- 7th April 2009, 12:01 #1

- Join Date
- Sep 2007
- Posts
- 143
- Helped
- 3 / 3
- Points
- 1,833
- Level
- 9

## input ip3

Hi all,

I would like to knoe why we are giving so much importance to the IP3 parameter even though the device is not experiencing that stage as per the normal theory.(It enters the saturation region once it attains the 1 dB compression point)

Could anyone tell me its importance in an amplifier's performance.Also I am attaching the respective Power Input Vs Power Output(Pin Vs Pout) schematic.Its the worst hand made diagram by me in the entire world though.. Pardon me for that..!!

Santom

- 7th April 2009, 12:01

- 7th April 2009, 15:20 #2

- Join Date
- Jun 2007
- Location
- Netherlands
- Posts
- 266
- Helped
- 8 / 8
- Points
- 2,782
- Level
- 12

## oip3 formula

IIP3 is a metric to judge the linearity of certain circuit. That's why it is important

Of course the circuit should not be set to work in saturation region

- 7th April 2009, 15:34 #3

- Join Date
- Jul 2005
- Posts
- 120
- Helped
- 10 / 10
- Points
- 2,154
- Level
- 10

## amplifier ip3

IP3 is a figure of merit.

It is also an indicator of how well an amplifier will work in real world applications compared to other amplifers.

- 7th April 2009, 15:34

- 7th April 2009, 15:46 #4

- Join Date
- Sep 2007
- Posts
- 143
- Helped
- 3 / 3
- Points
- 1,833
- Level
- 9

## oip3 definition

HI

Firstly,thanks for the post.I also read in the literature that it is used for measuring the linearity performance.But I dont understand that.It is however going to work only till 1 dB compression point right! Then why IP3 point ans also its realationship with linearity performance of the an amplifier..

Santom

- 7th April 2009, 15:46

- 7th April 2009, 20:25 #5

- Join Date
- Jul 2005
- Posts
- 120
- Helped
- 10 / 10
- Points
- 2,154
- Level
- 10

## ip3 intermodulation levels

Hi,

IP3 is a theoreretical number. The third order product will never intercept the desired product because of the compression or destruction of the amplifier or mixer or whatever.

But IP3 is a measure of a components linearity, because we know that the higher the incept point the more linear the the component is.

You can also determine how good the IP3 must be for applications such and multi-channel digitally modulated radio to insure ACPR is not a problem.

- 8th April 2009, 06:36 #6

- Join Date
- Jan 2008
- Location
- Central Florida
- Posts
- 389
- Helped
- 11 / 11
- Points
- 5,747
- Level
- 17

## derivation of oip3

Originally Posted by**santom**

By knowing the intercept (OIP3) of an amplifier you are now able to obtain specific values of the inter-modulation products expected for any input power of the amplifier.

With amplifiers if you increase the amplitude you will increase the inter-modulation distortion of the two frequency tones as well. If you increase the amplitude of the tones large enough there will come a point when inter-modulation distortion from the two tones of the amplifier will equal the amplitude of the two tones! Although this is theoretical, which means the amplifier will never increase the signal so high so that the inter-modulation distortion will equal the two tones. But even though the amp wont let you reach IP3 its nice to know for a given tone amplitudes what level will your inter-modulation distortion be at in reference to the two tones being amplified by the amplifier.

As an example. If the two tone amplitudes out of the amplifier is -10dBm and your amp OIP3 is 20dbm. There is a formula: OIP3=Pout+∆/2. That sheds a lot of light on what is happening. We know Pout is -10dBm and we know OIP3 is 20dBm. We can derive ∆/2 which is 30. With ∆/2 know we can find the linear power region of the amplifier by (OIP3- ∆/2) which is -10. So we know ∆/2 (30), linear power region of the amp (-10), now we can get ∆. ∆ will give the amplitude level of the inter-modulation distortion.

∆ = tone amplitude –((∆/2 )*2) = -70 or ∆ = OIP3 - (∆/2*3) = -70. So by changing the value of Pout we can get a new ∆/2 and ∆ and make a plot and know all the power levels of the inter-modulation distortions caused by the two tones vs. the power level of the two tones.

I attached a drawing and a table of all the know items we can derive for given tone amplitudes. Maybe you can make some use of it.

- 8th April 2009, 14:55 #7

- Join Date
- Jul 2005
- Posts
- 120
- Helped
- 10 / 10
- Points
- 2,154
- Level
- 10

## what is input ip3

That's what I said. You just had a fancy picture and used the "∆" a lot.

- 12th April 2009, 17:50 #8

- Join Date
- Feb 2009
- Posts
- 43
- Helped
- 3 / 3
- Points
- 1,484
- Level
- 8

## ip3 derivation

To All,

the IP3 definition applies in an region, where the increase of 1dB at the input produces a 3 dB increase of the IMD products. In the linear range, the output follows 1dB per 1dB increase at the input, for the second order product 2dB/dB at the input. Typical mixing is 6+8=14 MHz, or 7 +7.1= 14.1. The second order products can be avoided by a high or bandpass filter at the input. The third order products are 2 x f1 + (-) f2 because of the close spacing can not be filtered, A device with higher DC input power produces fewer IMD products. Fet's are typically better for third order products then Bip transistors for the same DC current.

The definition of the Third Order IMD Point, input or output , is really obsolete. It is a mathematical definition where the 1dB/dB line and the 3dB/1dB distortion products cross. This is purely a point, beyond any point of practical operation. This mathematical derivation assumes that there is only ONE non-linearity, typically Y21. At higher frequencies the frequency and voltage and/or current dependent transistor internal semiconductor capacitances adds heavy distortion and the simple equations are no longer valid.

The best way to get useful numbers is to quote the distortion products relative to the input.

These test should all done at 0dBm, -10 dBm ..... -40dBm for all the equipment to be compared.

The best way is the 3 tone test used for TV amplifiers and transmitters, looking at differential phase and gain. See more on the topic

http://www.rantelon.ee/eng/docs/parameetrid.pdf

For multi tone tests , which gives the most insight, look at

http://www.rfhic.ru/prod/pdf/W-2F1G20P_1.pdf

- 13th April 2009, 05:54 #9

- Join Date
- Jul 2005
- Posts
- 120
- Helped
- 10 / 10
- Points
- 2,154
- Level
- 10

## ip3 amplifier two tone

Everyone here is super smart, but I was trying to answer Santom actual question.

- 13th April 2009, 21:13 #10

- Join Date
- Feb 2009
- Posts
- 43
- Helped
- 3 / 3
- Points
- 1,484
- Level
- 8

## ip3 mixer definition

Yendori,

You are correct, and I gave some more insight in the topic, as I DON'T like the IP3 hysteria. I will prove through some examples that even for 1 RF transistor the math does NOT follow the physics