![]() |
| Fig.1 CMOS Inverter Construction |
And PMOS will let low voltage pass while NMOS will let high voltage pass. Therefore, the maximum output voltage (VOH=VDD) occurs when input voltage is low (Vin=0V) PMOS is on and pulls Vout to VDD while NMOS is off. Moreover, the minimum output voltage (VOL=0V) occurs when input is high (Vin=VDD) NMOS is on and pulls Vout to ground while PMOS is off. Thus,there is a transition region between VOH and VOL and there are three situations:
1. If input voltage (Vin) is low:
- When 0≤ Vin < VTN (threshold voltage of NMOS)
- Mn (N-channel) in cut off and off
- Mp (P-channel) in triod and Vout is pulled to VDD
- When
VTN≤Vin<Vo一|VTP| (threshold voltage of PMOS)
- Mn in saturation with strong current
- Mp in triode and VSGp (voltage on Mp) and current reducing
- Vout decreases via current through Mn
2. If Vin = Vout (mid point) ≈ 0.5 VDD
- Mn and Mp both in saturation
- Maximum current at Vin = Vout
3. If Vin is high
- When Vout + VTN ≤ Vin <VDD - |VTP|
- Mn in triode, Mp in saturation
- When Vin> VDD -|VTP|
- Mn in triode, Mp in cut off
And the CMOS inverter voltage transfer characteristic curve according to the situation above is shown in below:
![]() |
Fig.2 CMOS Inverter Voltage Transfer Characteristic Curve |
In order to simulate the curve above through MATLAB, the following equations will be applied:
1. The current equation of PMOS is :
- Cut-off region: i=0
- Saturation region:
FTK.jpg)
FTK.jpg)
FTK.jpg)
![]() |
| (2) |
- Linear region:
![]() |
| (3) |
2. The current equation of NMOS is:
- Cut-off region: i=0
- Saturation region:
![]() |
| (5) |
- Linear region:
![]() |
| (6) |
As for a no-load CMOS inverter: ipmos = inmos
(7)
And βp is the transconductance coefficient of PMOS which has the equation:
μp is the electron mobility in p-type semiconductor and Cox is the unit capacitance of gate oxide layer and (W/L)p is the aspect ratio of the PMOS Gate.
βN is the transconductance coefficient of NMOS which has the equation:
μN is the electron mobility in n-type semiconductor and Cox is the unit capacitance of gate oxide layer and (W/L)n is the aspect ratio of the NMOS Gate.
Therefore,
(1) When 0≤ Vin < VTN,
Simultaneous equations (3) (4) (7) can get the following equation:
Vo = VDD
(2) When
VTN≤Vin<Vo一|VTP|,
VTN≤Vin<Vo一|VTP|,
Simultaneous equations (3) (5) (7) can get the follwing equation:
(3)When Vo-|VTP| < Vin<Vo+VTN,
Simultaneous equations (2) (5) (7) can get the following equation:
consider the linearity current change of PMOS and CMOS and no-load CMOS inverter mode, can get the following equation:
(4)When Vout + VTN ≤ Vin <VDD - |VTP|,
Simultaneous equations (2) (6) (7) can get the following equation:
(5)When Vin> VDD -|VTP|
Simultaneous equations (1) (6) (7) can get the following equation:
Vo = 0V
Thus, the MATLAB code desiged by the experimenter to show the CMOS inverter voltage transfer characteristic curve is shown in figure below:
| Fig.3 MATLAB Code for CMOS inverter voltage transfer characteristic curve |
After running he program, experimenters can get the waveform as shown in the figure below:
| Fig.4 CMOS inverter voltage transfer characteristic curve |
According to the figure above Fig.4 is similar to Fig.2,which verify the simulation waveform simulated by the experimenter is successful.


0P0JQ8LTLNM`TF)M9ZW.jpg)










Hey, nice blog. Could you please let me know where you got the equations for the output voltage you used in the model.
ReplyDeleteThank you.
what is y_p and y_n ?
Delete