The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Measurements >> RF Measurements >> IP2/IP3
https://designers-guide.org/forum/YaBB.pl?num=1262010975

Message started by LeroyJenkins on Dec 28th, 2009, 6:36am

Title: IP2/IP3
Post by LeroyJenkins on Dec 28th, 2009, 6:36am

Gentlemen,

Can someone please help me understand IP2 and IP3 measurements? I've been in search of a good resource, not for how to measure these paramenters, but what the measured numbers actually indicate. I'm in search of a visceral understanding.

For example, I have a measured value of 77dBm for IP2 and 32dBm for IP3. Clearly these can't be actual measurments due to the magnitude, rather some projected value that somehow characterizes some sort of nonlinear behavior of my amp, correct?

Its not compression, gain compression is very straight forward, but what exactly do the second and third order intercept points mean in laymans terms???

Title: Re: IP2/IP3
Post by pancho_hideboo on Dec 28th, 2009, 7:13am

I can't understand meaning of your questions.
Please explain more clearly using correct terminologies.

IP2(IM2) have various type. On the other hand IP3(IM3) definition is unique.
What type of IP2(IM2) do you intend to evaluate ?

http://www.designers-guide.org/Forum/YaBB.pl?num=1193566207
http://www.designers-guide.org/Forum/YaBB.pl?num=1216663909/6#6

Title: Re: IP2/IP3
Post by LeroyJenkins on Dec 28th, 2009, 8:24am

The Second Order Intercept Point, also known as the SOI or IP2, is a measure of linearity that quantifies the second-order distortion generated by nonlinear systems and devices. http://en.wikipedia.org/wiki/Second-order_intercept_point

The third-order intercept point (IP3, IP3 or TOI) is a measure for weakly nonlinear systems and devices, for example receivers, linear amplifiers and mixers. It is based on the idea that the device nonlinearity can be modeled using a low-order polynomial, derived by means of Taylor series expansion. http://en.wikipedia.org/wiki/Third-order_intercept_point

Besides being a measure of nonlinearity i'm trying to understand what a high number means vs. a low number.

For example if i have two amps, one that measures 10 for IP2 and one that measure 1000 for Ip2 what is the difference? Does the first produce higher distortion? What about IP3?

Someone explain this to me like you woulld to a first grader, I will be very appreciative.

Title: Re: IP2/IP3
Post by pancho_hideboo on Dec 28th, 2009, 8:34am


LeroyJenkins wrote on Dec 28th, 2009, 8:24am:
For example if i have two amps, one that measures 10 for IP2 and one that measure 1000 for Ip2 what is the difference?
You don't seem to understand IP2.
Being different from IP3, there are many definitions for IP2.
Again see http://www.designers-guide.org/Forum/YaBB.pl?num=1216663909/6#6


LeroyJenkins wrote on Dec 28th, 2009, 8:24am:
Does the first produce higher distortion? What about IP3?
For IP3, it is easy.
See http://www.designers-guide.org/Modeling/modeling-rf-systems.pdf

Title: Re: IP2/IP3
Post by LeroyJenkins on Dec 28th, 2009, 8:42am

great, thanks!  :D

Title: Re: IP2/IP3
Post by Ken Kundert on Dec 28th, 2009, 8:44am

An explanation of intercept points is given in http://www.designers-guide.org/Analysis/rf-sim.pdf starting on page 35.

IP2 characterizes even order distortion and IP3 characterizes odd order distortion. IP2 and IP3 are conceptually similar, so I will only describe IP3. At low power levels, the third order distortion will dominate over higher level odd terms (5th, 7th, etc), thus at low power levels if you double the signal power the odd order distortion will increase by 8. If you keep increasing the signal power level then eventually the contributions of the higher order terms will dominate over the third order term, but IP3 is measured by extrapolating from low signal power levels, so this effect is ignored by the IP3 measurement. By definition the third-order intercept point (which is referred to as IP3) is the level where, based only on extrapolation, the power in the third order term is equal to the power in the fundamental (the first order term).

The actual number for IP3 does not represent a signal level you could hope to achieve from the circuit. Rather you only use it to compare the distortion of two designs. If all else is equal, the one with the higher IP3 is more linear.

-Ken

Title: Re: IP2/IP3
Post by pancho_hideboo on Dec 28th, 2009, 8:47am


LeroyJenkins wrote on Dec 28th, 2009, 8:42am:
great, thanks!  :D
Actual IP2 is not so easy compared to IP3.

Again see http://www.designers-guide.org/Forum/YaBB.pl?num=1193566207/1#1

IP2 described in http://www.designers-guide.org/Modeling/modeling-rf-systems.pdf is simple and superficial.

Starter of http://www.designers-guide.org/Forum/YaBB.pl?num=1216663909 also did not understand IM2,
although he read http://www.designers-guide.org/Modeling/modeling-rf-systems.pdf.


Title: Re: IP2/IP3
Post by LeroyJenkins on Dec 28th, 2009, 10:13am

Okay, so let me see if i've got this right.

In general something like an RF amp is considered a static nonlinear system, that is it can be modeled with an indefinate-order polynomial equation.

If we take the first 3 order terms the first one is the one we are usually interested in, this is the linear gain times the input. Every other higher order term in the equation is essentially distortion contribution.

The constant before the second order term is related to IP2, and the constant before the third order term is related to IP3.

As input power is increased, the output power increases linearly (ideally indefinately, but practially reaches compression). The issue is that in 'intermodulation terms' the output power actually increases quicker (because the slopes of IP2 and IP3 are 2:1 3:1 accordingly) so a system can become sucepcitble to distortion before compression.

The purpose of extrapolating IP2 and IP3 to an actual intercept point (although its not a realistic power reading) is to put some sort of quantifiable weight on how suceptible a system is to intermodulation distortion?

Please correct me where i'm wrong... as you can see i'm simply trying to paint a picture of this in my head.

Title: Re: IP2/IP3
Post by Ken Kundert on Dec 28th, 2009, 11:37am

I think you have basically got it.

-Ken

The Designer's Guide Community Forum » Powered by YaBB 2.2.2!
YaBB © 2000-2008. All Rights Reserved.