|
 |
phase noise doubler - RF Cafe Forums
|
yz
Post subject: phase noise Posted: Wed Apr 05, 2006 11:52 am
Lieutenant
Joined: Wed Apr 05, 2006 11:42 am
Posts:
1
Who can explain why phase noise after doubler increase by 20log2=6dB
and not 10log2=3dB ?
Top
fred47
Post subject:
Phase noisePosted: Wed Apr 05, 2006 7:18 pm
General
Joined: Wed Feb 22, 2006 3:51 pm
Posts: 104
Hi!
After
a doubler, there are intermodulation products added, as well as harmonic
products. See Steven Maas' book Nonlinear Microwave & RF Circuits
for more details.
Good Luck,
Fred
Top
IR
Post subject: Posted: Thu Apr 06, 2006 10:02 am
Site Admin
Joined: Mon Jun 27, 2005 2:02 pm
Posts: 373
Location:
Germany
Hello Fred,
Without reading the book so far, intermodulations
and harmonics are considered as signals and not noise, so how can they
increase the phase noise?
_________________
Best regards,
- IR
Top
fred47
Post subject: Phase NoisePosted:
Thu Apr 06, 2006 4:55 pm
General
Joined: Wed Feb
22, 2006 3:51 pm
Posts: 104
Sorry, yz, I misread your question,
and answered a similar but not identical question about why performance
is often worse than 20 log n.
The quick answer to yz's question
is that phase noise comes from undesired phase modulation of the carrier
by noise. The amplitude of the sidebands is determined by the modulation
index (approximately a linear dependence at low modulation indexes).
The modulation index doubles when the frequency is doubled. Since power
is voltage squared, the 10 log for dB becomes 20 log : log v squared
= 2 log v.
IR,
Signals and noise are ultimately indistinguishable.
If you could distinguish them in any way, noise could be removed from
a corrupted signal - an outcome we'd all stand up and cheer for! Considering
phase noise as resulting from incidental phase modulation by noise (usually
device noise) is a standard approach.
Sometimes principles are
easier to see when discussed in sine wave terms, which is why the distinction
sometimes gets made.
In this case (and again, my apologies for
answering the wrong question), sometimes you see worse-than-20 log n
performance due to the intermodulation products.
Peace,
Fred
Posted 11/12/2012
|
|
|
|
|
