Designer’s Guide Community

Julian18

Community Member

Offline

Posts: 55

question about input referred noise
Nov 29^th, 2007, 6:51pm

Hi, there:
suppose I have a circuit, and I do three simulations as described below
1) add a port as input probe, set its resistance, say 50 , and run noise analysis, now I can get input referred noise voltage from the result, call it NP.
2) replace the port with a voltage source in series with a 50Ohms resistor, run noise analysis, get the input referred noise voltage ,call it NV.
3) replace the voltage and series resistor with a current source in parallel with a 50Ohms resistor, run noise analysis, get the input referred noise current, call it NI.
Now can any one tell me what's the relationship btw these three quantities.

Thanks.

Back to top

IP Logged

aaron_do

Senior Fellow

Offline

Posts: 1398

Re: question about input referred noise
Reply #1 - Nov 30^th, 2007, 1:45am

Hi,

input referred noise is just the output noise power divided by the power gain. If you measure the output noise is V^2/Hz, and you want to refer it back to a voltage, divide by (V/V)^2. If you measure in terms of output current ^2 then divide by gm^2. Your 50 ohm resistor is just complicating things since you don't need it to find the input referred noise (unless its part of the circuit). In fact you have to subtract its noise contribution to find the input referred noise. For example, if you have a port and you measure NF, the IRN can be found from,

NF=(IRN+kT50)/kT50

where IRN is in V^2/sqrt(Hz)
hope it helps,
Aaron

Back to top

there is no energy in matter other than that received from the environment - Nikola Tesla

IP Logged

Julian18

Community Member

Offline

Posts: 55

Re: question about input referred noise
Reply #2 - Nov 30^th, 2007, 4:28am

aaron_do wrote on Nov 30^th, 2007, 1:45am:

Hi,

input referred noise is just the output noise power divided by the power gain. If you measure the output noise is V^2/Hz, and you want to refer it back to a voltage, divide by (V/V)^2. If you measure in terms of output current ^2 then divide by gm^2. Your 50 ohm resistor is just complicating things since you don't need it to find the input referred noise (unless its part of the circuit). In fact you have to subtract its noise contribution to find the input referred noise. For example, if you have a port and you measure NF, the IRN can be found from,

NF=(IRN+kT50)/kT50

where IRN is in V^2/sqrt(Hz)
hope it helps,
Aaron

Hi, Aaron:
thank your for your information.the reason of adding the resistor is that, I just want to see the relationship of these three quantities, as spectre reference sais, when input port is used as input probe, the noise associated with the port resistance will contribute to the total output noise, so I intentionally add resistors in two other situations to equate the noise source number in the circuit. and i expect that if we ignore the correlation btw the noise voltage and noise current, the following relation may hold:
N_current^2*50^2+N_voltage^2 = N_port^2
but i can not arrive at this relation from the simulation result. So I asked the question.

Any other information?

Back to top

IP Logged

didac Senior Member Offline There's a million ways to see the things in life Posts: 247 manresa,spain	Re: question about input referred noise Reply #3 - Nov 30^th, 2007, 5:21am Hi, Have you checked that both port & simulator temperatures are the same? Hope it helps,
Back to top	IP Logged