[time-nuts] Low cost alternate to Dual Mixer/DMTD

[WarrenS]

John 

Thanks, helpful and constructive points, 
I'll try and comment on them all.

> Ordinarily two oscillators won't track that well in
> response to shared environmental conditions alone.  
And I'm well aware of that. 
I started by measuring their interactions, each both on their own independent PS 
and across the room from each other.
One of the side benefits of that test was that in order to see their interactions 
I first wanted to get them stable enough, long enough to test them.
The units I'm using have NO power supply sensitivity, 
NO gravity sensitivity, NO temp sensitivity and 
No sensitivity to anything else fixable that I can measure and test for.
How I did all that is a different story.
   

>Fortunately injection locking is the easiest (and cheapest) malady to fix!
>Make no mistake, this is not something you can blow off in the general case.  
> It'll be good to see some comparative results once you have some additional isolation in place.
Believe me it is something I am well aware of and something that 
I have considered and tested for from the beginning of this project.
I have made many improvements to it already, but of course there always 
still more improvements that can be made, and  it is certainly possible 
that I did miss something. but from results that I'm now getting 
with Allan numbers etc, it would suggest that I've got it pretty well under control.
This is very much an ongoing project and I will try your and Bruce's suggestions 
to verify and improve what I now have.


> Have you tried modulating your DUT oscillator with white noise and FFT'ing
> the correction signal to visualize the actual loop bandwidth? 
No, Up to now, I have tested it with various function generator wave forms, 
and the noise from several different Oscillators that have widely different amounts of noise and isolation.
I'll see if I can come up with something more like you are suggesting.
One of my many other tested Osc is a Freq synthesizers from a separate buffered Low noise Osc.
I do believe that this Can NOT be PULLED or effected in freq by pretty much any external effect.
I just  used that to check how much I could pull the 10811s and as you (and Bruce) point out
it is very easy to cause giant effects to the 10811s when talking about, 0.1 ps and  1e-12 freq changes.
I now have it so that I can not see any effect at the 1e-12 range, 
and have for now concerned myself with other more important things that need improvement.


> Your setup sounds like a very reasonable approach to 
> better-than-TIC grade timing measurement, 
> It can certainly win the price-performance contest over other architectures.
    This started out as something very simple that I could put 
together with the parts that I had on hand.
I am very much into either trying for the simplest 'KISS'  design 
OR trying  for most accurate design.
To my amazement this has turned into BOTH, not only a simple configuration but 
one capable of better performance than any standard thing available that I've heard of. 


> Another challenge will be to find a way to determine your true residual floor. 
> You can't just feed both DUT and reference ports from a splitter and watch the
> residual phase error over time, correct?
True, From the test I've done so far the noise floor is at least a little below 
and probable way below my 10811 Oscillators, so testing just how low it 
really is way down on my 'need to do' list.
One test I did do was open the loop and feed the same Osc into both ports,  
one thru a phase shifter and see what the Open loop noise was.
That test caused me to change Phase detector from an active one that 
I started with to the passive one I'm now using.
I have come up with several possible addition ways to test it but for now
I'm content thinking that the tester's noise is WAY below the best Osc that I have to put on it.


>> The Only significant error outside the loop is the Osc its self
>> and Osc's internal EFC offset.

...
>> The EFC range being used during a measurement period is typically
>> less than 1/1000 of its range.
> As Bruce points out, that doesn't matter in the least.
I think you are mixing up several of  my unrelated comments and answers.
The "1/1000 ..." comment  JUST Had to do with the nonlinearly issue and why it is of NO concern.
The "Only significant error outside the loop" comment was addressing noise issues, NOT pulling or Osc interactions.
The "Osc it's self" comment which I stated is the only significant error source includes 
ANYTHING that effects the OSC like pulling, PS,  temp, G,  Time, noise, etc, etc.

> The injection locking is the very reason why you are seeing such 
> a small EFC correction range. 
That statement is way off. Yes it could be true BUT it is not this time.
The low EFC correction range is because the noise of these two Osc is in the 1e-11 to 1e-12 range.
and even with the noisiest OCXO that I have which is in the  1e-9 range, 
that is still only in the mV range out of a possible 10 volts.
I'll send you some plots off line of some of my test that show that I'm doing is at least very close.


thanks for the feedback
ws

*************

From: "John Miles" <jmiles at pop.net>

>> > If there is another contributor to phase locking then the full frequency
>> > instability wont be reflected by the EFC input fluctuations.
>> Don't get me wrong, in the beginning the oscillators pretty much
>> acted like they were sync locked
>>  because of poor ground, common PS, non isolated RF output,
>> etc,etc. That did not go away by luck.
>> What I'm saying is they do not now know that each other exist to
>> the level that I can test to, which is about 5e-13.
> 
> It's been my experience with the HP 3048A, which is basically the same thing
> as what you have built, that at least 20-30 dB of reverse isolation is
> necessary when two OCXOs (all 10811s and most of the Wenzel units I have)
> drive the LO and RF ports of a single mixer.  Make no mistake, this is not
> something you can blow off in the general case.  It'll be good to see some
> comparative results once you have some additional isolation in place.
> 
> Have you tried modulating your DUT oscillator with white noise and FFT'ing
> the correction signal to visualize the actual loop bandwidth?  My guess is
> that your effective LBW is somewhat wider than your design target, thanks to
> the injection-locking bugbear.
> 
> That said, yes, your setup sounds like a very reasonable approach to
> better-than-TIC grade timing measurement, especially with a digital back
> end.  It can certainly win the price-performance contest over other
> architectures.  It's great to see several different approaches actually
> being tried, as opposed to speculated upon!
> 
>> Something that some seem to of missed is that this is not the
>> standard RF circuit configuration with the standard open loop errors.
>> This is a closed loop Neg feedback "PID type" freq control system
>> where the errors inside the loop are reduced by the loop gain.
>> The Only significant error outside the loop is the Osc its self
>> and Osc's internal EFC offset.
>> The EFC range being used during a measurement period is typically
>> less than 1/1000 of its range.
> 
> As Bruce points out, that doesn't matter in the least; in fact, I'd submit
> that injection locking is the very reason why you are seeing such a small
> EFC correction range.  Ordinarily two oscillators won't track that well in
> response to shared environmental conditions alone.  Fortunately injection
> locking is the easiest (and cheapest) malady to fix!
> 
> Another interesting challenge will be to find a way to determine your true
> residual floor.  Because your measurement is frequency-based, you presumably
> can't just feed both DUT and reference ports from a splitter and watch the
> residual phase error over time, correct?
> 
> -- john, KE5FX
> 
> 
> 
>