Custom Search

More than 10,000 searchable pages indexed.

Greetings: There is so much good stuff on RF Cafe that there is no way to list or link to all of it here. Please use the Search box or the Site Map to find what you want - there is a good chance I have it. Thanks!

•−•  ••−•    −•−•  •−  ••−•  •
RF Cafe Morse Code >Hear It<
Job Board
About RF Cafe™
Copyright 1999-2015

Intermodulation products - RF Cafe Forums

Because of the high maintenance needed to monitor and filter spammers from the RF Cafe Forums, I decided that it would be best to just archive the pages to make all the good information posted in the past available for review. It is unfortunate that the scumbags of the world ruin an otherwise useful venue for people wanting to exchanged useful ideas and views. It seems that the more formal social media like Facebook pretty much dominate this kind of venue anymore anyway, so if you would like to post something on RF Cafe's Facebook page, please do.

Below are all of the forum threads, including all the responses to the original posts.

Post subject: Intermodulation products Posted: Sat Nov 29, 2008 12:40 pm


Joined: Tue Aug 26, 2008 5:18 am
Posts: 22
Location: UK
As an RF amplifier designer I have seen 5th order lower than 3rd order products.
My question is if we see 5th order higher than 3rd order products then what is the cause? Is it normal to see that?


Post subject: Posted: Tue Dec 02, 2008 10:28 am

Site Admin

Joined: Mon Jun 27, 2005 2:02 pm
Posts: 373
Location: Germany
This is all depends on the technology of the power transistor and the non-linear characteritic derived from it. This is very unusual behavior. At which back-off do you see this behavior?

I personally have not encountered in that, but I have seen in data sheets that IMD5 do not always increase with input power but can also decrease and then increase again.


Post subject: Posted: Wed Dec 03, 2008 3:43 am


Joined: Fri Feb 17, 2006 12:07 pm
Posts: 218
Location: London UK
Some years ago I was trying to relate the input>output transfer function to the harmonic levels for a TWT. I did some routine in MATHCAD that worked, using the Fourier Transform function they have. It seemed to work fine, although I never got to measure the ultimate levels during testing. I think if you can obtain a very accurate transfer curve, it would be possible to model it with a Fourier Transformer.
However, there is a complication I foresee:
The load match to harmonics can be so complex that it could grossly distort the measurements in a real circuit. The Fourier Transform would assume a perfect 50 ohm match, but in reality, the circuit you have could fortuitously show a much better S11 at the 5th harmonic than at the 3rd.


Post subject: Posted: Wed Dec 03, 2008 8:39 am


Joined: Tue Aug 26, 2008 5:18 am
Posts: 22
Location: UK
Device is FP1189 and I am checking IP3 with two tones at 10dBm below p1dB.


Tony Kurlovich
Post subject: Posted: Mon Dec 15, 2008 10:01 am


Joined: Thu Oct 19, 2006 6:02 pm
Posts: 7
It is important to make the distinction between apparent IM3, IM5 and real IM3, IM5. In a 2 tone test the spurs at plus and minus the difference frequency are considered to be IM3 and the spurs at plus and minus twice the difference frequency are considered to be IM5.

What we think of as IM3 is a sum of IM3 and IM5. It usually doesn’t matter because IM5 is usually smaller than IM3. An example.
Usual IM3, Usual IM5, ( A + A + B) or 2A - B
Usual IM5, (A + A + A - B - B) or 3A - 2B
IM5 that mimics IM3, (A + A + A - A + B - B - B) or 2A - B

I know this looks like slight of hand but it is real. Try a 2 tone simulation with a pure first order and a pure fifth order characteristic. You will get 2A - B. This condition won't happen in the real world but it tells what is hiding beneath that apparent IM3.

A common example is some medium signal Class-A amplifiers (where IP3 is >> PIdb + 9.5db) as they approach P1db have IM3 growing faster than 3db/db. This is usually due to the IM5 at that same spur frequency getting a bigger vote. Depending on the phasing, IM5 could subtract from IM3. This is unlikely in a broad band Class-A but is more probable in a narrow band and/or Class-B or Class-AB.


Post subject: Re: Intermodulation productsPosted: Thu Apr 09, 2009 10:40 am


Joined: Wed Apr 08, 2009 7:53 am
Posts: 12
Any way it is an uncommon behavior for a mid power Class A.. I´ve seen this effect on high power LDMOS Class AB amplifiers with high bias points, but not on med power Calss A..
I´ll start by doubting on the test bench.. If the SA is not set correctly, your measurement could be affected by the SA own intermodulation, and perhaps the DUT IM3 is being canceled by the SA IM3.. like a predistortion effect, and that is why you see the IM3 lower than it should on the screen... Check the SA setup.. Or change the cable length from the DUT to the SA to see if you have a different response.
Or perhaps you have big harmonics on the signal generators.. try to filter them..
Or perhaps you have low isolation on the tones combination at the input.. and so you are being affected by the reverse IP3 of the generators (an other predistortion effect).
With this answer i´m not trying to underestimate no one, this is just something that happend to me in the past.
Best Regards


Posted  11/12/2012

Your RF Cafe
Progenitor & Webmaster

Click here to read about RF CafeKirt Blattenberger… single-handedly redefining what an engineering website should be.

View the YouTube RF Cafe Intro Video Carpe Diem!
(Seize the Day!)

5th MOB: My USAF radar shop

Airplanes and Rockets: My personal hobby website

Equine Kingdom: My daughter Sally's horse riding website