Starman and the Falcon Heavy

Not exactly. The sun doesn't 'heat up and evaporate' the water, the zero air pressure causes the water to violently boil off into a gas. The gas then instantly freezes into chrystals, which is called Desublimation:

Definition: Desublimation is the phase change from gas to solid.
Content from External Source

So... does water Freeze or Boil in space?

But when you put liquid water in space — where it can no longer remain as a liquid — which one of these two things happens? Does it freeze or boil?

The surprising answer is it does both: first it boils and then it freezes! We know this because this is what used to happen when astronauts felt the call of nature while in space. According to the astronauts who’ve seen it for themselves:

When the astronauts take a leak while on a mission and expel the result into space, it boils violently. The vapor then passes immediately into the solid state (a process known as desublimation), and you end up with a cloud of very fine crystals of frozen urine.

There’s a compelling physical reason for this: the high specific heat of water.

Image credit: ChemistryLand, via
It’s incredibly difficult to change the temperature of water rapidly, because even though the temperature gradient is huge between the water and interstellar space, water holds heat incredibly well. Furthermore, because of surface tension, water tends to remain in spherical shapes in space (as you saw above), which actually minimize the amount of surface area it has to exchange heat with its subzero environment. So the freezing process would be incredibly slow, unless there were some way to expose every water molecule individually to the vacuum of space itself.

But there’s no such constraint on the pressure; it’s effectively zero outside of the water, and so the boiling can take place immediately, plunging the water into its gaseous (water vapor) phase!
Content from External Source

Ah, I see my mistake. Thank's for this, this is extremely informative
Not exactly. The sun doesn't 'heat up and evaporate' the water, the zero air pressure causes the water to violently boil off into a gas. The gas then instantly freezes into chrystals, which is called Desublimation:

The phase diagram for water shows if it wants to be a solid, liquid, or gas for any combination of temperature and pressure. Water starts at one point (say 37°C, 1 bar) for body temperature water at one atmosphere pressure (when it's liquid). Then it goes to essentially zero pressure, and a very low equilibrium temperature (when it's solid)

To get from from one point to another it has to follow a line. You can see it either goes from liquid to solid, or liquid to vapor to solid, depending on what decreases most rapidly, temperature or pressure.
Metabunk 2018-02-12 20-05-32.jpg
Youtuber Astronomy Live spotted Starman using 0.7 meter telescope on According to his calculations, the distance to Starman on the image was about 1.8 times the distance to the Moon.

Spotted Elon's Tesla and Starman in Deep Space!.jpg


Hey sweet, that's me! I actually over-killed it by going with the 0.7 meter, but I wanted to maximize my chances of detecting it. Yesterday I shot it again with a 0.5 meter scope at the same site when it was over 5 LD away from earth according to the orbit calculations, which included my data from the video as well as observations from other observatories.
It's at about magnitude 19 right now, which is still within the reach of larger amateur telescopes like this. Unfortunately today it's cloudy at Siding Spring, but here's a short two image gif from yesterday:
Get your last shots in now while the getting's good. It won't be coming this close to earth anytime in at least the next 40 years (as far as JPL has projected out so far).

*Edit to add, there are some important bits of evidence in the orbital elements that validate its identity as the SpaceX second stage, regardless of all the hoax claims. The perigee date and time indicates it was closest to earth at about 9:35 pm eastern time on the 6th of February (2:35 AM on February 7th by universal time which is close to terrestrial time or TT), about 6 hours after the Tesla was launched, right when the final burn to eject it from earth orbit was supposed to be occurring.

Secondly, the geocentric inclination is about 28 degrees, which just happens to be the latitude of Florida where it launched from. That's not a coincidence though; they launched it from Florida with a heading of about 90 degrees due east, so it should have had an inclination roughly equal to the latitude of the launch site. There may have been a very slight off-plane component in one or more of the burns so it's not exactly centered on LC-39A, but it's significantly less than a degree from that site.

Lastly, the perigee distance is less than 1000 km from the surface, which would be extraordinarily unusual for a natural object. Combined with the other factors it's clear this was an artificial object launched from earth, and its inclination and departure time perfectly match the SpaceX Falcon Heavy launch.
Last edited:
One more update, I tracked it again today. It's faint but it's still detectable to the 0.7 meter scope at Siding Spring.
With this I hope to have enough of my own data to nail down the orbit precisely enough to get very close to the official orbit published by NASA. I plan to use that to trace it back to its departure burn precisely enough to match it up with an Orbiter simulation of the launch and departure and show in a webcast that it's a perfect match for the position and timing of the Second Stage/Tesla. I've been getting a ton of negative comments on youtube from people who don't seem to understand that it's possible to identify it because "it's just a dot." While I can't necessarily expect most people to understand orbital mechanics, I hope to do a webcast this weekend educating them on the fact that it's a perfect match for the trajectory of the second stage and matches the position of the second stage burn witnessed by multiple people as it departed earth orbit around 9:30 eastern time on the day of the launch.
Sorry to bump this fairly old thread, but I recently "debunked" another picture from Starman on Reddit. I made happily use of the commonly known lens flare debunking techniques. Below you see the original picture of a "strange shape", and after it an image with the optical reflection construction around the centre of the frame.
The fun thing is, after a few times you start to recognise this easily in images claimed to be ETs flying through. Sometimes images are deliberately cropped to get rid of the light source, complicating proving the reflection.

(source: http s://



  • starman.jpg
    57.3 KB · Views: 535
  • starman2.jpg
    565.7 KB · Views: 523
Last edited:
Super, SUPER late here but was reading through this thread and just had to sign up to comment given that nobody had fully addressed the "debris" floating around in view of the cameras.

It's ice. Likely entirely or virtually entirely just ice. Specifically oxygen ice, mostly.

Though the camera angles are chosen carefully to avoid reminding the viewer of the fact too often, the car remained mounted to the top of Stage 2. (Obvious if you think about it, as it still had one burn to go at this point.) So it was also still in close proximity to the engine.

The Falcon 9/Heavy second stage uses cryogenic oxygen as oxidizer, and it is constantly warming and boiling off in the heat of the sun. Hence, small amounts of the oxygen are constantly being vented to space. The vent or vents are at the base of stage 2 right next to the engine nozzle. If you watch any Falcon 9 or Falcon Heavy launch with any kind of attention, you will see the occasional bit of fluffy oxygen ice building up down there, and occasionally breaking off as it's pushed away by the pressure of more oxygen building up behind it.

Obviously it doesn't break off with a whole lot of excess velocity, so it stays near the stage and/or payload for a while. When the stage has put itself into a roll, as is the case in this particular launch, it might seem to be rotating around the vehicle for a while; when an RCS thruster fires with a bit of the ice in view, it might seem to jerk off quickly to one side or another as the stage turns.

And, there's more. The RCS thrusters themselves use compressed nitrogen gas. It's non-cryogenic, but it can also produce ice debris. You can see this most predictably immediately prior to a re-light of the second stage engine. (Which means more so in a normal mission, where they just keep the cameras running the whole time, than one like this.) The RCS is used like old-timey ullage motors, to settle the propellants. The nitrogen ice produced by them comes out less fluffy and bulky than the chunks of oxygen ice do. More like a cloud of sparkly debris. And of course as soon as the second-stage engine fires, it's all quickly left behind. But there's no reason that RCS use during a coast wouldn't do the same thing.

Finally, I think it possible that even water ice may be a contributor at times. Water ice builds up on the rocket while it's on the pad, and most likely is largely long gone by the time the rocket leaves the atmosphere behind, but I think it's possible some might stick around for a while in the odd crevice, only to be released later on when enough of it sublimates.

So basically, there is no need to suppose the car itself is contributing to debris that looks like ice. It probably IS ice.
From the space denier's point of view, the fact that the clouds are the same in the livestream as in the satellite photos just means Elon had access to the same people who fake the satellite images, right?
Hello everyone. About the debris:

Turns out I've been observing and analyzing some of the debris trajectories seen in the 'Live views of Starman' video. It has already been attached in this thread, but just in case:

I have found that the movements of the debris seen in the cameras mounted on the Roadster are consistent with what is expected of rectilinear trajectories, as long as the counterclockwise rotation of stage 2 (in which the vehicle was mounted), the approx direction, and the the amount of time debris lasts on screen before leaving the camera's field of view are taken into account.
I went to the trouble of selecting debris that had a 'slow' movement (to make the rotation effects etc more noticeable), taking snapshots of them, and then overlaying and cropping them to get a path. Since I use the application's default image placement (IbisPaintX, it is an application that I use for drawing, but for this job it was useful, plus I was familiar with the controls) that I used, I can attest that they are well aligned and the sizes were correct.
Here some examples:
This lasted approximately 26 seconds. On the Starman video: 9:07 to 9:33

This approx 10 seconds. 1:11:58 to 1:12:08

And this 10 seconds. In the video: 59:18 to 59:28

If the above factors are taken into account, there is absolutely no problem with these. I did more like that, but it seems to me that three examples are enough.
Even making other more 'informal' observations (not tracing anything, following the movement of the particle with your eyes), nothing appeared to be out of place.

However, when I tried to do the same with the camera recording the stage 2 nozzle, some images worked within the set parameters. But one in particular seemed quite abnormal to me:
151 sin título_20201217002402.png
That lasted 7 seconds. 46:25 to 46:33

Entiendo que aquí el efecto de rotación debería ser menor, ya que está más alineado con el eje de rotación del cohete (de la misma manera que en un 'startrail' las estrellas más alineadas con el eje de la Tierra parecen moverse menos que el estrellas que no lo son, en el mismo período de tiempo), que supongo que pasarán aproximadamente por el centro de la boquilla. Sin embargo, mi principal problema es que si miras atentamente el camino, verás que obviamente se curva, pero inicialmente lo hace hacia un lado y luego hacia el otro. Incluso tratando de corregir el efecto ojo de pez de la cámara, el problema persiste. Como prueba de lo que digo, por mucho que intentes conectar todos los puntos con una línea recta o curva, no lo conseguirás y habrá desviaciones.

I would like to clarify that in my opinion there is too much evidence that this launch is real, and in fact all the trajectories that I have analyzed fit well, but this one left me a little lost.
I don't know if my methods are wrong or if I'm not taking into account some details. In any case, I would like to know someone else's opinion on this, what could be happening here?

Or if someone with experience wants to make their own montage of the sequence in some more precise way, they would be happy to see it.

Last edited:
I don't know if my methods are wrong or if I'm not taking into account some details. In any case, I would like to know someone else's opinion on this, what could be happening here?
Looks like maybe the camera is dropping some frames?
These frames may have gotten lost in transmission, and in that case, the video system repeats the frame that is already in the buffer; this looks like the moving object is pausing for a very short time and then jumping ahead.
Looks like maybe the camera is dropping some frames?
These frames may have gotten lost in transmission, and in that case, the video system repeats the frame that is already in the buffer; this looks like the moving object is pausing for a very short time and then jumping ahead.
Sorry if I did not clarify before, but I composed the image with around 70 random screenshots, so the spacing between each rubble at each point is not uniforme, and not be caused by a speed changing, due to the limited number of screenshots I could take and the randomness with the ones that I took. In fact, I don't think it's a loss of frames, since if you see the timestamp in the original video, the sequence looks pretty smooth. This is what I meant by that maybe someone can make a trace using all the frames of the video and obtain a cleaner trajectory than the one I obtained. I am currently working on android, and I have not found an application capable of generating the effect, but I know that complete trajectories can be drawn using a video in photoshop or 3d Max Studio.

I also take advantage of this comment to specify what I mean when I say that rubble marks cannot be joined with a line:
144 sin título.png
This lasted to 18:18 to 18:25 (7 seconds) in the video.
To compose this trajectory, I use also several screenshot, in this try to follow an order when taking them, but the important thing here is the trail.
Clarification: this debris has an irregular shape, it looks be like a small sheet of the glossy coating that the rocket engine has, so its shape varies in appearance as it rotates and moves.
Because of this I tried to use a curved line that crosses their centers to join the particles and trace their trajectory, in this way:
So this path can be described by a curved line (for that I use the modifiable elliptical ruler of the app), which is consistent with the rotational movement and duration of the debris in the camera's field of view.

However, when I wanted to do this with the other image (the one that struck me as abnormal), I was unsuccessful despite trying several times. Even with the naked eye if you pay attention, it seems to be noticeable that the path first curves slightly to one side and then the other.

I will be very attentive to the forum before any contribution, and I will try to continue investigating to see if I can advance something, in which case, I will report here immediately. Anyway, thank you very much for the answer, Mendel, and I am sorry if I did not clarify well what I did in those images. I hope now it is understood a little better :)
Oh, ok. When I watched the video, it had a noticeable stutter at one point, and I assumed that was what you had noticed. Your other pictures show a more regular sequence of dots.

I'm not really clear on what kind of trajectory you expect. If this trajectory is supposed to be straight, have you considered that lens distortion might play a role here?
Hello again.

151 sin título_20201218020525.png
What I mean is a very subtle, but noticeable change in the trajectory of this particular image.
A golden colored bar can be seen to the right of the image, which appears to be strongly distorted by the fisheye lens. Taking that as a reference (which is not absolutely reliable as I'm assuming it's straight) try correcting the fisheye until you get the bar to look straight.
152 sin título.png
However, as I already said, the problem still persists.
But what is the problem? I will try to be the most clear as possible
If you are floating still in space, and you toss a marble, it will have a straight path due to the law of inertia. However, if you are in space again, but this time you are slowly rotating to your left and throw a marble: from your rotating frame of reference in which you are 'still', the trajectory of the marble will appear to curve to the right ; But not because it has stopped moving in a straight line, but because your field of vision has rotated with respect to it. So, always using your frame of reference, in the first case: you should be able to describe its path with a straight line. In the second case: you should be able to describe its trajectory (as the marble moves in your field of vision) with a curved line.
We know that stage 2 of the Falcon Heavy is rotating because we can see the dynamic movement of the lighting, and we can see the Sun, Earth and Moon enter and exit the camera thanks to this rotation (in addition to the orbital movement, but in scales very short time periods of the order of seconds like the ones we are using is negligible). It is also confirmed by observing the curved path of many debris. We also know that the cameras are attached to stage 2, and therefore share their rotating frame of reference.
Since the rotation experienced by stage 2 is relatively 'slow', it is expected that particle trajectories that last a very short time on the screen can be described with approximately straight lines, and that those that last longer on screen can be described with curved lines (because they experience the effect of rocket rotation to a greater extent).

The problem with those two attached images is that the seen path does not seem to be fully described with either a straight line or a curved line.
Those are two of my best attempts, but no matter how long or curved the line is, it fails to naturally align with all the points that describe the trajectory.
I think the main cause of this is that the entire path seemed to curve first 'down' and then 'up':
Obviously I did it in a very exaggerated way, but approximately in the green segment is where the trajectory appears to curve slightly 'downwards', and in the celestial segment, where it appears to curve slightly 'upwards'.

My current idea is that the apparent deviation of the green segment comes from an abnormality in the motion of the particle, immediately prior to the tracing of the trajectory:

I originally didn't care about this as it seemed like the particle was just bouncing off the metal tube over the nozzle and changing direction due to this collision. So, to do the trayectory I start exactly after this direction change. But when you look closely at the clip, the movement seems too subtle to be due to a crash. It immediately reminded me of electrostatic repulsion effects.
It could be that both the particle (as mentioned earlier in this thread, it is probably ice) and the metal tube on the nozzle had electrostatic charge, and that as the particle approached, it was repelled, this being the explanation for the 'abnormality 'of the green segment of the path, and the blue segment would simply be the aforementioned effect of rotation.

However, I am not very informed about how the space environment affects the generation of electrostatic charges or about electrostatic charges in general, beyond the basics; I know that metals tend to lose electrons, and that water (although I am not sure if it is water ice, or even if it isn't ice) is also usually positively charged when we talk about electrostatics. So I will be reading about the subject and also thinking in what ways it could be that these objects had been loaded, to draw better conclusions about this possibility. The english is not my lenguage, and I'm sorry if at times I don't express myself well or what I write seems confusing, I hope this time my idea was clearly expressed. Again, thanks for your answer!
Last edited:
An electrostatic effect makes sense, especially around the rocket nozzle. While I know little about those effects in space, it seems reasonable that all those gasses exiting the nozzle would leave a residual effect.

It also seems possible that camera movement might produce the observed effect. What if the camera or mount had a slight wobble during tearless frames? Or maybe a thruster fired a tiny burst of gas? I don’t think we can know if either of those happened, but it might explain the shift in trajectory.

Since this was a live stream, the images captured were only from the nozzle camera. We can’t compare it to other cameras at the same moment to see whether the effect is visible on the others as well.
I have been reading about electrostatics and its impact on launches, as I have to say that the chances that what we observe at launch is due to electrostatic repulsion are not less...
It seems that there are many ways in which static charges can accumulate in a rocket (even by rubbing against clouds or by solar wind), so that it is a common problem that however can lead to serious problems, even damaging electrical circuits complete.
According to this article, the accumulation of charges on spacecraft is a very common problem that NASA tries to avoid at all costs. Said charges can be due to radiation or hot plasma in the atmosphere.
Many more scientific articles on the subject can also be found online, many in PDF format.
It should be noted that although there are tables of the tendencies of the different materials to obtain loads, even if we knew what both the rubble and the ship's nozzle were made of, Wikipedia seems to be quite clear in saying that the tables and trends help, but that the process that leads to these loads is very dynamic and not very predictable.
Literally sais: "The triboelectric effect is very unpredictable, and only broad generalizations can be made."

So we don't have the tools to say with certainty whether or not it is an electrostatic effect (Because the unpredictability itself could play for or against), but taking into account the relatively normal incidence of this effect around rocket surfaces, we have a potential explanation that we cannot rule out.
I might note, that Elon Musk "possibly" uses two cone-shaped engine thrust nozzles from his former early engines.... as a dinning room table support....w/a glass top.
(can't post pics... NDA)
Last edited:
And probably not necessarily info-worthy....,thrust engine cones are aluminum, and the current ones are cooled with circulating water, so as not to melt . Correct me if I am wrong. I am not a rocket doctor.