Re: Good Board Stack

["Thomas Droege" <tdroege2@earthlink.net>]

I knew this would bring a lot of comment.  ;^)

Arne writes:

"the nova finishes its outburst. You don't want to gamble on losing
the data, and so take flats more often, usually nightly since flats
are most commonly taken when the sky is too bright for observing
(either twilight flats or closed-dome flats) and so no important time
is lost"

As noted on this list, I have given up on light box/ dome flats.  The
problem with twilight flats is that you can't take them at the same
exposure as is used for the run, so there can be shutter effects.  Further,
there is almost no time during twilight that one can actually get a
twilight flat with a Mark IV, even with a short exposure.  Since the camera
is not super light tight, light leaks change the flat more than I would
like. 

To make flats, I pick out a period of time where the data is cloud free and
conditions are not changing - i.e. neighbors lights.  I pick 2-30 frames in
sequence, look at many of them, and then make the flat.   

What I do see is that the present scheme works just fine.  Any errors from
the flats are well down into the noise.  The flats generated from the sky
as we have been doing generate flat images.  I frequently take an image and
study it using N-S and E-W cross sections.  Unless there are clouds, the
images are flat well within the noise.  I don't see shadows or other
effects that would change the aperture photometry.  I do see clouds.  We
set a limit on how much cloud variation is tolerated.  This is much larger
than any other effect I can see. 

Should I be seeing dust donuts?  I have never seen one.  Or are they just
out of our range?  I think the sky noise is probably the culprit. It is
much larger with a Mark IV than most people are used to seeing.  Much
larger than the electronic noise, for example.  

Tom Droege


> [Original Message]
> From: Arne Henden <aah@nofs.navy.mil>
> To: <tdroege2@earthlink.net>
> Cc: tass <tass@listserv.wwa.com>
> Date: 1/28/2004 10:32:45 AM
> Subject: Re: Good Board Stack
>
> Thomas Droege wrote:
> > In the past, I have prepared new darks and flats about once a month. 
> > Looking at them, they repeat well within the noise unless I have done
> > something.  The plan for future running is to continue doing this.  My
> > thinking is to do a maintenance during the time of the full moon.  The
idea
> > is to dust/clean the lenses, and do darks and sky flats (before moon
rise).
> > I know that it is standard ritual to do flats every night.  These take a
> > lot of time, and don't contribute anything.  Likewise the cameras are
> > temperature stabilized and darks are constant.  What does affect both
are
> > the outside light conditions.  At my location these change constantly. 
> > There is no way to correct for when my neighbors decide to brighten up
the
> > night sky.  These are much more a factor than "good" darks and flats. 
So
> > why even try to do them every night?  The only reason that I can think
is
> > that "every one does them" so I will be criticized if I don't.  But they
> > are useless.  Comments, anyone?
> > 
>
> It is more of a cautionary measure than a necessary one.  If you have
> a spot in the system where dust will be close to focus, and that spot
> is open at some time during the night, then you have the potential to
> get new dust.  If you take flats once/month and the dust falls on the
> system the day after you take the flats, and then moves before you
> get a chance to take additional flats, you can't fully correct.
> In the Mark IV, the problem area is not the front lens surface, but the
> shutter mechanism.  You can get dust from paint flecks, fuzz from the
> cable, etc. that can make its way into the camera head.  If you haven't
> seen this, then you can gamble and do monthly flats.  Just be aware
> that you may lose a month of data.
>    For our Observatory, and whenever you visit another observatory, you
> only have one chance for some observations.  The comet moves, the
> CV fades, the near-earth asteroid becomes more distant and fainter,
> the nova finishes its outburst.  You don't want to gamble on losing
> the data, and so take flats more often, usually nightly since flats
> are most commonly taken when the sky is too bright for observing
> (either twilight flats or closed-dome flats) and so no important time
> is lost.
> Arne