Thought expt - correcting filter transmission with camera sensitivity and light intensity

[JCDowdy]

You are welcome as always.

 

That link is from NREL = National Renewable Energy Laboratory, of one of the ASTM standard solar spectra. The CIE (Commission Internationale de l'Eclairage), or International Commission on Illumination en anglais, has published several different air mass standard solar spectra. The AM0 is outside the atmosphere and the one I called most appropriate is the AM1.5 or 1.5 atmosphere which is a good general reference solar angle for temperate latitudes. If one is inside the tropics then the AM1 might be better but only around noon during summer.

[Cadmium]

Sorry if I sounded underwhelmed with the topic subject, which is a very interesting idea, and definitely would apply between the Bayer and non Bayer camera conversion.

Your topic is very interesting, and my two points (U-340 4mm, and lens transmission) were just two tangents that don't really apply that much to your point.

 

:)

[JMC]

Jonathan, could you attempt an effective transmission chart for some camera+lens+filter + UV-Flash?

 

Ok, this proved more complex than I thought it would - most likely a result of the morning coffees not being fully absorbed yet....

 

I have an ACS UV flash (EOS 600RT). As with the camera they do not share what filter they use to modify the flash, although they do say it is different to the one used for the cameras. Previously I have measured output from the flash, and I'm using that data here. I took the results from 300nm to 400nm and normalized it to the highest output, and re-plotted it to give a graph which shows output from 0-1 between 300nm and 400nm;

 

This looks different to the sunlight graph, as it starts relatively low at 400nm, before rising up and then dropping back down again at short wavelengths.

 

I took this data, and multiplied it by a normalized correction factor for the sensor, and the filter transmission graphs, to get the following;

 

They do look a little different to the sunlight corrected ones, but it's quite a subtle difference - the peak contribution is shifted towards slightly shorter wavelengths, and below 340nm it drops rapidly. The Baader U now has a 'flatter response' between 380nm and 360nm. I must admit I though there would be bigger differences in the final graphs, but looking at the sunlight and flash graphs I can see why they look like they do.

 

Of course I have not considered IR contributions as mentioned before, and there is an IR output from the ACS flash. Also I would not compare absolute numbers from the flash and sunlight - with it being normalized it does not take into account light source intensity, just relative contribution.

[JMC]

Sorry if I sounded underwhelmed with the topic subject, which is a very interesting idea, and definitely would apply between the Bayer and non Bayer camera conversion.

Your topic is very interesting, and my two points (U-340 4mm, and lens transmission) were just two tangents that don't really apply that much to your point.

 

:)

No problem Steve. I didn't think you sounded underwhelmed. I'd love to test with and without Bayer filter. I hope to get an Eos 5DSR converted to multispectral, but retaining the Bayer filter, at some point. Once I do I'll be able to compare them.

[Andrea B.]

ACS annoys me with their lack of disclosure. Clearly they are skewing results towards the 380 nm region. I'd rather have no internal filtration or clear glass and use it with external filtration rather than be constrained like that with no information about what bandwidth I was shooting in!! Why do they do that? Besides which, all it takes is one enterprising soul to pop that filter out and measure its transmission and spill the big "secret". There should be NO secrecy in these matters. [rantlet over]

[Cadmium]

The exposure results Bob has been getting with his monochrome converted UV camera are impressive, demonstrating the increased sensitivity of a non Bayer camera.

Regardless of the price of such a conversion, I wonder how many people would want to lose the Bayer UV false colors most of the time.

Personally, I like the false colors, but it is a headache to deal with in many situations, wind...movement... and high ISO just doesn't do it for me.

Also, it seems like Bob's monochrome UV shots, even as artistic and wall hang-able as they are, don't seem to get too many comments,

and my guess is that monochrome UV just doesn't grab people quite the same as the false color version,

or perhaps some people don't even recognize exactly what it is they are looking at, we are so use to seeing false colors in UV.

[OlDoinyo]

This may seem an old-fashioned question, but could curves such as these be constructed using film rather than digital sensors? I would think film still has better bandwidth below 330 nm, although optics to explore that notion are not easy to come by.

[Cadmium]

I don't know about film, but I think Bob was able to do hand held UV movies with his monochrome UV conversion.

[OlDoinyo]

I don't doubt that he could do movies, but in which wavelength band, exactly?