[time-nuts] Question for my new GPSDO

[Bob kb8tq]

Hi

Stepping back a bit here …..

All oscillators have phase noise. It’s going to vary with the offset from carrier.
What you get at 10KHz will be much better than what you get at 1Hz. As long
as the EFC does not make this any worse, there is no need to get very crazy.
Putting the added noise 40 db below the intrinsic noise vs “only 6 db” will make
no measurable difference.

There are a lot of ways to come up with a phase noise estimate for an oscillator.
Measuring the device you have is one way. Looking at the spec sheet for that 
oscillator is also a pretty good way. Even looking at multiple spec sheets for similar
oscillators will give you a reasonable idea of what is probably going on.

A tone on the EFC line will generate FM (as opposed to PM) noise. In the full 
blown case you get multiple sidebands. For a noise voltage, the first sideband
should be all that matters. The key point is that there is a 1/F relation between FM
and PM noise. Effectively you get a lowpass filter for free. 

The more sensitive the EFC is, the more modulation you get from a specific level.
Just as with drift from the reference, lower sensitivity is better in this case. 

Most real oscillators have some lowpass filtering in them as part of the basic
circuit design. Just what point this is at may be very hard to determine. It’s usually 
a good guess that modulating an OCXO at kilohertz rates is not going to work very 
well :)

So there are a lot of variables to look at before one ever starts wondering about
lowpass filters. Understanding the components comes first. Past that it’s looking up 
various formulas and playing a bit with spread sheets. 

Bob

> On Nov 3, 2019, at 8:07 AM, Attila Kinali <attila at kinali.ch> wrote:
> 
> Hoi Tobias,
> 
> On Mon, 14 Oct 2019 11:49:29 +0000
> Tobias Pluess <tobias.pluess at xwmail.ch> wrote:
> 
>> I have now designed the EFC circuit such that it is easily possible to use 
>> different DAC and voltage reference models. I have also reverse-engineered 
>> the circuit which is used on the Oscilloquartz GPSDO. They seem to use two 
>> cascaded Sallen-Key lowpass filters with approx. 1 Hz corner frequency to 
>> integrate the PWM signal, so I have included this circuit as well in my 
>> design. This then allows later to test different DACs as well as the PWM.
> 
> One important and probably obvious point to notice here is, these 
> filters produce some considerable amount of noise. In order to get
> out best performance, the noise should be limited by the opamp.
> To achieve this, the resistors have to be small (this is the reason,
> why my design uses 2k2 resistors). But in order for small resistors
> to have the desired corner frequency, the capacitors have to be large.
> Unfortunately, large capacitors have drawbacks: high eqivalent series
> resistance (ESR), high dielectric absorption (DA), high leakage, and
> high microphonics. 
>