help: about the effect of pad and bonding wire.

[givensoo]

Now i am making a UWB lna, but if we put the Power supply PAD and the bonding wire Pad(of course include the bonding wire) on and simulate. the result is terribly bad.
the resone i guess is that : an inductor(bongding wire) connects to the source of the MOS, so it becomes a source degeneration structure.
But a lot of papers i read haven't mentioned this(i think it is a huge problem). so what can i do to steer clear of this trouble?

btw: the lna just has a single input not diff for the system's structure.

 

[ce]

I'm not sure about your topology, but decoupling the source of your MOS in a common source amplifier should do the trick. Generally differential structures don't suffer that much of bonding parasitics in the power supply path.

 

[givensoo]

Take the following topology as an example:


In ideal situation M3's source connects to graound.But it dosen't.It connects to the bonding wire and pad.Then throught the bomding wire to the ground.
The question is the bonding wire introduces inductor which makes M3 as a source degeneration structure. So what can i do to steer clear of this trouble?

 

[ce]

You should add more decoupling between your supply.

 

[test_out]

1.You should include the pad and wire bonding in the first step of simulate.
2. you can use wire-bonding as the degeneration inductor. (Ex, when you simulate, instead of using inductor, you use wire bonding for inductor function).

Hope this will help.