https://open-source-silicon.dev
Join Slack
Hello everyone. I'm trying to make a plot using the internal xschem plotter. I have attached the i...
# xschem
e
Hello everyone. I'm trying to make a plot using the internal xschem plotter. I have attached the intended relationship to be plotted and the xschem postfix representation of the ( relationship) expression. I just want to confirm if the postfix representation of the expression is correct because I'm not getting the expected plot. The xschem plot and the expected(reference) plot are also attached here. @Stefan Schippers @Tim Edwards
l
looks equivalent to deriv(id) / id to me. Maybe • use a smaller timestep and • scroll your graph to the right one pixel (or set the graph's xmin=2*stepsize) to exclude the values near zero? (deriv is weird on first value)
• Could also use an ammeter.sym to make sure you're grabbing the right value, since the symbol referencing is weird. • Putting this at top of your ngspice control block should improve accuracy for small currents:
Copy code
.option GMIN=1e-18
.option method=gear
• maybe also add epsilon in your plot to avoid dividing by zero (or start plot with larger current). If your ammeter is named `vm`:
Copy code
i(vm) 1e-15 + deriv() i(vm) 1e-15 + /
s
The expression for 
λ
divides 
gds
by 
Idsat
not by 
Id
If output resistance is needed I would simply calculate 1/(d(Id)/d(Vds)):
e
Thank you very much @Luke Harold Miles. I'd try that out.
Yes @Stefan Schippers thank you. I wanted to estimate the value for \lambda in the gf180 SPICE file. Since I need Idsat, please how do I get the value for Idsat?
4 Views
Open in Slack
PreviousNext
Powered by

What is Linen?

Linen is a search-engine friendly community platform. We offer integrations with existing Slack/Discord communities and make those conversations Google-searchable.

Powered by