Skip to contents

Measure a simulated plot

Source: R/measure.R
spice_measure.Rd

Scalar measurements of a plot with the .measure grammar of ngspice-47 (SPICE3f5 has none): rise and fall times and delays (TRIG ... TARG ...), level crossings (WHEN), values at a point or at an event (FIND ... AT= / FIND ... WHEN), derivatives (DERIV, a Dune extension), and window statistics (MIN, MAX, MIN_AT, MAX_AT, PP, AVG, RMS, INTEG). The engine does the measuring, exactly as for the deck's own .measure cards (which spice_run() returns in $measurements); measurements read the plot only and never change a simulation.

Usage

spice_measure(plot, spec)

Arguments

plot

A spice_plot of a .tran, .ac, .dc or .sp analysis, from spice_run() or spice_read_raw().

spec

A character vector of measurements, e.g. "trig v(out) val=0.1 rise=1 targ v(out) val=0.9 rise=1", "when v(out)=0.5 cross=last", "max v(out) from=1n to=5n". Its names, when given, name the results; unnamed ones are called measure (or measure1, measure2, ... for several).

Value

A tibble with one row per measurement:

  • name (chr), analysis (chr: "tran", "ac", "dc", "sp") and plot (chr: the plot's label);

  • value (dbl): the result – a time for WHEN and delays, a value for FIND/MIN/AVG/..., the scale value for MIN_AT/MAX_AT; NA when the measurement failed;

  • trig, targ, at, with, from, to (dbl): the extra numbers ngspice prints beside the value (delay ends; where a MIN/MAX occurs; the extremum of MIN_AT/MAX_AT; the window of AVG/RMS/INTEG/PP), NA where they do not apply;

  • error (chr): why the measurement failed ("out of interval (WHEN)": the event never happens in the window), NA otherwise.

An event that never happens is a row with error set, not an R error. A malformed spec is an error of class spice_error_measure.

Details

Each spec is what follows .meas <analysis> <name> on a card; the analysis is the plot's own. Values take SPICE suffixes (2n, 1meg). Crossings interpolate linearly between points; AVG integrates by trapezoids over the window's points, RMS and INTEG by Simpson's rules with the window edges interpolated, all as ngspice does.

Mirror of the Python package

This is spice.measure(); the deck's measurements are Result.measurements there.

See also

spice_run() (its measurements field).

Examples

res <- spice_run("rc step
v1 in 0 pulse(0 1 1n 1p 1p 20n 40n)
r1 in out 1k
c1 out 0 1p
.tran 10p 10n
.meas tran trise trig v(out) val=0.1 rise=1 targ v(out) val=0.9 rise=1
.end")
res$measurements
#> # A tibble: 1 × 11
#>   name  analysis plot        value    trig    targ    at  with  from    to error
#>   <chr> <chr>    <chr>       <dbl>   <dbl>   <dbl> <dbl> <dbl> <dbl> <dbl> <chr>
#> 1 trise tran     Transien… 2.20e-9 1.11e-9 3.30e-9    NA    NA    NA    NA NA   
tran <- spice_plots(res)[["Transient Analysis"]]
spice_measure(tran, c(
  t50 = "when v(out)=0.5",
  slope = "deriv v(out) at=2n",
  vmax = "max v(out)"
))
#> # A tibble: 3 × 11
#>   name  analysis plot            value  trig  targ    at  with  from    to error
#>   <chr> <chr>    <chr>           <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <chr>
#> 1 t50   tran     Transient A… 1.69 e-9    NA    NA NA       NA    NA    NA NA   
#> 2 slope tran     Transient A… 3.68 e+8    NA    NA NA       NA    NA    NA NA   
#> 3 vmax  tran     Transient A… 1.000e+0    NA    NA  1e-8    NA    NA    NA NA