Idea for flats - a digital photo frame, will it work?

[Uranium235]

I just thought of a method of simplifying the taking of flats for refractors. A digital photo frame, load it with a white jpeg and there you go! Portable, tough(er) and battery operated. Cheap too.

Its still light isnt it? So it doesnt really matter what frequencies it contains or lacks.

Probably just a daft idea...

Edit: Just checked at Argos, but theyre all mains operated. But a 7" panel for £35? Thats almost worth a punt.

[ollypenrice]

The spectrum can matter. I have tried screens that turned out to give no Ha and others that turned out to give no blue. At the time I was still religiously shooting flats for each filter, so this mattered.

However, if your filters are parfocal and clean (which may mean in an electric wheel) or if you are using OSC, then I doubt that it does matter. I no longer bother with multi filter flats and use Lum flats for everything.

This rather suggests that an incomplete spectrum for your panel would have no discernible effect.

However, if your filters are not fairly parfocal or are not very clean then you'll need individual flats - and for that a complete spectrum from your panel.

Olly

[Uranium235]

Thanks Olly, its just a thought Id had since I did the dreaded reverse polarity on my EL panel inverter, which is now kaputski.

Im a stickler for clean optics, and I like the idea of one flat for all - but im not sure how that would work in narrowband. If I cant find another inverter, I will give the photo frame a go and if it doesnt work, I'll just use it for the family photos.

[ollypenrice]

Thanks Olly, its just a thought Id had since I did the dreaded reverse polarity on my EL panel inverter, which is now kaputski.

Im a stickler for clean optics, and I like the idea of one flat for all - but im not sure how that would work in narrowband. If I cant find another inverter, I will give the photo frame a go and if it doesnt work, I'll just use it for the family photos.

Try it. If there is no shift in focus maybe illumination is illumination regardless of bandpass.

Olly

[kirkster501]

...or you can use an iPad for flats. Works well for small refractors but not big enough for a 8" reflector.

[russellhq]

I know that some light pollution filters create a colour cast to flats which may mean one or 2 of your channels are not getting fully exposed i.e. CLS filters.

I guess you could set your iPad to a pale blue/green/red to compensate?

[IanL]

I have found through three iterations of building a lightbox that spectrum certainly does matter for OSC flats. For the first two attempts I originally tried 'white' LEDs, then I tried tri-colour RGB LEDs (three LEDs in one package). In all cases the light output looked reasonably white to the naked eye.

In both cases I ended up with a red channel that was basically noise and not much else; I could barely even make out the vignette from the FR even at full stretch whereas it was obvious in the green and blue channels. Unsurprisingly, the calibrated and stacked images ended up with noticeably worse noise in the red channel as a result. The problem is that you have to use one exposure length for all three channels and you may find your source dominates in one or two channels at the expense of the third.

For the final iteration, I spent a lot of time looking at manufacturer's data sheets to find a combination of separate R, G and B LEDs that should produce a similar amount of illumination and added some miniature potentiometers to provide some adjustment to each channel. I still end up with a colour cast on the final image (but that is really easy to kill off in PixInsight) and a much more evenly matched noise across the three channels.

There are other ways to skin the cat, obviously if you are shooting mono through filters, you'd just adjust the exposure length of each set of flats to try to get a similar level (though even then it would probably be desirable to have similar and short exposures so you don't have to worry about flat darks). I suppose it might be possible to take three flat exposures for a OSC of different durations and then extract the three channels and recombine them, or maybe just use the least noisy channel as a luminance flat (but I'd assume this would be at half the resolution of the OSC image, don't know what problems that would cause).

For 30 quid it might be worth a punt as it would be a more robust device than an EL panel, and a lot less expensive to drop than an iPad or a laptop screen. Worst case scenario if it wasn't suitable, you could use it to display your magnificent astro-images to visitors instead.

[michaelmorris]

I've had similar issues with colour cast on my homemade light box. which is illuminated with LED lights. I use a Canon DSLR for imaging and Deepskystacker (DSS) for image alignment and calibration. DSS uses the pixel luminance data for flats calibration on each channel seperately. So uneven wavelength outputs in the light box shouldn't be a problem as long as the peak in each channel is between 1/3 and 2/3 of the maximum.

One for DSLR/OSC imaging option might be to take the flats in monochrome, then convert them to RGB format before applying them as flats. However, I'm not sure this is possible as DSS requires that all the elements are in the same format (e.g. 24-bit RAW files) and I.m sure whether I can save an imported CR2 RAW file and retain the format after altering it - any ideas how this could be done?

[Uranium235]

A quick update: Ive now got an 8" photoframe for 35 quid. Its screen is just about big enough to cover the 80ed dew shield. Ive loaded it with a white jpg and it seems ok, apart from the corners where it seems to darken a little.

I havent had chance to test it yet, but if the results are favourable i will post details.

[Uranium235]

dup post sorry

[m37]

I have one of these http://www.ikea.com/gb/en/catalog/products/00202324/ which is now surplus to requirement. I'm thinking put it behind something semi-opaque and dial in the colour you need. The wheel lets you select any tone of rgb colour you want.