Time for some new kit

[Mike]

I'm selling my EQ6 and SXV-H9. Have decided to get a H-EQ5 as they have impressed me. I think they perform better than the EQ6's from what I've seen. I will also be getting a DSLR, probably the 1000D. I will then have a set-up that is more portable and practical and also one-shot colour so less messing around. Hoping to have it all in time for Kelling Heath.

[MarkS]

Cool!  DSLR imaging is a great way to get multi-megapixel imaging at a reasonable price.

Mark

[RobertM]

And a whole new learning curve...

[Mike]

And a whole new learning curve...

True, but it's been so long since I did any serious imaging that I would have had to re-learn things anyway.

Anyone know if the 1000D is a good choice of DSLR? Any other recommendations?

[mickw]

You'll probably need this at some stage -

http://www.ghonis2.ho8.com/rebelmod450d1.html

[RobertM]

And a whole new learning curve...

True, but it's been so long since I did any serious imaging that I would have had to re-learn things anyway.

Anyone know if the 1000D is a good choice of DSLR? Any other recommendations?

Depends on your budget.  EOS1000D body only is approx £210 on EBay whereas 450D ~ £310, 550D varies a lot ~£410+ and if you really want to push the boat out there is the EOS 5D full frame camera...

[Mike]

Yes but any ideas on which of the Canons would be most suited for astrophotography? Bearing in mind I have an ED80 and ZS66.

[MarkS]

Mike,

First of all get the proper SkyWatcher flattener for the ED80 - the William Optics ones aren't such a good match.

As for the camera, do you intend to use it solely for astro? Do you intend to mod it?  Will you mod it yourself - if so stick to the cheaper end of the scale - you don't want to wreck an expensive camera (speak to Mick - he has experience!).

Some things to consider:
EOS 400D and 1000D are only 12bit
450D, 500D and 550D are 14 bit
none of the above suffer from the amp-glow which plagues the 300D and 350D

Regarding Quantum Efficiency http://www.sensorgen.info/ has some useful info and so does this thread http://forums.dpreview.com/forums/readflat.asp?forum=1019&thread=34516718&page=1

The figures are "reverse engineered" from data provided by http://www.dxomark.com/index.php/Camera-Sensor/All-tested-sensors using the methodology documented here: http://www.sensorgen.info/Calculations.html

Summarising some QE figures:
350D  27%  - this is the camera I am still using
1000D 32%
400D  is probably similar to the 1000D
450D  30%
500D  36%
550D  40%

The figures above have been calculated on a consistent basis so they give a good indication in the gradual improvement in QE.

I used to think that small pixel sizes invariably meant noisier images but I am gradually coming to the conclusion that small pixel sizes are not an issue as long as QE does not suffer and as long as read noise reduces in proportion to the area of the pixel.

I think the important criteria are the following:
1) Read noise (measured in electrons, not ADUs)
2) Dark current - especially over a long duration imaging session of hours
3) bit depth (14bit is preferable to 12 - it means there's less need to bracket exposures to prevent the highlights saturating)
4) Overall QE
5) QE for H-alpha wavelength in particular
6) other useful features such as liveview (which helps focusing) and such as shooting video

The problem is that no-one publishes all these figures.

For the 1000D, 450D, 500D, 550D Gary Honis has done some experiments of how noise increases over a long imaging session: http://ghonis2.ho8.com/rebelmod550duncooledcomparison.html
So I reckon the jury is still out on the 500D/550D - their noise seems to climb faster and higher.

I think I would go for the 1000D (fairly cheap) or the 450D (14 bit) if I hadn't already bought a cheap second-hand 400D off eBay.  This 400D will shortly become my "workhorse" camera - my tests show it has a better overall QE than the 350D; much lower dark current (after 2 hours imaging) with no amp glow; the read noise and H-alpha sensitivity is only marginally worse than the 350D.  It would be good to thoroughly test the 1000D and 450D at some point also.

Hope this helps.

Mark

[MarkS]

I'm not sure I've previously published the results of my 3 way test between the 300D, 350D and 400D - so here they are - all at ISO 800:



The 3 camera simultaneous thermal test was done at room temperature 21C - I took a continuous sequence of 5min frames over a period of 2 hours - though the battery expired on the 300D partway through.  The graph scale shows electrons/pixel/second against frame number.



One other thing - when taking a sequence of images with the 400D (and probably other cameras as well) it is very important to switch off the rear LCD during exposures otherwise it heats up the CCD and the dark currrent goes through the roof!

Mark

[Mike]

Thanks Mark. This further leans me towards my choice of the 1000D as the 550D was the other option. The 1000D certainly seems to be one of the less noiser cameras at higher temperatures and it's a decent price.

[RobertM]

Mike,

The 550D has a lot of video circuitry on the sensor so that may well explain the extra noise to a degree.

Although ideally you want a camera with low read noise and high dynamic range you can work around that.  Use low iso (higher read noise) where LP is bad (you can't use low read noise to advantage) and use higher iso (and lower read noise) for dark sky sites where read noise might be a contributing factor.  If Nikon hadn't buggered up their D7000 by monitoring the shutter with an IR lamp it would have been the best camera by miles (ultra low read noise through the range up to max ISO).

Hopefully I'll have time tonight to get on with the 1000D/450D comparison and that should give you a bit more information.

Robert

[Carole]

it is very important to switch off the rear LCD during exposures otherwise it heats up the CCD and the dark currrent goes through the roof!

Hmmm didn't know about that, maybe that's why I have such a lot of noise in my images.  Must find out how to do it.

Carole

[MarkS]

it is very important to switch off the rear LCD during exposures otherwise it heats up the CCD and the dark currrent goes through the roof!

Hmmm didn't know about that, maybe that's why I have such a lot of noise in my images.  Must find out how to do it.

I only accidentally found this out by experiment.  I questioned why the noise was rising so fast from frame to frame.   I let the camera cool then repeated the expt with the LCD off and it made a huge difference.

Mark

[Carole]

I questioned why the noise was rising so fast from frame to frame.   I let the camera cool then repeated the expt with the LCD off and it made a huge difference.

Yes I noticed that the warmer the camera got the paler some of the images got too and did the same thing on a long run, shut it down to give it a break.  I was told to switch off live view because of noise, but I thought that was live view on the laptop and always switch that off now, but hadn't thought about the LCD screen on the actual camera which I don't really need).  

Sorry we've highjacked your thread a bit Mike, but it's all useful info for you if you're going over to DSLR. 

Good tip.

Carole

[Mike]

H-EQ5
QHY5
Canon D1000

All ordered and on the way