View Single Post
  #9  
Old 17-02-2013, 10:49 AM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Quote:
Originally Posted by Paul Haese View Post
Hi Ray,

just want to say that an f4 system is not ideal for doing galaxy imaging. It is just too short in focal length. Personally I would forget that and look more at a system around f8. Probably an RC or a CDK is going to serve your purposes best. Use a camera with large wells sensors like the 11000 will give you great stars and allow nice deep imaging.

Anyway just my opinion and good luck with your selection.
thanks very much for the input Paul. I am convinced that the small pixels compensate exactly for the short fl, so will continue down this path - maybe give you the opportunity to say @I told you so@ at a later date -

Quote:
Originally Posted by Poita View Post
10" @ f4 is going to make for pretty small galaxies surrounded by a lot of black.

At 9.25" @ f10 I find I sometimes wish for more mag and always wish for more aperture.


Something like the upcoming 16" RC might be a better choice.
thanks for the comments Peter. I put your system into the spreadsheet and, assuming you use 9 micron pixels, what I am proposing will have almost exactly the same image scale and resolution as you currently use - plus a bit more aperture. Since you feel a need for better resolution, I guess that my assumption of 2 arc sec seeing may be a bit pessimistic and I will investigate the effects of a slightly longer fl.

Quote:
Originally Posted by gregbradley View Post
Thanks for your reply Ray.

Hi Greg - thanks or the continuing discussion

With the 694 which is in regular camera terms a micro four thirds chip so if it were in a DSLR it would be a 2X crop factor compared to a 35mm chip like STL11.

So 1 metre focal length with that chip gives you the equivalent of a 2 metre focal length in terms of field of view or crop factor. 2 metres is a reasonable focal length for most people's seeing and to get a reasinable image scale on the larger galaxies (the smaller ones will still be small).

Yes, this is exactly the basis of the design - a 1m fl f4 system with 4.5 micron pixels works out to be geometrically equivalent to a 2m fl f8 system with 9 micron pixels. Werner posted an interesting image on IIS from a 10 inch f4 scope using one of the configurations I looked at - although the camera was an 8300 and not a 694, it gives some idea of how much resolution and image scale is available with small pixels on an f4 scope http://www.iceinspace.com.au/forum/s...d.php?t=103350

The well depth issue is a bit vague and shows up in some images and not in others. It seems to vary with the setup, the sub length and the focal length.

What seems to happen is a bright star saturates those shallow wells easily. Then as the star has a spread function the next pixels out overflow then the next and then the next. I realise the overflow is taken away. I don't know how perfectly these anti blooming channels work or if the microlenses have light scatter as well.

The other thing that may be at work here is crosstalk between pixels - unfortunately all of these issues are pretty much hidden from anyone buying a camera.

So the effect is on a bright star the number of pixels out from the centre that overflowed is much higher on these small well chips than the massive wells of the 16803 type chips. Hence the bloated looking overexposed stars. A better way of putting it would be these chips overexpose bright stars more easily resulting in nasty looking halo effects like say a Horsehead image with Alnitak in the frame. I assume the dynamic range calculation is simply read noise divided into well depth and the assumption there is it will display a wide dynamic range because the read noise is so low. That's a different concept to overexposure. DSLRs dynamic range varies with ISO. Its usually best at lowest ISO and falls off as you boost the ISO.

My understanding is that overexposure results when you run out of headroom. If you have a lot of noise you need to integrate longer to get a given SNR and you will run out of headroom on bright objects. If you only have a little bit of noise, you can get the given SNR without so much integration time, so are less likely to overexpose stars. the close tie up between noise and well depth is captured in the dynamic range - well depth by itself does not tell the whole story.

Couple that with a fast F4 and you will probably need to do short exposures to prevent blowing out the highlights ie. the bright stars.

I looked more critically at the SX website images - they are all taken with small scopes, so I think that the blooming effects are most likely due to the inherent PSF of a small aperture.

The Bunyip scope - 12 inch (12.5 inch?) Newt has shown tremendous images and tends to support your argument except it is 12.5 inches or so and I would contend 12 inches is more of a sweet spot for most people's imaging situations.



694 versus KAF8300 is a different argument. 694 has smaller pixels, higher QE, slightly lower noise, but even lower full well capacity (the main worry for me - it may pan out to be not a concern though). 694 is smaller than 8300 chip in size (I think).

On paper 694 may be a winner if the small wells turn out to be no worry.

Another point as well. I bet the QE of these chips varies with the angle of the light hitting them. If too sharp an angle performance drops off on most chips heavily. Hence the microlenses on DSLR chips.

could possibly be, but from what I have read, I don't think the light cone angle is much of a problem until you get to about f2

I look forward to your results as it will be an interesting cutting edge system. If you go for 10 inch I would go for super high quality mirrors.
Mark Suchting? Are you making the scope?

I will be modifying a commercial scope - I don't have a machine shop and modifying existing bits is a much more efficient option than trying to build something from scratch. And at my age, I may not have time to wait for bespoke optics - I think twice about green bananas .

I notice the latest incarnation of CDK has 6 or 7 fans. 3 or 4 at the back and 3 at the side (I think just in front of the primary, no doubt to get rid of the thermal layer on the mirror more efficiently. That would be worth looking at.

yes, I noticed that too. I have been toying with the idea of front face cooling on my current 12 inch Newt and it may be an option for the new scope. I recall reading a very interesting article from an old BAA journal on front surface forced air cooling that shows how effective it can be in getting rid of the boundary layer.

Greg.
comments in the above text


Regards ray

Last edited by Shiraz; 17-02-2013 at 11:09 AM.
Reply With Quote