This page (revision-25) was last changed on 02-Feb-2017 13:19 by PeterYoung

This page was created on 12-Apr-2007 18:31 by UnknownAuthor

Only authorized users are allowed to rename pages.

Only authorized users are allowed to delete pages.

Page revision history

Version Date Modified Size Author Changes ... Change note
25 02-Feb-2017 13:19 394 bytes PeterYoung to previous
24 07-Jul-2010 22:42 336 bytes PeterYoung to previous | to last
23 16-Jun-2010 22:15 359 bytes PeterYoung to previous | to last
22 16-Nov-2009 22:14 332 bytes PeterYoung to previous | to last
21 22-Jul-2008 11:12 299 bytes David R Williams to previous | to last

Page References

Incoming links Outgoing links
SandBox
SandBox

Version management

Difference between version and

At line 1 changed one line
Hey!
About the errors in eis_prep.
At line 3 changed one line
I am the sandbox!
A new version of eis_prep, due to Mike Marsh, has been uploaded in ssw.
It incorparates a corrected expression of the errors.
At line 5 changed one line
You can play in me!
In the old version, it was assumed that the fractional error of the
intensity in cgs units was equal to the fractional error of the intensity
in DN, rather then in photon units.
That is, given d, p, and I the intensities in DN, photons and erg/cm2/s/sr/A,
it was set in eis_prep
At line 7 changed one line
Just click on "[Comment?|]"!
err(I)/I = err(d)/d
At line 9 changed one line
If you are a registered user already, please login first then click on "[Edit|]"!
instead of
At line 16 added one line
err(I)/I = err(p)/p
At line 12 changed one line
----
Actually, p is almost proportional to d, with a slight dependence on the
wavelength: that is p = f(d) ~ K*d, with K almost constant.
Errors are computed in DN as
err(d) = sqrt(d+2.5^2)
being 2.5 the value of the read out noise. It follows then that
err(p) = sqrt(p+f(2.5)^2) ~ sqrt(K*d+(K*2.5)^2)
being f(2.5) the read out noise in photon units.
Thus, the correct expression of the fractional error
err(I)/I = err(p)/p ~ sqrt(K)/K * err(d)/d
differs from the previous one by the factor 1/sqrt(K).
This generated incorrect error values; the new values can be up to 50%
larger than the previous ones.
This page (revision-25) was last changed on 02-Feb-2017 13:19 by PeterYoungTop