabilities:  given both Mutec and Etherregen may prefer the square wave signal  (and there is only one output) from the BG7TBL OCXO. Can Mutec MC-3 be fed the square wave, and act as the distributor through one of the word clk outputs to the Etherregen; as well as using the reference clock for  USB reclocking?

 

thanks

Short answer, no. 

 

I believe you can get that clock custom made with 2 square wave outputs, both 75ohm. There was a discussion about it in this thread. 

[LowMidHigh]

1 hour ago, bernardl said:

At this point, knowing what we know, if budget were not too much of a concern, would there be any reason not to select the Cybershaft OP21A as external clock for the EtherREGEN?

 

- it has the lowest reported noise of any clock

- it has multiple outputs with selectable impedance

- it has a dedicated external linear power supply

- the company has a great reputation

 

Am I missing something?

 

Cheers,

Bernard

 

 

Bernard,

 

You'd better do a more thorough research:

- Selectable impedance doesn't yield the correct impedance either way. It's a faulty solution, hands-down.

- It generates a sinuous wave while audio equipment expects a square wave.

 

Perhaps the Cybershaft has the lowest phase noise, but when connected to your equipment, it will suffer from the aforementioned deficiencies. Also bear in mind that both Cybershaft and Mutec swore by -116db is the lower limit for sound improvements to could be perceived. Then marketing took over, and both companies introduced a -120/121db to up the ante. In essence, I suspect, we're in the flat region of the phase noise/SQ curve.

 

I suggest you take a deep breath and study the various offerings again. In my opinion, Mutec Ref10 is the best option, and the SE is the one if you have the have the "Best" 

 

Good luck!

 

 

 

 

[LowMidHigh]

33 minutes ago, bernardl said:

 

Thanks.

 

Would you have a reference to a document showing that the Cybershaft clocks emit sinusoidal signals and not square waves? I wasn't able to find such information on their site.

 

Cheers,

Bernard

 

 

It used to be on their website, and Kinji wrote to me as much.

 

When I first embarked on my ref clock journey, I gave his products a close look. Going by his own words, the products didn’t cater to the audio market specifically, but to any application that required a high precision clock in an affordable price. He emphasized that point when I inquired about a return policy: “Can't guarantee audio relate performance, so no returns.” As you can deduct from the impedance selectors and impedance convertors, delivering the purest signal isn’t a top priority for him. Furthermore, to my chagrin, he conflicted himself in our exchange too many times to build trust. 

 

Anecdotally, later on, when he realized demand in the audiophile market was building up, he started gearing his clocks accordingly: More outputs, better encasement, and initiating a war-of-tug with Mutec vis-à-vis phase noise. 

 

[LowMidHigh]

1 hour ago, bernardl said:

 

Thanks.

 

Have you ever tried one of his recent high-end clocks vs Mutec?

 

I read recently a comparison performed in HK where the conclusion was clearly in favor of the Cybershaft vs the Mutec Ref10 (or perhaps Ref10 SE120 not sure).

 

Cheers,

Bernar

 

I've seen the synopsis of that A/B.

I take this kind of exercises with a grain of salt. (I don't insinuate insensirty, but I can't rule out expectation bias). And I would certainly not spend $7,000 based on thier opinions, especially since the upgameship of the East has profoundly affected people's outlook nowadays.

(Even if Cybershat had the phase noise edge, its suboptimal selectors and shape of wave would likely render the signal more jittery. John S and another contributer above, with testing equipment, poisted as such).

Personally, since Mutec accepts returns and it hails from the pro industry with several complementary products, I would go with them. Mind you, the 8 ports can become very handy. I'm using 5 and I can see 2 mores put into service in the future. Lastly, presale, I communicate quite extensively with Chris, their CEO, and felt absolutely comfortable with my decision.

[LowMidHigh]

3 hours ago, bernardl said:

 

Thanks.

 

Would you have a reference to a document showing that the Cybershaft clocks emit sinusoidal signals and not square waves? I wasn't able to find such information on their site.

 

Cheers,

Bernard

 

Again, I just want to point out the business practices of Cybershaft, which omits such basic information from their literature. 

[LowMidHigh]

29 minutes ago, zerung said:

@LowMidHigh  That statement....care to expand your views?

 

This is solely my observation and opinion and reflects no prejudice whatsoever.

 

I reside in Thailand and I run into many people from Chinese, Hong Kong and Taiwan. I also read English issued Chinese publications. There’s a palpable sprit of competition on their behalf, with the aim of superseding anything that is western. Think China 2025. It’s a fascinating and foreboding process that stokes chauvinism and involves politics, big industry, commerce, education and culture perceptions, all which are beyond the scope of our discussion. What’s germane is that oriental consumers—in our case a Hong Kong based forum—are likely to show preference to the Japanese Cybershaft over the German Mutec. 

 

[LowMidHigh]

24 minutes ago, zerung said:

@LowMidHigh Sorry I dont buy this.

You need to be careful with your views.

I live in HK and if anything, there is a strong bias towards European things....(which is definitely changing), if anything else. Which to me is based on an individual preference, driven by analytical analysis of sound (Using the instruments on the sides of ones head) and not necessary objective inference based on a  test data's....

To push the whole of Asia off the map with a swathe of your 'Non Prejudicial' statement is alarming.

 

No one has pushed the whole of Asia off the map, and I don't have to be careful with my views, unless promoting violence or hate. 

 

The only interesting tidbit in your rebuttal is this:

"...is a strong bias towards European things....(which is definitely changing),"

The words in bold are actually in agreement with the process I've been witnessing.  Thank you.

[LowMidHigh]

2 hours ago, R1200CL said:

If your serious about purchasing a Keces, which look very beautiful with that display, I would suggest you do some private emails with Alex first and discuss a JS-2 purchase. 
Also not fully understand your whole system, there may be possibilities to move power supplies around in order to make sure no moat is broken. All this can Alex help you with out in an email exchange. 
 

I have no idea how good the Keyes is, and never heard about it, but those displays etc is adding up to the cost and not the SQ. Also just to bring in another very good PS is the Paul Hynes ones. But I understand you’re after a dual PS output.

 

Further it seems your budget allows for one AfterDark PS plus’s either of the following 

LPS 1.2 (Here is a cheap one). 

Farad

SR4T

That is opening up for more flexibility.  Just saying. 

 

Keces is a solid company with robust engineering and R&D. Personally, I consider the contribution of LPS is subtle. and distinguishing among the various products is nary impossible, given a blind test. 

[LowMidHigh]

5 minutes ago, R1200CL said:

I totally disagree.

 

https://www.wilsonamplifiers.ca/blog/50-ohm-vs-75-ohm-which-is-best-for-you/

 

https://www.microwaves101.com/encyclopedias/why-fifty-ohms

 

Even Esoteric uses 50 ohm. 
http://www.area51musicfest.com/products/catalog/pdf/g01_en-p.pdf

 

The article you cite contradicts your asserssion:

 

"For home applications, they do a perfectly fine job of transmitting signal up to 50 feet of cable with home or small building installation topping out at 5,000 sq ft." 

 

The most important thing is to match impedances The rest is splitting hair and nitpicking. 

 

 

[LowMidHigh]

5 minutes ago, R1200CL said:

https://sound-au.com/articles/coax.htm
 

Are these clocks 10 MHz ?

 

The fact also remains is that most if not all OCXO data sheets for 10 MHz  stats 50

ohm. 

 

Mutec Ref10 is furnished with both. The balance is tilted towards 75ohm, which reflects the majority of slave devices out there, e.g, MC3+USB and the ER. 

[LowMidHigh]

10 minutes ago, R1200CL said:

My understanding is a master world clock uses a different frequency than 10 MHz. Just because studio has used 75 ohms, cause cables as cheap, shouldn’t be our reason to standardize on 75 ohm. 
 

Anyway, most people looking for a clock for the EtherRegen will look at the cheap Chinese one, or hopefully AfterDark clocks. His clocks is 50 ohm. He just changed one clock for those that has the 75 ohm EtherRegen. The SU-1 used 50 ohm input. 

 

So using Mutec as a reference isn’t relevant these days.  And do we even know if there’s a 50 ohm or 75 ohm inside. Not that it really matter. 
 

Another important discussion is if we should standardize on square vs sine. Or does it matter at all. 

 

Sorry, but you should do more research. The MC3+USB data sheet explicitly states 75ohm for its clock input BNC. 

 

John S. discusses in one thread the difference between the 2 wave shapes at length, expressing the superiority of the square one. For that reason, I suppose, both the MC3+USB and the ER expect square input. 

 

Theoretically, you should match both impedences and wave shapes between the master and the slave. Feel free to do A/B or A/Bx in your copious free time. Would appreciate your report. 

[LowMidHigh]

18 minutes ago, AfterDark. said:

 

Hello! PYP,

 

For matching the impedances for 50ohm to 75ohm, we recommend to use Cybershaft 50ohm matching device for this. This is certified to work with EtherREGEN 75ohm edition.

 

Best Regards,

Adrian

AfterDark.

 

 

 

 

Perhaps that gizmo had remedial properties, but the signal is still degraded.

This is true to the T splitter to a lesser extent.

 

All in all, you achieve the best feed possible to a 75ohm ER by matching impedance and using a square wave. 

[LowMidHigh]

38 minutes ago, MasterWarzombie said:

Your words reassure me a little because I made the choice of newbie to buy the AfterDark Queen Square Wave clock in 75 ohm to match with my ER 75 ohm even if I am well aware that it is a cheap clock in comparison with some higher end models. I had an eye on the Sotm solution but the price was disproportionate for the moment with my first experience.

 

I think that under some circumstances spending on an expensive REF clock is unjustified. For example, if your environment is acoustically sub-optimal and that can’t be changed due to an aesthetically sensitive wife.  

 

Put differently, re-clockers, audio-grade switches and Ref cloaks are the icing on the cake—and some cakes simply don’t deserve it. 

 

Furthermore, those costly clocks come to maximum bear only when one’s system consists of multiple devices that can be tethered to said clock. Spending upward of $6,000 on a REF10 SE120 to merely connect an ER to it is rather a waste. My opinion of course.

  

[LowMidHigh]

7 hours ago, owldiscourse said:

Just ordered an ER and and an Afterdark 75 ohm clock. So it's best to get two LPS 1.2 vs one JS-2 then? I'm embarrassed to say I have no experience with LPS other than an iFi I picked up to run my switch. Never really noticed much difference with that one but never did extensive A/Bs either.

 

A question I have is that I'm using a very good power filter - the Shunyata Everest which has isolated duplexes. Would I still benefit anyways from using (2) LPS? I have enough outlets to plug the clock and ER.

 

PYP nice setup with Mola Mola btw!

 

As PYP alluded, you're better off with LP-1.2 x 2 (if the LP-1.2 has the oomph to power the AfterDark).

 

[LowMidHigh]

29 minutes ago, MasterWarzombie said:

Etherregen should preferably be used with a square wave, but this is still an audiophile debate.

 

it's a fact that isn't subject to a debate.

John S. covered it at length in one related thread. 

[LowMidHigh]

2 hours ago, ambre said:

 

HI EverytimeHigh, 

 

NO.  We are still waiting for the clock papers by Uptone & Swenson company.  Repeatedly stating ...before buying clock products, please wait for the clock paper to come out.......to make an informed choice!

 

In the Audio industry like Merging Nadac also Master Clocks brands Mutec, Tascam, Antelope .....all use a Sin Wave 50 Ohm and less 75 Ohm (mostly for video) in their studios and consumer applications.

 

Suppose there are more or less conflicting issues and starting points for design / approach /and proper implementation of a 10Mhz of 25 Mhz Master Clock in a switch, re-clocker like Mutec MC3 USB  and maybe a new ISO_OXCO _Regen with 10Mhz input?   Also the necessity to use a high quality clock cable... to minimize various kinds of jitter, ground plane issues with a known or unknown causal effect? 

 

Strange enough many people including myself and many studio professionals prefer /  like the sound of a clocked  Mutec, SOTM and EtherRegen switch? Maybe it's technically spoken ...worse.... not better?  Or do we like this kind of jitter or other contaminants?

 

We all are hearing apparently positive differences...... by using  a really cheap Chinese BJ7, an Afterdark Giesemann, a Cybershaft, a Mutec SE 120 Master Clocks? And not to be underestimated the effects of a good BNC Clock cables. Like the Afterdark (50-Ohm) and the more (very) expensive SOTM, Chord etc.  BNC-cables?  

 

Whats is going on ?

 

Regards, Andreas 

 

Are you aware the ER expects a square wave? If not, do your research and all the pieces will fall in place. 

 

As for SOTM championing their sinuous wave, IMO their marketing materials are full of fluff. I'll take John's opinion over theirs.  But for each their own

 

 

 

[LowMidHigh]

3 hours ago, R1200CL said:

There are different ways to obtain a square, so it’s debatable 

 

You may degenerate the signal so much, you’re better staying away from a square. 
So question is how to determine if a square is done right ?

 

A convoluted way of thinking, as poor implementation doesn’t obviate the fundamental principle of matching a transmitter to receiver characteristics. 

 

And since you keep hinting your doubts in the manufacturers’ charts and ability to produce satisfactory waves, perhaps it's on to  back those doubts with measures or at least A/B (or better still, A/Bx). 

[LowMidHigh]

If anyone is interested in a premier review of the clock's wave shape, here's prior posts by John and Alex:

 

 

 

[LowMidHigh]

 

1 hour ago, Superdad said:

I cleaned up the distracting drivel.  (The privilege of being a moderator in our own forum. ) Carry on...

 

 

I didn’t understand his main argument: “You have to clean it up at the source.” My source is a router, through which I stream my music in a continuous fashion. So what do I do instead of re-clocking it with an ER that minimizes jitter and phase-noise? Should I get a yet-to-appear audiophile router?  Contact my internet provider and insist they “do something?” 

 

I don’t like when people are kicked off from forums. “First, they came for the communists, e.t.c.” But people who consistently divert from the thread’s subject should be banned from posting there—but only there.  

[LowMidHigh]

12 hours ago, R1200CL said:

Sine is obviously the best to use, and rather have the receiver convert the signal to square. 
 

Creating a good square isn’t easy. Also square demands better cables.

 

Please keep using sine 50 ohm as your standard, and offer 75 as an option.

 

If you change something in your clocks, I think searching for help, and create a better square output stage is where efforts should be. 

 

 

 

Why would they want to shoot themselves in the foot? 

75ohm/square is the industry standard for audio equipment (with only Japan not falling in line, I believe). 

 

it doesn't matter how loud and long you're going to bark at that tree--that standard won't change. Get over it and move on.