VCO Phase noise: consequence for mixing...

[Chromatik]

Hi, there!
I'm new on this board, and maybe I'm veryy close to rules, so sorry in
advance!
My question, despite its simplicity, is rather head breaking: I'm
trying to understand and quantify the consequence of phase noise, due
to a VCO, when injected in a mixer.
As I think this phase noise could be assimilated as a coloured noise,
so a gaussian SPD, I don't really imagine what could happen going
through mixing process: a white noise??
Then, is it possible to quantify the consequence of this phase noise on
the mixer's sensibility?

Sure you could teach me some, and it'll be very kindfull.
Thanks in advance!

Mike

PS: sorry for my unperfect english!

 

[John Woodgate]

I read in sci.electronics.design that Chromatik

I'm new on this board, and maybe I'm veryy close to rules, so sorry in
advance! My question, despite its simplicity, is rather head breaking:
I'm trying to understand and quantify the consequence of phase noise,
due to a VCO, when injected in a mixer. As I think this phase noise
could be assimilated as a coloured noise, so a gaussian SPD, I don't
really imagine what could happen going through mixing process: a white
noise?? Then, is it possible to quantify the consequence of this phase
noise on the mixer's sensibility?

Phase noise is just a type of frequency modulation [delta(f) =
delta(phi)/delta(t)]. Considered as frequency modulation, the spectrum
has an additional gradient, proportional to delta(f).

I expect you can se that mixing doesn't change the shape of the spectrum
of a frequency-modulated signal, so it doesn't change the spectrum of
phase noise.

The effect on mixer sensitivity depends on how the signal is
demodulated. An AM detector may be quite insensitive to small amounts of
phase noise, whereas an FM detector (or similar) cannot distinguish it
from the wanted signal.

 

[Chromatik]

Hi, Mr Woodgate,
Thanks for your answer,
It's true, my question was not so clear enought, when I've talked about
phase noise injected in a mixer. Let's consider the problem again:
In a PLL, the input signal (with phase noise, due to the elements
before and emission) is mixed with the VCO's - filtered input
controled- signal, like in this scheme

input signal
¦
Mixer----->----- filter --.-----> output signal
^___ VCO________¦

Due to the loop, and of course non-linearities, I think - but I'm not
sure- that the output signal would have a "white noise" component.
I'll try it with a Matlab sim...
.... what do you think about it?

Thanks again for your help

Mike

 

[John Woodgate]

I read in sci.electronics.design that Chromatik

Due to the loop, and of course non-linearities, I think - but I'm not
sure- that the output signal would have a "white noise" component.
I'll try it with a Matlab sim...
... what do you think about it?

I'm afraid I'm not into the finer points of PLLs. But several people
here are.

 

[Chromatik]

OK, so I might wait a little
No problem, thanks !!

Mike

 

[maxfoo]

Hi, there!
I'm new on this board, and maybe I'm veryy close to rules, so sorry in
advance!
My question, despite its simplicity, is rather head breaking: I'm
trying to understand and quantify the consequence of phase noise, due
to a VCO, when injected in a mixer.
As I think this phase noise could be assimilated as a coloured noise,
so a gaussian SPD, I don't really imagine what could happen going
through mixing process: a white noise??
Then, is it possible to quantify the consequence of this phase noise on
the mixer's sensibility?

Sure you could teach me some, and it'll be very kindfull.
Thanks in advance!

Mike

PS: sorry for my unperfect english!

downconverting improves phase noise that's why offset PLLs are so popular...

 

[Chromatik]

Hi, maxfoo !

For you, what are the difference with a "classic" PLL and an Offset
PLL?
I mean in a "classic" PLL, the input signal is mixed with the looped
VCO signal (like in my little scheme, 4 messages up), so phases are
compared...
If you have a link or docs on these Offset PLL, I'll be very kind !

Thanks a lot in advance!
Regards,

Mike

 

[maxfoo]

Hi, maxfoo !

For you, what are the difference with a "classic" PLL and an Offset
PLL?
I mean in a "classic" PLL, the input signal is mixed with the looped
VCO signal (like in my little scheme, 4 messages up), so phases are
compared...
If you have a link or docs on these Offset PLL, I'll be very kind !

Thanks a lot in advance!
Regards,

Mike

Don't have any specific links, but IIRC Microwave Journal magazine had an
article on offset plls back in sept 2004. think you can get the article from
their website if you register with them.
Offset PLLs are widely used in GSM.