Low Jitter 20MHz oscillator

No affiliation, just an old customer.

Jitter is a function of the time span over which you measure it. You can draw a graph of RMS jitter vs measurement interval, in other words the jitter when you use the oscillator to time out various intervals, like if you divided it down to lower frequencies.

Cycle-to-cycle jitter dominates the low end of the graph. Most cheap crystal oscillators have a few picoseconds RMS jitter cycle-to-cycle and for timing intervals roughly up to a millisecond. If the osc is used to generate longer intervals, RMS jitter will start to increase roughly linearly with the interval, with the corner in the 1 ms range maybe. A cheap XO might have a few ps RMS jitter for short times, and maybe 5 or 10 ns per second. A really good SC-cut ovenized oscillator might have a few picoseconds jitter at 1 second.

The Fox Xpresso oscillators have low cycle-to-cycle jitter, like 1 ps.

Thermally insulating an XO with foam or something greatly improves long-interval jitter.

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation 
Picosecond-resolution Digital Delay and Pulse generators 
Custom timing and laser controllers 
Photonics and fiberoptic TTL data links 
VME  analog, thermocouple, LVDT, synchro, tachometer 
Multichannel arbitrary waveform generators

Any reasonably good crystal oscillator would do, provided sufficient power filtering. There is nothing special about few ps jitter at 20 MHz.

Vladimir Vassilevsky DSP and Mixed Signal Designs

You can easily slap one together with a 74HCU04. Just make sure it's the "U" version. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| Phoenix, Arizona  85048    Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.

Conner-Winfield DV75C series has maximum integrated phase jitter 1ps RMS over 12kHz to 20MHz. Period jitter 5ps RMS maximum. Also very good temperature stability for a low-power non-ovenized oscillator (which I happen to need).

Some info on jitter specs:-

Making an oscillator is easy. Making a low jitter oscillator isn't. A few uV of power supply noise would mess up the jitter specs of a 74HCU04 based oscillator. Minuscule temperature variations would probably do the same.

Jeroen Belleman

I doubt the "Minuscule temperature variations", and bypass capacitors are there for a reason >:-} ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| Phoenix, Arizona  85048    Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.

Xtal base clock for 20 mhz should be ok..

Digikey has lots of XC type osc that are surface mount that fits your needs.

Jamie

How do you plan to measure the jitter?

Why does it matter?

--

John Larkin Highland Technology, Inc

jlarkin at highlandtechnology dot com

Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser drivers and controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation

Hello Jim,

What do you mean with "slap together with a U version" ? I believe an integrated oscillator will not gain benefit with an additional U version of anything.

Best regards, Habib

John,

Measuring jitter is really the point (method based on something which ressemble on eye diag or something like that ... i don't know at the moment) the best is that i would not measure anything like that, just based on component spec.

I need a CMOS low jitter oscillator (few ps in absolute way) only on specs for feeding the clock input of an16bits AD converter (is that so hard ?)

Some guys said that a basic 74HCU would be sufficient ... i did'nt read anything such a thing on any 74HCUxx datasheet.

I remember i had used IDT chips feeding a 100MHz PLL clock input for a PPC (AMCC PPC440) with something like 1 or 2 ps absolute jitter. I'm expecting that it's not so hard to find the equivalent on 20MHz.

I will manage the point this week.

Thx anyway for the topic. Habib

r with a 74HCU04.  Just make sure it's

A 74HCU04,an xtal, 2 caps, 2 resistors and you have a very nice oscillator

-Lasse

Really good point, thank you. But for the silabs solutions i think these guys are too obfuscated ... and i'm too in a hurry.

Best Regards, Habib.

Ahh ok a simple RC oscillator on 20MHz ! mmhhh ... i would not experiment such a basic thing when stable, cheap and reliable integrated oscillators exits for years with a jitter as low as 1ps.

Thank anyway, Habib.

Habib, Your debut if off to a very shaky start. Most of the world uses 74HCU04 as the core of their crystal oscillators. And those two-pin ports on Microchip's uP's... guess what's inside ?:-) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| Phoenix, Arizona  85048    Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.

You're clueless... with a mouth to match :-( ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| Phoenix, Arizona  85048    Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.

You're French... what can you say ?>:-} ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

not an RC oscillator a crystal oscillator, the R's are for bias and drive the C's for the xtal

-Lasse

Jim,

I will never experiment a RC or Quartz + 74HCU or anything like that to achieve a low jitter oscillator. I leave this sort of thing to some flashers beginners. No offense.

And the fact is i'm French has nothing to do with this.

Habib.