SX Spectrometer; any opinions?

[rl]

Hello to all

I've had some fun over the last 10 years with a CCDSPEC spectrometer looking at  novae/ supernovae but would like to upgrade to something with more resolution (R = 400 for the CCDSPEC) which would allow looking at double star orbital velocities. The obvious way to upgrade is Shelyak...they're pretty much the industry standard and the modular construction does allow some flexibility in terms of facilities and cost.

While at Astrofest last weekend I noticed Starlight Xpress are producing a spectroscope. Superficially, at least. it is quite a clever compact design that integrates a medium resolution spectrometer with an Argon/ Neon calibration source, Flat Field illumination, variable slit width, electronic camera focusing. the dispersion element is a toric concave grating which in effect combines dispersion with focusing the spectrum image. The final image is about 31mm long covering near-UV to near-IR. This is longer than most CCD chips so the camera port is motor-driven allowing travel up and down the image for selection of the wavelengths of interest. The camera port has a coarse adjustment lockable with a grub screw. The fine adjustment is motor driven.

https://www.sxccd.com/wp-content/files/Handbook-for-the-SX-Spectrograph-PRO.pdf

As far as I can see the only disadvantage (other than price) is that the grating is fixed which fixes the resolution at 1500 - 2000 which is a useful step up on the CCDSPEC. It is claimed that the optics correct for some off-axis aberrations which are noticeable on the CCDSPEC.

The basic frame is machined from solid to maintain stability. 

A guide camera is included which works a little unusually; there is a 90:10 beamsplitter cube which allows the guide camera to acquire the target. Once on the chip, you move the target to a calibrated pixel by tweaking PHD2, wherupon you should be on the slit, but the slit is not visible in the guide camera. I guess the beamsplitter inserts a 3-magnitude penalty for your target aquisition but that's probably not that much worse than a poorly reflective slit mirror. FOV might be an issue if your mount is not the most accurate; the FOV is fixed by the size of the Lodestar chip. One very nice feature of my old CCDSPEC is a transfer lens on the guide camera port which halves the scale and doubles the field; a separate guide cam also leaves you with a choice of chip size.

The optical design has been around since 2016 so there must be a few out there by now, but the latest incarnation with everything USB-controllable looks really neat and is designed for remote operation which might be an important consideration for me in the future. It strikes me as a well-thought-out box of tricks that should be really easy to use provided getting the target on the slit isn't an issue.

The handbook does not quote the best input f-ratio but it is apparently at its best at f/7 and will work to about f/4. Ideally I'd use it with my 8" f/4.5 or 12" f/4 Newtonians

The price is about £4k.

Anyone using one out there prepared to share their experiences? Good or bad? How does it compare to the Shelyak options?

Thanks in advance, RL

 

[robin_astro]

It had been around a long time now but like you, I am still waiting to see any serious results from one. An all reflective design has significant advantages.

The things that put me off is the obvious astigmatism inherent in this type of design which broadens the spectrum which is not good for SNR with faint objects and for looking for structure in extended objects eg comets and galaxies. This is also present to a lesser extent in the Shelyak (Christian Buil's) UVEX reflective grating design where it is tamed by using a cylindrical prism.  I also much prefer the WYSIWYG mirror slit (which both Baader and Shelyak and the old SBig SGS use) over  the beam splitter which is highly dependent on precise alignment and focus between guide camera and slit without being able to see directly what is actually happening at the slit.  If that is out even by a pixel you can lose a lot of light without even being aware of it.  The built in guide camera design could also make the instrument become obsolete, rather like the SGS that used SBig's two chip guider/imager cameras.

Cheers

Robin

Edited February 5 by robin_astro