The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> IIP3 of differential pair using Cadence
https://designers-guide.org/forum/YaBB.pl?num=1193668569

Message started by Julian18 on Oct 29^th, 2007, 7:36am

Title: IIP3 of differential pair using Cadence
Post by Julian18 on Oct 29^th, 2007, 7:36am

Hi, there

I want to see the IIP3 of a differential pair amp, the circuit is shown below, but when I changed the resistance of the port, the IIP3 also changed, and the values are btw -2dbm to 29dbm (1K Ohms to 0.1Ohms ) , so how can i choose the right value for the IIP3 of my ckt? ( The load of differential pair is resistor.)

By the way, I also have simulated the situation where the resistor in parallel to the port is cut, and this time i also got different IIP3 values as I changed the resistance value associated with the port.

So what is the right setup and what is the right value of IIP3? Thanks

Title: Re: IIP3 of differential pair using Cadence
Post by Ken Kundert on Oct 29^th, 2007, 2:04pm

I don't think you want the port at the output. Try getting rid of it.

-Ken

Title: Re: IIP3 of differential pair using Cadence
Post by aaron_do on Oct 29^th, 2007, 6:23pm

Hi,

I think you don't need pres either. Is that for matching? You should find your IIP3 in terms of voltage or volts squared. What i do is I use the PSS and PAC analysis, and use the output currents to test IIP3. The IIP3 will be a power value and you should note it down. After that, use the PAC form to plot the input voltage versus input power. You can then easily read off your IIP3 in terms of voltage using the power you noted down earlier.

I'm not sure if there is a direct way of sweeping an input voltage instead of input power to get IIP3.

Oh yeah, you don't need to change the resistance of the port. If you do, it shouldn't affect the IIP3 (in volts), but it may affect the IIP3 in power since the resistance of the input port directly affects the voltage at the input node.

cheers,
Aaron