Skinwalker Ranch: Laser Changing Color

[Mick West]

In this clip, Travis Taylor describes an experiment done to determine if the color of a laser cannon (a multi-beam laser) shone on a white target is chaning color.

He starts justifying this by referencing a previous "blobl" anomaly in that location, which I'm pretty sure is just an internal camera reflection as shown here.

Source: https://www.youtube.com/watch?v=P_oUEAuK9MM


For testing testing the color, he uses a hobbyist spectrum analyzer by Theremino. This is a combination of a spectrum splitter (a diffraction grating, sometime a bit of a DVD), a web camera, and some software. The hardware is truely DIY, and we don't see it except in near-darkness. It's just a wooden box with a slit at one end and the innards of the camera with a bit of DVD taped to it at the other end.
https://www.theremino.com/wp-content/uploads/files/Theremino_Spectrometer_Construction_ENG.pdf

The diffraction grating takes the light source through the vertical slit and then spreads this out horizontally into the component colors. This is recorded as video with a web camera (a very cheap NexiGo N60, in this case) and the result is fed into the software - something like this:

Here Travis is hand-holding the analyzer hardware (the wooden box) to the eyepiece of a telescope,

pointing at this white board.

He note it "got brighter in the green, and dropped off in the blue". As we go from:

To:

But notice what's missing in the second image? There's no actual spectrum. If you point a spectrum analyzer at a bright light source that's not visibly changing in intensite, then you would expect something like:

But instead, there's just nothing. Not a different arrangement of red, green, and blue, but nothing.
What this means is Travis Taylor is moving the spectrum analyzer hardware so it's not lined up with the telescope eyepiece.
The green never gets brighter. It all drops to almost nothing. There's a massive change in the brightness of everything, so the light coming from the white board is diminished or absent, and other color components take over - i.e. we shift from the bright laser light to very dark ambient light. That very dark light has a tiny bit more blue.

The trick here is if you just look at the graph you don't realize you are measuring darkness. But the key is to look at the spectrum. Here I'll boost the levels a bit so you can see what's happening.

This is when he's pointing it at the board.

We see the full spectrum, blue, green, red, as you would expect.


Compare it to this. Basically darkness, there's a little light coming through only blue makes it to the visible video. He is 100% NOT pointing the analyzer correctly. So he's not measuring anything.

 

[FatPhil]

Compare it to this. Basically darkness, there's a little light coming through only blue makes it to the visible video. He is 100% NOT pointing the analyzer correctly. So he's not measuring anything.
View attachment 69098

100% concur with this analysis. Notice the lack of scale on the lower graph, there's no way of knowing whether anything has gone up or down because that graph auto-scales such that the largest peak is at the maximum height (the ones where this doesn't appear to be the case have some noise right at their extremities which is squashing the middle down). So the only thing we're left with to deduce absolute changes from is the upper spectrum, which says "everything dropped massively", and your conclusion follows immediately.

I might be a bit more charitable to them on this one, it does appear to be just incompetence at gathering useful data - which is harder than it seems, you have to think of all possible way your data might be wrong and eliminate likelyhood for each of those in order to be confident your data is reliable - rather than a deliberate intent to mislead. But then again - why's he holding the sensor? Is a clamp too expensive? I remember taking photos of a lunar eclipse through a telescope, and my "camera mount" was a toilet roll inner cut to a length such that the telescope primary's focal plane would converge approximately where my camera's sensor was, and such that I could manually feel that the two were roughly lined up, as it was all under a hood made of my university gown. Not even any tape: there was no point, I could only move one tripod at a time, and the elevations kept needing to change too, they were getting detached constantly. A way more complicated problem than these ufo experts are trying to solve. Got some photos good enough that the guy at the developing shop complimented me on them. Did I mention that I was poor student at the time? Can we gofundme them a clamp if they're on such a limited budget?

 

[Mendel]

But then again - why's he holding the sensor?

Dramatic necessity. The actors should be seen to do something.

 

[FatPhil]

Dramatic necessity. The actors should be seen to do something.

>BOOM< Headshot!

 

[JMartJr]

why's he holding the sensor? Is a clamp too expensive?

Possibly this: a clamp holding a sensor is not as interesting on television as having one of your characters holding it and pointing it at things is. Using a hand-held sensor shows one of your characters DOING something, the clamp holding it is passive.

it does appear to be just incompetence at gathering useful data - which is harder than it seems, you have to think of all possible way your data might be wrong and eliminate likelyhood for each of those in order to be confident your data is reliable - rather than a deliberate intent to mislead.

Capturing useful data is perhaps not the goal, making good television is.* "Good television" here meaning "attracts viewers and keeps them watching," rather than an artistic of aesthetic judgement. They are not incompetent investigators, so much as they are competent and successful entertainers.



*Especially when capturing good data is likely to show that nothing is happening, which is boring. A TV show about a team of investigators on a ranch where nothing interesting happens and the results of all the tests are normal is not going to have much of an audience, and so is not going to make much money!

 

[Calter]

It's hard, because the way the experiment is carried out screams incompetence, while the format of the show screams intentional manipulation of events, but I would argue that what's happening here is just done intentionally to have content. I don't think they would have set up the lasers and have a whole camera crew filming at night if they thought the experiment had a chance to not yield content (the whole footage would have been useless if they just said "we shined a laser at this board and turns out light is behaving exactly as expected, we'll get them next time").

If they wanted to dramatize the events, while doing a somewhat reasonable experiment, they could have filmed stuff like turning on the lasers or dramatically placing the sensor in the correct position, but as mentioned already, the experiment would always lack the most dramatic point where they actually measure the change they want to see.

If capturing good data were anywhere near the goal then the experiment would have been repeated several times (to make sure it's actually a thing), and if it does happen consistently then it should also have been replicated elsewhere to confirm that what they are observing is an effect caused by what they think it's causing it (if they could replicate it anywhere else then it's obviously not some anomaly in that specific place). It's not a good format for TV but that's the price to pay for gathering good data.

Someone that is an astrophysicist should 100% know that, if you want to know how much something weighs you don't just put it in a scale once and call it a day, you put it several times and get the average weight (even if the 10 times measured the exact same thing every time). If you want to know what temperature something is boiling at you don't just boil it once, you boil it several times while keeping track of the atmospheric pressure. When you want to measure something, you measure it several times if possible (which in this case, it is)

 

[NorCal Dave]

Possibly this: a clamp holding a sensor is not as interesting on television as having one of your characters holding it and pointing it at things is. Using a hand-held sensor shows one of your characters DOING something, the clamp holding it is passive.


Capturing useful data is perhaps not the goal, making good television is.* "Good television" here meaning "attracts viewers and keeps them watching," rather than an artistic of aesthetic judgement. They are not incompetent investigators, so much as they are competent and successful entertainers.



*Especially when capturing good data is likely to show that nothing is happening, which is boring. A TV show about a team of investigators on a ranch where nothing interesting happens and the results of all the tests are normal is not going to have much of an audience, and so is not going to make much money!

Could not agree more! I love that Mick and others can show the nonsense that passes for "science" on this silly show, but in the end it doesn't really matter. It starts as entertainment that along the way convinces people that this stuff is real. Unfortunately.

My brother doesn't really believe in UFOs or Bigfoot, but he loves the show and thinks weird stuff does really happen out there. I was at a public place last week within earshot of another family. One guy was carrying on about how the US Government has been studying UFOs and such for the past 15 years at a ranch in Utah. Weird stuff really happens there. My wife gave me the eyes like "Are you hearing this?" I let it pass.

I love the way this whole laser cannon-wormhole-UFO-hotspot set has the same eerie vibe as the landing pad from CE3K. One would think a wormhole would be just as hole-like during the day, but of course then you can't see fricking sharks with lasers on their heads, and the homage to CE3K puts the viewer in the right frame of mind:

 

[JMartJr]

One would think a wormhole would be just as hole-like during the day, but of course then you can't see fricking sharks with lasers on their heads,

Similarly with "ghost hunter" shows or "bigfoot hunter" shows. If either exist, they exist in the daytime too, But got to shoot at night -- makes it easier to control what the viewer sees, and is SO much more spooky-looking!

 

[Ravi]

Incompetence or deliberately clumsy operation of the hard & software, indeed. If only they had not clicked "use auto normalize" and turned off the scale bar..

 

[FatPhil]

I'll be brief, this post isn't intended to be more than vaguely coherent, it's just a dump of the last 30 minutes of my tuesday afternoon.

I watched this vid (@Mendel - you might like how this ties in with the flame shadow concept you mentioned a day or so back):

Source: https://youtu.be/watch?v=1o8ktldjcog


Aside - 200 squid for a fairly functional spectrometer?!?! @Mick West - tempted by a new toy?

Therein, I noticed it had the same spectrometry software as the SWR wonks. I googled it, it's from here:
https://www.theremino.com/en/downloads/automation
It's a windows exe, but source is available:
https://www.theremino.com/wp-content/uploads/files/Theremino_Spectrometer_V3.1_WithSources.zip

I wondered if it would be easy to add a vertical scale to the graph, so looked through the code. It should be possible:
In Sources/Module_Spectrometer.vb, in Sub AddFilter() does grab the MaxValue (to be found at x=MaxValueX) as it preprocesses the data.

The quickest hack to get the maximum value (in meaningless units) out to the user is to hijack

Sub ProcessCapturedImage()
    ...
    Form_Main.Label_MaxPeak.Text = "Max: " & nm.ToString & " nm"

which could be augmented to include MaxValue as well as nm (which is derived from MaxValueX).

And hey presto - their absolute peak values will be revealed!

There's certainly a way of actually getting them as actual scales on the graph itself, but that might require working out their coordinates and other stuff that's well outside my wheelhouse.

 

[Calter]

Worth noting that the manual does state that it is not useful as a way to measure intensity, so I don't know if there's much value in modifying the code so that it shows a vertical scale.

The weaknesses of a spectrometer based on webcam are:

• Dynamics Limited. You must be familiar with the instrument and use the right amount of light and appropriate exposure settings. If the light and the sensitivity is too high, they appear artifacts that distort the spectrum. If they are too low, the lower rows disappear.
• Strong dependence on the characteristics of the WebCam. With a few WebCam results are much better than with others. Some drivers may be completely unsuitable and even color settings and sensitivity can cause significant variations in the quality of the measurements.
• Accuracy and resolution limited. The resolution of the camera and the non-linearity caused by the deflection grating and by the objective lens, limits the accuracy to about one nanometer.
• Only wavelengths can be measured, not the amount of light. The Webcam (and all cameras) have a response not linear so quantitative measurements are impossible. You can not measure the amount of light, but only appreciate relative and approximated differences.

However, the features are good for teaching and for small laboratories.

I still think the program is intuitive enough that it is fairly obvious that if you aren't seeing the light you are meant to be measuring inside the orange box at the top then you know you aren't recording what you are meant to be recording.

Having played a bit with the program, I'm now curious about the orange bar on the right

I can't find anything about what it represents, but playing around with the exposure it seems to go down with the less light there is, only really going as low as there when it's basically pitch black, which to me is further evidence that they were just recording the ambient light most of the time (though maybe I'm just misunderstanding the bar). I am somewhat confused as to why the bar doesn't jump higher during the brief period it actually records any light, so I'm not 100% sure what the bar actually does, but maybe it just wasn't exposed long enough for it to change much.

Compare this with the Steve Mould's video where the orange bar is full even when just recording the frequency of the sodium streetlight

Source: https://youtu.be/1o8ktldjcog?si=z8zL48jPZVm3cmiG&t=585

Lastly, if anyone recognizes what makes the orange box at the top blink then let me know, as far as I can tell it's also related to how full the orange bar is, the fuller the bar is the less the box blinks.

 

[FatPhil]

Worth noting that the manual does state that it is not useful as a way to measure intensity, so I don't know if there's much value in modifying the code so that it shows a vertical scale.


I still think the program is intuitive enough that it is fairly obvious that if you aren't seeing the light you are meant to be measuring inside the orange box at the top then you know you aren't recording what you are meant to be recording.

Indeed - note my explicit "meaningless units" above. Note, however, that my "value at max" is not far in meaninglessness from the "is a peak"/"is a dip" detection that's already a feature (and even the "max" detection itself). Running up a sequence of slightly increasing readings in a direction that corresponds to greatly increasing sensitivity *isn't* heading towards a real maximum, it's the opposite, but the software will label the local max it finds as a peak. The "feature"'s more to satisfy those who are willing to ignore the image part, and only trust a number. It's reading 14.8, right, that proves there's a ghost - I saw it on a ghost documentary - they showed me the meter, it was 14.8. Proof![*] I was working on the principle that there would be several orders of magnitude separating their two cases, and being out by a bit and a lot and a bit more for bad luck really wouldn't matter. Not a global solution to anything, just a SWR solution.

Having played a bit with the program, I'm now curious about the orange bar on the right

Me too. I can't remember exactly how it behaved now, but if it was at a fixed wavelength, then a wild-arse-stab-in-the-infrared could be that there's some NIR LED on the camera in their enclosure that they forgot to disable/mask/remove.

(A harder cutoff could be created using an IR-pass package/filter in front of it, to deliberately make it invisible (which could be why they overlooked it).

EDIT: image url: https://toshiba.semicon-storage.com.../knowledge/e-learning/discrete/chap5-3_en.png
via: https://toshiba.semicon-storage.com...wledge/e-learning/discrete/chap5/chap5-3.html

For infrared LEDs used in television remote controls etc., GaAs (gallium arsenide) is the material used; for red/green indicator LEDs, GaP or InGaAlP is used; and for blue LED, InGaN or GaN is used.

/EDIT

[* Honestly, I saw that argumentation in a woo-woo clip covered in a debunking show a couple of decades back, and I will never forget that number. My g/f and I to this day have "14.8" as an in-joke for "here's a person who doesn't understand the technical equipment and data he has in front of him, whilst pretending otherwise", or more generally for "presence of data, but absence of information".]

 

[Ravi]

Aside - 200 squid for a fairly functional spectrometer?!?! @Mick West - tempted by a new toy?

That's a neat spectrometer. Can't find any diagrams explaining the internals, but I consider that there is likely an imaging lens inside, imaging the slit (via a grating) on the detector, hence the tube. One can also make one using (curved) mirrors, making it even better at using it at lower and higher wavelength. These mirrors also are not extremely expensive anymore.

 

[Ann K]

Indeed - note my explicit "meaningless units" above.

In the lab we used to call them "Arbs".

 

[Leifer]

Not as dramatic, but I wonder if the laser will slightly change spectrum, when heating up. Many industrial lasers have a cooling system in use, such as a peltier.

 

[Mendel]

Not as dramatic, but I wonder if the laser will slightly change spectrum, when heating up. Many industrial lasers have a cooling system in use, such as a peltier.

Obviously that's not much of a concern in an entertainment laser system, such Skinwalker is using. So it feels possible. The question is, would the size of this effect match the observation?

 

[Ravi]

Not as dramatic, but I wonder if the laser will slightly change spectrum, when heating up. Many industrial lasers have a cooling system in use, such as a peltier.

Of course it is cooled, but it will not matter in the laser colour/wavelength as the laser medium is fundamentally coupled to the laser wavelength. Thus, say the laser head warms up, it will reduce it's efficiency (or stop working), reducing the power, but the wavelength will not change.

 

[FatPhil]

Not as dramatic, but I wonder if the laser will slightly change spectrum, when heating up. Many industrial lasers have a cooling system in use, such as a peltier.

If you're using lasers because you care about the wavelength, then, yes, you care about the temperature. Higher temperatures lower the energy/frequency. To the human eye, the effect would be minor, fractions of a nanometer (relative to hundreds) per degree kelvin, but I'm not the best to evaluate how minor that is, as I have defective colour vision. They also decrease the power of the laser, so you lose efficiency. A more mundane effect is simply that there's enough waste energy that needs removing from the system that if you don't actively cool them, you'll simply unsolder your components, or let out the magic smoke.

EDIT: Sauce for my "fractions" above:
"The wavelength of a laser module shifts about 2 nm with every 10 degree rise in temperature." -- https://www.s-laser.com/info/what-is-the-effect-of-temperature-on-the-wavel-52099560.html
"Without some form of external wavelength stabilization, the higher the laser's temperature becomes, the longer the wavelength becomes. For DFB lasers, for example, this is typically around 0.1nm/°C." -- https://www.arroyoinstruments.com/blog/how-does-temperature-affect-the-wavelengths-of-lasers/