UpTone Audio is now an official MUTEC dealer! New REF10 Nano clock in stock today — $1,799

[Superdad]

As we have become drawn into the world of external 10MHz reference clocks the past few years--for EtherREGEN and other devices--we have shared our knowledge, measurements, and observations here on this forum.  

 

We have followed the offerings from our friends at Cybershaft, AfterDark, and others.  And for our own consideration we have obtained OEM quotations from industrial OXCO module manufacturers. So we have a rather good sense of what it costs to produce a high quality complete clock-box. It truly is not a trivial endeavor, especially if one is to build a properly buffered/isolated multi-output unit with a really good square wave (sine wave is easier so you see those a lot; inexpensive boxes often have very poorly shaped square wave output due to techniques used).

 

In 2019, the German studio/pro-sound engineering firm MUTEC--headed by the brilliant Christian Peters--introduced the REF10, with 8 isolated outputs and fantastic phase-noise performance (at the outputs, not just of the OCXO module) of -145dBc/Hz (@10Hz offset). Nearly immediately it became THE standard for state-of-the-art reference clocking. But with a retail price of $5,000, it not only remained out of reach for many, but also challenging to justify unless a person had multiple components accepting an external clock. [I mean, our own $680 EtherREGEN is a bargain for what we pack in, but I never could advise someone to spend $5K on a clock just for our switch. ]

 

Now while MUTEC's many digital clock, distribution, reclocking, and format conversion products are quite popular for use in recording studios and other professional environments, Mr. Peters surely noticed and catered to the high-end audiophile market as well. Noticing some shifts--with increasing numbers of components accepting an external clock, and Asian firms jumping in with multiple low price offerings--MUTEC has responded with the much more affordable, 4-output REF10 Nano. 

Utilizing virtually all the same refined circuits of its larger brother, the Nano delivers close to the same guaranteed phase-noise performance, and the same exceptional square waveform.

Aside from having 4 outputs instead of 8, the primary difference between the large REF10 and the new REF10 Nano is the power supply.  The REF10 has a built-in full linear power supply, whereas the Nano has a built-in hybrid SMPS/linear supply.  The Nano's internal power supply can be supplanted by an external +15VDC supply of ones choosing as there is a DC input jack (a non-standard jack size for which they include a solderable mating plug).

 

That brings me to our news today:

Since MUTEC announced the REF10 Nano, I and others have been posting about it and watching for release of units into the field.

I recently entered into discussions with the owner of MUTEC's exclusive wholesale-only importer, TransAudio, and worked out pricing and representation arrangements.  For a few weeks all of MUTEC's export shipments of REF10 Nanos were stuck in German customs (odd since customs hold-ups usually occur only at the destination country), but TransAudio’s first delivery of Nanos arrived this week. I promptly purchased three units! 

 

So not only is UpTone Audio now an authorized dealer for MUTEC, but we are the VERY first dealer in the USA to have stock!

Here are the units which just arrived today. Two with black faceplate, one with silver face.

 

The price for the REF 10 Nano is $1,799. These are going to sell very fast so contact me right away if you wish to order.  

 

Our acclaimed JS-2 choke-filtered, dual-output, 5~7.4 amp linear power supply can be configured to have a 15V setting, but if you want a fine supply to power both the REF10 Nano and our EtherREGEN without defeating the EtherREGEN's active-differential isolation "moat," then you will want to wait for our forthcoming (Q1/Q2 2024) triple-regulated, isolated dual-output, 2A/4A JS-4 (est. $1,350).

If you order an in-stock JS-2 at the same time as a REF10 Nano, we will ship the 14-pound JS-2 box freight-free (the Nano's carton is light but large/long, so I doubt we can do free-freight on that).

 

By the way, UpTone also is authorized for all the other fine MUTEC models that TransAudio Group stocks here in the USA.  Those are--at new lower pricing:

REF10 SE120 — $5,499

REF10 — $3,599

MC3+USB — $1,299

MC3+ — $949

(Black faceplate units are the most readily available.)

 

WHY BUY FROM UPTONE?  Easy: Expertise, friendly personal consultation and service/support.  I think most everyone here knows that is what we are about.  Of course if you have a brick-and-mortar storefront dealer near you that carries MUTEC, then I encourage you to purchase from them. 

 

Also, a shoutout to @Clockmeister: Aside from being a literal rocket-science engineer, he runs Coherent Systems in the UK, and they are a MUTEC dealer--probably with more stock of the REFE10 Nano than anyone in Europe right now. So if you are on that side of the 'pond' please give them a call.

——————————————————————————————————

Here are more pics and details:

 

 

[Superdad]

...

[Superdad]

15 minutes ago, mfaoro said:

Congratulations! What a Christmas present!

Happy to put one under YOUR tree! Just say the word…

[Superdad]

5 minutes ago, agladstone said:

Can you also provide the specs for the other Mutec products? 

Sure, when I am back at my desk in the morning. But all MUTEC product specifications can be easily viewed on their website:

https://www.mutec-net.com/produkte.php

 

By the way, sales have been going great. Already sold the first 3 units and have another 2 arriving to our offices early next week. 
 

And we are offering a professionally terminated, ultra low-loss Belden 4794R high bandwidth 12G SDI 75-Ohm cable—your choice of length—to pair with it.

 

So feel free to contact me about ordering—the REF10 Nano or its big brothers—at their terrific new lower prices. 

[Superdad]

3 hours ago, The Computer Audiophile said:

 Very nice. 

And a very thick 8mm diameter—so a bit stiff.

[Superdad]

8 hours ago, Oggo said:

Are the outputs of the Nano truly isolated from each other...

 

Indeed the outputs of the REF10 Nano are fully isolated from each other. I have tested and confirm their claims. Plus I can see the tiny transformers and the buffer circuity before each output.

Unlike several other brand multi-output clocks, the MUTEC's outputs are actually isolated. 

[Superdad]

17 hours ago, Schafheide said:

P.S. Re the recommended burn-in time - does the temperature of the device enclosure have any effect on the burn-in time ie EtherRegen (very hot) vs Mutec Ref10 Nano (room temp) ??

Not really.  Most initial long-term change on circuit boards are due to "forming" of the capacitors.  That just takes time, can be subtle, and often has some bumps (sounds good initially, then nasty for a bit, then really smooth as it settles into a final state--all in the range of first 10 hours to 200 hours).

Then for clocks, you are always just looking at measurable stability from hours of warm-up--after each power-on cycle from cold. 

[Superdad]

29 minutes ago, Qstik said:

I've had a very odd experience.  I ordered a fairly inexpensive 2ft length of the above 8mm Belden 4794R cable from Blue Jeans Cable based on Alex's choice of that cable with the Nano he now sells.  It comes with quality BNC connectors.  It was intended as an experiment to see if it was better than the 3.8mm Belden 1865A 75 ohm cable as supplied by AfterDark with their Emperor Double Crown 75 Ohm sine MC into the external clock input of my Gustard R26 NOS DAC.

 

The recommendation of Belden 4794R applies to square wave clocks such as the Mutec.  For your sine wave clock you will be better served by a semi-rigid cable such as the one that Cybershaft offers (Cybershaft direct is less expensive but their web shop pages are down right now--so you can see their cable at AfterDark's page here:

https://www.adark.co/collections/cybershaft/products/cybershaft-50ω-bnc-cable-premium-2-version).

 

 

29 minutes ago, Qstik said:

I'm using a high quality 75/50 ohm matching pad to mate to the 50 ohm input impedance of the R26 external clock input, and yes, the matching pad makes a big improvement.

 

Perhaps that pad is also acting as some sort of filter as well because for sine wave clocks impedance matching does not matter at all. You would be better served by instead inserting one of the Mini-Circuits 11MHz low-pass filters (either BLP-10.7-75+ or BLP-10.7+, does not matter) at the input of whichever devices you are feeding the clock to.

 

[Superdad]

4 hours ago, Qstik said:

However, if a DAC, for instance, is expecting a 10 MHz sine wave external clock signal…

Sorry, no such thing as a DAC “expecting” a sine or square wave.  As long as the clock circuit receives a clock of the correct frequency it will trigger correctly. 
In general, a really sharp (steep rise) square wave will be better than a sine. But a good sine (with amplitude modulated harmonics filtered out) does work just fine as well. 
Deeper conversation about this subject requires diving deeply into the design, architecture, and performance of the receiving device’s clock synthesizer’s PLL.

[Superdad]

2 hours ago, Qstik said:

1.  By saying that a DAC "expected" a sine or square wave clock signal I meant that its clock receiving PLL circuit was designed, optimized and/or worked best with either sine or square wave signals.

Sorry, but it just does not work that way.

Here is a primer post of John's that may be helpful:

 

2 hours ago, Qstik said:

2.  I can understand that a square wave clock signal with a very sharp rise time might be best for minimizing the effects of ground plane noise while a sine wave clock signal might be best for minimizing time domain jitter.  Put another way to John's point, sine wave clock signals tend to be worse for ground plane noise sensitivity.

That's conflating a bunch of complex concepts!  You might do well to reread our paper on the subject:

Considerations regarding 10MHz external reference clocks...

 

2 hours ago, Qstik said:

But my question to John is why wouldn't a 10 MHz bandpass filter be best for 10 MHz sine clock signals?  I would expect ground plane noise to be possible above and below the clock frequency.

I think because the characteristics of the filter itself are best kept away from the signal you are trying to pass.

 

2 hours ago, Qstik said:

3.  Google tells me that the wavelength of a 10 MHz electrical signal is 98.36 ft.  Since the cable I ordered from BJC is two feet in length; it is very nearly an integer ratio of 50:1.  This is probably bad from a reflection standpoint.  I may just order a 3 foot length of the same Belden 4794R SDI cable to try as an experiment.

Honestly this discussion of cables and outputs--especially since you are using sine wave clocks and not a Mutec (square wave) which is the topic of this thread--belongs in the other big thread on clocking and the EtherREGEN.  [I might move over to there all these posts.]

Thanks for you enthusiasm for the subject Mike.  

 

BTW, I'm happy to report that REF10 Nanos are selling well for us. Been getting a new shipment in nearly every week and it seems that so far UpTone has sold more REF10 Nanos than any other USA-based firm.  And that's without ANY advertising! 

 

[Superdad]

1 hour ago, Schafheide said:

Has anyone else noticed this?

Yup. Clocks like to take their time to get warm and stabilize.  I just leave mine on 24/7. 

[Superdad]

15 hours ago, Schafheide said:

...but my concern (in leaving it on 24/7) is the life of the electrolytics (this, of course, also applies to the LPS).

Your thoughts?

 

There is likely much more stress on components from power on/off cycles--and heat/cool cycles--than from continuous use. Quality aluminum electrolytic capacitors will last decades.  I have power amps that are more than 40 years old and have rarely been powered off.  Still perfectly fine.  

Your Mutec clock is likely to outlive you. 

[Superdad]

6 hours ago, Oggo said:

@ Alex:

 

What would be your recommendation soundwise - Mutec REF10 (@ 3,600 $) or Mutec Nano + JS-2 LPS (@ 2,760 $ + extra costs for optional cables for AC and DC)?

Depends upon your system application and needs, but the performance and refinement of the big REF10 is hard to beat—especially at the new lower price (for years it was $5K). 
Call or write if you would like one. Can ship right away. 

[Superdad]

4 hours ago, PYP said:

Good review of the Mutec REF10 Nano:  https://www.alpha-audio.net/review/mutec-ref10-nano-in-time/

What a great price for such quality.  

 

Indeed, everyone is quite pleased with the REF10 Nano.  I can barely keep them in stock.

 

But there is something seriously wrong with Alpha-Audio/Jaap Veenstra's measurements of the REF10 Nano.  He comes up with a horrible -72dBc/Hz at 10Hz offset. Mutec specifies (and @Clockmeister has verified) -142dBc/Hz at 10Hz offset.

Really makes me question his other recent jitter/phase-noise measurements.

I am not a member of the Alpha-Audio forum so am unable to post a comment/question there. Perhaps someone else will, or perhaps I'll send Jaap (and Christian Peters at Mutec) an e-mail asking for clarification.

 

[Superdad]

2 hours ago, treitz3 said:

When I power up my amplifier, the Mutec resets itself and then I have to unplug the ER for about 10 seconds to reset it, otherwise the streamer gets no signal. After that? No issues.

Hi Tom:

Well the latter half of what you report is no surprise—since a clock has to be already be connected and running at the moment power is applied to the EtherREGEN and it does not tolerate even the shortest glitch or loss of clock; such necessitates reapplication of power (though it really does not need a 10 second pause—3 seconds is fine).

 

The real question is why your Mutec clock “resets” it itself and why is such caused by power-on of your power amp.

Perhaps somehow the voltage surge or droop—or some sort of static discharge—occurs when you turn on your amp and that causes a glitch in the clock output. Even a 3 microSecond glitch will cause the EtherREGEN to lose connection. The ‘A’ side might recover but the ‘B’ side will not—until you remove and reapply power.

 

Perhaps you can experiment with where your power amp is plugged into the AC mains.

Feel free to post a diagram of your system and perhaps we will spot another possible source of your issue or have remedy to suggest. Are you using the other outputs of your fancy REF10 SE120 with any other gear?

 

[Dang, I would have been happy to have sold you that top clock!  But just today I took our first REF10 SE120 order—from someone well known here—and he’s also getting an MC3+USB.]

[Superdad]

6 hours ago, Clockmeister said:

Yes I feel the measurement of that particular figure of merit is somewhat off, however the reviewer understood the purpose and performance of Nano 10 very well.

Sorry, but I think that regardless of his subjective praise of performance, publishing a review of a clock—that is certified at -142dBc/Hz—and claiming it is only -72dBc/Hz is an egregious error, one that should not stand.

Seriously: Every 10dB difference is an order of magnitude (10x) difference; So I won’t begin to calculate the unfathomable 70dB discrepancy he claimed. 

Jaap V./Alpha-Audio Lab generally seem to go to some lengths with their measurements (he recently built a shielded box to eliminate impingement of environmental RFI on tests) so this really surprises me.

But it seems clear that despite usage of an expensive Wavecrest analyzer, he is either using the wrong tool for phase-noise measurement or does not have any decent reference clocks connected for cross-correlation.

 

Honestly, I liken it to reading a sports car write up wherein the reviewer let all the air out of the tires, reported the car was very slow, but liked its performance nonetheless. 

[Edit: P.S.   Alpha-Audio’s measurements of the REF10 Nano claim 28 picoseconds of total jitter. Picoseconds?! It should be more in the range of single digits of femtoseconds. Remember, 1ps = 1000fs!  Something is seriously wrong there.]

[Superdad]

5 hours ago, Clockmeister said:

I was being dipolmatic (odd I know 😁)

You're getting soft Tony! 

 

I just sent a long yet respectful e-mail to Jaap Veenstra at Alpha-Audio about his very odd measurements--and I copied Mr. Christian Peters of Mutec.  

Hoping we can get this cleared up before some $100 Chinese clock vendor starts comparing themselves to Jaap's erroneous -72dBc/Hz / 28ps measurements... 

[Superdad]

4 hours ago, treitz3 said:

The pause in power did not happen with the AD clock. I just turned on the amp and everything worked as it should have. The only change I have made was to swap out the clock to the Mutec. This was when the Mutec presented the problem. I do keep the clock on 24/7 and will continue to do so.
The plug for the amp is one a different outlet than the one the clock is, and there is a PS in between the clock and the other outlet. In theory, this *should* prevent any power loss on a power up of the amp. The two outlets are on the same 20 amp circuit though.

Wow Tom, that's quite a diagram of your whole complex system! Nice job.

The RG400 Pro "power supply" you list at the head of everything: Is that one of those Richard Gray giant inductor units?  If so, I'd like you to try removing the Richard Gray box entirely. (Because of the effect of the Richard Gray box on the whole wall circuit, just plugging the Mutec into the wall may not be enough.) I suspect that something is happening when you power on the amp--even though it is not attached to the RG inductor box.

Also, you can try changing clock cables, though I doubt that is the cause.

 

That's all I can think of based on your diagram.

Cheers,

--Alex C.

[Superdad]

On 2/25/2024 at 8:44 PM, JayDog said:

The Aurender also has the capability to stream Tidal as well as an internal HD. As mentioned above, it has USB output only. Would it potentially make an improvement run the Ethergegen to the Aurender N10 and then use a Mutec MC3+ via USB from the Aurender to output to the Lyngdorf integrated? The Nano would then be clocking both the Etherregen and the MC3+. The Lyghdorf integrated has a multitude of input options and can accept any of the outputs from the MC3+. 

 

Hi Jason:

Sorry to be slow in reply (got in two glorious sunny days of spring skiing with my wife and friends--just before this massive storm hit us here in California!).

 

I think your best SQ is still going to be with your Lyngdorf TDAI-3400 on the ‘B’ port of the EtherREGEN—with no other Lyngdorf inputs used (lest you run risk of defeating the EtherREGEN’s active-differential isolation “moat”—as you would with say a USB connection between Lyngdorf and some equipment upstream/‘A’ side).

 

And as for possible implementation of your early Aurender N100 and a Mutec MC3+USB:

Hard to say if Aurender USB > MC3+ via AES (or S/PDIF) > Lyngdorf TDAI-3400’s S/PDIF input would sound better than just Aurender USB > Lyngdorf TDAI-3400 via USB cable.  

The MC3+USB would give you an opportunity to utilize another of your Mutec REF10 Nano’s outputs, and perhaps the Lyngdorf’s AES/SPDIF input will sound better than its USB input—especially with a fantastic fully isolated and reclocked signal from an MC3+USB. 

But all the above is very much dependent upon both the Aurender N100 USB and the qualities of your Lyngdorf’s various inputs.

 

On 2/25/2024 at 8:44 PM, JayDog said:

Down the road I will add a linear power supply for both the Etherregen and the Nano. Any thoughts about this potential set up?

Well you have my comments above about signals and gear, but for a nice linear power supply for EtherREGEN and REF10 Nano I'll urge you to sit tight for the production release (3 months) of JS-4, our triple-regulated, dual-isolated output, 2A/4A, 8 voltage settings from 3.3V~15V. It is going to be spectacular...