A while back there was some discussion about using ND filters for bright star photometry. Reservations were expressed concerning the non-uniform attenuation of light across the spectrum.
I've used a Hoya ND4 filter on a few nights to date. The setup is a ZWO ASI294MM camera, 200mm Canon lens, and V filter. The most recent time series, taken last night, comprised 400 images over 7 hours of the southern EW eclipsing binary NT Aps, with an amplitude 8.2 - 8.7 V (VSX).
The filter allowed me to perform photometry on sharply focussed 60 second exposures. The mount (Star Advenurer Pro) was guided in RA only, but I suspect there is differential flexure, since the field drifted a fair bit across the night, so I'm not sure guiding is an advantage. The mount is a star tracker, designed to drive a DSLR camera for wide-field astrophotography. I've loaded it up with an astro camera, filter wheel, heavy camera lens, guide scope, guide camera, and a bunch of cables.
The link below should take you to the curve produced by AstroimageJ, which plots normalised relative flux.
In my experience with this setup, the precision in time series is 'good' if I can use 'longer' exposures, say 60 sec as in this example.
Roy
When I tried to access the light curve via the link, I had to login.
Sorry, looks like the link may not work for readers of this post.
It's not critical. Just wanted to let people know that the ND filter works for me.
Roy
…
Maybe this link will work:
NT Aps 240523 AIJ Jight Curve.png
Roy
The check star chosen for the photometry of NT Aps was HD 127635, V 8.401, B-V 0.365, at 14 37 39.83 -75 32 58.61. Note that NT Aps is not in the AID.
The light curve of HD 127635 (from my photometry last night) shows obvious delta Scuti variability with an amplitude of about 0.06 V mag.
A search of the GCVS and the VSX at the position of this star reveals no 8th magnitude variable. The VSX lists two nearby EW type eclipsing binaries from ASASSN-V, of 13th and 14th magnitudes.
A TESS QLP light curve is that of a delta Scuti pulsator with multiple frequencies, and a maximum amplitude of around 0.08 mag.
The periodicity evident in a VStar plot of the TESS data is about 0.16 d. I haven't yet done a proper period analysis.
Roy
Correction. When the VSX is searched at the position of NT Aps, the star _is_ listed and has an AUID.
Roy
Nice light curve, Roy!
ND filters work fine for photometry - BSM_NH2 has a couple in its dual filter wheel setup, if anyone wants to experiment. The main issue is in whether the ND filter modifies the sandwiched photometric filter passband. If it does, you have to obtain transformation coefficients for the combination. Since the reason for using the ND filter is to decrease the throughput, you then have to take significantly longer exposures of a standard field to obtain those coefficients. It is also useful to see if the ND filter is uniform across its surface area. One way of testing this is to raster a bright star across your field of view on a very clear night and see if the star stays the same brightness.
Arne
Arne,
You wrote:
"It is also useful to see if the ND filter is uniform across its surface area. One way of testing this is to raster a bright star across your field of view on a very clear night and see if the star stays the same brightness."
I'm trying to work out how you might do that. If by 'brightness' you mean measured magnitude by differential photometry, target and comp stars would need to have a small separation, but then both would be effected by changes in transmission across the filter.
If instead you mean raw ADUs, how would you 'subtract' the effect of any vignetting to isolate the effect of varying transmission?
If images are callibrated, I would have thought flat fielding should eliminate any problem with varying transmission across the filter. I think of the elimination of vignetting by flat fielding as an analogy.
Roy