Jump to content
Computer Audiophile
One and a half

Teac 10MHz clock : CG - 10M

Rate this topic

Recommended Posts

lmitche   
3 minutes ago, kevin gilmore said:

I found it on ebay, new in box, never soldered to etc. I got lucky, keep looking.

 

It sounds better than the frequency electronics ones which are DDS. There are 2 different kinds of these, some are programmable the others are not. They are all pulls from cell phone tower electronics packages, so no idea how much life is left.

 

The lamp is part of the rubidium physics package. Which also includes a heater for the detector. The oven for the crystal is

something else.

Thanks!

Share this post


Link to post
Share on other sites
4est   
8 minutes ago, kevin gilmore said:

I found it on ebay, new in box, never soldered to etc. I got lucky, keep looking.

 

It sounds better than the frequency electronics ones which are DDS. There are 2 different kinds of these, some are programmable the others are not. They are all pulls from cell phone tower electronics packages, so no idea how much life is left.

 

The lamp is part of the rubidium physics package. Which also includes a heater for the detector. The oven for the crystal is

something else.

Oh, OK. Thanks for the info. I am glad to see your posting about this, and John S's reply. I too have found external clocking to be an improvement at times despite the protestations that an external clock cannot be "better".

Share this post


Link to post
Share on other sites
lmitche   
52 minutes ago, Forehaven said:

If you get one of these Larry, how/what device do you plan on utilizing the clock?

It was more of an exercise in curiousity than anything else.  It started when I owned a Brooklyn DAC with a word clock input. That was sold long ago.

 

For now my approach is to use the lt3045s to deliver ultra clean power to boards with clocks.  I'm also working to eliminate as many devices with clocks as possible.

Share this post


Link to post
Share on other sites
mourip   

Glad to see this topic being explored especially with John and Kevin joining in.

 

There is also a Mutec REF10 thread here. I recently evaluated and purchased one and then posted my findings. I was able to thin out my equipment string by adding it and am really pleased with the outcome.

 

The DIYer in me would love to find a less expensive option for use in my headphone rig...

Share this post


Link to post
Share on other sites
On 10/21/2017 at 5:46 PM, mansr said:

Why do people think the long-term stability (years) provided by atomic clocks is at all relevant to audio applications?

It is not except that it can give peace of mind regarding effective life expectancy of the clock circuit/module itself;  other than that, it gives more than some red herring statistics to throw in for the makers of Rubidium oscillator based clocks as well as others.  "ppb" statistics sound way cool and more impressive than "ppm" stats ignoring the fact that Rb circuits tend to have much higher phase noise than their OCXO counterparts when the OCXO modules are of sound design and high manufacturing quality.  For microwave/satellite/advanced telco transmission implementations long-term stability is useful and adhered to as a key criteria from what I've been told as it is more germane to those industries' needs for master clocking.

 

Short-term stability, temperature stability (which affects overall performance and stability as a whole), phase noise and Allen Variance are "the" criteria to evaluate a clock circuit and/or external master clock or external word clock (those being different things) for its likelihood of good performance in an audio or video setting.

 

My 2 cents...

Share this post


Link to post
Share on other sites
On 10/22/2017 at 1:49 AM, JohnSwenson said:

The linked article is way out of date and not even looking at the same things.

 

There are some REALLY good clock synthesizer chips on the market now, which did not exist when this article was written. These chips lock on to the reference clock and can synthesize multiple frequencies while adding extremely low phase noise of their own. There are two types on the market, the jitter attenuating type and the type that assumes you have an extremely low jitter reference. IF you have an extremely good reference the second type will give you better results, but the first type can take a not so great clock and turn it into an output that is much better than the reference!

 

IF the DAC uses one of these internally then feeding it a really good external reference IS a good thing. Of course building that ultra low phase noise OCXO in the DAC to begin with would be even better, BUT having an external connection into one of these chips gives the user more flexibility. The DAC can come with a very good but not insanely expensive clock, which dramatically decreases the price of the DAC. When the user wants to upgrade they can add the insanely expensive reference clock without having to buy a new DAC.

 

I can guarantee that NONE of the devices in that linked article had anything close to what is available today. Most of those "master clocks" that were covered are actually pretty lousy clocks. Their primary goal was to have lots of flexibility in input types and output types so they could be used in many different setups. The circuitry they used was even poor for the time. As far as phase noise goes they bear extremely little resemblance to the reference clocks that are being discussed these days, which JUST output 10MHz, but do it EXTREMELY well. 

 

That article goes into great length discussing how a master clock can get degraded by running it through a microphone cable. Nobody here is talking about running a 10MHz reference clock through a microphone cable. The people buying these are going to be using highly optimized extremely wide bandwidth cables which have very small impact on such a signal. Such a cable can be had for $50 from the professional RF companies so it is not particularly expensive to transfer the output from the reference clock to the DAC.

 

All this is predicated on having a DAC which uses one of these modern extremely good clock synthesizers. If it doesn't, then yes, adding an expensive reference clock is going to be useless.

 

John S.

John,

 

Absolutely agree with all of this! The clocks tested are dated technology know for having bad phase noise characteristic and therefore injecting high jitter values into the audio (or video) chain.  They are not even close to what is available today.  I noted with interest this an other key 'facts' that would be proven wrong using a very high quality external clock like the Esoteric Grandioso G1 or G-01, Sforzato/BVA implementations, various Cybershaft products, etc...

 

There is one extremely valid point in the article however that I am in 100% agreement with, that being the point about the need to use impedance matched tees and terminators at every point of connection where a digital cable is attached to a component. I have doing that as of late with noticeable results.  In the world of 10-Base2 (the old days) and other more modern communications or networking standards where coaxial cable standards are utilized, this is absolutely followed as best practice.  In my system, I did notice a difference after the tees and terminators were applied and allowed to break in.  Took me a bit by surprise how noticeable the difference was but it is there for sure.

 

Studios with multiple channel needs are not the only ones who need master clocks or word clocks.  Each time over the last 10+ years with separate Transport & DAC and even some single-box players that I've added a high-quality external clock, the sound quality has improved as well as sound-staging, tonal and imaging accuracy as well as pacing.  A: B tests have also been run with a few very cynical audiophile friends....all point to the betterment of the final MUSICAL result when a high-quality external clock is used.

Share this post


Link to post
Share on other sites
rickca   
14 minutes ago, One and a half said:

This vendor from Spain lists the CG-10M clock for EUR999 and there's the new streaming DAC and DAC NT-505, UD-505 as well. Barely 12 months for the UD-503 to be superseded.

I found one vendor selling on amazon.com for $2500.  It's for CG-10M-S.  Does the Teac CG-10M have different quality clocks at different prices, or this just a huge discrepancy?

https://www.amazon.com/Generator-Reference-CG-10M-S-Domestic-products/dp/B076NG2NMG

 

Can you actually buy this product yet anywhere?  

Share this post


Link to post
Share on other sites
7 minutes ago, rickca said:

I found one vendor selling on amazon.com for $2500.  It's for CG-10M-S.  Does the Teac CG-10M have different quality clocks at different prices, or this just a huge discrepancy?

https://www.amazon.com/Generator-Reference-CG-10M-S-Domestic-products/dp/B076NG2NMG

The -S suffix is usually a silver edition as this nomenclature is taken from the NT and UD devices.

 I haven't seen one on Black (yet). There's only the one. The price originally was YEN148000, EUR1120, so the EUR999 is not far off the mark. USD1300 or 1500 perphaps but not USD2500 and the Amazon part is 100V if it's direct from Japan. The clock is very new, and maybe hard to deliver on short notice.

Share this post


Link to post
Share on other sites
rickca   
2 minutes ago, One and a half said:

The -S suffix is usually a silver edition as this nomenclture is taken from the NT and UD devices.

 I haven't seen one on Black (yet). There's only the one. The price originally was YEN148000, EUR1120, so the EUR999 is not far off the mark. USD1300 or 1500 perphaps but not USD2500 and the Amazon part is 100V if it's direct from Japan. The clock is very new, and maybe hard to deliver on short notice.

Good point about buying a Japanese domestic product.  Thanks for confirming there's just the one model in either black or silver.

Share this post


Link to post
Share on other sites
mourip   
12 hours ago, rickca said:

I found one vendor selling on amazon.com for $2500.  It's for CG-10M-S.  Does the Teac CG-10M have different quality clocks at different prices, or this just a huge discrepancy?

https://www.amazon.com/Generator-Reference-CG-10M-S-Domestic-products/dp/B076NG2NMG

 

Can you actually buy this product yet anywhere?  

 

That vendor is trying to capitalize on lack of early US availability.

 

For not much more than that you can probably find a discounted retail Mutec REF10 that runs on US voltage from a US vendor with a US warranty and has 8 outputs so that you will never need a tee connector for clocking more than one device.

 

Share this post


Link to post
Share on other sites
mourip   

If you mean the REF10 compared to the Mutec USB they are different animals. The Mutec is a great "swiss army knife" and can re-clock and also provide a wordclock but is not a reference clock. The REF10 is a reference clock and is used by other devices to make their wordclock more precise. The Ref10 is much more expensive. ~1K versus ~3K but is made to a very high standard with LPS instead of an SMPS ...besides being used for a different function.

 

Take a look at the REF10 and Mutec 3 threads here on CA.

Share this post


Link to post
Share on other sites
Fyper   

No I don't mean the Ref10 but the MC3+USB. More and more DACs have a 10M input (no need to use a word clock between the ref clock and the DAC. Therefore you can compare reclocking and master clocking with same DAC.

Since the price tags of the MC3 and this TEAC is close, I'd imagine that the reclocking done by the Mutec would give a result comparable to the master clocking of the TEAC.

Another way to say would be : if the TEAC Ref clock is used as a master clock on the MC3+USB, I wouldn't expect an improvement compared to the MC3+USB alone.

I'd think the REF10 is way better than the TEAC.

 

Share this post


Link to post
Share on other sites
24 minutes ago, Fyper said:

 

I'd think the REF10 is way better than the TEAC.

 

Well, it should be, since it's 4 times the price of the CG-10M. This is the issue, I could afford the CG-10M, but the REF10 would take a lot longer. Now if Mutec could produce a knocked down version, not so many ports, try 2 of each 50 and 75 Ohm output? 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×