Help to tackle on process change in LNA design

[suling]

Hi, all, had designed a 2.45GHz LNA using differential cascode structure with inductive loading in CMOS process. Had done a cascaded simulation with Mixer. However, the interface loading changed quite a lot as process change which cause the gain to change. Also, the frequency peak shifted upwards or downwards by 150MHz wrt typical process. Wonder if someone can advise me on method to tackle on process change in such a case.

Thank you in advance.

 

[kspalla]

Problem is confusing if we talk on LNA why to talk about mixer.
Why not the process change in mixer than LNA?
Be clear and share some more data or information to help you more

 

[frannk]

Sounds like you need to set the output load lower using resisters so your mixer loading won't affect the LNA performance. By doing this, you'll also increase your bandwidth somewhat by lowering the Q of the tank. However, you'll probably need to use more current now to increase the LNA gain to make up for what's lost.

 

[suling]

Hi, thank you for all the advices. The process change is lesser now. Had changed slightly on the resonant tank in the interstage matching and added a loading resistor in between LNA & Mixer.