Spurious Measurement - RF Cafe Forums
Post subject: Spurious Measurement
Thu Sep 16, 2004 12:32 pm
I maintain software for a station
that tests RF modules. However, I am new to RF, so I have a question
regarding a calculation. If I have a known carrier (f) and a known
spurious signal that is f+x away, what is the proper calculation
for dBc between these two? Can I just subtract the dBm value of
each from each other? I have searched but have been unable to find
the info I am looking for.
Unread postPosted: Thu Sep 16, 2004 1:53 pm
You can subtract the dBm values of each as you
mentioned. There is another important term in RF related to spuriouses
and especially to intermodulations, it is called IP3 and stands
for 3rd order Intercept Point. This parameter is especially measured
in non-linear devices such as Mixers and Amplifiers and is calculated
P1= the power level of the
fundamental (desired) signal (in dBm units);
Delta= the difference
between the level of the third order product: 2f2-f1 or 2f1-f2 to
the level of the fundamental signal (in dB);
is in dBm value.
If your station supports PLLs' measurements
then this term is not relevant, and you just have to subtract the
values of the two signal levels.
Hope this helps,
Thu Sep 16, 2004 4:27 pm
Thanks... That is what I tried,
but differing results than a different test that written by someone
else. His may because instead of two known signals, he is doing
his test across a "region", or spectrum. On this information he
is performing his calculations.
Thanks again for your help
and if I need to know more (obviously I do)I will be back....
Unread postPosted: Fri Sep 17, 2004 1:45 pm
Joined: Sun Aug 03, 2003 2:02 pm
Location: Erie, PA
might be a little late to the game here, but for what it's worth
* Itay's remarks on IP3 product calculation are correct when
the tones generating the intermods are well below the compression
point of the system, but begins breaking down when within about
10 dB or so of the P1dB (for just two tones - lower as more tones
* Since you did not specifically address 3rd-order
intermods as being the cause of the spurious signal, then here are
a couple possible reasons for disagreement of your results with
- If the two signals are not CW (have some
bandwidth), then the dBc value measured previously might be the
difference between the integrated powers of the two signals.
- The previous measurement might account for any attenuation in
the system (a filter, maybe) that introduces a difference in the
two signal levels other than simply the amplitude delta displayed
on a wideband spectrum analyzer.
- Your measurement system could
be generating or adding power to the spurious signal. Add 10 dB
at the input of the spectrum analyzer and see if both signals decrease
by 10 dB (they should).
What is a typical discrepency between
your measurement and the other guy's?
- Kirt Blattenberger
Unread postPosted: Tue Sep 21, 2004 7:07 am
In the readings I have seen so far, the delta, typically, has
been about 10dB. One thing I noticed about his algorithm is that
because he doing interpolation across a spectrum region (e.g., 4MHz
to 9MHz above & below the carrier) he is adding in a linear
factor to help with his calculations.
In my case, I am just
looking at the carrier vs. a particular spurious signal at a particular
frequency (a test requirement). This may possibly be the explanation
for the delta between his readings and my readings. I just would
not expect such a large delta. Also, to answer your inquiry, I am
using a wideband spectrum analyzer (Agilent E4440a).
for lack of a better expression, I am a newbie to RF signal analysis.
But the math should work (I think).
Tue Sep 21, 2004 12:14 pm
Joined: Mon Apr 12, 2004 4:50 pm
Hwrd69, Are you making your measurements though a directional