Problem in corner case design of opamp

[gavanbash]

Hi everybody.
I designed a new purposed opamp which it has 90db gain. in this opamp I use a positive structure in order to increasing opamp but in the various corner case it dosen't work.
How can I design my opamp in order to work in every corner cases properly?
I don't know what should I do in design procedure.
How can I solve this problem.
Thank you

- - - Updated - - -

I have this results in different corner cases

 

[dick_freebird]

If you mean that you built in some positive (feedback?)
for gain boosting, you need to be sure that you have
not made it hysteretic, which would overcome a small
signal stimulus in simulation.

I don't see a problem in what you showed, other than
that the devices would need to run with pretty low
Vgs to be in saturation (which the "region" indicates
that you are).

But you say nothing about what the op amp, as an
assembly, is failing to satisfy. Maybe the FETs you
display are not even the problem.

You should treat each corner case as a "failure"
and do a failure analysis, only using simulation
instead of bench equipment and analytical lab.
Get down to the root cause of {whatever you're
unhappy about} for each case, collect the cases,
figure out any common thread(s), and commence
to designing, pass 2. 3. 4.

That's where you earn your paycheck. Simulation
is just punching the timecard.

 

[gavanbash]

Thank you for replying.
I used positive feedback structure in my topology but I can't understand hysteric. could you explain it please?
another thing is that you said I have to find out root cause in other corner cases but I don't know how can I do that.
I'm really confused about corner cases design procedure do you know any books or paper which I have to read it?
Thank you for your replying.

 

[chaitugvsnk]

Hi Gavanbash,

can you pls let us sknow the Op-amp ckt configuration you are using. May be thee CKT picture with W/L used might help us to understand better. also let us know what corner case is it failing.