From: Bastien ROUCARIES <roucaries.bastien@gm...>  20071216 22:01:27

Le dimanche 16 d=E9cembre 2007, Z a =E9crit=A0: > Hi again, > > This time I connected an AC power source to a rectifier bridge, and 2 > blue 20 mA LEDs in series (from library) as load. Transient simulation > runs propely at 300 mW AC source power, but I get a Jacobian singular at > 301 mW. > > At 250 mW it was ok until I connected an 1 pF capacitor. (actually, it > is a placeholder for a varistor) > > Sometimes the simulation becomes ok, if I change the number of > timesteps, or the time interval of the simulation. It is more stable, if > I start simulation from 0 msec. > > If I increase the capacitance to 1 uF, the simulation is stable. > According to the diagrams, the LED current becomes stable a little bit > below 1 mA. > > I would like to know, how I could eliminate these Jacobian singulars. I > would run the simulation above 300 mW and below 1 uF.. > > I have a similar circuit with 350 mA leds, 3W power source and > capacitors around 1 mF, and there I have the same errors in the region > 500..2000 uF. Any help appreciated. Try to add resistor in parallel with capacitor (about 1Mohm) Normally it he= lp Regards > Thank You for your reply in advance. > > Szalai Zolt=E1n =2D=20 "ROUCARIES Bastien" roucaries.bastien@...= com =2D= =2D DO NOT WRITE TO roucaries.bastien+blackhole@... OR BE BLACKLISTED 