Symbiotics at the UV end

[Peter Somogyi]

Hello,

Let me publish a few result examples for UV end (defining a new method for myself), with my best care of response (2 ref stars where appropriate - averaging where needed), using Vega from the calspec library for IR curve + of course local ref.stars in the near. (This means ref.stars + target stars all everything divided by a Vega IR = instrumental response corrector, but used only a low order spline later "Automatic 2".) Using Vega was not extremely necessary, but I can see bad effects (minor ripples, UV end deviation) that it has removed.
I was doing exposures up to the level H-alpha allowed (ideally I'd overexpose even H-beta, but seeing no common procedure sending spectra in this way, and just overexposing H-alpha don't seem to be of much improvement).

Let me show a few interesting results:

asdb_v694mon_20170331_841.png (7.78 KiB) Viewed 3536 times

In the above case, there's a sudden cut at Balmer end. It's not because the object was low (23 degree in my case), that drop is visible on the spectrum end itself visually! (I can see people cutting it here, I'd suggest not to do so even when noisy.)
I would also introduce the practice to send in some noise in the end till meaningful, even if speculative... it's needed at least for SNR estimate, and the definite end is hard to specify. Here in this case, for U band evaluation this hard cut at Balmer end should be something important...

AG Dra is something reproducing here to me, now showing a bit more (including noise as a watermark):

asdb_agdra_20170331_984.png (6.97 KiB) Viewed 3536 times

For fullness, a few main targets:

asdb_tcrb_20170401_018.png (6.75 KiB) Viewed 3536 times

BF Cyg got quite bluish, and using Vega (+ ref star: 1 Sge) did make the continuum much more straight till blue end:

asdb_bfcyg_20170401_133.png (7 KiB) Viewed 3536 times

However, below 3650A I'm also uncertain, due to high gradients arise on the IR curve.

Cheers,
Peter

PS: all was made with a 300/1200 Newton + Alpy 600 + Atik 414 EXm + internal calibration frames.

[Francois Teyssier]

Nice to see the Balmer jump!

François

[Andrew Smith]

Glad you got a satisfactory method worked out Peter, which is giving good results.

Regards Andrew