Frank

You have probably gathered that this is not a simple topic.

1st, I would say that IIP3 is not restricted to mixers, all components 
that utilise semiconductors have non-linearities, and so they could all 
have an IIP3.  For example, low noise amplifiers have an IIP3, but you 
can't easily compare an amplifier to a mixer.  Also, filters, especially 
crystal filters, are non-linear, as are inductors that use ferrite such 
as torroids.

IIP3 means the third order intercept point, referred to the input, due 
to intermodulation distortion (IMD).  IMD can ONLY occur if there are 
two (or more) signals present.  Imagine only one signal at the input to 
a system (it doesn't matter what the system is), then no matter how 
strong that signal is, there cannot be any intermodulation.

Now imagine two signals at the input to our system, and that system is a 
radio receiver.  Let's assume their amplitudes are equal - this is 
unlikely in practice, but it makes the maths a lot easier and this is 
how IMD is best understood.  Lets assume that the signals are quite 
strong, and are at 27 and 28MHz, i.e. 1MHz spacing.

If we put a spectrum analyser on the input of the receiver we would see 
just the 2 signals.  Now, if we put the spectrum analyser elsewhere in 
the receiver, for example after the first gain stage, we would see more 
than two signals, each separated by 1MHz.  These extra signals are the 
Intermodulation products, caused by non-linearities - in this case in 
the Rx amplifier.

The closest IMD products will be at 26 and 29MHz, and will be at equal 
amplitude.  These are the third order IMD products, because they are 
procuced by the 2nd harmonic of each of the two input signals mixing 
with the fundamental of the other.  We will ignore the other IMD 
products for this discussion.

The level of the IMD products will be less than the two input signals. 
Now, it gets more tricky - if the level of BOTH of the input signals is 
increased by 1dB, then the levels of the 3rd order IMD products will 
increase by 3DB.  So, the difference between the levels of the two, 
real, input signals (27 and 28MHz) and the IMD products (26, 29MHz) has 
decreased.  The IIP3, (3rd order input intercept point) is the 
theorectical level at which the level of the two input signals and the 
IMD products is exactly equal.

The IIP3 is always theoretical, because in practise the system 
(amplifier, mixer or whatever) will have gone way into compression 
before this point is ever reached.  Therefore, IIP3 can never be 
measured directly - it has to be determined from measurements made at 
lower input levels.

I had to make a few assumptions in the above statements, in order to 
explain the principle.

I hope that helps.

regards

Grant G8UBN

Reply via email to