IDAS Type 4 LRGB... are violet stars a good thing?

[Stratis]

Hi chaps,

While waiting for the seemingly-interminable bad weather stretch to clear up, I've spent some time looking into my imaging setup. I'm using a QSI 583 mono CCD, so rely on LRGB for my colour data.

The filter set I currently use is the Astrodon E-series, which is alleged to be colour-balanced specifically towards the 8300 sensor, but I ended up doing a lot more narrowband than broadband and haven't had much opportunity to test that.

I've recently run across a somewhat esoteric filter set from IDAS, which comes with a most unusual feature; the red filter actually permits some very low-wavelength blue light along with the red:

IDAS claim that this gives hot stars a 'violet' colour by boosting the red channel on a hot blue-white star...

My question then, is this a good thing or not?

On the one hand my scientific brain feels this is somehow 'corrupting' my colour data... but on the other hand, it is true that the human visual system perceives the upper blue frequencies as purple, not the pure blue that a standard B filter would give you.

I'd appreciate the insight of my elders on this one, if anyone has any experience with them

[MartinB]

That's interesting.  I suppose you could experiment by adding a bit of red to a blue channel and seeing what impact that had on the blue stars.  Have you got any colour data to play with?

[Stratis]

That's interesting.  I suppose you could experiment by adding a bit of red to a blue channel and seeing what impact that had on the blue stars.  Have you got any colour data to play with?

I thought of having a go at that, but if you look at the response curve, only the very highest blue frequencies are permitted by the red filter, so simply adding blue to red won't work, as that would effectively make *all* blue stars purple rather than those with a bias towards the upper blue/ultraviolet range.

I guess that's the problem they're trying to address, a standard B filter treats all 'blue' light the same, and thus cannot actually make that jump to violet that the rods and cones in the human eye actually show us. 

[MartinB]

I thought of having a go at that, but if you look at the response curve, only the very highest blue frequencies are permitted by the red filter, so simply adding blue to red won't work, as that would effectively make *all* blue stars purple rather than those with a bias towards the upper blue/ultraviolet range.

I had a try and it soon dawned on me that it wouldn't work for exactly that reason. 

[riklaunim]

I have those filters, mainly for IR false color images. Adding blue light to red filter is very risky as atmospheric dispersion or chromatic aberration may spoil the image.

[ollypenrice]

From the graphs the unusual red filter doesn't go any deeper into the violet than the clear or the blue filters so I'd have expected well corrected optics to cope with it in the same way that they would L and B, no?

I think my fear would be that the stars might look a tad magenta, and just not what we're used to. Magenta stars shout 'Wrong!' to me.

There are stars with blackbody curves peaking outside the visible blue but that is not what these filters are about. I won't be buying any myself just now but if there are any images to ponder I'll ponder!

Olly