I want to know about composite transistor. In literature composite transistor is considered equivalent to a single transistor. Can someone tell me how? If both the transistors of composite transistor are in subthreshold saturated region then how can they be equal to a single transistor.
Hi,
If both transistors are in subtreshold, than the current throgh M2 can be written as:
I2 = I0*exp((VGS2-Vth)/nVt) and the current through I1 as:
I1 = I0*exp((VGS1-Vth)/nVt)(1-exp(-VDS)/Vt)
Knowing that I2=I1=I, it can be proven that the two composite transistors can behave as one (if n is almost 1, and the transistors have same Vth)
It is not equal. How two oranges are equal to one orange? They don't. In your case, the layout and the operation is different. Run simple ID vs VDS and ID vs VGS characteristics and you'll see that the two cases are not equal. Even the input capacitance is higher in case of the cascode comparing to a single transistor.
In the literature they make the approximation: self-cascoded ~= a single transistor. But keep in mind that in fact it is not a case. Maybe you could obtain the characteristic of the cascode with one transistor, but I'm not sure.
In literature it has been mentioned that using self cascode output swing is increased compare to a normal cascode. Its voltage drop vdsat is equal to a single transistor. The cascode device size in self cascode is kept larger.
Yes the output swing may be increased comparing to a normal cascode as a normal cascode may be biasied (the upper transistor) higher than in self-cascode. The VDsat drop equal to a single transistor that should not be true rather as there are two transistors in series. Yes, the cascode device size in self cascod should be larger.