iraf.net

I'm sorry to ask again, but this time I'm really confused.I want to do an accurate sky subtraction and used fxcor and specshift to cross correlate my 14 sky spectra and then shift them. Now I wan't to combine them to an average sky spectrum. For this I thought scombine might be the perfect routine since it really combines the wavelengths and not the pixels.But when I apply the routine my output looks very weird. The sigma rejecting methods don't work. I still have cosmics in my spectra and the weirdest thing is that it is to...I don't know how to say but it is to small. I'm trying to send a picture to show what I mean. I tryed almost everything and I thought it does basicly the same as imcombine and im combine works fine even without sigma algorythms. [img:3d899bbba4]http://polly.wuestenfuchs.eu/wp-content/uploads/2010/02/sgi10776.gif[/img:3d899bbba4]This image (I'm sorry that it is so big) shows how the combined spectrum looks in comparison to one of the single sky spectra. The tiny one is the combined!
The command I used is: scombine sky_*.fits sky_2.fits group="all" first+ combine="average"
[img:3d899bbba4]http://polly.wuestenfuchs.eu/wp-content/uploads/2010/02/sgi10800.gif[/img:3d899bbba4]And here is what I get with the avsigclip and lsigma = 3, hsigma = 3. May be I just don't understand that algorithm.Thank you.
Nora

Hi,I know it has been some time since you posted your question. But if you would like some help you could give me some data you have problems with and the parameters and command you type. To do this put your images in the anonymous ftp area at iraf.noao.edu/pub. Be sure to post that you did that so I know to look for it. You can post parameters, the output of "dpar <task>", in the message or put text files in the ftp area. If you have gotten past this problem that is fine.Frank Valdes