John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Member
Joined 2004
Paid Member
State of the are for close in phase noise is the Oscilloquartz BVA http://www.oscilloquartz.com/files/1363164953-Br_ OCXO 8607.pdf . Don't rush to place your order. The lead time was 6-12 months and they are reputed to cost in the $10,000 range. They still have higher phase noise at higher frequencies than a Wenzel blue top ($1,500, 8 weeks if you are lucky). The possibility that either would sound better is very small but it would help justify a $100K DAC.

Richard, the simple fix would be to get the critical oscillators for your effort from Mouser or Digikey: the Crystek oscillators are more than good enough and available as individual parts. If the frequency is not a ratio of 22.5692 or 24.576 its not an audio clock. The 3rd clock in the box is likely for the USB interface and not likely to have any influence.
 
Disabled Account
Joined 2012
clockPlots.jpg



OK thanks Demian.


-RNM
 
AX tech editor
Joined 2002
Paid Member
The easiest would be to phase lock a good RF generator to the existing clock and then FM modulate it with an external generator. Swap its output for the internal clock going to the DAC. Any "synthetic" jitter will not really represent what happens in a specific architecture.

Actually his AP will do all that to a few mouse click. The dScope does it for me.

Jan
 
Do you guys need something to build in your spare time? '-) However, there may be subtle errors, so use your engineering skills to find them. OR you could buy a used Para JC-2 and have it all done for you!

John,

Is there a particular reason why you would use MOSFET follower for the output, instead of JFETs ?
And why such high values for the source resistors (R13, r14) and output resistors (R64, R67) ?



Thanks,
Patrick
 
State of the are for close in phase noise is the Oscilloquartz BVA http://www.oscilloquartz.com/files/1363164953-Br_ OCXO 8607.pdf . Don't rush to place your order. The lead time was 6-12 months and they are reputed to cost in the $10,000 range.

Richard, the simple fix would be to get the critical oscillators for your effort from Mouser or Digikey: the Crystek oscillators are more than good enough and available as individual parts. If the frequency is not a ratio of 22.5692 or 24.576 its not an audio clock. The 3rd clock in the box is likely for the USB interface and not likely to have any influence.

Currently, the BVAs are pure unobtainium, even for us at work, and we have bought a
substantial number of them in the past. AFAIK they have sold the farm to Microsemi.

Yes, the Crysteks are quite good, especially if one considers price, size and dissipation.

For my home lab, I have bought one of those Lucent GPS disciplined sources from ebay;
they have 2 double ovens for redundancy and one GPS receiver. The price oscillates
between $150 and $200. At the moment, we are in a low price period.

The oscillators are at 5 MHz since that is a sweet spot; I have done a doubler/distributor.
< http://www.hoffmann-hochfrequenz.de/downloads/DoubDist.pdf >
Sorry, no kits from my side.

regards, Gerhard

(and no, I won't comment on using that stuff for audio.)
 
Member
Joined 2002
Paid Member
It doesn't have 10MHz

Yes, it doesn’t. (*)
I read that you abandoned the idea, this is the good news.
Top notch modern Cesium frequency standards offer at their 10MHz output SSB phase noise worse than most of the clocks you showed on your post #68534

Exctract from Microsemi 5071A specs
SSB Phase Noise 10 MHz :
1Hz - 100dBc
10 Hz - 130dBc
100Hz - 145dBc
1000Hz - 150dBc
10000Hz - 154dBc
100000Hz - 154dBc


Extract from an older Datum FTS 4040A specs
SSB Phase Noise 10 MHz :
1Hz - 99dBc
10 Hz - 130dBc
100Hz - 140dBc
1000Hz - 150dBc
10000Hz - 150dBc
100000Hz - 150dBc


George
(*). You were interested on a used one. HP6061 is the (only?) one ending on e-bay. I was confident though that you would have focused on the important issues.
 
Member
Joined 2002
Paid Member
I do have a rubidium clock. Fairly fast to get a lock in the PLL locked crystal oscillator included. But phase noise is higher than my best reference oscillators. But I just couldn't pass up the chance to get a piece of equipment that used to be almost unobtainable.

More than a decade ago, during the investigation phase of purchasing a Caesium clock for disseminating time in a cal lab, tight budget turned us to Rubidium clock. In both cases phase noise was questioned (it would be very convenient for us if we could double the utilization of this unit as a local reference standard for frequency calibrations. We decided that we shouldn't).
But Richard may ask the seller for the phase noise of this Rubidium clock generator
10 000M Ultra Precision Low Noise RB Word Rubidium Atomic Clock Generator HiFi | eBay
and compare to some of these
High Reliability Ruggedized Oscillators | Time & Frequency References | Timing & Synchronization Systems | Products

Now there are some that can be programmed to be used as a CD clock. But a good crystal oscillator should be better in that application.

That was the clear message in jxc post!

See Fig. 35, Fig. 36
Introduction to Quartz Frequency Standards - Oscillator Comparison and Selection
and these
http://www.endruntechnologies.com/pdf/Osc-Guide.pdf
GPS-Derived 1/5/10MHz Low-Phase-Noise Outputs


What about Fermi Velocity? Scott?

Ohm's Law, Microscopic View

I hope we will not turn back to April 2014 (all the then participants are still active and kicking, ready for a new round :D )

For my home lab,...
The oscillators are at 5 MHz since that is a sweet spot; I have done a doubler/distributor.
< http://www.hoffmann-hochfrequenz.de/downloads/DoubDist.pdf >

Respect !

George
 
Last edited:
Ohm's Law, Microscopic View

I hope we will not turn back to April 2014 (all the then participants are still active and kicking, ready for a new round :D )

Respect !

George

I haven't seen jn for a while so I pass.:)

I always thought charge was a property not a thing that stands by itself. I think your link will do for the wires we solder down on our circuits.
 
C=q/V ? You know one of those "First Principle" things! :)

What do you think q is? Charge without a carrier is an eccentric concept.

An object is negatively charged if it has an excess of electrons, and is otherwise positively charged or uncharged.

That is the top and bottom plates of a charged capacitor.
 
Last edited:
why? are you now designing liquid He cooled MC preamps?

if you even skim the paper you will find an equation with abs temperature K^-3 dependency for an effect that is ~1-2% at 610 mK

what's 2%*500^-3 (hint: I had to change my 10 digit display calculator to engineering exponential mode)
 

You really think that charging a capacitor is not doing work on the charge carriers and storing energy in the field? Charge and electrons are used interchangeably at times, sloppy sometimes maybe.

Electric charge is the physical property of matter that causes it to experience a force when placed in an electromagnetic field.

When two materials are in contact, electrons may move from one material to the other, which leaves an excess of positive charge on one material, and an equal negative charge on the other.
 
You really think that charging a capacitor is not doing work on the charge carriers and storing energy in the field? Charge and electrons are used interchangeably at times, sloppy sometimes maybe.

Do you really think Fairies move charge around as you stated before?

Do you really think that Santa Claus gave you all that stuff?

Or do you just manage to mis-read just about everything?

Where do you get the idea that charging a capacitor does not involve work or anyone even hinted that?

It is clear that you confuse charge with electrons. Current is the movement of charge it is not the movement of electrons. You accuse me of being confused about electron drift.

Now George didn't want all this repeating so this will be my last post on this issue. You can have the final word from whatever ethereal plane you are floating by at the moment.
 
Last edited:
Status
Not open for further replies.