How does Uncox vary when temperature increases?

[KGF KING]

Hi All,

I would like to know how does the Uncox vary when the temperature increase? Because as far as i concerned the Uncox should be lowest at SS corners and highest at the FF corner and i thought the pattern will be the same for the temperature as well where lowest at the -40C and highest at the 125C. But from the calculation (betaeff/(w/l)) i get highest at the low temp and low Uncox at the high temp but at the same temp the Uncox increases from SS to FF?

 

[sidun.av]

Hi,
I don't really understand what do you mean by saying this:

i get highest at the low temp and low Uncox at the high temp but at the same temp the Uncox increases from SS to FF?

Broadly speaking, you are analysing a product of 2 variables:
μ₀C_ox, where C_ox = ε_ox/t_ox.

• When you are sweeping the temperature, your μ₀ plays a major role, because it has a negative temperature dependency:
  ( https://lampz.tugraz.at/~hadley/psd/L4/mobility.php )
• When you are sweeping the process corner, t_ox plays a major role, because the oxide thickness changes with the process (should be thicker at ff and thinner at ss hence achieving lower capacitance at ff and higher speeds).

So, when you are sweeping both process corner and temperature, your final value of μ₀C_ox is the combination of both factors I mentioned above.

Hopefully, that helps.

 

[KGF KING]

Hi Sidun,

i tried to calculate Uocox in cadence to design my opamp, so i'm getting these numbers when i tried to divide betaeff/(W/L) so i was thinking whether the value i'm getting is acceptable or not and my expectation was UnCox for NMOS for FF125 and FS125 should be same right?

 

[sidun.av]

Hi Sidun,

View attachment 197213
i tried to calculate Uocox in cadence to design my opamp, so i'm getting these numbers when i tried to divide betaeff/(W/L) so i was thinking whether the value i'm getting is acceptable or not and my expectation was UnCox for NMOS for FF125 and FS125 should be same right?

Your way of calculating the μ₀C_ox is the indirect calculation through your beta and W/L. I've explained in my previous post how μ₀C_ox depends on temperature and process corner. If you have doubts, you can find the t_ox values for the corners in your process documentation and take values of μ₀ vs temperature and calculate it yourself.

my expectation was UnCox for NMOS for FF125 and FS125 should be same

Why do you think so? It's not clear for me why these should be equal, because you are varying both corners and temperature (t_ox and μ₀) and both are process/temperature dependent parameters.