how to improve ip3

Dear all,
I want to design microwave amplifier.And i do not know how to improve IIP3 and can not ensure how much it is OK?Anybody can give me a hand?
Thank you.

Hi
I think ...
1st optimize devices dc point for max iip3 and power consampion constraint

The bias bypassing can improve IIP3 performance without increasing current consumption.
The impedance of bypassing circuit should be lower than 25% of the input impedance of the transistor at particular frequency spacing, preserving the gain performance of the amplifier.

In general any mismatch due to noise matching improves IIP3.

Increasing emitter/source inductance reduces gain and improves IIP3 (excessive inductance could make the amplifier unstable).

Hi,

Give me you email

David

David, All documents you can post on this forum please

regards

Paul.

Hi,

GUYS I HAVE PROBLEM TECHNICAL I CAN"T UPLOAD SO I CAN SENT TO EMAILS

David

Hi, vfone
I don't quite understand your saying.
what means bias bypassing? why could this improve IIP3 performance without increasing current consumption?
In general any mismatch due to noise matching improves IIP3, why?
Is there any materials about this?
Thanks.

Thanks all of you for help.
and thanks for DDavid kindness,my E-mail:John_li3000@sohu.com

A first order rule-of-thumb is that IP3 is roughly 10dB above the compression point, or P1dB. I'm assuming you understand the definition of what IP3 is and what it means, and its relevance. If you don't, just let us know and we'll clear it up for you.

Therefore design the amplifier to have a compression point roughly 10dB below the IP3 target.

There are additional tricks you can use to further improve the IP3. These tricks generaly are narrow band in nature, may not track well over process, and may be difficult to apply. However, they are worth looking at. Here's a document I wrote that explains these tricks:
http://www.electriceng.com/Files/IP3Nulling.pdf

Greg

greg,

Thanks for the link of your IP3 paper. It is excellent. A compact and design oriented approach, and a really good collection of important things in this topic.
g

Thanks g579....

Its interesting how you joined before the very first microcomputer was invented.... I didn't know this board was up-and-running on vacume tubes :)

Greg

greg,

I do not remember the precise date, but somewhere in 2002 the board has changed its style. All members who joined earlier got the joining date of 01.Jan 1970.
I do not know the real reasons (probably some technical reason), but it happened so.Administrator perhaps remember, but it is now history.

Bias bypassing means all the filtering of the LNA that are on the bias circuit (base/gate and collector/drain).
To improve the linearity, the fluctuation of the base/gate and the collector/drain shall be stabilized with low impedance at so called video frequencies or baseband frequencies (between DC and usually up to 40MHz). Playing with these decoupling filters, you can get for the same DC current, different linearity intercept points.

In general in LNA design the Noise matching is different than the Gain matching (conjugate of S11 is different than conjugate of Gamma optimum).
Gain matching would affect the IIP3.
Experimentally was proved that in general the Noise mismatch would improve IIP3. A compromise between these two parameters shall be done.

Hi,

I sent to you some papers but the best way to improve IP3 by atleast 3dB it's using balanced amplifier (3dB if you are using with two amplifiers only)

DAVID

Thanks for your help me,David.And Thanks for Vfone's reply.Of course,thanks all of you for attention.Welcome for more Suggestions about this issue ,let us go on...