|
|
Post by MartinT on May 20, 2020 10:56:39 GMT
Who has one? Could you relate your experience with and without using the master clock? I'm talking about the external 10MHz reference clocks out there designed to improve the sound quality of streamers, DACs and reclockers. Typical brands in widespread circulation include Mutec, Cybershaft, Teac, SOtM, Antelope, SRS etc.
As with all aspects of hi-fi, the claims and counterclaims make for some fascinating reading. Typical measurements quoted are jitter, phase noise, Allan deviation. There is further complication with some clocks and devices using 50Ω connectors and co-ax cabling while others use 75Ω. Clock technologies used include crystals, oven temperature control, rubidium, caesium, GPS.
Use this thread for discussion of all master clocks.
|
|
|
|
Post by John on May 20, 2020 13:34:49 GMT
I think for you this is a natural progression
|
|
|
|
Post by wannarock2 on May 20, 2020 14:24:02 GMT
My Katana DAC uses the low jitter clocks from Nihon Dempa Kogyo Co. the - NDK NZ2520SDA. The power is filtered by 2 low-dropout regulators in a series along with a high frequency filter. Noise tested in UV. Buffers with power filtered 3 times. Jitter at 390fs. www.ndk.com/images/products/catalog/c_NZ2520SDA_e.pdfPs. If anyone can explain the above to me in layman’s terms it would be appreciated. (ツ) EDIT: oops, sorry i see in the OP’s original post this is for external reference clocks. still early here. coffee has not kicked in.  |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 20, 2020 14:29:35 GMT
Still a lot of room for improvement though 390 femtoseconds is fine, at what frequency and how was that figure arrived at? how is it specified C2C, TIE? or bath tub?measurement? , however I would use an NHK over a Crystek clock any day.
The real key is looking at the amount of harmonic noise that is generated by the clock that creeps into the rest of the system.
|
|
|
|
Post by MartinT on May 20, 2020 16:35:36 GMT
The Crystek 575 is a really interesting oscillator. Loved by vendors for its great paper specs, less liked by DIYers who feel they can improve on them. My LKS DAC came with a 575 and the Coherent low phase clock that replaced it improved the sound quality quite substantially. Yet the EtherREGEN also uses a 575, so it seems you can't get away from them. Is it the Crystek or poor implementation? I know what Tony would say  |
|
|
|
Post by MartinT on May 20, 2020 19:43:58 GMT
Here's a question to which I haven't found a definitive answer. Is a squarewave or sinewave master clock signal better?
|
|
|
|
Post by The Brookmeister on May 20, 2020 20:20:49 GMT
Who has one? Could you relate your experience with and without using the master clock? I'm talking about the external 10MHz reference clocks out there designed to improve the sound quality of streamers, DACs and reclockers. Typical brands in widespread circulation include Mutec, Cybershaft, Teac, SOtM, Antelope, SRS etc. As with all aspects of hi-fi, the claims and counterclaims make for some fascinating reading. Typical measurements quoted are jitter, phase noise, Alain dispersion, drift. There is further complication with some clocks and devices using 50Ω connectors and co-ax cabling while others use 75Ω. Clock technologies used include crystals, oven temperature control, rubidium, caesium, GPS. Use this thread for discussion of all master clocks. Is this thread created because Tony is trying to sell you his clock? LOL thats so funny. |
|
|
|
Post by MartinT on May 20, 2020 20:50:59 GMT
Err, no, I created it because I'm fascinated by the technology and by how clocks can improve sound quality. I want to hear more about them, especially real user experiences.
|
|
|
|
Post by wannarock2 on May 20, 2020 21:24:46 GMT
Here's a question to which I haven't found a definitive answer. Is a squarewave or sinewave master clock signal better? trick question! having said that i like the sine wave up until they equalize |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 20, 2020 21:25:18 GMT
Jury's out on whether a TTL or clipped sine is more suitable for the task. Theoretically clipped sine should be more suitable due to its distinct lack of ultra leading square waveform therefore reduction in harmonic distortion further up the frequency ranges. Again theoretically not spewing RF all over the delicate digital electronics inside your dac/cd/streamer Therein lies the rub as a near perfect leading edge square waveform with minimalistic pre and overshoot should produce a more desired jitter figure 2 and reduction in wideband noise (not just at 20Khz lol) however in some instances it produces unwanted harmonic noise well into the low single figure Ghz that's billions of hertz which is really totally undesirable.
So do you go for the first principle route and try and find a way of dealing with the overshoot, post ringing and harmonic issues. Or have not such an accurate clock but with far less spectral noise to contend with and use a more aggressive cornerstone 2nd order PLL to deal with the jitter timing effects
Or do you fly in the face of convention and come up with a hybrid unit that spans the gap.
Though that may take a wee while and a not insignificant amount of funding for very respectable test and measurement set up plus a few years of research
Mind you we could ask MCRU for a long term loan of the BR black hole device pretty sure at least six individuals on this forum would happily except he could also throw in a new Solution 7 series pre amp as well pretty convinced on current form that will not be a problem  |
|
|
|
Post by MartinT on May 20, 2020 21:59:15 GMT
trick question! having said that i like the sine wave up until they equalize I didn't mean it to be a trick question, but I see you fell for it anyway! The sound of a squarewave versus a sinewave, when used for a master clock signal, will not match how the music sounds when emerging from the DAC. It's more about the jitter at the threshold point used by the DAC's clocking electronics. A squarewave gives a more defined 'edge' for accurate triggering but contains far more noise. A sinewave has much more slope and therefore potential for more jitter when triggering but injects a lot less noise into the DAC. |
|
|
|
Post by wannarock2 on May 20, 2020 22:25:42 GMT
^ Cheers! Thanks for the explanation.
|
|
|
|
Post by MartinT on May 21, 2020 20:45:49 GMT
I think for you this is a natural progression Short of redesigning my system and buying much more expensive components (which I'm not going to do), I agree, John, a master clock is just about the final icing on the cake. Apart from fiddling with cables and such, which is fun. I fancy experimenting with the idea soon. After all, I have two components which can use a master clock and my DAC already has a superclock. Everything in the digital path is important, right? You and I have certainly vouched for that truism. |
|
|
|
Post by John on May 21, 2020 20:52:36 GMT
Yes we at times pushed each other with this
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 22, 2020 8:47:28 GMT
My I suggest starting off with trying a SoTM clock if you are seriously looking at the devices you have listed above. AE are very obliging with their loans, it will give you a respectable base line from which you can work from. Another high regarded make is Drawmer a UK pro audio company that produce quality products at sensible pricing. Having used these myself in the past I can safely say the work very well. Drawmer D clock Information |
|
|
|
Post by MartinT on May 22, 2020 10:35:12 GMT
I will start lower scale and lower cost than that, so that I can establish a baseline of performance using an external clock. It may well sound worse than with using the existing internal clocks, but that is the learning experience for me. Experimentation is fun, so I'll see what I can obtain at low cost on the used market. It doesn't reflect on what I may end up with.
On the sinewave versus squarewave debate, I am reminded that the Coherent clock in my LKS DAC outputs a sinewave. It obviously has low jitter and phase noise but, importantly, doesn't output all the higher frequency hash involved with a squarewave. I shall keep that in mind.
|
|
|
|
Post by John on May 25, 2020 15:01:47 GMT
|
|
|
|
Post by MartinT on May 25, 2020 15:06:11 GMT
I'm way ahead of you, John, but I'm not looking at components for audio.
|
|
|
|
Post by MartinT on May 29, 2020 12:52:44 GMT
|
|
|
|
Post by MartinT on Jun 22, 2020 9:58:41 GMT
Members with high-end clock solutions, I'd love to have your thoughts on what you're hearing.
Do all clocks make similar changes to the sound? What are you using the clock into?
I only have my own experiences and what I've heard at Tony's to compare with.
|
|
📶NAS > Gustard x26p DAC > Benchmark LA4 Pre-Amp > NAD M22-V2 Amplifier > Revel F228Be Speakers
