Fomalhaut 8/9/2010

sheeny (Al) · #1 09-09-2010, 08:00 PM

Clear night here last night, but Eta Carina was too low in the sky, so I had a play with Fomalhaut trying to get to grips with exposures on the ST8.

I took several spectra of Fomalhaut at differing exposures. I think this is from a single 20s exposure (I know... I should write these things down

...). It's the longest exposure without clipping.

C11, SA100 Grism, 0.5 FR, ST8i.

Play was declared shortly afterwards due to camera driver gremlins streaking on the field...

Al.

Heian (Mark) · #2 09-09-2010, 08:27 PM

Nice one Al,
this seem a silly question, but why do you have to take a 20sec exposure on such a bright star? I'm assuming it's software settings

I know with my dsi3 + 120mm refractor exposure times for me are in the order of 1 sec, or shorter.

cheers

Mark

sheeny (Al) · #3 09-09-2010, 09:36 PM

Quote:

Originally Posted by Heian

Nice one Al,
this seem a silly question, but why do you have to take a 20sec exposure on such a bright star? I'm assuming it's software settings

I know with my dsi3 + 120mm refractor exposure times for me are in the order of 1 sec, or shorter.

cheers

Mark

Just trying to come to grips with exposures on the ST8, Mark. I've been trying to maximise S/N ratio, and trying to get a handle on what is the right level of exposure. The histogram when capturing in CCDSoft doesn't appear to be linear

... and the maximum value achieved is well short of what I expect.

The 20s exposure peaked at about 27000 after I subtracted the background, but the spectrum was the right shape, longer exposures were clearly clipped (at 32k intensity) - the spectrum had a flat top.

At this stage I don't think I can use the histogram to judge my exposures like I'm used to with the DMK.

I have to admit not having my head properly around what's happening with the histogram yet...

I've attached some raw spectra: 5s, 20s and 60s. The 5s spectra shows the noise, the 20s one has less noise, but the 60s one is a different shape due to clipping or the non-linearity of the ABG chip (unfortunately my ST8 is ABG

).

Al.

Terry B · #4 09-09-2010, 09:40 PM

Excellent work.

Merlin66 (Ken) · #5 09-09-2010, 10:30 PM

Al,
Good to see more SA100 spectra!!
Stacking multiple 10sec or 20sec will dramatically improve the SNR.
Good practise for the Spectra-L200....

You'll soon be able to compare results a 1A resolution....
Ken

Heian (Mark) · #6 10-09-2010, 01:06 PM

Al,
I'm unsure if my method gives me the best possible result, but with the DSI3, I will take 40 or 50 x 1 sec exposures, let the meade software automatically stack them and use that as the image for further processing. I also adjust the threshold level for the software to accept an image to add to the stack, upto 80% value.
The dodgy images are just discarded automatically.

Seems to work, and the "noise" , as shown by the jagged line in your 5s exposure, is avoided.

Maybe I've just had some good luck with software selection

cheers
Mark

sheeny (Al) · #7 10-09-2010, 05:04 PM

For the SA, I would say your method is probably the way to go, Mark, because you can presumably get sharper images by stacking the short exposures (each one less affected by seeing). Even video stacked in Registax or Avistack I think is the way to go with the SA.

You just don't get the opportunity to reject any bad seeing in a long exposure.

My aim wasn't so much to get a super spectrum of Fomalhaut but to experiment with exposures in the ST8, because I suspect it isn't linear being an ABG chip. In the back of my mind, too, as Ken alluded to, I will soon have a working L200 which will require guiding on the slit, and relatively long exposures so I was doing a bit of guiding practice at the same time.

Al.

Merlin66 (Ken) · #8 10-09-2010, 05:13 PM

Torsten Hansen is achieving results with his SA100/ DMK31 very close to the theoretical limits. Stacking an AVI certainly pays off.