How is the ADL5380's 3rd Order Intercept (IIP3) measured?

Document created by analog-archivist Employee on Feb 23, 2016
Version 1Show Document
  • View in full screen mode

Could you clarify how the IIP3 values were calculated for the ADL5380? It says
on the graphs
‘INPUT IP3 (I AND Q CHANNELS)’. Does this imply a vector sum of the I and Q
channels before measuring the third order intermods? Or are the I and Q channel
IIP3 values so similar that they cannot be distinguished on the graph?


The 3rd Order Intercept (IIP3) testing is done by applying two signals (i.e.
f1, f2), say with 1Mhz or 10Mhz difference and with the same applitude via a
combiner to the input of the DUT and in this case the RF  input of ADL5380.
There are two 3rd Products at 2*f1 – f2 and 2*f2 – f1 and the IIP3 =
Pfundamental – (IMD/2)

The data measured on the datasheet is looking at one leg only, either I or Q
and not the combination.  They are not vector combined. 

If you were to look at one leg either I or Q, the data should be the same. 
Independently there is no difference between the two paths from the perspective
of amplitude and frequency.  The difference lies within the phase and if both
signals were ideal then the IIP3 of the independent leg should be the same as
the vector combination.