help on High Swing Current source

[allennlowaton]

Please help me with this guys.
I 'm trying to grasp about this high swing current source.
Below is the diagram of the high swing.



The resulting A-wave:


My HSPICE code:


I have concerns on the following:

1.I can't understand why is it that the resulting amplitude of output current don't reach the 100uA of the reference current.

2.As I alter the channel length, there is a significant increase in the amplitude, though I'm still trying to derived the computation.

3. As I make the channel length of the M4 equal with the other three, the output current drops to almost half of the original.

Any ideas will be appreciated.

 

[dick_freebird]

This mirror style embeds a fundamental mismatch in the
Vds of M1 unless you balance things exactly right. The
lost headroom is what accounts for the fidelity (lack of).
However having another "output stack" as feedback to
the upstream reference, as I alluded in the other thread,
could give you a fairly low headroom, good match.

 

[allennlowaton]

thank you for the reply.

i guess i need to read more about this subcircuits.
i'm doing simulations about this. and i know that these are subcircuits that are part of a greater system.
I'm wondering if anyone can me a view outside of this circuit, a broader view or applications perhaps.