VPulse taking finite time to rise & fall

[Animesh Maurya]

I want to create a voltage pulse with following parameters

V1=0
V2=5
TD=0.25
TR=0
TF=0
PW=0.25
PER=0.5

Here TR & TF is defined 0sec, but still the pulse is taking finite
time to rise & fall. Although in reality the pulse will obviously take
some time for rise & fall. But how can I realize the ideal condition
in Spice.

Thanks

 

[Jens Tingleff]

In article <58eab14a.0308252341.1ef1f08b@posting.google.com>, Animesh Maurya
says...

I want to create a voltage pulse with following parameters

V1=0
V2=5
TD=0.25
TR=0
TF=0
PW=0.25
PER=0.5

Here TR & TF is defined 0sec, but still the pulse is taking finite
time to rise & fall. Although in reality the pulse will obviously take
some time for rise & fall. But how can I realize the ideal condition
in Spice.

The ideal condition is only useful as a mathematical abstraction (and then only
if you define what happens at the discontinuity). In Spice, there is only one
value for each time value, so your zero rise time cannot be part of the system.

For a specific Spice simulator, there will be some options which control the
smallest possible step and the precision of the solution. By manipulating these
options, you should be able to get smaller and smaller steps around the
discontinuity. By observing what happens to the solution and applying some
knowledge of the system, you should be able to make conclusions about what the
actual solution looks like.

Hope this helps

Jens

Disclaimer: I'm discussing "normal" transient solvers. There might be some
steady-state solvers and/or envelope simulation solvers which have more abstract
approaches to solving these problems.

Key ID 0x09723C12, j.tingleff@ieee.org/jens_tingleff@yahoo.com
Analogue filtering / HIPERLAN / Mdk Linux / odds and ends
http://www.imaginet.fr/~jensting/ +44 1223 211 585
"I don't think you *can* dig your way off a planet.." D Adams

 

[Animesh Maurya]

"Kevin Aylward" <kevin@anasoft.co.uk> wrote in message news:<uwH2b.1005$nu6.696@newsfep1-gui.server.ntli.net>...

Ho humm. You havent really though about this have you?

Anyway, set the tf, tr to 0.1ps, its an engineering approximation to
zero that is unlikely to cause erroneous results in the real world. You
may also need to set the max time step to < 0.1ps, but why bother. Your
only sking for trouble.

In reality, analogue design is al about making decent approximations.
With all due respect, if you don't know when to do this, then you work
is really going to be cut out for your.

Kevin Aylward
salesEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

Well you were right that I never thought about what I said. What you
said, I did is making a way to the problem .

Setting tr & tf to 0.1p causes

ERROR -- Time step = 865.3E-15 is too small in Transient Analysis at
Time = .25 . Minimum allowable step size = 2.000E-12.
The device which is changing too fast is Q_Q1.

Rather I tried 0.01u which worked out to the desired result.

I think you are bit annoyed, but you must try to understand that
neither
Iam a Spice professional nor the Electronics, presently Iam only
engineering student who have just started learning both.

Animesh Maurya

 

[Kevin Aylward]

Animesh Maurya wrote:

"Kevin Aylward" <kevin@anasoft.co.uk> wrote in message
news:<uwH2b.1005$nu6.696@newsfep1-gui.server.ntli.net>...

Ho humm. You havent really though about this have you?

Anyway, set the tf, tr to 0.1ps, its an engineering approximation to
zero that is unlikely to cause erroneous results in the real world.
You may also need to set the max time step to < 0.1ps, but why
bother. Your only sking for trouble.

In reality, analogue design is al about making decent approximations.
With all due respect, if you don't know when to do this, then you
work is really going to be cut out for your.



Well you were right that I never thought about what I said. What you
said, I did is making a way to the problem .

Setting tr & tf to 0.1p causes

ERROR -- Time step = 865.3E-15 is too small in Transient Analysis at
Time = .25 . Minimum allowable step size = 2.000E-12.
The device which is changing too fast is Q_Q1.

That's the sort of reason why one doesn't do it. As a general rule,
*always* make a circuit *real*. If a real circuit cant do it, as I said,
your asking for trouble asking a simulation to do it. This should be
"common sense". There is no point, in trying to switch signals much
faster than 1/10 of the maximum speed you anticipate the circuit can
handle.

Rather I tried 0.01u which worked out to the desired result.

I think you are bit annoyed, but you must try to understand that
neither
Iam a Spice professional nor the Electronics, presently Iam only
engineering student who have just started learning both.

I agree, we all have to learn sometime. However...my view is that some
things should be discovered or known about on ones own. You want to at
last get a basic understanding of what's going on. Read up a bit on
numerical solution of equations.

Kevin Aylward
salesEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.