# Retard-o SPICE EXP Question

Discussion in 'CAD' started by Mike Rocket J. Squirrel Elliott, Nov 19, 2003.

1. ### Mike Rocket J. Squirrel ElliottGuest

Should be obvious to me, but my lack of skill interpreting SPICE
helpfiles is, after all, the stuff of legend. It's tough being a legend.

I want my voltage source to climb exponentially from 0 volts to 50 volts
after a 10ms delay and have a risetime constant of 1s.

But EXP 0 50 10ms 1

Creates a waveform that starts at 0 volts, then, after 10ms, plummets to
-50v, and climbs exponentially back up to 0.

How do you get it to sit at 0v for 10ms, then climb exponentially to +50v?

2. ### Mike EngelhardtGuest

Mike,
EXP(0 50 10ms 1 1e308)

--Mike

3. ### Mike Rocket J. Squirrel ElliottGuest

1e308? Some magic number?

4. ### Mike EngelhardtGuest

Mike,
It's just a time that is longer than I thought you'd have as the stop time

--Mike

5. ### Mike Rocket J. Squirrel ElliottGuest

Yeah. A reasonable guess ... it's -- what? -- 3e300 years?

6. ### Paul BurridgeGuest

Mike obviously believes in giving plenty of leaway.

7. ### Mike Rocket J. Squirrel ElliottGuest

For good reason! There's nothing more annoying than the premature
termination of a long transient simulation of those circuits you're
designing to last well beyond the projected age of the universe.

8. ### Mike EngelhardtGuest

Mike,
It's an easy-to-remember approximate upper limit of the
magnitude of a 64bit floating point number. Most any
reasonable number encountered in numerical computation would
be between -1.7e308 and 1.7e308. Numbers beyond that
range can be grouped with infinity. Yet, 1e308 is much longer than
the estimated age of the Universe. So 1e308 is slightly less than
infinity but much longer than forever.

--Mike

So many singularities, so little time.