There were previous educated discussion regarding the SolarRadiation vs Illumination (
https://groups.google.com/forum/#!newtopic/weewx-user). Upon that, even been different observations, one may consider a good approximation/estimation of "Illumination / SolarRadiation =~ 119-120" multiplier.
Thus, under Belchertown skin generated 2 plots, that are similar but just value "multiplier" as the following (low value observations due to very cloudy day):
Belchertown 'graph.conf':
[[chart1]]
title = SolarRad & UV
[[[radiation]]]
name = Solar Radiation
color = "#ffc83f"
zIndex = 1
[[[maxSolarRad]]]
name = Theoretical MaxSolarRad
type = area
color = "#f7f2b4"
[[[UV]]]
name = UV Index
yAxis = 1
yAxis_min = 0
yAxis_max = 16
color = "#90ed7d"
zIndex = 2
[[chart2]]
title = Illuminance
type = spline
yAxis_min = 0
[[[illuminance]]]
name = Illuminance
yIndex = 1
[[[maxSolarRad]]]
name = Theoretical MaxIlluminance
type = area
color = "#f7f2b4"
yAxis = 0
yAxis_label = "Illuminance (lux)"
observation_type = 119.0*maxSolarRad
[[[UV]]]
name = UV Index
yAxis = 1
yAxis_min = 0
yAxis_max = 16
color = "#90ed7d"
zIndex = 2
Although accomplished to estimate Illuminance from SolarRadiation, it's a bit "non-sense" to have both similar graphs plotted. So, searching for a solution to have both at same graph with multiple yAxis.
Found the following posting in Highcharts demo, but could not reproduce (or translate) into Belchertown yet.