Try increasing gamma if dark sections aren't distinguished

Try increasing gamma if dark sections aren't distinguished

Sunday, November 16, 2014

Photographic Light Pollution Filter Comparison: Hutech IDAS LPS-P2 vs Chroma Loglow

Having recently moved I've noticed some improvement in my skies from bortle white to bortle "off white".  At zenith on a good night, my sky quality meter reads 18.12 mag/arcsec^2.  There is a red glow to the north up to an altitude of 40 degrees and a grey glow above that to 70 degrees.  The sky is darker above and to the south.   The milky way is impossible naked eye.
I do all my imaging from these skies with an LPS filter, narrow band filter, a combination of both.  Emission nebulae are great, galaxies are difficult, but can be very rewarding (especially if there are HII regions),  reflection nebulae are very difficult and dust is impossible.
As a testament to the fact that I purchase too many filters, I was given a Chroma Loglow filter to test.  So I've compared it the the Hutech IDAS LPS-P2 filter which many (including me) consider the gold standard light pollution filter.  A quick visual inspection showed more light passing through the Chroma than the IDAS.  I initially thought the Chroma had more of a reddish cast, but realized this was on a fluorescent bulb.  Searching around for an incandescent bulb yielded similar, more neutral color balance for each with perhaps a slight blue cast to the IDAS.

Iris Nebula, a reflection nebula surrounded by dust shot through heavy light pollution:
Reflection nebulae are usually blue because the scattering is more efficient for blue light than red (this is the same scattering process that gives us blue skies and red sunsets) (Wikipedia).
The black void above, below, right, and left of the nebula is caused by intervening dust.  darker skies allow the glowing dust to be imaged.

Dumbbell Nebula, an emission (planetary) nebula, at maximum elevation with minimal gradients due to light pollution:

Here are the details of the filter test:
It should be noted that the filters were used on different nights, though the conditions were similar.  My first target was M24, low in the south, but in the direction with least light pollution, using an FS 60C at F 4.2 and a Starlight Xpress H9C camera.  The star and background signal was much higher with the Loglow, though I did not note any difference in depth or gradients in this dense star field (comparison images not shown).  The seeing was slightly better when the IDAS was used.  The combined image can be seen here in a prior post.

Dumbbell Nebula
Next up was M27, an excellent target, as it has strong OIII and Ha emissions transiting at zenith in an area where gradients and light pollution are minimized.  Using an FS 102 at F 6, I compared a Baader IR/UV filter to the IDAS and the Loglow.  There were no appreciable light pollution gradients.  The nebular emission signal was enhanced relative to the stars with both the IDAS and the Loglow.  I could not appreciate any significant difference between the IDAS and the Loglow.  I've attempted to minimize processing to color balance and matching digital development:




stack of 9x120 sec exposures, color balanced by eye followed by digital development in maxim.

Iris Nebula

Last was the Iris nebula, a reflection nebula surrounded by dust in the dreaded northern muck, again with the FS 102 at F6.  As there appeared to be a significant color difference in the reflection component, the images were color balanced with x-calibrator.  The reflection component appeared significantly more blue with the IDAS and Loglow compared to the IR/UV filter.  I am not certain whether this is the result of grey light pollution contaminating the IR/UV images or enhanced color by the IDAS and Loglow due to rejection of signal in the middle of the color spectrum.  In any event, I found the effect aesthetically pleasing.




stack of 19x5 minutes color balanced with x-calibrator, ddp in maxim.

An aggressive histogram stretch in maxim showed the worst gradient with the IR/UV and best with Loglow, though the IDAS was close.




I did shoot some luminance with the two filters, but the background signal was so high with 2 minute subs, that it reached the non-linear region of my camera, making flats ineffective.  Between hot spots and dust specks it was difficult to make any comparisons.

I compared the stellar, nebular, and background signal for a representative section of the M27 stacks and found the following:
For all of the quantities, the Lowglow had higher signal than the IDAS; the IR/UV the highest signal of all. 
However, for the ratio of stellar signal to background and nebular signal to background, the IDAS had the highest ratio, followed by the Loglow, and then the IR/UV filter.

M27 background1,2802,2302,890
nebula (color balanced)177927333388
nebula (unbalanced)167227453400

iris background3,5005,3767,257
iris star19,000 21,000 28,000
excalibrator green0.930.991.03
excalibrator blue0.760.811.16

Takahashi Fs 102 @619 mm Hutech IDAS LPS-P2/Choroma LoGlow, Starlight Xpress H9C/H9
M 27 34x2 minutes RGB, IRIS ~100 x 2 min luminance, 84x5 min RGB
9/15-9/24/14 Neport Beach, CA

Bottom Line

The two filters were comparable in terms of reducing light pollution gradients and enhancing nebular emissions.

The LoGlow seemed to be slightly better than the IDAS-P2 in terms of light pollution gradient reduction.

The LoGlow also passed more signal while the IDAS-P2 had a higher signal to background ratio. 

clear skies

Wednesday, November 12, 2014

10/23/14 partial eclipse animation

finally finished the animation of the partial eclipse in hydrogen alpha.
imaged over about two hours.
a "usb fault" caused me to miss the very beginning while trees interfered with the end.
here's the full disk animation (large file) over 2 hours:

though the sun spots were dramatic in white light, they were less prominent in hydrogen alpha at this scale.
other than the moon passing by, there was not a whole lot happening during the time period of the animation at large scale.

Here's the sunspot, AR 2192 in hydrogen alpha:

Here are 150% enlargements of two small sections that showed some activity:

Sunday, November 9, 2014

solar active region AR 12192: the monster revisited

here's a wider view of the monster sunspot grouping i caught just before the recent partial eclipse.
according to one source, this grouping, known as AR 12192 is one of the largest sunspot groupings in recorded history

fortunately, it arrived at a time when i was working on my sunspot technique and had made great strides, imaging structures i'd never been able to catch before.  specifically, the honeycomb-like granular structure on the bright surface and the and the "penumbral filaments" at the edge of the dark spot (best seen around the large spot to the right). 

"Granules on the photosphere of the Sun are caused by convection currents (thermal columns, BĂ©nard cells) of plasma within the Sun's convective zone. The rising part of the granules is located in the center where the plasma is hotter. The outer edge of the granules is darker due to the cooler descending plasma." (Wikipedia)
These convection currents contribute to the sun's magnetic field.

the penumbral filaments at the edge of the dark spot aren't that well understood (at least by me).  they seem to involve an interaction between magnetic fields of different orientations and convective flow.
magneto-convective cell reference

the sun rotates once every 27 days or so
will be interesting to see if this grouping comes back around.  

the key to the improved image was taking a huge number of frames over a short period and discarding all but the very best.  to do this i had to use a different camera with a smaller field of view and faster frame rate.  unfortunately, the field of view was smaller than this large grouping.

compare the detail (especially the fine filaments) on this small high frame-rate field to the right side above:

i tried a mosaic in an attempt to get a wider field and high frame rate, but it came out to too patchy

for reference here's a prior image at the same scale:

IMAGE improvement DETAILS
the following measures gave improved images:
1. switched cameras from DMK 51 to zwo ASI120MC which allows a much higher frame rate.
-DMK is great for wide full disk animations, but limited to 12 fps-i'd misread the framerate as 60 fps-that's the output avi :(
using a partial frame with the zwo got me to an average of 54 fps which overcame the handicap of the color camera
2. much smaller stack: took only the best 20 or so images at the upward curve to the left of registax stackgraph quality indicator.
3. kept the imaging time very short--20 seconds
4. put the solar filter (baader film) on the scope inside and immediately pointed it at the sun when i brought it outside.
-the charcoal black carbon fiber tube heats up very quickly with direct sunlight. rather than insulate, i figured the film would reflect the energy immediately, keeping the tube out of direct light. 

i think the biggest factor was limiting the stack size
which means i can reprocess some of my older images :)

did not recollimate
tried an off-axis mask prior to these changes with no improvement
no change in focus technique

Sunday, October 26, 2014

solar selfie from 10/23/14 eclipse

why call it a silhouette when i can be trendy?
if you look around while the sun is partially eclipsed, you may notice strange things in the shadows: a blurring of the sharp line between light and dark, strange shapes in light.  leaves in the trees may create a pinhole camera effect giving projections of the eclipsed sun on the ground.
i couldn't find an example in the immediate area so i made one myself ;)

I know, you're thinking that's just my first web space casting a shadow.  Here's a more dramatic example i encountered indoors created by holes in the blinds:

Thursday, October 23, 2014

eclipse preview and a monster spot

as usual, the sun has upstaged the moon, producing the largest sunspot folks have seen in years at the same time as the partial solar eclipse.  it's so big folks are simply referring to it as the monster.  larger than jupiter and easily visible without magnification (eclipse glasses only).
here's a section of it:

and here's the full disk in hydrogen alpha mid way through the partial eclipse:

more to follow...

Sunday, October 19, 2014

solar surface animation, eclipse alert 10/23/14

OK first up
there will be a solar eclipse visible from the US on thursday 10/23/14.
needless to say, as it's only a partial eclipse, the sun will still be blindingly bright.
protective eye-wear or projection is a must for viewing.

not much happened in my full disk Ha animations.
close-ups of prominences however gave interesting movements.
so here are some close ups of the disk surface.
not quite as dramatic as a prominence lifting off, but still interesting.
first, here's the full field (click on image for full size):

high contrast grey-scale

here's a circular filament around an active region:

here are what look like classic magnetic field lines from one sun spot to another:

a filament arching off the surface:

eruptions around a sun spot, which, i guess is why they're called "active regions":

imaging details:
7/5/14 newport beach, ca
DMK 51, 2.5x Powermate, Lunt 60 PT B1200
2 hours of imaging at ~12 fps every other minute
best 300 frames every minute
7.8 ms exposure
the images were spoiled by dust on the sensor
finally salvaged them by creating artificial flats
though you can still see some faint shadows of dust spots in the animations
(right side of last)

Monday, October 6, 2014

Saturn 2014

forgot to send out this year's best shot:
getting some nice color in the bands
a hint of the north polar cloud, but no hexagon
can see Cassini's division and the maybe Encke minimum

the images are taken by using a video camera, stacking thousands of short long exposures
allowing processing software to select the images least distorted by seeing
stacking thousands of images in this way gives a much more sharp image

in theory one can upsample the video by 2x to get better magnification
so i decided to do a test, comparing and upsampled image
with an image using a 2.5x barlow to give more magnification
but requiring longer exposure.

which will be better?
larger magnification with longer exposure
or shorter exposure eliminating seeing effects?

2.5x barlow:

upsampled 2x:

compare the Encke minimum and colored bands
close, but i think the barlow wins
though the seeing wasn't really sufficient for either