Re: diode models

On 07/23/2016 02:42 AM, al davis wrote:
> As Felix said, the "native" diode model doesn't implement breakdown.
That's OK.  I like the idea of using as much native as I can, so I can be ready to help develop gnucap later.

For now, using HV diodes, I found this:
https://leicesterraspberrypi.wordpress.com/projects/modelling-a-diode-for-use-in-spice-simulations

Running that script with an input file like:

============mcc_4kv_diode_R3000-V-I.dat==================
1.4	009
1.6	025
1.8	060
2	099
2.2	135
2.4	180
2.6	220
2.8	260
3	310
3.2	350
3.4	400
3.6	430
3.8	470
4	500
============mcc_4kv_diode_R3000-V-I.dat==================

where the units are volts and milliamps,
gives the parameters:

john@toolbench64 [pulser-simul-cultshock]./diode_model_curve_fit.py mcc_4kv_diode_R3000-V-I.dat
Fit converged in 217 iterations with the following parameters
IS = 1.67253115538e-08
N = 3.98407657291
RS = 4.16606158811

and it draws an I-V curve with your points and an analytic function fitted to them to use as visual sanity check.

John Griessen