Anyone have any experience in shooting the sun with a 650 nm filtered IR DSLR plus a solar filter? I'm wondering if any details could be extracted using this body and a solar filter for my lens that I purchased for the eclipse. I realize it's not a narrow pass hydrogen alpha filter but it does block wavelength shorter than 650 mn.
Solar photography with near IR DSLR
Posted 05 September 2017 - 04:35 PM
Can you specify what a "650nm filtered IR DSLR" is?
A daytime modded IR photography camera that blocks 650nm and shorter?
Specific model camera? 650nm is an interesting choice for daytime IR. Astromod camera sometimes have a h-a (656nm) sensitive IR block placed on, so I just want to make sure there is no confusion.
Bottom line: It will work. But not as good as a stock camera. Note below on resolution loss.
I tested this theory real quick with Baader Solar film, a Hoya 720nm IR pass filter and a "full spectrum" DIY modified camera (Sony a5000).
White light still possible with the above.
But FWIW, a "solar filter" ---although it would be nice if you specified,
but the typical "white light" filter
attenuates all wavelengths down to safe levels from the Sun.
The Sun is really bright.
You will pick up the deep-reds allowed past by the 650nm filtering.
If you looked at the transmission line of the actual filter, as the percentage of transmission (from 0) increases, the bright Sun will be beaming in more and more of whatever spectral line it can.
even if the 650nm was a sharp cut-off (0 percent transmission below 650nm), there are many deep-reds above it.
Similar to doing narrowband with ANY COLOR CAMERA,
you will have a resolution loss.
Unless you have light-leak across the pixel color filters on the camera sensor,(camera specific issues),
the red light is only registered by 1/4 of your camera pixels (1 red 2 greens 1 blue for every 4 pixels).....
So you will have loss of resolution compared to someone with the same camera (not IR filtered) doing "white light" with a comparable filter and lens,
since the camera will be registering the attenuated light across the detectable spectrum,
registering the light through the blue and green filters in front of the other 75% of pixels on the camera sensor.
The unknown is the particular pixel filters of your specific model camera and their transmittance of deep-reds and NIR light that the specific solar filter will allow to pass.
You will need to shoot RAW stills and extract the color channel data before the data is demosaiced,debayered/an interpolation algorithm is run to predict the color for each of the pixels.
Not every software can do this.
Then, you can compare the red, blue, and green channels separately and decide if there is any data in the blue and green channels besides noise...
Some cameras (if the deep reds/NIR leaks) may do better in the green channel if you overexpose with your setup,
and then find isolating the green channel (50% of pixels filtered) as the best way to double your potential resolution....
You could also make up for resolution loss by dithering and drizzling techniques. But then again a regular camera maybe easier.
Posted 13 September 2017 - 09:22 PM
Here'a a shot I took with a 300/4 PF Nikkor with a TC-14 E II extender and solar filter (Spectrum Telescope from Highpoint) on a crop sensor modified Nikon body with a built-in 665nm IR filter. It cuts off frequencies below 665mn but still allows longer frequencies unlike a true H alpha filter which has a narrow bandwidth around 656nm. I'm wondering if the "grain" in the image is the solar chromosphere detail or an artifact.
Edited by alwilder, 14 September 2017 - 08:07 AM.
Posted 14 September 2017 - 11:22 AM
I'm wondering if the "grain" in the image is the solar chromosphere detail or an artifact.
You will have a little grain due to only 25% of the pixels registering the best data as described above.
I always like to compare with Solar Dynamic Observatory satellite data sets:
Of course the datasets are from different wavelengths than "white light," but some of the data is pretty similar and can give you a good idea of what you've captured,
as well as orienting the Sun correctly (North pole top of frame)...
I guessed that you took the image around Sept. 8, ~8:15UTC?
"Courtesy of NASA/SDO and the AIA, EVE, and HMI science teams."
Edited by t_image, 14 September 2017 - 11:24 AM.
Posted 14 September 2017 - 06:27 PM
I just ordered a B+W #486 UV/IR cut filter that cuts frequencies beyond about 710nm. Combined with the camera's built-in IR filter that cuts visible frequencies below 650mn, I'm hoping to get something interesting.