Thursday, February 11, 2016

The Spectre of the iPhone Spectroscope

UPDATE: Feb 27th 2016 I have got them calibrated now. The iPhone6 version calibration is still a bit rough. Lining up on the Helium Lines seems to produce the best results. The camera response is markedly different for the iPhone over the Cannon 550D. Now that I know what we are looking at here, next time I image I can make sure I get the image in a bit better focus. I have also found a way to reduce camera shake by delaying the shutter open until 3 secs after you press the shutter button. Whilst you can see visually the same lines, when extracted using Visual Spec the camera response seems a bit different.

UPDATE: Feb 22nd 2016 Well here is the comparison - again I am still very new at this and am still learning how to properly calibrate the specturm.

Colour, the final frontier. Of course if you want to be a "real astronomer" the visible spectrum is only a small part of the story. Much of our knowledge of astronomy comes from spectra and radio astronomy.

At the 2014 NACAA Conference I was drooling over Ken Harrison's amazing Spectroscopes after his talk and made the comment, it would seem so unfair to buy one off the shelf without going through the pain [Learning Experience] of trying to build one myself. So true to my word I had a go at the basic process of "getting colour".

I wasn't too concerned at this stage about quality, just some rational experimentation with the process to gain a deeper understanding. So I went off to the local hardware and bought a few bits and pieces and broke up a cd (carefully) and peeled the film back. At this stage it was unclear to me whether Barry Manilow CDs or Justin Beiber CDs would provide the best high or low resolution diffraction grating. Carefully pulling apart a three blade razor and using two of them to create a slit, challenged my dexterity. After tinkering away for a couple of sessions over a few weeks, I managed to come up with this monstrosity. Splits the light, can see colour, but probably absolutely useless for real science, but fun to play with and demonstrates the principle of a spectroscope well.

The first place to start this journey is with Ken Harrison's book Astronomical Spectroscopy for Amateurs. It is well written but looks quite daunting when you first pick it up. However if you work through it bit by bit, its a great book with everything you need to know.

Feeling a little more confident, I invested in the next logical step - a Star Analyser 100. This is the entry level standard for beginners, and produces fine results as we'll see.

After passing it around the family to have a look at the cool effects looking at the ceiling lights, I lost the plot and departed from Ken's careful, meticulously presented steps, with the outrageous thought - I wonder how this baby would go on an iPhone! I am constantly amazing at STEM events and star parties how the "younglings" immediately are so amazed by what they see through the eyepiece they want to whip out the smartphone and take an image home with them.

Surely it couldn't possibly work. After all the iPhone 6 sensor is only 4.8mm by 3.6mm, it has a focal length of 29mm and is f2.2. But its an 8M pixel camera (said my evil twin subconscious), your Fingerlakes PL11000M is only 11M pixels so its only 3000 less pixels, how bad could it be....hang-on whats the pixel size ...ah 1.5um versus 9um, interesting. So the sensor is 3264 x 2448, interesting ..... iPhones do take good photos......on a sunny day.....not in the night sky. This went on for a while!

So in the end there was nothing else to do but try it, and learn from your spectacular mistakes!

The next problem was how do I keep the shutter of the iPhone open for long enough to take star photos on a guided telescope - this is not the moon, a great iPhone target normally, but with the standard phone settings thats all the iPhone seemed good for. So channeling my teenagers I thought, there must be an app for that. To my amazement I could not only find one, there was a choice of several apps and 645 Pro could even do it with an Kodak Ektrachrome 64 film "feel to it". NOTE: to those born after the 90's, can you imagine only getting 36 images on one roll of film and not being able "to delete the bad ones" until after you had paid $25 and sent it off to the developers and had it returned to your letter box. In those days the lens ONLY pointed away from you - THE HORROR! Anyway I digress. 645 Pro basically turns your iPhone into a simulated DSLR and enters the workflow of the photography before any JPEG compression. You can set ISO and shutter speed, bracket exposures do all sorts of things that you can do on a DSLR.

With my trusty new app, my camera adapter, my star analyser, a 25mm eyepiece, I was ready for action.

I started on Canopus, after the bright star align was completed, but was more interested in Betelgeuse. After slewing and removing the eyepiece and inserting my newly built contraption, I was sure it was the world's first ultra-low resolution Spectroscope. A quick google search showed I was in fact two years behind the times. However to my amazement I had colour and with some detail, I got "lines" as well. I messed around trying different settings and moved over to Betelgeuse and took some more and put the 2 x barlow in front of the Cannon 550D and tried that for comparison purposes. (See image - top of page)

As an experienced "normal" (although I know you are wondering by now how "normal" that is) astrophotographer I was keen not to overexpose the image - Hmmmm - I have no idea what the well depth of an iPhone sensor is, lets just take as many as we can and see how we go.

I must say the results amazed me. Punching the air in victory, my evil sub-conscious dredged up an "I am invincible" [from the Bond film Golden Eye].

By now it was approaching midnight, and quietly tiptoeing around my back yard I packed up the telescope and headed inside. Damn ... I was so excited, I forgot to change the camera settings to save in dRAW/TIFF - back to the drawing board! Well at least I have some nice completely useless JPGs, but what an exercise that was, one I know I will be able to use again and again.

Friday, February 5, 2016

Certain of uncertainty for near earth asteroids

In my continuing efforts to de-mystify the art of Asteroid Astrometry, I thought I'd follow up last week's article on about 2016 BE with a deeper examination of the Uncertainty parameter when its listed in orbital elements.

This week there is some attention on 2013 TX68 which will possibly make a record close pass of 11,000 klms or possibly be 40 times further away than the moon on March the 5th. I can see that puzzled look on your face ;-) 2013 TX68 is also a Virtual Impactor in 2017, a term we discussed last week.

So firstly lets get some perspective on this uncertainty thing.

2011 CF66 was also listed as a virtual impactor for Feb 2nd 2016, it didn't hit us, no-one was worried if it would, and in fact no-one has any idea where it actually is. It is only a tiny asteroid about 3-9m in diameter and wouldn't have done any damage even if it did. In fact there are 20 other virtual impactors listed in the Risk Table this year, the next one might approach on Feb 18th, is 2009 VZ39, and is slightly smaller than 2013 TX68. 2009 VZ39 is also in the daytime sky and not observable for follow up and further confirmation. I only highlight this to emphasise the point here - all asteroids once they are discovered need to be tracked for sometime, to improve the precision of the orbit before any pronouncements about where they are going to be at a certain point in time. The difference between 2013 TX68 and 2009 VZ39 for example is that 2013 TX68 was observed for 31 data positions (astrometry) over 3 nights where as 2009 VZ39 was observed on only one night with 8 astrometric data positions. If you look at the orbital elements for 2013 TX68 the uncertainty parameter is listed as 7, where as for 2009 VZ39 there is not even enough data to start that calculation. For 2011 CF66, there is a 1.1e-8 chance of a collision between 2016 and 2114, so its mathematically possible, but highly unlikely.

NASA/JPL produced this nice graph with it's press release this week which illustrates the point well.

Image Credit: P. Chodas (NASA/JPL)

What you see is a graphical representation of the "error bars" or the zone of uncertainty, based on the orbit elements that we currently know. This asteroid will be picked up again in future surveys and the zone of uncertainty will reduce further.

Uncertainty Parameter is quite a complex calculation, but it essentially always starts off being a "9" and reduces over ensuing months as more data is collected. Uncertainty is a table of the "Runoff" of arcsecs per decade and Level 7 just means essentially there will be less than 33,121 arcsecs of "runoff" over the next decade. You can think of this as being: in a decade the place to look will be 33,121 arcsecs bigger than the range of uncertainty we need to factor in when we look for it now.

Many of the "click-bait" bloggers and conspiracy theory followers regularly confuse Uncertainty with the Torino Scale - they shouldn't, as even WIKIpedia has a great explanation of the Uncertainty factor, however they google "Asteroid rating 7" and get a hit on what the Torino Scale is and confuse the two, without doing any further investigation. The Torino scale is a risk weighted table and ALL current Virtual Impactors are listed as Torino Level 0. The virtual impactors that are listed have "potential" collisions over a range of years and are only there because we largely don't have enough data yet to remove them.

Asteroid Apophis (99942) 2004 MN4 briefly shot up the Torino Scale to a record high Level 4, before subsequent radar imaging and 4452 observations over 10 oppositions reduced its Uncertainty Parameter to 0 and its Torino Level to 0, with an asterisk that it needs to be carefully watched. Asteroid Apophis will be a naked eye object on April 13th 2029 when it makes a very close but HIGHLY CERTAIN pass of the earth. This level of certainty is only refined by many hundreds of hours of dedicated work from professional and amateur astronomers.

Lastly, Uncertainty should not be confused with don't know, don't care, have no idea what we are doing. Its just a case of not YET having enough data, more needs to be collected.


Custom Search