1/3 vs 1/4?

[robmichael81]

I'm in the market for a astro camera and not sure on which one to go for. There are so many options. I thought I was going to go for the ZWO asi120 however I started to notice there were option such as 1/3 and 1/4. What does this mean?

If anyone can give me an idea on what would suite me best I would appreciate it.

Basically I'm looking for an entry level, mainly planetary (but would like it if it could take basic DSO as well), camera that eventually, as time goes buy and my budget (wife) allows, will probably be used just as a guide camera as my interest and equipment develops.

Rob 

[Bizibilder]

The values of 1/3 or 1/4 refer to the size of the sensor -  one third or one quarter of an inch across (about 6-7 mm in new fangled units).  I have an ASi120mm mono camera and it is excellent!  However if you go for mono then you will need a filter wheel and filters if you want to take colour (RGB) images.  A one-shot colour version of the camera is available that allows you to take colour images without the use of filters etc.

This is an image taken with an ASI120mm mono and a 120mm refractor last night when the seeing was not particularly good.  http://stargazerslounge.com/topic/244979-the-moon-24-may-2015-4-pane-mosaic/

[ecuador]

Obviously the mono camera with a filter wheel will give you superior results (with more effort) compared to a one-shot-color. I say "obviously" because otherwise there would be no reason to prefer a mono camera

The size of the sensor (1/2", 1/3", 1/4") determines the pixel size and, more importantly, the field of view.  A CCD camera will take advantage of just a small part of your telescope's field of view, so you probably don't want to go too small.

[Andy Mitchell]

One thing to watch out for is when the sizes in given in inches, this is the equivalent diameter of ye olde vacuum tube sensors. These were specified by the diameter of the glass tube, and only part the tube was photosensitive. Typically a 1/4" sensor has about 4mm active diagonal, 1/3" 6mm, 1" 16mm etc. Rather smaller than you expect.

Hardly done any planetary stuff, but I here people are doing good stuff with the Xbox360 web cams which have a small CCD sensor. You remove the case and replace the lens with an adapter, total cost about 20 quid

[JamesF]

The size of the sensor (1/2", 1/3", 1/4") determines the pixel size and, more importantly, the field of view.  A CCD camera will take advantage of just a small part of your telescope's field of view, so you probably don't want to go too small.

The sensor size has nothing to do with the pixel size.  They're completely independent.

James

[JamesF]

One thing to watch out for is when the sizes in given in inches, this is the equivalent diameter of ye olde vacuum tube sensors. These were specified by the diameter of the glass tube, and only part the tube was photosensitive. Typically a 1/4" sensor has about 4mm active diagonal, 1/3" 6mm, 1" 16mm etc. Rather smaller than you expect.

I've read that the sensor diagonal is typically around two thirds of the stated size, give or take a bit depending on the aspect ratio of the sensor.  I think that pretty much ties in with the examples you give.

James

[DaveS]

Got in before me.

Yes this thing come up with monotonous regularity in video / cine forums. I think it's well past time this antediluvian measure was stamped out. The only reason for keeping it is as a marketing weasel measure to bamboozle the unwary. 

[JamesF]

It's pretty bizarre, isn't it?  Marketing departments presumably now have to work out what size tube this sensor would have fitted in were it to be used in a way that is pretty much certain never to happen.

And then you get the really weird sizes, like 1/1.2".  Ok, so I know what it is now, but the first time I saw that I was somewhat confused.

James

[ronin]

And then you get the really weird sizes, like 1/1.2".  Ok, so I know what it is now, but the first time I saw that I was somewhat confused.

Still not sure what they mean by that.

I tend to find out what the sensor is then search out a data sheet that manages to do that strangely bizzare thing of giving the size in mm.

Like who uses measurements like mm ? :grin:

[JamesF]

I tend to find out what the sensor is then search out a data sheet that manages to do that strangely bizzare thing of giving the size in mm.

Like who uses measurements like mm ? :grin:

That does really seem to be the easiest way to do it.  That or multiplying the pixel size up by the resolution.

In fact there's another utterly bonkers thing about it.  Pixel size?  Measured in thousandths of a millimetre.  Linear dimensions of sensor?  Millimetres.  Pretty much anything else to do with the sensor will be given in SI units.  And then the quoted size for sales purposes?  (Sometimes very odd) fractions of an inch.  What?!

James

[pipnina]

The values of 1/3 or 1/4 refer to the size of the sensor -  one third or one quarter of an inch across (about 6-7 mm in new fangled units).  I have an ASi120mm mono camera and it is excellent!  However if you go for mono then you will need a filter wheel and filters if you want to take colour (RGB) images.  A one-shot colour version of the camera is available that allows you to take colour images without the use of filters etc.

This is an image taken with an ASI120mm mono and a 120mm refractor last night when the seeing was not particularly good.  http://stargazerslounge.com/topic/244979-the-moon-24-may-2015-4-pane-mosaic/

Careful! If you start talking about mono vs RGB CCDs Olly will show up with his RGB VS RGB+Ha image again!    (Not that that's a bad thing, he just does it surprisingly often)

[pipnina]

That does really seem to be the easiest way to do it.  That or multiplying the pixel size up by the resolution.

In fact there's another utterly bonkers thing about it.  Pixel size?  Measured in thousandths of a millimetre.  Linear dimensions of sensor?  Millimetres.  Pretty much anything else to do with the sensor will be given in SI units.  And then the quoted size for sales purposes?  (Sometimes very odd) fractions of an inch.  What?!

James

I think it's for the people who... Aren't of the younger generation. (I know my parents still use OZ and LB to measure weights)

[robmichael81]

Thanks for all the advice,

I'm guessing (as a newbie) that the colour ZWO asi120 should do just the job? I know the mono will probably give better results but I have so many limitations in my modest equipment that I think the mono and filter wheel would be far to much at this stage.

Rob

[riklaunim]

You could always use mono with one filter (infrared passing like red/orange visual filter, Baader IR-Pass etc. but keep it bright) and play with Moon and planets in mono only

Jupiter is going away. Saturn is low and will benefit from mono imaging with infrared filter. There is a bit of Venus and will be more by the end of the year and any cloud detail would be in UV or some in IR. Uranus is also nearly only photographed with broad infrared filters to catch cloud bands.

[VikkiFord]

I'm a noobie too and so far these are the best I've captured using a ZWO ASI120MC. I'm sure I can do better but need the shorter days back to carry on playing   

Hope they help. Jupiter was a 130p with 2X Barlow, the others were 200pds with 2X Barlow except the moon which was a 'luck pushing' 3X.

Best regards.

Vikki.

[VikkiFord]

To give you a comparison, this photo was taken with a Canon EOS 1200D in a single shot. It lacks the sharpness of the ZWO for some reason and I suspect depth of field may be the culprit. Or not so good seeing for a single frame. Someone did comment that the ZWO picture did not have the resolution of the Canon picture but the ZWO did produce the, overall, better image. I'm going to try stacking some Canon images next time I get the chance. Come back the longer nights!

Vikki.

[JamesF]

I can't see why the ZWO image should necessarily have less resolution than the 1200D image if the optical systems were otherwise the same.  The ZWO sensor has smaller photosites than the 1200D, so it shouldn't give lower resolution and perhaps more at the same focal length.

One thing that strikes me about the 1200D image is the blurring of the limb in the top right of the image.  I'd guess that might be due to the exposure time, but it could also be the barlow not being sufficiently well-corrected further from the optical axis.  Were I to guess, I think I might plump for the latter explanation.

Stacking frames from the 1200D would definitely help, but life might be easier if you started without a barlow at all.  At the focal length of the 200PDS you should find the entire lunar disc covers at least two thirds of the height of the sensor so you'll get a nice size image and you won't have to fight with the software to make a mosaic as well as all the other processing.

James

[VikkiFord]

He was probably referring to the difference between the 18MP Canon and the 1.2MP ZWO. Personally, I think the ZWO looks great. Really looking forward to seeing what happens when I stack 30 or so Canon frames. Ah! And those pics had pronounced heatey-ripply-thingy distortion too.

Here's a full pic I did with the 1200D. It looks nice but, I don't know, it kind of lacks too. I might be expecting too much of a full shot though.

Mosaic? There's an interesting thought 

Bear in mind I've not had many good nights with this so my learning is being hampered by Mr and Mrs Cloud (and their offspring).

Vikki.

[VikkiFord]

Actually, James, is this boiling down to arcsec per pixel and resolving power of the telescope? I think I'll go with a new thread on that instead of causing a derailling of this one.

Apologies to Rob 

Vikki.

[ecuador]

The sensor size has nothing to do with the pixel size.  They're completely independent.

James

 Of course it does! You divide the length of the sensor by the number of pixels (to get the pixel pitch size to be exact). How else would you calculate it? For example the ZWO asi120mm uses a 1/3" sensor, so it is 4.8mm in width and the horizontal resolution is 1280 pixels, which gives you (4.8mm/1280 =)  3.75μm.

[JamesF]

 Of course it does! You divide the length of the sensor by the number of pixels (to get the pixel pitch size to be exact). How else would you calculate it? For example the ZWO asi120mm uses a 1/3" sensor, so it is 4.8mm in width and the horizontal resolution is 1280 pixels, which gives you (4.8mm/1280 =)  3.75μm.

Then you mean "the sensor size and the number of pixels along each edge determine the pixel size" rather than just "the sensor size determines the pixel size".  There are 1/4" sensors (for exampe) with all sorts of different pixel sizes.

But in fact as the inch measure is only an approximation then it doesn't even reliably tell you the sensor dimensions.  For example, another 1/3" sensor from an astro camera has a horizontal resolution of 1024 pixels, giving a pixel size of 4.8/1024 = 4.69um pixels by your calculation above.  Only they aren't that size at all because the horizontal dimension of the sensor is actually slightly under 4.8mm.  A 1/2" sensor also from an astro camera which allegedly should have a horizontal length of 6.4mm has 1280 pixels horizontal resolution, giving a pixel size of 5um, but actually they're bigger than that because the horizontal length is greater.

There are more factors that mean the measure is unreliable too.

James

[ecuador]

Then you mean "the sensor size and the number of pixels along each edge determine the pixel size" rather than just "the sensor size determines the pixel size".  There are 1/4" sensors (for exampe) with all sorts of different pixel sizes.

But in fact as the inch measure is only an approximation then it doesn't even reliably tell you the sensor dimensions.  For example, another 1/3" sensor from an astro camera has a horizontal resolution of 1024 pixels, giving a pixel size of 4.8/1024 = 4.69um pixels by your calculation above.  Only they aren't that size at all because the horizontal dimension of the sensor is actually slightly under 4.8mm.  A 1/2" sensor also from an astro camera which allegedly should have a horizontal length of 6.4mm has 1280 pixels horizontal resolution, giving a pixel size of 5um, but actually they're bigger than that because the horizontal length is greater.

There are more factors that mean the measure is unreliable too.

James

I don't know why you complicate things. The OP asked about the 1/2", 1/3", 1/4" sensor "sizes" (types actually). So, when you compare two sensors with the same resolution, the pixel size is determined by the sensor size, so you would know a 1/3" VGA sensor has smaller pixels than a 1/2" VGA sensor. Determining EXACTLY how much the difference is, that is of course another story - the inch "types" are good for comparisons but not that good for calculation, as they are meant to be about 1.5x the diagonal of the sensor, but each manufacturer makes a very "loose" interpretation. But while various 1/3" sensors might vary a bit in size, they are all larger than the 1/4" sensors etc.

In any case you can't say "completely independent" when it is the opposite. The only problem is that the dependency is not very straightforward to calculate due to the aforementioned variation between manufacturers.

[JamesF]

I don't know why you complicate things. The OP asked about the 1/2", 1/3", 1/4" sensor "sizes" (types actually). So, when you compare two sensors with the same resolution, the pixel size is determined by the sensor size, so you would know a 1/3" VGA sensor has smaller pixels than a 1/2" VGA sensor.

So now you introduce the fact that the resolution must be the same, something that wasn't mentioned at all in this thread up to and including the post in which you said the inch measure determines the pixel size.  Therefore even if we allow that the sensor type determines exactly the physical dimensions of the sensor, it still doesn't determine the pixel size on its own, only in combination with the resolution, so the statement isn't true.

If "the sensor type determines the pixel size" without further qualification as you said initially then you should, requiring no further information, be able to tell me which of the 1/3" and 1/4" sensor cameras I have on my desk right now has the bigger pixels.

James

[ecuador]

It is sort of like talking to a wall. I said sensor size determines the pixel size and I guess there was an implied parenthesis (for a known resolution). Instead of noting that the pixel size ALSO depends on the resolution, which would have been I guess a useful comment for people who would not know it and perhaps I should have explained it more clearly, you instead responded with the entirely incorrect statement "The sensor size has nothing to do with the pixel size. They're completely independent.", which is what I was trying to correct. I don't see what you are trying to debate, I merely corrected a false statement.

If you still don't get what I am saying perhaps I can formulate it for you: when z=f(x,y), we say z is a function of (depends on) both x and y. z and x are definitely not "completely independent".

[JamesF]

It is sort of like talking to a wall.

You're right. It is.  So I shall post no more on the matter.

James