I would like to make an ideal diode model and have tried to achieve this using different settings in the .model statement, but there is always a voltage drop. I would like to make an ideal diode with no voltage drop even at mV values.
\$\begingroup\$
\$\endgroup\$
1
-
\$\begingroup\$ You should really take time to read the manual. It may be spartan, but it can help you most of the times. ltwiki.org also has valuable information. In this case, you could have gotten the answer straight from the description of the diode. \$\endgroup\$– a concerned citizenCommented Feb 4, 2018 at 7:25
Add a comment
|
1 Answer
\$\begingroup\$
\$\endgroup\$
4
You can use this diode model
.model Didl D(Ron=0.0001 Roff=100G Vfwd=0)
-
1\$\begingroup\$ Somewhat related - but not really - if you add "Ilimit = " to the parameter list, you can make an ideal diode a current limiter ! \$\endgroup\$– efox29Commented Feb 3, 2018 at 14:10
-
\$\begingroup\$ Will this increase with accuracy (to being ideal) as Ron decreases? You can set is using exponential notation i.e. Ron = 1E-10. \$\endgroup\$ Commented Feb 3, 2018 at 15:18
-
\$\begingroup\$ @loudnoises for sure you can use 1E-10 \$\endgroup\$– G36Commented Feb 3, 2018 at 16:51
-
5\$\begingroup\$ Setting too low, too high, or a too great difference between the magnitudes of the values, is not recommended, the same way null values can be problematic. To avoid this, you can simply set
Vfwd=1m Ron=1m Roff=1Gand you're set. It's unlikely these values would interfere too much with your circuitry. If you want to avoid possible 'hiccups" due to the ideal sharp knee in the transfer function, you can also addepsilon=1m revepsilon=1mto force a convergence-friendly quadratic region.ilimit, though useful, should be used with care, as it can cause problems due to the forced limitation. \$\endgroup\$ Commented Feb 4, 2018 at 7:30