Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic 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.

Register Log in

time step too small in ltspice

Status
Not open for further replies.

AbhimanyuSingh

Member level 1
Joined
Jul 7, 2010
Messages
36
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Location
chennai
Activity points
1,563
Is there something dodgy in the circuit. The time step of the circuit is of the order of 10e(-19). The prob there is with the diode. But what is it and what to do next?
 

Attachments


KerimF

Advanced Member level 4
Joined
May 17, 2011
Messages
1,461
Helped
374
Reputation
756
Reaction score
370
Trophy points
1,363
Location
Syria
Activity points
12,314
This happens even if the circuit has nothing wrong, electronically speaking.
If you attach in a zip the files that can run your circuit, there is a chance that someone (perhaps I) can help you overcome the convergence problem.

Kerim
 

mtwieg

Advanced Member level 5
Joined
Jan 20, 2011
Messages
3,468
Helped
1,235
Reputation
2,476
Reaction score
1,207
Trophy points
1,393
Activity points
25,987
I'm almost certain that if you tie one or both sides of of V1 to ground with high value resistors (like 10meg) then that will make LTspice happy.

I'm used to dealing with strange errors in ltspice and I come across this one a lot. Spice wants every node to have some DC path to ground, and sometimes diodes don't provide that well (when they're switching, anyways).
 

KerimF

Advanced Member level 4
Joined
May 17, 2011
Messages
1,461
Helped
374
Reputation
756
Reaction score
370
Trophy points
1,363
Location
Syria
Activity points
12,314
this is the .rar of the circuit.
I tried first to analyze the circuit if it is a working one.
Did you intentionally left the output of opamp U2 unconnected?

Important note:
I noticed you have used the automatic node names (like n002 and n014) in the formula of B1. This is risky because most of the circuit nodes will likely change their name (automatically) anytime the circuit is updated. It is better that we give the important nodes fixed names as we see it convenient, like In, Out, Fbk... etc. so that their names won't be changed when the circuit needs to be altered.

Kerim
 
Last edited:

andre_teprom

Super Moderator
Staff member
Joined
Nov 7, 2006
Messages
9,181
Helped
1,141
Reputation
2,301
Reaction score
1,116
Trophy points
1,403
Location
Brazil
Activity points
53,328
...The time step of the circuit is of the order of 10e(-19)...
Seems to be excessivelly small.
Note that simulation float point values have a limited magnitude size.
At numerical analysis, its impact is at convergence of iteration steps.

+++
 

dick_freebird

Advanced Member level 5
Joined
Mar 4, 2008
Messages
6,915
Helped
2,025
Reputation
4,054
Reaction score
1,865
Trophy points
1,393
Location
USA
Activity points
55,373
The most common cause is insanely low node capacitances.
Elements which are bistable or singular also play hell. Try
looking at the diode in isolation if you are convinced it's the
problem and not the victim, especially for signs of an I-V
"blowup", junction capacitance approaching zero or worse
yet changing sign, etc.

Ehat this error mesage purports is that your last timestep
had some error, that no amount of downranging timestep
can pull in - because the error is already embedded to the
matrix.

This being the case you might look at the final timepoint
for unreasonable voltages / currents, and figure out who
is pushing garbage.
 

Status
Not open for further replies.
Toggle Sidebar

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top