Superclock in SL-1210

[Marco]

Marco, the crystal is divided down so far it's likely to make 0.0000000001% difference of the square root of f_ck all. It's a specious claim at best. What we really need is a demo- heads up, back to back. ;-)

Indeed - I totally agree. Perhaps, since you know Bob, that could be arranged? I'd be WELL up for that!

However, I'd still love to find out what crystals both T/Ts use - for the simple satisfaction of knowing if the one in the Monarch is technically superior (even if not in practice). If it isn't any better, we can discount that as being the reason why the Monarch might sound better than a Techy.

Marco.

[sq225917]

If it sounds better it will be down to removing the electronics, and the integrity of the platter and plinth, certainly not 10ppm better xo accuracy. ;-)

[Marco]

Indeed! However, it would still be nice to know the crystal compliment used in both

Marco.

[WOStantonCS100]

Indeed! However, it would still be nice to know the crystal compliment used in both

Marco.

Agreed.

[ar-t]

WOW!!!!!!!!! Lots of good questions, and thoughts for discussion. I will try to answer them all, in order. If I miss yours, bang on my ear. It is only a senior citizen moment.

[ar-t]

If what you say about sourcing crystals is true, then I have a big question for Inspire about their Monarch turntables. They say, "...such as a new
quartz chip of greater spec..." So, either someone has already done what you say only has a slim chance of being possible or the Inspire folks aren't being honest. I would love to know which one it is.

I would never say they are dishonest. The guys in marketing tend to get all excited about little things. It is outside of their purview, and things that are insignificant, to the technical guys, seem like a big deal to the non-technical guys. And they tend to blow them out of proportion.

Besides, it makes for good copy.

[ar-t]

Thank you Patrick, that's very useful information. The crystal type as far as I can see is a HC49 low profile can.

So, if we assume for the moment that the crystal cannot be improved upon, we are left with two scenarios:

1) Lift the crystal from the PCB and make a temperature controlled environment for it, say between 40-60C.

2) Lift the crystal from the PCB and put it into a better/more stable oscillator circuit.

The question is, will either of these give us lower jitter than the crystal in the AN6680 circuit fed with a very clean power rail?

Temperature control will do nothing for jitter.

Stability has nothing to do with jitter.

Jitter is, for the most part, a function of the parameter we call "Q". There are 2 values: the unloaded, which is the Q of the crystal, and the one that really counts. Which is the loaded Q.

Loaded Q is the Q of the crystal, when it is actually in the circuit. All oscillator circuits will lower the Q. Laws of physics at work. You can take a high Q crystal, and load it the wrong way, and it will perform poorly.

However...............

There are lots of other things that enter into it. I have lots of high Q crystals that do not perform very well. Lots of things enter in, most involve the presence contaminants, inside the crystal can.

As I said "If it was that easy................"

[ar-t]

That's an obvious and demonstrable claim, most simple Xo come in one of two specs, 30ppm or 20ppm, other than that you really would be specifying an uber part in a oven and a decent tcxo is going to cost a lot more than Bob could afford to put in that deck. The simple way to find out is to look at the part currently in the 1210, if it's the 30ppm piece then It can be improved, if it's a 20ppm then it really can't.

I view all such 'marketing' claims with healthy skepticism. I'm with ART on this one.

Ah, the old "Ours is better, because it is x ppm, and yours is y ppm" argument.

OK............it is useful spec, in terms of absolute accuracy. But, in this case............

Let's look at this. A crystal is designed to operate into a specified load, and we will say it is 18 pF, for discussion. What this means if you stick it into a circuit, that presents the crystal with a 18 pF load, then it will oscillate at the right frequency.

Sort of. Within reason.

20 ppm. 30 ppm.

What this means "If you stick in a 18 pF circuit, it will oscillate at the specified frequency, within +/- 20 ppm.

Translated: 4 MHz clock.............20 ppm..........will be off no more than +80 Hz, or - 80Hz.

When it comes to your TT (and you have absolute pitch), that note that is supposed to be 440 Hz will be off as much as 0.0088 Hz. I am not a musician, so I will leave it to someone else to tell us how many cents it is off.

There, ppm demystified.

Has no bearing on jitter, contrary to what many of our Asian buddies will tell us. Their schtick is "Good quality!" Ok, but, being polite, their measure of quality is much different from ours. When it comes to crystals/oscillators, this is their sole measure of quality.

Other than...............yeah, price. Which is price, not quality. Try telling those guys!

BTW..............those fancy ovenized clocks, kinda sorta like what someone will stick in your DAC (for an extra $10,000!) start at around $500. Give or take a few bucks. Yes, they have good phase noise. And those guys will make pretty much any frequency, and one, if you only need one. All it takes is cash. And lots of it.

[ar-t]

Pat,

Your comments regarding sourcing crystals, and their use, is helpful. Industry is slowly taking us all down to the lowest common denominator. Fortunately there are still those who are dedicated to making things better.

Perhaps you would care it illuminate us as to why you think there are no regulators capable or regulation up to 100 MHz.

Regards
Paul

AH! Good question.

OK, let's take something most DIYers have run into: LM317. Your basic regulator, that operates on feedback. Let's look at the data sheet.......

Output z, at 1 kHz, is 3 milliohms. Yep, pretty good. But, the feedback is already starting to fall off, which means it will rise on a decade per decade rate. At 10 kHz.........yep, up to 30 milliohms. 200 milliohms, at 100 kHz.

Hmmm...........is that right? Yes, because............that cap you need to make it stable starts to enter in. So, the rate will slow down. At least until the cap hits its self-resonant frequency. (Which is something that also deserves discussion, but not today.) When we get to 1 Mhz, it is now 1 ohm. Doesn't seem like much of a regulator, to me.

If we extrapolate, using the 1 MHz value, and discounting the effect of the cap (which in this case will probably be past its self-resonant frequency), we get 10 ohms, and 100 ohms, at 10 & 100 MHz, respectively.

Let's look at something that does not use feedback, so there will not be falling loop gain, vs frequency: the simple emitter follower.

Some nerd will probably point out that an emitter follower does have internal feedback, and blah, blah, and more blah. (If you want to argue that, I bet some guys at DIY Audio want to argue with you. I don't!)

Ignoring all of that............does its impedance stay constant?

No...........

Some of you may have noticed the output will "go inductive". Which means it will rise, just like an inductor does. So, we chose to say "it has gone inductive."

Where does it start to become inductive?

100 kHz.............maybe 200 kHz.............

Google "alpha transition frequency". That is the point where a follower becomes inductive.

[ar-t]

Thanks, Simon



So if that's the case, and as you say likely what Inspire have 'improved' in the new crystal that they're using, it's not going to account for any huge improvement in sound.... Is it? Therefore, whatever is making the Monarch sound so good (as per the reviews), if you are right, obviously isn't the new crystal.

The only way to achieve a genuine upgrade in that area, it would seem, is to have a top-notch bespoke crystal designed, as Partick has outlined. And based on what he's said, that ain't going to happen anytime soon!

What needs to be done, so that we're dealing in facts, not fiction, is for the crystals used in both the Monarch and the Techy to be identified, and that information reported here. Who's up for that job, then?

Marco.

Hey, it may very well sound better. They probably tweaked something else. Saying it has better stability, crystal, etc., sounds more high-tech than.........well, something simpler, like buying a cheap, industrial-grade cap, that improved performance. We all know that only audiophile grade caps can make it sound better!

I think they call that "artistic license".