Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

question about spectre simulation

Status
Not open for further replies.

jiangxb

Full Member level 3
Joined
Apr 1, 2007
Messages
153
Helped
25
Reputation
50
Reaction score
4
Trophy points
1,298
Location
hangzhou, china
Activity points
2,075
spectre simulation book

hi

i encounter a question when simulating a circuit in spectre, and the error report is as follows.

========================
Error found by spectre at time = 1 as during transient analysis `tran'.
No convergence with minimum time step. Last acceptable solution computed at 0 s.

The values for those nodes that did not converge on the last Newton iteration
are given below. Also given is the manner in which the convergence
criteria were not satisfied in the following form:
Failed test: | Value | > RelTol*Ref + AbsTol

V(M0:int_s) = 32.1378 pV, previously 6.79829 pV.
residue too large: | 81.3739 uA | > 836.571 pA + 1 pA
V(NETZ010) = 5 V, previously 5 V.
residue too large: | -113.27 pA | > 1.49678 fA + 1 pA
I(V1:p) = -4.84012 uA, previously 0 A.
update too large: | -94.7231 nA | > 48.4012 pA + 1 pA
V(C2.c1.hn1) = -15 V, previously 0 V.
update too large: | -300 mV | > 150 uV + 1 uV
========================

how should i do to solve this question?
 

update too large spectre

It looks like u need to relax ur convergence criteria, as the simulator couldn't find a solution within the specified criteria even when using the minimum allowed time step.

An easy solution is to use Moderate or Liberal setting when choosing the transient analysis, a better solution is to change the reltol,abstol and relref settings in order to avoid this problem( however using the second solution u need to be careful in order to get the required accuracy out of the simulator).

Check Ken Kundert's book , Designer's Guide to SPICE and Spectre,in chapter 4, u will find a good explanation of how transient analysis work and the effect of different settings on the simulation accuracy and convergence difficulties.
 

    jiangxb

    Points: 2
    Helpful Answer Positive Rating
simulation update too large

if you had set any initial conditions on your cap or any such conditions, try removing them. or probably if you have switches be sure your timing is right.
the problem is that spectre was not able to solve the equation of your circuit with the values you had supplied.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top