Focal ratio

[Capricorn]

The Keck telescope is a 10M morror with a focal length of 17.5 M giving it an f1.75...now whats the arcseconds per pixel on that?

Cannot be answered, it depends on the pixel size.

If a meter across one answer, if a mm another, if 0,1mm another...... :D

Strange statement:rolleyes::rolleyes:

Search for Keck and imaging, they probably specify it somewhere.

Are you counting both operating as one scope giving an effective diameter of 85 meter and what is the wavelength being sampled.

[Capricorn]

Organization California Association for Research in Astronomy

Location Mauna Kea, Hawai'i, USA

Coordinates 19°49′35″N 155°28′27″W / 19.82639°N 155.47417°W / 19.82639; -155.47417

Altitude 4,145 meters

Wavelength Optical, near-infrared

Built Keck I 1993, Keck II 1996

Telescope style Reflector

Diameter 10 m (33 ft) each

Angular resolution 0.04 to 0.4 arcseconds for individual telescopes, depending on target and instruments used

Collecting area 76 m2 [1] each

Focal length 17.5 m (f/1.75)

Mounting Alt/az

Dome Spherical

Website W. M. Keck Observatory

Angular resolution isn't as small as I would have thought.

Won't bother buying one.:D:D:D:D

[ncjunk]

Cannot be answered, it depends on the pixel size.

If a meter across one answer, if a mm another, if 0,1mm another...... :D

Strange statement:rolleyes::rolleyes:

Search for Keck and imaging, they probably specify it somewhere.

Are you counting both operating as one scope giving an effective diameter of 85 meter and what is the wavelength being sampled.

Oh dont be picky! Most of the ccds we all use are between 5.4 and 8...thingies...dont have the right symbol..

So 5 or 8 would do, the diference wouldnt be too big between them.

I think its 0.00942 arc seconds per pixel for a 285ish chip.

Thats a great focal length! Now what about a focal reducer? What is the lowest f number you could achieve...i did start on focal length...moving slightly off topic now.

I assume the 0.04 angular resolution is the achievable resolution with ideal seeing?

Nope, its the maximum resolution of the optics..mirror...based on the properties of light, size of the mirror and light path. Sub 0.1 is achievable on 1.2 meters and higher.

[the_laughing_crow]

So could some one tell me about the geometry of the "light rays" reflected from the secondary towards the eyepiece?

Are the convergent, divergent or parallel?

[ncjunk]

Dont think its quite that simple...here is the wiki link for refractors.

http://en.m.wikipedia.org/wiki/Refracting_telescope

You can see on one design lens is convergent and eye piece is divergent.

Thats a lot of reading for a simple question! Looking at a few more links..

[the_laughing_crow]

At the minute I'm really just trying to get my head around focal ratio in a reflecting telescope.... and I think a key to my understanding of different focal ratios is understanding how light is reflected from the secondary towards the eyepiece.

Once Ive got that nut well and truly cracked I'll move on to making sure I understand refracting scopes

[ncjunk]

Well here is the reflector link

http://en.m.wikipedia.org/wiki/Reflector_telescope

A standard newtonian uses a parabolic primary and a flat secondary. This gives a similar light path to the refractor, in a way.

[Steve 1962]

Neil

Seeing as you asked - The Keck would give 0.06 arcsecs per pixel on an Atik 383L! The field of view (according Ron Wodaski's CCD Calculator) is 2.7 x 3.6 arc minutes with the same camera.

As others have said - the scale of the Keck is all about resolution and image brightness (as well as magnification)

I can't see that anyone has mentioned this directly - but (ignoring barlows etc.) the way I understand this is that cameras effectively have a fixed focal length, so the longer the focal length of a scope, the more the image is magnified for any given camera.

This is because magnification = focal length of scope/focal length of eyepiece or camera

However, because the size of the aperture or primary mirror of any given scope is fixed - the amount of light entering the scope is fixed - so the longer the focal length, for any given aperture and any given camera, the bigger, but dimmer the image will get.

This means that a 200mm f10 Newtonian will provide more magnified, but dimmer images than a 200mm f5 Newt for the same camera.

In summary - for a given aperture and camera -

longer focal length = higher f ratio = higher magnification and dimmer image (good for lunar and planets)

shorter focal length = lower f ratio = lower magnification but brighter image (good for wide fields and dim DSOs)

Take Cover! Uncoming!

[Steve 1962]

Sorry the Keck thing had already been answered while I was typing.

If the secondary mirror of a Newt is flat, then the light rays that enter the scope parallel until they hit the primary must then travel all the way from the primary to the secondary and onto the eyepiece / camera in converging paths - otherwise you'd never achieve any focus without an eyepiece (as is the case with a camera at prime focus).

Duck!

[FraserClarke]

How can size of image = (focal length) ^2

Does a cow at 10 feet away make an image the same size as a ladybird at 10 feet?

Because that is what the equation says!

The point is to demonstrate the important scaling relationships. I missed out the constants (the intrinsic size of the cow and ladybird) which give you an absolute image size. For the same object, the size of the image depends only on focal length.

Image size depends on the square of the focal length, because images are, by definition, areal features, not linear features. If you magnify an image by 2x, the area it covers if 4x larger... it depends on the square of the focal length.

[FraserClarke]

The Keck telescope is a 10M morror with a focal length of 17.5 M giving it an f1.75...now whats the arcseconds per pixel on that?

That's the primary mirror only, and there is no prime-focus on Keck. The f/ratio at the Cassegrain focal stations if f/15. Focal length there is 150meters, so the plate scale is 1.3"/mm. Professional CCDs typically have 15-micron pixels, so you'd get 0.02"/pixel; if you just put the detector in the telescope focal plane.

That's pretty useless, because the image quality at Keck is about 0.5" on a good night so you're massively oversampling the image. So, professional telescopes basically never put the detector in the telescope focal plane. They use instruments to re-image the light (and pass it through filters etc) onto the detector. These bring the f/ratio back down again, typically trying to get the pixel scale to be a good match to the image quality. So, if you want 0.2"/pixel on Keck, you need to make your instrument change the f/15 telescope beam down to f/1.5. And to do that you need to pay a lot of money to a good optics designer and a good optics manufacturer, because f/1.5 optics are really hard to make well!!

That brings up the point that Capricorn made about putting eyepieces/barlows etc into the system, and ruining the simple f/ratio arguments for telescopes. The thing here is that the extra optics changes the f/ratio of the SYSTEM, and it is the f/ratio of the system that matters... If you put a 4x barlow into your system, you've changed for f/5 telescope into an f/20 telescope, and the image will be 16x fainter (because the effective focal length is now 4x longer).

There is no difference between telescopes and cameras -- they are fundamentally the same thing.

[haitch]

I've just read this thread and have got to grips with most of it (I tend to understand scopes by mentally drawing the light path in my head).

The one thing is has really taught me is that I made the right decision to stick with visual observing - I do this to relax!

Have fun wrestling with your CCDs/DSLRs/Webcams

[FraserClarke]

At the minute I'm really just trying to get my head around focal ratio in a reflecting telescope.... and I think a key to my understanding of different focal ratios is understanding how light is reflected from the secondary towards the eyepiece.

Maybe the links provided already gave you enough to answer your questoin? But just in case, attached is a ray diagram of a basic newtonian telescope (top). You can see how the parallel rays from the star (on the left) come in to the telescope, bounce off the parabolic mirror on the right and start converging to a focus. The flat mirror in the middle of the tube brings the rays out to the side so that you can access them more easily (the idea being that the flat mirror is less of an obstruction than you sticking your head in front of the telescope!).

The two drawings below show telescopes with the same aperture, but different f/ratios (f/5 and f/10), and how (and hopefully why) the different focal length affects the magnification (scale) of the image. The blue and green rays represent light from two stars.

Telescopes.pdf

[dph1nm]

Surely for a scope of a set aperture the same number of photons are being reflected at the same rate from the primary onto the secondary regardless of how far away the secondary is.

To go back to the physics of this, the answer is it really doesn't matter what focal length or ratio you have, you still get the same number of photons from a given object for a given aperture.

For imaging it really make no sense to worry about focal length, what matters is the pixel scale in arcseconds on the detector. True, you can alter this by changing the focal length, but you can also use a detector with a different physical pixel size, or bin up the existing pixels.

All this talk of faster imaging at shorter focal lengths is a bit of a con anyway - all you are really doing is throwing away information (resolution in this case) and claiming your image looks better.

NigelM

[the_laughing_crow]

I completely stopped following this thread by accident.

Many thanks to all the very helpful posts. My understanding has come on leaps and bounds. But the most recent post really cracked it for me, it put the whole thing into the perspective which I had in my head anyway and helped my translate that into why some people consider faster or slower scope better for different purposes.

So a special thanks to you dph1nm