The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Input referred noise of a MOSFET
https://designers-guide.org/forum/YaBB.pl?num=1381913748

Message started by Hrishikesh on Oct 16^th, 2013, 1:55am

Title: Input referred noise of a MOSFET
Post by Hrishikesh on Oct 16^th, 2013, 1:55am

Hi..

Through an ac simulation, I see that the input referred noise of a MOSFET reduces with increasing frequency, stays constant for some range, and again rolls off at about 1-10 GHz. I understand that low frequency noise is dominated by flicker noise and the frequency independent part is thermal noise. Can anyone tell me why it should further decrease at high frequencies?

Title: Re: Input referred noise of a MOSFET
Post by raja.cedt on Oct 16^th, 2013, 2:51am

How did you simulated, please show me the spice setup. Please try to simulate y21(short circuit current gain).

Thanks,
Raj.

Title: Re: Input referred noise of a MOSFET
Post by harpoon on Oct 17^th, 2013, 2:33am

sounds like there is some low pass filter in the simulation ...

as raj mentioned, post the schematic/spice netlist and we may be able to help ...

Title: Re: Input referred noise of a MOSFET
Post by Lex on Oct 17^th, 2013, 4:08am

Could be in model parasitic gate resistance/capacitance.

Title: Re: Input referred noise of a MOSFET
Post by Hrishikesh on Oct 17^th, 2013, 10:03pm

Thanks for the responses. The netlist is very simple. I have an 50x1 NMOS with the drain connected to 5V supply through a 10K noiseless resistor. The source is grounded and the gate is biased at 0.8V. I am doing an AC simulation by sweeping the frequency. To get the input referred noise in V/sqrt(Hz), I am dividing the total noise at the output by the amplifier gain.

It appears that the effect is due to the Cgd of the NMOS. Because, the roll off frequency reduces as I connect a capacitor between the gate and drain. I can also see a zero present at that frequency. But I always thought that the input referred thermal noise is 4kT*2/(3gm) V2/Hz. It would be great if any of you can throw some light on the effect of Cgd.