spectro-aras.com

Not sure if these should be here as they are a first go at using the SA seriously. I have still some work to do to establish an instrument response and calibration that I am happy with but this is what I have managed to date. The image scale is ~10.5 A/pixel and I have used a linear calibration based on the first order and the Ha emission feature. For some reason I don't seem able to get a good stable calibration from A0V star HIP 12640 but I used it for the instrument response. I suspect I am not yet getting good focus.

I would welcome comments and if the files are of the corrct format and have all the information reguired.

Thanks Andrew

Sorry here they are.

Hi Andrew,

The shape of your DN Tau spectrum is significantly different to mine so there may be a problem with the instrument response. I am using a different star for response correction though. See my post here
viewtopic.php?f=28&t=540
I took a quick look at DN Tau last night and it appeared essentially unchanged from my other measurements (apart from during the event on 20121209) Do you have GSC 01829-00018 in the same field? If so you could measure it and compare it with my measurements of the same star (attached).


Cheers
Robin

Hi Andrew

Robin Leadbeater wrote:
The shape of your DN Tau spectrum is significantly different to mine so there may be a problem with the instrument response.

I have now compared all three spectra with my initial survey spectra
viewtopic.php?f=28&t=502
and it is only DNTau which shows a large difference. Actually the spectrum you posted for DN Tau looks more like an non response corrected spectrum. It could be worth rechecking the data reduction for this spectrum

Cheers
Robin

Thanks Robin - Always not a good idea to differ from the master! I attach my spectra of GSC 1829:18. I would say it is qualitativly similar to the ones you posted in that the general shape is similar. Mine has larger swings in the response. I think I underexposed the DN Tau spectra with the individual images being quite weak. I should have binned 2x2 or increased the exposure.

I tried with and without normalisation and got the same results. I have been processing from about 8 am and it is now 5pm so I will call it a day!!!

I will revert to trying to validate my system and processes but wanted to get a shot at them before they set behind a very large tree in the next door garden! Strange at it may seem I think it is mmuch harded calibrating and getting a good responce curve with teh low res SA rather than say the LISA where the sharper features a simpler to locate and remove the risiduals in teh response curve.

Also do you do flats for the various hour angles you have been using or just one set? I have added a Brightstar rotator to my rig so it is quite simple to change angle from inside in the warm.

Thanks for the comments and help.

Regard Andrew

Please ignore the last zip file - I have been at this too long and have got my stars mixed up - I will post GSC1829-18 and DN Tau both with response corrected tomorrow (I hope).

Regards Andrew

Sorry about the mixup with the files here are DN Tau and GSC 1829-18 with the correct processing (I hope). I think they are a reasonable match with your work Robin although I will continue to work to improve the instrument resonse and calibration.
Robin, would you mind posting you spectra of HD29646 so I can look at your resolution? I would also like to have any tips you might have on focusing.

Thanks Andrew

Hi Andrew,

This version of DN Tau is much closer to mine. Your GSC1829-18 is similar but has a slightly different slope. This could be the result of using a different star for instrument response or different atmospheric extinction. See attached (blue are your spectra, pink are my spectra from 20121211.
Independent of this though, the ratio of DNTau to GSC1829-18 should be repeatable between observers for a measurement taken at the same time. I will reduce my spectra taken on the same night and we can then compare them this way.

My resolution is low at only about 60A FWHM at H beta, partly because of the low dispersion (17.5A/pixel) I decided to sacrifice resolution for SNR for this project because as I understand it, the use of the Star Analyser spectra is to accurately measure the shape of the continuum. Other spectrographs like the LISA/LHIRES are better at measuring the spectrum at higher resolution but with the disadvantage of lower cadence and no simultaneous comparison star measurement. My observation of the fast transient suggest there is a risk of not seeing all the variation unless the cadence is faster than around 2 min per observation.

Cheers
Robin

Thanks for taking the time to look at my work Robin. I am pleased we have a similar result.

The differences are probably due to a number of factors including extinction both in the reference and target stars. Do you compute a response curve for each set of observation or do you stick to the same one, taken say at a low air mass? Choice of reference star might matter depending on how close the star is to the reference (e.g. Pickles) spectra. In any case small offset errors get magnified by the response curve especially at short wavelengths.

Given the wide spectral range I agree with you that the use of a comparison star (differential spectroscopy) in the same field is going to be necessary.

To establish a common approach I would like to see the following specified:
Reference star for instrument response and if this is fixed or done each set of observation
Position Angle (PA) of the spectra wrt the zero order for each field
Comparison star for each target
Maximum exposure length by target “cadence”
Preferred observation pattern (e.g. is it better to follow DN Tau all night or for say an hour with observations of RY RN mixed in.)

I am sure this list is not definitive so please add others.

On resolution I increased my plate scale to 10.5A/pixel to try to balance the Chromatic Coma and the seeing. I have very poor seeing most of the time here in the Cheshire gap (3-6 arc secs) and I estimate R~50.

I am about to lose the targets behind a large tree so unless I get a clear night in the next week I will have to wait for the next season to get any more observations.

Thanks again Andrew

Hi Andrew,

Here are your spectra (blue) compared with mine for the same night (pink) and in December (green)
I noted a significant offset (+50A) in your wavelength calibration relative to mine (I have shifted your spectra to compensate for this in the graphs)
There are differences but they tend to be broadly similar in both DN Tau and GSC1829-18 asmight be expected for differences in atmospheric conditions or in instrument response due to using a different reference.
Remaining differences could be down to variations in the background subtraction for example (In my observations at least, the background is higher than the signal at the extreme ends of the spectra where the differences are greatest)

I suspect we are now exploring the limits of accuracy in terms of producing quantitative data using the Star Analyser

Cheers
Robin