The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> Setting for loop analysis in non-high impedance no
https://designers-guide.org/forum/YaBB.pl?num=1128007111

Message started by georgejor on Sep 29^th, 2005, 8:18am

Title: Setting for loop analysis in non-high impedance no
Post by georgejor on Sep 29^th, 2005, 8:18am

Dear all,

I have a question about loop analysis, hope anyone can help me. Thank you very much!

I want to break a close-loop circuit to perfrom stability analysis (gain & phase margin) now. Normally, i should break the node where is the high impedance point (e.g. between OTA's output and input).

But the node that i want to break is between Drain of M1 and Source of M2. If i break it and insert inductor, capacitor and ac-source into it, the couple effect between drain and source is ignore. So, what can i do?

I had tried to duplicated the M2 and connect it to M1 in parallel, so that Drain of M0 can still see the source of M2, but Ids in M0 is double. Seems not correct, right?

Please help me! Thanks!

Best Regards,
George

Title: Re: Setting for loop analysis in non-high impedanc
Post by sheldon on Sep 29^th, 2005, 8:24am

Georgejor,

Without a schematic, it is difficult to understand
where M1 and M2 are in the circuit. Some comments:
1) Spectre supports stability analysis and will calculate
the gain and phase margin directly for you. See the
documentaiton for details.
2) In general, "opening" the loop using inductors and
capacitors is not a recommended method for simulating
op-amp stability, see Ken boon, "The Designer's Guide
to SPICE and Spectre" for details.

Best Regards,

Sheldon

Title: Re: Setting for loop analysis in non-high impedanc
Post by georgejor on Sep 29^th, 2005, 8:48am

Thank you for your reply!

Here is the diagram of the circuit. acutally, it is a circuit in paper "AREA-EFFICIENT LINEAR REGULATOR WITH ULTRA-FAST LOAD REGULATION"

I perform an analysis as follow:

In this setting, the couple effect of M0 drain and M4 source is ignore.

So, it is modified as follow:

But the Ids of M0 is doubled!
So, how can i do? Thank you!

Title: Re: Setting for loop analysis in non-high impedanc
Post by georgejor on Sep 29^th, 2005, 8:49am

May i add one more thing, i am using HSPICE instead of Spectre. Thanks