It says here that:
Camera settings also can be determined from incident-light measurements, for which the exposure equation is
where:
As far as I know, illuminance E has units of lux, not lux-sec which is more commonly denoted by H.
So, the question is: where do the extra seconds units come from?
Haters of AI will be pleased to know that ChatGPT was unable to help with that question …
On the left side, the per second term t is cancelled by the light quantity on the right that is your lux second.
Thanks. Yes, that is mathematically correct even though the SI definition of illuminance does not include time. Based on that, calling “E” illuminance is technically incorrect
Actually the equation balances only if E is lux, not lux-sec:
N^2 / t has units of t^-1 because N is dimensionless
ES / C = lux.ISO/(lux.s.ISO) also has units of t^-1 contained in the constant C which is lx.s.ISO
ergo, Wikipedia’s statement that illuminance is lux-sec is incorrect. I am no longer puzzled … 
In terms of ISO, ES would represent exposure-time in lux-seconds; on the left, shutter speed.
In Wikipedia’s equation, ES is illuminance times ISO, not exposure-time in lux-seconds.
N^2/t = ES/C … for ISO = 100, N^2/t = E/2.5 …
For N=8 and t=(1/125) E= 20,000 lux (1858 fc) = about 13 EV …
It seems that we talking at cross-purposes.
P.S. just found this later in the same article:
" Illuminance is measured with a flat sensor; a typical value for C is 250 with illuminance in lux."
Ipso Facto.
S and C are time integrated, the units lux seconds per square meter and lux seconds. To me it balances in terms of time.
At my age, I find it difficult to verify terse one-line responses.
I fold …
be glad it is only about photo exposure…
could be worse like
Looks like it’s been fixed! (see original post)
Did you do that?
Yes, after some unit analysis to double check… I believe it makes better sense now.
Well done!
