Camera Labs

Thanks for the info Bob
As you can see, I'm still on the thread and still keen in Astrophotography.

I did a pixel peep and noticed some red dots having 'trail' or 'flarring' ,
Could this have been due to the earths' rotation does not sync with the stars rotation ?
What about clouds , would these affects the shot ? Did your test on 120sec, involves clear skies ? or these are no-issues ? In your earlier thread you mentioned about outside air temp being cold, but the sensor temp could have heated up hence some colour shift in the shots, so effectively if one lives in tropical areas where the avg temp for night is 28-30 Deg Cel, does it have any effects on the capturing , ie to say, I presume the sensor would definately be more 'hotter' , and thus the colour would be more 'warmer' ?

DavidL

_________________
Nikon D7000 and a bunch of Nikkors

Hi DavidL,

Clouds weren't around that night and pixel peeping the stars in the centre of the image shows that the mounting did a good job tracking the stars. I think the main effect you are seeing is in each of the corners where the brighter stars look a little like fat "T"s with the tail of the T pointing away from the centre of the frame. This is due to aberrations in the lens which are pretty much to be expected at f/1.2.

As for the sensor temperature, and assuming I am right in thinking that the colour shift seen is due to heating, I think that night-time temperatures in the tropics don't rule out DSLR use at all but may mean that if you are going to try a sequence of 2 minute exposures then it might be prudent to leave a couple of minutes between each exposure to allow the camera to settle down. On the positive side that leaves a little more time for star-gazing.

Living as far south as you do, you at least have an opportunity to see the galactic centre at this time of year. That sight is permanently denied to those of us who live so far north.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Interesting question - long answer! As you know CMOS sensors, such as are used in my 40D, are supposed to be immune to the "pixel bleeding" that some CCD sensors suffer from along their "rows".

This phenomenon of "square stars" affects the brighter stars in my images and can be seen in both the processed RAW and original JPEG files. This can't be put down to atmospheric "seeing" as that would tend to create round stars. One of the nice features of the IRIS software is that you can read the camera's RAW files using 1x1 binning. Here is a much magnified portion from the centre of one of the original images.



The biggest difference between this and the processed images is the way the processing has converted the square stars into squares with longer diagonal "extensions". As for why the bright stars are square this seems to be referred to in the trade as "crosstalk". Here is an example of the way sensors are constructed, courtesy of Fuji Corporation.



If some light is scattered by each microlens then it can find its way across to other microlenses and hence to the adjacent pixels. My take on this is that the proportion of scattering from microlens to microlens adjacent in the horizontal, vertical and diagonal directions is such that the resultant "shape" is likely to be square. If the incident light isn't white then not all of the sensors in the Bayer matrix will respond to this scattering and that is nicely shown in the largish bright star in the upper right corner of my image above with its lattice like structure. Of course, once the image is de-mosaiced this lattice structure disappears.

As with so much information on the Net, figuring out the correct search string is key. Once I realised that doing a Google search for "microlens crosstalk" was the way to go things became much clearer. In particular, I like the article Crosstalk challenges CMOS sensor design. This also addresses the issue of electrical crosstalk.

Gordon (and Thomas), here's a question for you. Does this crosstalk and the resultant loss of spatial resolution around bright objects affect the way you test lenses?

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Nice article but no explanation, why crosstalk would have a square proliferation pattern.
But apart from the geometric shape of crosstalk, it certainly influences resolution tests, as they are built on measuring contrast. So "bleeding" from bright areas to dark zones in a resolution-chart will reduce contrast and thus resolution.
On the other hand, if you know the parameters of cross-talk for a certain sensor you could compensate for it. And if you only subtract a little from cells surrounding a very bright photo-well. That is how every contrast-improving algorithm works.

What astonishes me more in the large 1x1 binning pic is that many small stars seem to leave a diagonal imprint: one bright pix lower/left and one bright pix upper/right). That a tracking error of the telescope-mount?

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

Hi Thomas,

Thanks for your feedback regarding lens testing and contrast enhancement. As for why the shape goes square (as opposed to cross shaped for example) I don't really know. The reason I included the illustration of the sensor is that it doesn't "feel unreasonable" that light can reach diagonally placed microlenses not only directly from the central microlens but also indirectly from the horizontally and vertically adjacent microlenses. The proportions obviously determine the exact shape.

Yes, I noticed the trailing you refer to running along the 45°/225° diagonal. Between exposures I stopped the mounting's drive motor for a few seconds each time and I can see the shift in stellar positions between each image. That orientation actually runs roughly along the 135°/315° diagonal. That means there was probably a misalignment of the polar axis, a task I have to do before each session, rather than a rate error in the speed at which the polar axis was rotating.

Assume the polar axis is misaligned by 1°. If I have my mental picture correct that means an image shift of 2° in 12 hours or about 1/3rd of a minute of arc every 2 minutes. From the picture the observed 2 or 3 pixels on a 40D sensor with an 85mm lens implies an error of about 1 minute of arc per 2 minute exposure if my sums are right. That implies a polar axis misalignment of about 3° which is disappointing as I would have hoped to be within about half that value. It ain't easy!

The misalignment is a shame because although this is only seen by pixel peepers it does imply a slight loss of sensitivity to the faintest stars.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Hi Bob, yep, any sensor artefacts like these will ultimately impact lens results, although probably only at the very finest degree. I'm happy with our methodology though as I've always said we're testing the system - ie lens AND body - as oppose to the lens alone. I realise this means the tests are therefore only 100% applicable to the actual body used, but I believe there's sufficient material there which can be applied to other bodies to make a useful decision. And I do often return to some lenses and retest with higher resolution bodies when they become available.

As for the star images, that's an interesting explanation... but it's still annoying they have this square look. I was a little disappointed with my comet shots which looked more electronic than I'd like... At first I thought it was because of a single long exposure and that stacking several shorter exposures might help, but now I'm not so sure...

Hey maybe digital isn't the way forward after all! Back to hyper-sensitising film in the freezer!

Gordon

PS - only kidding - film ain't the future for me anymore! Although I do know an aurora photographer down here who swears by it for what he claims is a much higher sensitivity to red wavelengths...

That's why you can buy the EOS 30D with the IR filter removed. It would be great to see someone modify the 40D (with its remote live view functionality) by removing the filter and replacing it with optical glass. On the face of it that would be expensive but maybe not so bad compared to dedicated CCD cameras with the same pixel count.

I wonder if that square look is correctable with dedicated astro processing software. Should be possible but close double stars may confuse the issue.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Normally with a rectangular geometry of the photo-sites the directly adjacent cells should get the highest crosstalk and the diagonally adjoining cells only less. So normally you'd expect a circular proliferation pattern of crosstalk...

Btw. If you want to test crosstalk with a stationary subject, take an LED of say 3mm diameter and shoot it with a 24mm lens from 24m distance. That should just cover one photo-cell

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

But if it isn't crosstalk (optical and/or electrical) what is it?

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Bob,

Wonderful photos! I have always had a lot of respect for astrophotographers. It looks like it takes an amazing amount of technical knowledge, practice and patience to walk away with just one image. Great to see the Milkyway is still there!

_________________
Leica M9, Nikon D300s, Leica X1, Canon S95, Zero Image 6x9

Hi Steve,

Thanks for the encouragement. However, I am all too aware that I am just a beginner in the field and that there is a lot of improvement still needed. Maybe this thread will help others avoid some of the prattfalls!

I saw a web page recently where the author stated that the three things needed for successful astrophotography were:

1. Patience
2. Experience
3. Patience

Couldn't have put it better.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Hi folks,

Another cold, crisp and moonless night in the early hours this morning so I thought I would use the remote Live View facility of my EOS 40D to see what a really crisp focus of my EF 70-200mm f/4L IS USM lens could achieve.

The set-up is as described before in this thread with the camera piggy-backed on top of the telescope and the Hutech sodium street light filter in place. The following shot is made with a stack of three frames, each of 60 seconds duration at 200mm and f/4 with a similarly exposed dark frame subtracted from each image. The exposure time of 60 seconds is relatively short for capturing any nebulosity, especially at f/4, but I wanted to minimise any tracking errors so that I could see how sharp the lens was at 200mm. Consequently I dialled ISO 3200 into the 40D. Here's the result after post-processing in IRIS and noise reduction in Photoshop. The colours have been significantly enhanced.



During processing I chose to leave some of the background sensor noise visible in order to extract as much detail as I could from the Orion Nebula. The bright star at the bottom left is the left-most star in Orion's belt (Northern Hemisphere viewpoint) and below it, if you click on the thumbnail above and download the full size image, there are signs of further faint nebulosity.

As hoped, the lens passed the test well with no significant aberrations.

Bob.

EDIT: I have also posted a 100% crop from this shot here.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Another great shot Bob. If I may ask, do you live within or away from the city? I live in the city and its impossible to see any stars because of the air pollution and all the lights reflecting off the clouds. When there are enough low clouds, I've seen the sky a dull grey-orange and at 9pm at night! I've only been able to see stars in Malaysia on the resort islands.

_________________
Graham
Flickr

Hi Graham,

I live in a large village so while I don't have to contend with city lights I do have to contend with the local sodium street lighting, two examples of which are only about 100 metres away. That's why I use a Hutech LPS-P2-FF filter. It's not a perfect solution but it does make a very significant difference.

The rather vain attempt by the industrialised nations to illuminate the Universe (which is very big) is the subject of a campaign by the International Dark-Sky Association as well as a number of national pressure groups. The really stupid thing is that the authorities, and hence the tax payers, could save significantly on electricity charges if they installed better directed lighting.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Hi Bob, another greart shot! And another example illustrating the virtues of the 70-200mm f4 - it really is a quality lens...