Hi there,
In the near future I'd like to jump into serious photometry and see what I can do with my Fraunhofer. Of course, it won't be as easy with my slow focal ratio and mediocre colour correction as with better equipment, but this is ok to me. For learning and the one or the other useful scientific result on my way to mastering photometry with my equipment, it's perfect. I'm going for UBVRI filters with the V and R filter first. But there are several manufacturers of these out there, Astrodon, Custom Scientific, Baader, FLI, Andover Corp., LOT. Also there are different - let's call them - "interpretations" of the UBVRI passbands, Johnson-Cousins, Bessell, Kron/Cousins. Andover claims that the Kron/Cousins filters are better suited for CCDs than the Johnson/Bessell. What does Kron/Cousins actually mean: does it mean the same as Kron-Cousins (if this "interpretation" exists at all) or are these two separate "interpretations", Kron and Cousins, which are both equal, thus, Kron/Cousins. On Google Images I saw plots showing Johnson/Bessell. Does this mean the same as Kron/Cousins? This all appears that there are the following "interpretations": Johnson-Cousins (as a combination), Johnson, Cousins, Bessell, Kron. Is this correct?
I tried to make sense of it all by looking through some scientific papers. According to Bessell's "UBVRI Passbands" from 1990, it seems that initial confusion was done in the 1950ies by Johnson himself as he did some bewildering airmass extinction conversion to zero airmass that wouldn't be done this way anymore today. Also did he not properly match to his own invented passbands over time. I did not understand that much of the rest of the paper, though, and am still confused by the whole range of "interpretations". But as Bessell points out to some mistakes that others have done in their investigations and can explain them rationally, it seems to me that Bessell did a good job with his passband adjustments and matches to the UBVRI system.
Would it be reasonable to buy from the Bessell UBVRI set? Despite it's better match for photomultiplier tubes?
Thanks and kind regards,
Lou
Hi Lou, Bessell has also been working on the fact that, at Johnson's time, the PMT was used instead the present "photon count" sensors (CCD or CMOS). PMT like used by Johnson, are sensitive to the light flux, the intensity (energy), not to the photon count (but here I should note that there is a particular mode of use of the PMT that also achieve photon count, not used by Johnson). That means the Johnson's filters shall be redefined for the present photovoltaic sensors, this is what Bessell did with his "photonic filters" definition.
I think the said "Bessell" commercial filters presently available follow that "photonic definition" not the Johnson's one and it should be better to use them with CCD or CMOS. But ok, this is not all what defines the full response of the system, each CCD/CMOS has its own response, the optics is also involved and finally the atmosphere too. In any case a color "transformation" of the data is needed to get the correct result.
Clear Skies !
Roger (PROC)
Lou,
The majority of amateur CCD photometrists use Johnson (BV) and Cousins (Rc Ic) filters. The final tuning of the measurements to match the standard system is done by applying transforms. The Johnson U filter is rarely used by amateur CCD photometrists, and even then, only for bright objects.
Johnson/Cousins filters are available as interference filters or as colored glass (Bessell). The interference filters are more popular now for two reasons: 1) They have a better transmission, i.e. you'll get a higher SNR (compared to a colored glass filter) with the same exposure time, and 2) the green glass used in the Bessell V filter is hydroscopic and tends to get hazy, sometimes after only a year or two of use.
Photometric filters are expensive, and a full set (BVRcIc) filter can eat up a significant part of the total budget for a photometric system. The V and B filters are, by far, the most useful. My advice would be to buy interference V and B filters. Later, if you stay interested in photometry, you could get the Ic and maybe the Rc. Forget the U.
Phil
Hi Phil,
To get the historical chronology right, Bessell applied the photomultiplier passbands of Johnson's 1950ies system to the silicon detectors around the 1990ies and used coloured glass - so this would be the original Bessell UBVRI - and a later update to Bessell after the millenium would be the interference Johnson-Cousins filter, who have equal curves as Bessell, but with higher transmission and longer durability. About right?
Concerning your comment that the majority of amateur photometrists use the dielectric coated Johnson-Cousins system UBVRcIc for nowadays silicon detectors, how does this compare to the Andover filters where there is no combination of Johnson and Cousins (Johnson-Cousins), but only the individuals (Johnson OR Cousins, while Johnson being equal to Bessell and Cousins being equal to Kron)? Compared with the Astrodon filters the Andover Johnson/Bessell seem to be the equivalence. But does this mean that generally all Johnson OR Bessell filters from after the millenium are equal to Johnson-Cousins? And why isn't the Kron/Cousins system preferred as Andover and LOT claim them to be more suited for silicon detectors? Sorry, this is all very misleading and I guess I am not the only one with these questions.
Thus, thank you all for your clarifications,
Lou
Lou,
"To get the historical chronology right, Bessell applied the photomultiplier passbands of Johnson's 1950ies system to the silicon detectors around the 1990ies and used coloured glass - so this would be the original Bessell UBVRI - and a later update to Bessell after the millenium would be the interference Johnson-Cousins filter, who have equal curves as Bessell, but with higher transmission and longer durability. About right?"
I think Bessell filters are usually considered to be Johnson-Cousins filters since they were designed to match the original J-C system (bases on photomultipliers) when used with CCD's. So now we have two ways of making Johnson-Cousins filters for CCD's- Bessell (colored glass) and interference filters.
"Concerning your comment that the majority of amateur photometrists use the dielectric coated Johnson-Cousins system UBVRcIc for nowadays silicon detectors"
That's not exactly what I said, but I think this comes from the same misunderstanding. I said that most amateur photometrists use Johnson-Cousins filters. To me, this would include Bessell filters. I'm sure there are plenty of people using Bessell filters. I just think that when they get tired of trying to rub off the haze from their Bessell V for the third time, the next V (or the next set) they get will likely be interference filters.
I can't comment on the Andover or Krohn filters since I am not familiar with them.
Phil
Hello Phil,
The way you (and Roger) explain Bessell's Johnson-Cousins filters as coloured glass filters and nowadays non-Bessellian Johnson-Cousins filters as interference filters, both with a photonic definition (adjusted for silicon sensors), makes much sense. Thank you both very much. It seems however that "Bessell" as a substitute for "coloured glass" and interference filters has began to mix with a filter set example provided by the Baader set, which is advertised as having central wavelengths and transmissions ranges after Bessell and with an AR-coating being dielectric. So these Bessellian filters from Baader are interference Johnson-Cousins filters, as well. At least if I am not wrong with the linguistic assumption that: dielectric = interference.
Lou
Hi Lou,
Yes, I was speaking about the 2011 definition not the 1990 one.
CS
Roger
Andover is not a common vendor in the astronomy photometric filter market. Their descriptions of "Kron-Cousins" and "Johnson-Bessell" don't make sense, and their basic data on the individual filters don't give bandpass shape or even central wavelength. I think you are getting confused by paying attention to this one vendor.
The photometric filters for astronomy are called Johnson-Cousins. These refer to the actual spectral response curves, where you define a central wavelength and how the filter transmits light with respect to wavelength. Johnson had a complete set of UBVRI, developed in the 1950's, and other researchers had competing bandpasses at that time that were similar but slightly different. In particular, the Cousins UBV bandpasses were essentially the same as what Johnson was using, but the RI bandpasses were slightly bluer than Johnsons. Gerry Kron in the north had filters that were very similar to the Cousins filters in the south, and so sometimes in the literature you will see the R,I filter bandpasses described as Kron-Cousins.
As professionals started to use these filters, they settled on using the Johnson bandpasses for UBV, and the Cousins bluer-than-Johnson bandpasses for RI. These decisions were largely based on the availability of colored glasses to form the response shape, and on the availability of standard stars and photometry of other stars in the literature.
So keep in mind: Johnson-Cousins refers to the bandpass shapes, central wavelength of the transmission, etc.
Bessell, on the other hand, was a researcher who combined colored glasses in a "prescription" to arrive at close approximations to the Johnson-Cousins bandpasses. So a "Bessell filter" just means the glass prescription to create a filter with the right bandpass central wavelength and shape to mimic the Johnson-Cousins system. Remember that the real bandpass is a combination of the transmission through the filter as well as the spectral response of the detector, so Bessell made several different prescriptions, depending on whether the final detector was a photomultipler tube or a silicon CCD/diode.
Since Bessell used colored glasses in his 1990's papers, any filter description in the literature or vendor catalogs that states "Bessell filters" means colored glass filters that mimic the Johnson-Cousins passbands. There are a couple of different varieties of these, including ones that put cover glass on both sides of the V filter to reduce/eliminate the hazing that can occur. Also, some of the glasses that Bessell prescribed (and also that Johnson prescribed) are no longer available, and so substitutes have been made. This is especially true for the U filter.
Other vendors start with the Johnson-Cousins bandpasses, and mimic them using dielectric (interference) coatings. As mentioned earlier, these can be very efficient, and not hygroscopic. My preference are the dielectric filters, as they can be tuned to match a CCD response very closely to give very good approximations of the Johnson-Cousins system, and are not substantially more expensive than the colored-glass filters. There are some limitations. For fast optical systems (say, f/3), the incoming light cone will shift the interference pattern and the filter bandpass will change. Interference filters tend to have more out-of-band transmission, and have to be very carefully "blocked" by the vendor.
So I would probably stick with one of the more common vendors, such as Astrodon, Baader or Custom Scientific. Ordering through a camera vendor, such as SBIG or FLI, is another option that gives a similar result. One of the nice things about using filters with refractors is that their (often) poor color correction is irrelevant, if you restrict the bandpass with a filter. You may see a different focus for, say, a blue filter than a red filter, but each filter should be nicely focused without any kind of a color halo. We use both dielectric filters as well as colored glass filters for the AAVSOnet telescopes with good results.
Arne
Hi Arne,
Thank you for posting this filter description. I think it is the clearest I have seen anywhere.
Clear skies,
Dennis
Thank you Arne for this detailed and summarised explanation of the history of the filter development. Indeed, I've never seen such a good explanation up to now!
Concerning coloured glass filters, I came across a few descriptions, from Astrodon among others, who state that coloured glass prevents large halos compared to non-coloured glass filters. Astrodon has improved their non-coloured glass filters for this kind of disturbance. So it seems, at least, that coloured glass filters with up-to-date improvements are as good as improved non-coloured glass filters and you can't go wrong with any of both - not that anyone claimed otherwise. Or are there any voices with different experiences and can't recommend any of the above mentioned filter suppliers?
Lou