reply to post by Anokorok
Here is an interesting question; given that SOHO is at L1,
(
map.gsfc.nasa.gov...)
...what is the source of these cosmic rays on the dates in question? For the angles to be crossing the planar surface of the CCD to be able to
initiate bleed out like this (if that is even possible) then there are a surprisingly consistent and oddly uniform ( energetically ) number of cosmic
rays coming from sources above and below the Galactic plane ( remember our solar system sits tilted 30 some odd degrees off from the galactic
plane)
While I am not exactly positive, it would appear that SOHO is using a multi-channel 1024x1024 SITe CCD Model TK1024AF CCD
(
www.surplusgizmos.com... )
although it could easily be one of it's ancestors, but according to nasa this is a slow read out and processing that undergoes extensive in orbit
calibration calculations to correct for radial or "barrel" distortion common to this type of camera.
(
www.cv.nrao.edu... ).
The end result is only something like 10 pictures an hour. (so a 'fast' cosmic ray shouldn't leave such an odd shape)
Also according to the technical specs for the SITe-CCDl-TK1024AF
(
astrosun2.astro.cornell.edu... )
the design and construction of the device make it vary hard to explain bleed over that remains so symmetric through more than a tiny fraction of
angles . And even then it is almost impossible to explain both fuzzy random "exhaust trails" (groups of blobs not symmetric) and groups of sharply
defined white-out ( especially once again considering the device calibrations)
Further more Nasa's own site on SOHO LASCO Comet Finders' Page,
(
sungrazer.nrl.navy.mil...)
describes how to not misinterpret objects and how to correctly identify: Cosmic ray 'noise', stars planets other known objects and debris trails and
utterly it fails to mention or show this "well known...bleed through" , although it does show:
planets and other bright objects bleeding out in a horizontal ( noticeably missing any fuzzy 'exhaust plumes') fashion consistent with the hardware
parameters...
( same site as above under #1: /planets.gif )
Cosmic ray noise ( streaks and points , but again no combination of blurry 'exhaust fumes' ( which have clearly been through processing )attached to
rigidly defined non fuzzy pixel blocks( areas obviously not processed the same way under the soho internal 'barrel' corrections) :
( same site as above under #2: /cosmicray.gif )
space junk( notice how the software automatic process masking around the edges on the ccd image and over the arm bare a strong resemblance blocky
pixelish wise to the large white 'dustpan' areas in question )
( same site as above under #3: debris.gif )
Yet oddly during one of the heaviest saturation events in the history of SOHO the "Bastille Day Event" not a flying dustpan or wood boring bit to be
seen (
science.nasa.gov... )
In fact the object most like the one in question is on the comet hunting guide:
(
sungrazer.nrl.navy.mil... )( a page which also has more cosmic ray pictures but no shovels , dust pans , etc)
So what do we have?
Well we have at least two process sets being applied to odd areas on the image that do not seem to match the given parameters for the device or it's
calibration when taken together (fuzzy 'exhaust' areas in direct connection with crisp pixelation areas), indicating an automatic (predisposing
preprogrammed) software masking of an artifact anomaly.
Frankly I think the underlying blobs and whatever they are trailing are the 'anomaly' that the soho software tries to 'correct' and they either create
the sharply defined bleed out or, given it's crispness that is added as a mask in another layering step.
Oh, and about not seeing any of these side ways (rotated as oddly stated) well no , But a symmetric bleed out shape IS AN ABSOLUTELY CERTAINTY it
would just be a different shape, variations on a left/right triangle should certainly show up given the right cosmic ray impact angle, in fact if one
never sees this it is a good indication that that software masking and not hardware overload is the first suspect
edit on 4-2-2011 by Silverlok because: bad links
edit on 4-2-2011 by Silverlok because: forgot stuff