calculate Cox in mosfet in cadence

Status
Not open for further replies.

aredhel_vlsi

Member level 4
Joined
Aug 21, 2009
Messages
72
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Activity points
2,013
calculate cox

Hello all,

(this is my first post)

Could someone help me calculate the gate oxide capacitance Cox per unit area in a pmos (W/L=52/2) in cadence? I know that Cox= eox/tox , but how do we know these parameters? I work on 0.18 um.

Sorry if it is a silly question, i'm a newbie, and I desperately need this info!
 

how to find eox in a mosfet

Hi,

are you using Analog Design Environment for your simulations?
If yes just choose:
Results -> Print -> DC Operating Points
and then click on the MOS you want to know Cox.
Now read the parameters like Cgg, Cgs, Cgd, etc...
Try to calculate Cox following the theory and the operating region in which you are.
For example you can follows equations found in this paper:
**broken link removed**

I hope this help you.
CBs
 
cadence cox

aredhel_vlsi said:
Could someone help me calculate the gate oxide capacitance Cox per unit area in a pmos (W/L=52/2) in cadence? I know that Cox= eox/tox , but how do we know these parameters? I work on 0.18 um.

Do you need Cadence?

eox=e*erSiO2

e=8.854e10^-12 F/m

er of silicon dioxide=3.9 to 4.5

You still need to find the oxide thickness for the process.
 
oxide capacitance 0.18 um

Thank you all so much about your answers !

crystalballs said:
are you using Analog Design Environment for your simulations?
If yes just choose:
Results -> Print -> DC Operating Points
and then click on the MOS you want to know Cox.

crystalballs, this is a good idea, but I have a problem refered to the use of program:
I use Analog Design Environment.. Obviously I should run a dc analyses. I chose the DC Operating Points, but I can't see anything..I also selected the mos but it isn't indicated by a color. I know it works, as I have tried it before at another computer.. Perhaps I have done something wrong during simulation... Do you know any good manual for cadence, I have found some in the web, but they have the very basic simulations and options. I haven't seen anything like that in a cadence manual :|
 

Hi again ! I have estimated the Cox I wanted, to go on temporarily, but I used the er=3.9 to estimate it according to the theory.

Can I ask, how do they symbolize the er in the files of spectre? I've found something like 'lint', and 'etab'. The lint I found is from 2.9 to 3.9 and looks suitable, but i don't know what it is.
 

I don't think you mess with the relative permittivity, if it's
embedded in the Cox value. Or, you can give an epsilon
and a Tox instead. You have to drill down into the model
docs for the particular model type you are using.

Now, one variable in all of this and a big one at short
channel is how to figure what portion of the drawn gate,
is really active channel (Cox) and how much is underlapped
by the source and drain LDD / drain extension / side
diffusion (which would properly go to Cgdo, Cgso and not
figure into the k'=u0*Cox (or more advanced, but similar
MOS current equations).

Start with the model level / type as called out and go hit
the Cadence manuals for Spectre. If you don't have them,
I have seen them in Google searches (often hosted at
universities).
 

Re: how to find eox in a mosfet

hi, crystalballs
I have a question on checkin the Cgs. what I see from the operating point results of the cgs or cgd are some negative values, as shown in the figure attached. i have difficulty in understanding this. How should I know the value of these capacitances?
Thanks.

jimiblues

 

Hi,

I have similar issue. I am going to use in Spice NMOS like tunable resistor with these parameters.

**broken link removed**

So the Eq. for calculation of R is 1/{(µ Cox W/L)(Vg − VT)}, where W and L are the channel width and length and VT is the threshold voltage of the MOSFET, µ is the free electron mobility in the channel and Cox is the gate oxide capacitance per unit area and MOSFET is tunable via Vg.

Unfortunately, I don't have any experience with this, but I guess µ, Cox, VT should be constants, according to process always different. Am I right? How to calculate R for example for W/L = 20u/1u?

Sorry for amateur question and thanks for any help and suggestion.
 

Re: calculate cox



Hi
This is simple thing. just do
"Results -> Print -> DC Operating Points
and then click on the MOS you want to know Cox."
Then note the value Beff(beta effective)=un*Cox


Thanks
Sandy
 
Status
Not open for further replies.

Similar threads

Cookies are required to use this site. You must accept them to continue using the site. Learn more…