Ideas for dealing with Cosmic Rays?

Indeed - all my Burberry gear and shell-suits are now in the attic. Must get them out in time for Euro 2012.

michaeloconnell wrote: A good thick sheet of lead in the imaging train should do the trick.
Give Don Goldman a call and see if he has them in 50mm square pieces.

mine shaft might also work, unfortunately rather limits the FOV .

stepryan wrote: mine shaft might also work, unfortunately rather limits the FOV .

No light pollution though...

Actually, just looking back through a few hundred images taken in the last few days, they're plagued with cosmic ray hits

Dave

I was tempted to respond when you first posted but when I thought about it I realised that I didn't know too much about cosmic rays effects on CCD images - I can recognise an obvious one but I thought these rare and very obvious artefacts.

I ran a test over four hours (4 x 1hr darks - CCD operating indoors) - in the early hours after you posted (My imager hasn't moved since we last discussed CCD characterisation!).

In three of those hours I see very little evidence of CR strikes - but in one hour - there were twenty to thirty - clearly during that hour there was a flurry.
My CCD chip is small compared to yours (my Atik 16 IC uses a Sony ICX424AL chip which is only 5.79mm (H) x 4.89mm (V)) - so in such a flurry you will see more with your chip.

I clearly have a lot of anomalous signal in one of my 1hr frames. What surprised me was that there wasn't a huge ADU count associated with these. They were a small multiple of the average pixel count - but obviously anomalous. I can see why these bother you if you are blinking images (my assumption) and want to see a difference between one frame and another with the view of detecting small deltas.

I expected CR events to be "high-energy" events resulting in pixels being maxed out.
Maybe I'm reading this wrong - but that isn't necessarily the case (in terms of ADUs being maxed out).

After some reading - I believe that we can receive, for example high-energy protons (cosmic origin), and this can result in large - or maxed out ADU counts - or even displaced molecular structure in the CCD substrate resulting in damage to a pixel (hence increase in hot/cold pixels over time). However, that's not what I saw in this very brief experiment (and I need to think about this more). CCDs in space will tend to receive these effects in large quantities.

The flurry that I saw, I believe, is because when a high energy proton strikes the atmosphere we get a chain of reactions which results in, usually, multiple muons - many of which can strike our sensors (and some residual other stuff).
These are sometimes streaks - but usually just pixels with unusually high (but not necessarily overwhelming values).

I got one event of note. One little trail which looked like a little hair - straggly.
From what I read this is most probably due to "low energy electrons" - beta radiation - and probably locally sourced - which is multiply scattered (Compton effect) - I can't pretend to understand this too much. I gather that these are called "worms". These can result from local radiation sources like the BK7 glass in front of the sensor array and from hotspots in near-by concrete. There are reports that some observatories have a greater "CR Strike rate" when the telescope is positioned in particular directions which are caused by radioactive hot-spots in local concrete.

In practical terms - if the source is cosmic - then there really isn't anything one can do. If the effects are local then one can shield the measurement instrument (for example with tantalum foil). However, this is expensive and and doesn't address the problem that much.

There are algorithms that generally eliminate effects due to CR strikes - I'm surprised that the software you use doesn't address the problem.

Glad you posted your question - made me think.

Mark C.

Mark - you're right. I'm assuming they are cosmic ray hits but they could be something else.

Like you, I've seen many a (what I assume to be) CRH - straggly long line of hot pixels. On reflection, what I'm seeing now is certainly an increase in these CRH's but also a lot of random hot pixels, which are gone in the next frame.

Funnily enough, I've not noted it before so I doubt it is anything to do with my equipment. Perhaps a neighbour did something recently?

The one thing that did strike me was the increase in solar activity and I wondered if that could be a cause.

As for the software, it probably can do something but for the science stuff, we try to do as little as possible to the image other than standard calibration. For pretty pictures though, sure, the problem can be easily gotten rid of.

Thanks for posting - more food for thought!

Dave