Anyone can get 0.2dB NF for cascode LNA?

[stevie-g]

Hey guys,
I'm looking for any cascode CS-LNA circuit design which can get 0.2dB NF at 2.5GHz in simulation.
Do you have it?

 

[vfone]

In a cascode topology the noise introduced by the cascode device (upper transistor added to improve stability of the amplifier) it will be additional to the initial noise.
So, a cascode always will get higher noise figure than a single device (or at least very close).

The pHEMT ATF-38143 from Avago has one of the lowest NF on the commercial market, but has higher NF than you are looking for.

 

[stevie-g]

So, do you mean that I need to sacrifice the upper transistor to get better noise? Change to a new topology?
What topology you can recommend?
Or, is there any other possible way without changing the topology?

In a cascode topology the noise introduced by the cascode device (upper transistor added to improve stability of the amplifier) it will be additional to the initial noise.
So, a cascode always will get higher noise figure than a single device (or at least very close).

The pHEMT ATF-38143 from Avago has one of the lowest NF on the commercial market, but has higher NF than you are looking for.

 

[Abhishekabs]

For 0.25dB NF NEC transistors are good, but at this frequency & 0.2dB NF it is hard. you may need to make superconducting LNA

Added after 51 seconds:

I mean you need to use cooling techniques.

 

[stevie-g]

I see...
Besides this method? Any more?
Besides make the transistor smaller and tune the input match, any other techniques to reduce the noise?
Or is there any technical paper can recommended?

For 0.25dB NF NEC transistors are good, but at this frequency & 0.2dB NF it is hard. you may need to make superconducting LNA

Added after 51 seconds:

I mean you need to use cooling techniques.

 

[Abhishekabs]

What is your application? specifications?
Speaking of only 0.2dB NF LNA is Vague.
Rather you may get a PhD if you design 0.2 dB NF LNA. for your frequency.
It is not easy.

NF= KTΔF
Here K is constant, so reduce T or ΔF that is bandwidth to get low NF.

 

[stevie-g]

Ya... I know it's a bit tough.
I thought there got any new topology I don't know.
Anyway, thanks man!

 

[tomhive]

Ya... I know it's a bit tough.
I thought there got any new topology I don't know.
Anyway, thanks man!

What is the BW you are looking for.You can use SPF3143 it has lowest noise figure.