AnsweredAssumed Answered

SolvedWill system IIP3 change with tone spacing settings?

Question asked by wangyiliu99 on Dec 13, 2012
Latest reply on Feb 18, 2013 by wangyiliu99
Hi, all

this is an updated post. 

Several weeks ago, I raised one question asking about the system IIP3, how the value would change if the input two tone power, frequency spacing changes. 

till now, I am clear that IIP3 will NOT change with input power level. Since IMD3 product will change three times as fast as main input signal, according to its calculation, dBc will drop 2dB if input increase by 1dB, IIP3 will not change. I referred to agilent application note 5965-7920E.

I am still quite puzzled by the tone spacing. Given the test setup is well set, RBW(smaller than the tone spacing), mixer level, RF input attenuator, etc., signal generators and power amplifiers are high quality products. How would the IIP3 change with the tone spacing?
I did the measurement, in the measurement, tone spacing is changed from 1MHz to 2, 5, 10 and 20MHz, 1 , 2, 20 has relatively the same IIP3 values, however, IIP3 decreases when the spacing is 5 or 10 MHz. Actually I am aware that one of my SG is not so good especially that it generates quite strong spurious at 10MHz off the main signal. So would it be like that, theoretically the IIP3 would be the same no matter the tone spacing is? (Given the RBW is smaller than the tone spacing)

I searched quite a lot and failed to find anything. So your help would be highly appreciated!


former post
Hi, all

I am not sure whether I should ask this question on this board or not. If not, please kindly advise me where I should. Thanks.

I am writing to ask a general question about the IIP3 (3rd order intercept point) of the passive filter measurement. I have checked the agilent application notes about the TOI measurement. Still I was unable to solve my problem.

The measurement setup is exactly like the one in Agilent's application note, two signal generators as input source, then the low pass filter, then isolator, then to the power combinar, then to DUT, finally to spectrum analyzer.

My questions are 
(1) If I only change the two input signal frequency spacing, say from 10MHz to 1MHz, while other settings being kept the same, how would the IIP3 of the measurement setup change? Improved or worsened?
(2) Similar to question (1), if I measure my DUT(band pass filter), how would its IIP3 value change if the input two signal frequency spacing is changed?

It is possible to do several measurements and do the comparison. However, before getting down with actual measurements, I would like to get some advice from here. Are there any application notes concerning my question, or any theory reference available? Please forgive me if my question is so naive. 

Your help would be highly appreciated!


Edited by: wangyiliu on Feb 14, 2013 11:47 AM

Edited by: wangyiliu on Feb 14, 2013 11:57 AM

Edited by: wangyiliu on Feb 16, 2013 12:15 AM

Edited by: wangyiliu on Feb 21, 2013 3:58 PM