[time-nuts] Re: Phase noise with a lock-in amplifier.

[David Kirkby]

Poul-Henning Kamp wrote:
> In message <4262A4F2.6010508 at onetel.net>, David Kirkby writes:
> 
> 
>>So if the two inputs are fed to a mixer have different path lengths, the 
>>inputs to the mixers will *not* be identical.
> 
> 
> Right, but I think there are much easier ways to measure it.  For
> instance, send the signal into one of the modern high-speed A/D
> converters (65MHz, 14 bits) and do a FFT on the result.

Well, putting something together with those sorts of chips is not 
trivual, whereas commerical lock-in amps are on eBay quite cheaply.

Also, the resolution of the fast A/Ds is limited. You say 14 bits, and I 
think that is about the state of the art at that speed. Compare that to 
audio A/D's, which have 24 bits, and cost a fraction of the cost.

> (If you want to try this, check out the "USRP" which Matt Ettus built
> for the GNUradio people: http://www.ettus.com/)
> 
> 
>>I'm pretty sure if you low-pass filter the output of that mixer, you 
>>will see the oscillator phase noise shifted from 10MHz down to near DC.
> 
> 
> Mixing highly correlated (or as in this case: identical) signals is
> far from a trivial undertaking.

That might well be a problem.

Sometimes I have these 'barmy' ideas. Sometimes they lead somewhere 
useful. Perhaps this one does not. But when I get some time I will hook 
up a commerical lockin and a mixer to an oscillator and see what I 
measure. I have a Marconi sig gen that goes to 2.2GHz. I assume the 
phase noise on that at 2.2GHz will be a lot poorer than the reference 
oscillator in that sig gen (10MHz). It will be intersting to see what 
differences are observed on a lockin.

I guess some of you must think I am mad.

-- 
Dr. David Kirkby,
G8WRB

Please check out http://www.g8wrb.org/
of if you live in Essex http://www.southminster-branch-line.org.uk/