The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Low temperature simulation of a simple NMOS
https://designers-guide.org/forum/YaBB.pl?num=1128383962

Message started by sphere83 on Oct 3^rd, 2005, 4:59pm

Title: Low temperature simulation of a simple NMOS
Post by sphere83 on Oct 3^rd, 2005, 4:59pm

HI guys,

I was simulating a very basic NMOS transistor using spectre analog environment in order to obtain the Ids/Vgs plot. I used parametric analysis on Vds and dc analysis on Vgs to obtain the plot at 300K. Everything seems fine.

However, when i tried to switch the temperature to 4K ( as i need those simulations to run at 4K ), there were errors. The error messages were:

Error found by spectre during initial setup.
M0.my_nch: The junction reverse saturation current is too small = 4.15 C.

Warning from spectre during initial setup.
M0: 'Dvt0' = -15.0839mV is negative.

What do the messages mean?

Please help, urgent.

Kindest regards,

Sphere

Title: Re: Low temperature simulation of a simple NMOS
Post by ywguo on Oct 3^rd, 2005, 9:05pm

Sphere,

It is obvious that the temperature is out of range. Why do you need run simulations at 4K? It is impossible to run a circuit near absolute zero degree.

Best regards,
Yawei

Title: Re: Low temperature simulation of a simple NMOS
Post by sphere83 on Oct 4^th, 2005, 4:33pm

Thanks Yawei for your prompt reply!

It is going to be used to amplify signals from single electron transistors operating at 100mK to 4K. However, the major concern here is the thermal noise and flicker noise. The Id characteristic curve is similar at 4K and at 300K( not same, but similar i think ) and thus can be used.

Is there a way I can set the spectre analog environment to simulate those transistors at 300K but make it think that it has a noise temperature of 4K or less? I'm a real noob in Cadence spectre but i'm willing to learn hard! ;D Kind people out there please point me to the rite direction!

Cheers

Sphere

Title: Re: Low temperature simulation of a simple NMOS
Post by Ken Kundert on Oct 4^th, 2005, 6:52pm

The issue is not with Spectre per se, but rather with the model. The model is not designed to operate at those temperatures. You could try different models, or you could write a simple model in Verilog-A.

-Ken

Title: Re: Low temperature simulation of a simple NMOS
Post by sphere83 on Oct 4^th, 2005, 7:40pm

Thx Ken ;D WIll try it out !