Decisions decisions

[Kenny]

Samsung Galaxy Tab 3 8.0" bought last December for £179 broke. Amazon unable to repair so giving a full refund. I could a) buy a new one now only costing £134 or b) invest in a DSLR Camera for astrophotography. Hmmmmm. Comments (particularly re option b)?

p.s. Carole - reading your DSLR guide. I can get a 450D for £138...

[Carole]

Well Kenny, it's certainly a good way to get into deep sky Astro-imaging and cut your teeth on it all.  The only thing is you would have to get yourself a mount like the HEQ5 to enable long exposure, so this bumps the cost up somewhat.

Also what scope would you be using?

You could get the Moon on your Dob with a DSLR but that's about all really, planets are too small. 

Carole

[The Thing]

You could get the Moon on your Dob with a DSLR but that's about all really, planets are too small. 

It's relative, if the DSLR pixels are the same size as the planetary camera (and they are generally ish) and you can get the same optical magnification (i.e. a barlow or two) the the image scale in arsecs/pixel can be the same, you just have to throw away the surrounding empty frame on the DSLR image.

Using AstroPhotoTool there is a Planetary function that takes either a movie or high frame rate JPGs using LiveView (depends on the Canon Digic processor version) and you can use the LiveView 10x feature which effectively give you a region of interest ROI function like a CCD, i.e. it chucks the surrounding pixels away and just gives you the important bit with the planet.

Does a 450D have LiveView? I know a 305D doesn't.

[Kenny]

Would I also be able to use a DSLR for wide field photos without telescope magnification?

[Carole]

Would I also be able to use a DSLR for wide field photos without telescope magnification?

You can Kenny, but unless you have a tracking mount you won't be able to take exposures for more than about 20secs before trailing starts to show, unless of course you want to actually take star trails.

you can get the same optical magnification (i.e. a barlow or two)

Forgot about using a barlow, but I still think they would be quite small.

Carole

[RobertM]

A small portable tracking mount for wide field is a very useful thing to take on DSCs and holidays.

With the small pixels in these cameras you will see trailing well within 10 seconds (without tracking) depending on where in the sky the camera is pointed and field of view.  I remember seeing trailed stars with Fays 650D and a camera lens after 5 seconds (zoomed in).

Don't forget star trails - they don't need any tracking and are useful practice.

Robert

[The Thing]

Forgot about using a barlow, but I still think they would be quite small.

Carole

Only relatively to the size of a DSLR frame, the planet can actually occupy the same number of pixels in each case - been there, done that. However the webcam route seems to produce better images at the end of it.

[MarkS]

Samsung Galaxy Tab 3 8.0" bought last December for £179 broke. Amazon unable to repair so giving a full refund. I could a) buy a new one now only costing £134 or b) invest in a DSLR Camera for astrophotography. Hmmmmm. Comments (particularly re option b)?

Very difficult to get apps to run on a camera

[Kenny]

Camera won.

Canon 450D digital SLR with original Canon 18-55mm kit lens plus a Sigma DC 18-200 mm lens.

[Carole]

Well done Kenny, look forward to watching your progress.

Carole

[Kenny]

May be a slow start. I haven't used an SLR in over 20 years, and that one had film in it!

RTFM.

[The Thing]

YouTube - bound to be some stuff on there.

[Fay]

in general with a lens at 10mm  you should be able to do 60 secs without trailing, 15mm lens = 40 secs, 20mm lens= 30 secs. push the ISO and you can cut down the length of subs

[Kenny]

Thanks Fay. That's very useful. I'd heard of rule of 600 which is similar. I'll post some of yesterday's attempts.

[Mac]

Your sort of correct Fay, but it can be as low as 4 seconds before bluring,
Heres a copy of a post I made on another forum, just adjust the maths a little,

You are nearly correct in that its 600/fl as a general rule not 500, also you forgot one very important thing,
That general rule is for a full frame camera, the 7D isnt a full frame,.
Some maths to help you get your head around why.
Your sensor size is 5184 x 3456 pixels on a  22.3mm x 14.9mm chip,
which gives a pixel size of 4.3um or 0.0043mm

if you take a star as being a point source then it will produce a point source on your chip,

if we assume that that the star completely fills one pixel and to create blur has to move onto the next pixel,
this equates to a movement of one pixel.
¦....¦....¦....¦
If the above image represents three pixel, if the star as a POINT source is on the far left of the pixel,
all the time the star is in that pixel it will be detecting its light, once its moved onto the next pixel,
you can declare that the image is now blurred as the star has now moved.

Your field of view on the camera is 96 deg x 73 deg @ 10mm, so each of the pixels on your camera covers,

73 / 3456 = 0.0211

96 / 5184 = 0.0185 degrees. (assuming movement only on long side)

The stars in the sky rotate (well we do technically) 360 degrees in 24 hours (technically 23H56m) which is 86400 seconds.

so how long in seconds before one star moves 0.018 degrees

in one second the star moves 360/86400 0.004 degrees

the number of seconds take to move 0.018 degrees (one pixel) is 4.4 seconds.
THIS IS FOR STARS ON THE CELESTIAL EQUATOR. i.e. 90 deg from the pole
You then need to add some trig equations to calculate exactly how much SLOWER the stars will move depending on where you are
pointing your camera.
i.e if you are pointing your camera at the celestial equator then you will get maximum movement, i.e. 0.004 deg per second
The higher up towards the pole star the slower the stars appear to rotate,
If you are pointing your camera at the pole star. i.e. the pole star is centered exactly on your chip, then the diagonal of your camera will have a FOV of 120 degrees corner to corner, The corners being 60 degrees from the polestar, with 90 degrees being the celestial equator.
These stars move  @ 360/86400Deg * sin(60) = 0.0036
Time = 0.0185 / 0.0036 =
At this distance the stars will take slightly longer before they move one pixel, this is 5 seconds.
So after 30 seconds your star would have moved about 6-7  pixels.
Hope this helps.
Mac.

[MarkS]

Forget the rule of 600 - I see this rule repeated time and time again in forums.  It dates back to the days of film, trying to avoid obvious trailing on a photographs printed at some standard size.  But we tend to look at images at full size on big monitors where the trailing becomes obvious.

So, I prefer to think in terms of star trail length which is give by own rule of thumb formula:

Star trail length in pixels = focal_length * exposure_time / pixel_size / 15

where:
exposure_time is in seconds and pixel_size is in microns.

If I used the rule of 600 on my 600D at 20mm focal length it would suggest an exposure time of 30 seconds.  But the 600D pixel size is 4.3 microns so this would give a star trail length of :

20*30/4.3/15 = 9 pixels

Do I want star trails of 9 pixels?  Not in general.  That's why I don't use the rule of 600.

To find out the star trail length produced by the "rule of 600" for your own camera then use this:
star trail length = 40/pixel_size

Then you can make your own judgment on it.

Mark

[Kenny]

Thanks all. Very useful clarifications.

Interesting experiment this evening with the Canon 450D and Nikon lens set at 18mm. Steady camera mount, no tracking. ISO-1600. f/3.5. 30 seconds exposure was far too long (unsurprising) so didn't bother this evening. 15-20 seconds gave the best looking image, starting to pick up clues to some Messier objects, but still a surprising amount of star trailing (6-8 pixels on the full size image). 10 seconds wasn't enough, leaving a very grainy dark frame with much less detail than the longer exposure images.

The focus may have been marginally off which would obviously compound the problem. Unlike the other evening where I was able to use this Moon for sharp manual focus, I really struggled this evening to focus on stars (or Jupiter).

[MarkS]

The focus may have been marginally off which would obviously compound the problem. Unlike the other evening where I was able to use this Moon for sharp manual focus, I really struggled this evening to focus on stars (or Jupiter).

Are you using the kit lens (which has no focusing scale)?  Don't rely on the auto focus, even with a bright object like the moon, it does not perform reliably enough.  So set it to manual focus.  The best tip I can give you is to set the zoom (if it is a zoom lens) at one extreme or the other (e.g. 18mm) and then stick on a couple of small strips of masking tape - one on the focusing ring and the other adjacent to the focusing ring so that the act of focusing moves one past the other.  Draw a focusing scale on these in pencil. 



Find the precise focus point on the scale you have drawn by taking a series of test shots on a tripod.  You can do this in daylight on a distant object or on a trailing star field at night.  Once you have found focus then tape the ring precisely in that position so it can't be accidentally knocked.

As you perform your test focusing, always move the focusing ring in the same direction (either clockwise or anticlockwise) because the lens mechanism has a fair amount of backlash so if you mark an exact focus point and then approach the same marked point from the other direction then you won't be exactly in focus.  I actually draw an arrow on the masking tape to indicate the direction to turn the focusing ring.

This might all sound like a palaver but seriously, focusing is by far the biggest issue that folk have, precisely because they don't tackle it in a careful systematic manner.  The kit lens can produce quite extraordinary results when used carefully.  By the way, if you want to use the camera at mid zoom then tape the zoom ring into a fixed position before using the same procedure for setting focus.

I tape over the AF/MF switch also so it reminds me to remove any tape fixing the lens in position before using it in autofocus again

The other big tip is to use a dew band around the lens because generally speaking the lens will dew up very fast in typical UK imaging conditions.

Mark

[RobertM]

For focus I rely on the fact that nearly all lenses produce different colour fringing intra and extra focal.  I focus till that colour fringing is minimised.  Admittedly finding a suitable star might be a challenge in a wide field but live view zoom makes all the difference.  If I can't use a mask (Bhatinov) then that's the method I use.

HTH
Robert

Just to qualify, in using this method the star has to be close to the centre of frame.  If you use one near the edge then that will likely be a slightly different focus position ie less than 'perfect'.

[Kenny]

Thanks.

So far have avoided using the kit lens for night photography as a couple of reviews said it wasn't very good. I'm using the Sigma DC 18-200mm lens which is only manual focus. It has only a very basic focus scale, nothing you could accurately use so your idea of using tape and focusing in the daytime is definitely worth a try. I can't see much through the viewfinder or Liveview, certainly not enough to focus on anything but really bright objects (like Jupiter last night) but even that was a challenge. I was more or less going to infinity then left a bit. p.s. also taking a shot then checking the focus by zooming in with the playback mode.

What is a dew band?

I took a few frames of Orion and Canis Major last night. Having problems stacking them with Registax (another story) so tried manually stacking them as lighten layers in Paint.net. Interestingly, and annoyingly, the stars don't align across the whole frame. When I align the stars in the central region, the further out to the edge you look the greater the alignment gap (very noticably). Not sure if this is an error I've introduced. Going to have another go.

[MarkS]

What is a dew band?

I took a few frames of Orion and Canis Major last night. Having problems stacking them with Registax (another story) so tried manually stacking them as lighten layers in Paint.net. Interestingly, and annoyingly, the stars don't align across the whole frame. When I align the stars in the central region, the further out to the edge you look the greater the alignment gap (very noticably). Not sure if this is an error I've introduced. Going to have another go.

A dew band is one of these:
http://www.firstlightoptics.com/dew-prevention/astrozap-dew-heater-tapes.html
They need a 12v power supply

Lenses have distortions - that's why the alignment is failing as constellations pass through the field of view.  Most stacking programs will do their best to correct the distortion they find.

Mark

[Carole]

I took a few frames of Orion and Canis Major last night. Having problems stacking them with Registax (another story)

Kenny, have you tried stacking in DSS (Deep Sky stacker), Registax is really for stacking AVIs I didn't know you could stack DS images in it.  DSS is free:
http://deepskystacker.free.fr/english/download.htm

I thought there was a more recent version than this, but this is all I could find.

In this you can load your lights, darks, flats and bias.

Carole

[Kenny]

I took a few frames of Orion and Canis Major last night. Having problems stacking them with Registax (another story)

Kenny, have you tried stacking in DSS (Deep Sky stacker), Registax is really for stacking AVIs I didn't know you could stack DS images in it.  DSS is free:
http://deepskystacker.free.fr/english/download.htm

Oooh. Thanks. Will give that a go later today.

[Kenny]

What is a dew band?

A dew band is one of these:
http://www.firstlightoptics.com/dew-prevention/astrozap-dew-heater-tapes.html

What size would fit a dslr.

Do you attach it to tbr front or nearest the camera body?

[MarkS]

It wraps around the cylindrical lens body, as close to the front lens element as possible (since you are trying to heat the front lens element to prevent its temperature dropping below the dew point).  If your lens body is 3" in diameter buy the one for a 3" telescope.  If it is 4" in diameter buy the one for a 4" telescope etc.

You need a 12volt supply to drive it.  A dew controller is an optional extra so you can control the amount of heat it delivers but there's nothing wrong with running it full power.  You may well need full power in any case to push sufficient heat through the plastic body of most modern lenses.

Mark

[The Thing]

What is a dew band?

You can make them quite easily if you have a soldering iron. PM me if you want to have a go as I have instructions and a box of bits left over from my DewItYourself dew heater kit business (now closed) which you are welcome to have a rummage through. That way you can make one just the right size.