measuring iip3
I making a 2.1 Ghz LNA and i am using cadence RFDE with ADS, The thing is that i cant figure out how to do the IIP3 measurement using this software. I can get the fundatmental tone but i can figure out how to generate the 3rd harmonic using the two tone test...any suggestions?
Use two signals at input with same amplitude and a small (1 MHz for example) frequency difference and in harmonic balance specify the two fundamentals and the max order (for example for each fundamental you can use order 5 and as max order use 4).
I hope it can help.
Mazz
The third carrier due to nonlinearity of the system will appear at those frequencies.
2*w1-w2 and 2*w2-w1 and 2*w1+w2 and 2*w2*w1
these are "3rd Order Harmonic Contents" and 2*w1-w2 and 2*w1-w2 are particularly important that are very close to wanted carrier.
IIP3=OIP3-Gp
So, when you measure IIP3, you should observe these components at the output and simply extract the Power Gain of the LNA to find IIP3.
Hey,
Thanks for the help...the thing is i cant understand how to give two inputs of different freqs ..... i am using the P1_tone to give the input in cadence RFDE ADS, when i put two P1_tones in series i dont get an ouput for the third harmonic....
F1 and F2 frequencies should be close as possible to observe in the band of interest.For instance F1=2.1GHz and F2=2.105GHz and so the difference will be 5MHz that will be observed in the band.
The 3rd order components in this case will be
at 2.095Ghz and 2.11GHz ..
You will check the difference between these components and one of the main carrier.This will give you nonlinearity FOM of the system.
hi
you take two sourses put combiner and sum port connect to the you DUT
David
you can directly use P2_tone compont,set two carriers and powers.In ADS,you can use this component instead two P1_tone components.
Hope it can help you
Ref : obtaining IP3
This is a dangerous thing, as many stages, specifically complex arrangements do NOT follow the required 3db per 1db input increase law. Only ideal or idealized stages follow this law.It is a MUST in my opinion to test this behavior in 5 dB steps starting with - 40 dBm input and then step up and make sure that S11 is as close as possible to 0 !. The main reason why this does not always work in practice is that the devices generate sufficient harmonics which as an example, through the Miller effect, re-appear at the input..
Take a look at
RF/Microwave Circuit Design for Wireless Applications, Ulrich L. Rohde and David Newkirk, published by John Wiley & Sons, April 2000, ISBN 0-471-29818-2
Mixer and amplifier chapter