(This post was updated on 15.11.2014 with a revised version of the animated image from the Flight Dynamics team.)
The animated image below provides strong evidence that Philae touched down for the first time almost precisely where intended, a strong testament to the precision of the flight dynamics teams who planned Philae’s journey down to comet 67P/C-G.
It subsequently rebounded and, after touching down a second time, came to rest where it is now – a still unconfirmed location likely outside of these images.
Philae touch-down site seen by Rosetta. Credits: ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0.
The animation comprises images recorded by the navigation camera (NAVCAM) on board Rosetta as the orbiter flew over the (intended) Philae landing site on 12 November.
The images were taken at
- 2014-11-12 T15:30:32 UTC
- 2014-11-12 T15:35:32 UTC
The landing time was 2014-11-12 T15:34:06 UTC onboard the spacecraft, between the timestamps of the two images.
The first image is thus 3 min 34 sec before touchdown. At this time, Philae was approximately 250m above the surface.
The second image is 1 min 26 seconds after first touchdown. The touchdown is seen as a dark area. This dark area could be considered as strong indication that the lander touched down at this spot (possibly raising dust from the impact).
Both images are shown below, together with a third image, which is the same as the second, but also shows a green square; this is the predicted touch-down point after re-construction of the lander descent trajectory using various NAVCAM, OSIRIS WAC and OSIRIS NAC images.
They were taken from a distance of about 15 km from the surface, giving an approximate scale of 1.3 m per pixel.
You can also clearly see the boulder identified in the ROLIS images just to the top right of the circle. In all images there are pairs of pixel artefacts (not the lander!).
The individual images are provided below; the animated GIF above makes use of the first two.
Editor’s note: Many thanks to Emily Lakdawalla of the Planetary Society for helping with the image processing of the image initially used with this blog post.
Also, Flight Dynamics have clarified that the green square in the image below is actually the computed touchdown point based on data taken during the deployment and descent to the surface, not the originally intended landing point (as stated in the previous version of this post). Nevertheless, it’s still very close to the latter, again a testament to the excellent work done by Flight Dynamics.
Rosetta NavCam view of Philae first landing site – 2014.11.12.15:30:32 UTC Credit: ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0
Rosetta NavCam view of Philae first landing site – 2014.11.12.15:35:32 UTC Credit: ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0
Rosetta NavCam view of Philae first landing site – 2014.11.12.15:35:32 Credit: ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0
Discussion: 116 comments
To the flight dynamics team:
It was already said multiple times by many mission leaders and cannot be repeated too often:
My great respect: A fantastic job !
And a “Well done” to the search team!
Hope, Philae can be located … good luck
Again me!
Excuse me, but I think many are reading the blog and are hungry to get information about the fate of Philae.
Only the quick info please: Signal yes or no ???
Signal received!
Hi Homo.
https://www.reddit.com/live/tw0cnch7nxjx/
the flight engineers didn’t expect philae to be bounced as it touches the surface ?
They expected it would rebound – that’s why Philae had harpoons to secure it. Unfortunately, the harpoons didn’t fire as planned.
Or they were fired, but against a rocky terrain, not a (relatively) soft ice ground…
Yes, but no, a system with dampeners, gas ejector and harpoons were supposed to keep the lander on site.
The harpoons did not fire and therefore the Philae bounced
They did actually. Philae had harpoons that were meant to be launched into the surfaces when it would touch the surface of the comet, but the harpoons didn’t work, so the lander bounced away, but landed then again later. Philae’s still working and they got pretty much the data they wanted, so the mission is amazing success anyway 🙂
The bounce was expected. The lander was supposed to fire harpoons to anchor itself to the comet. For some currently unknown reason, that didn’t happen. Also, a thruster that was supposed to fire and hold the lander down did not fire.
How come both the harpoons and the grounding thruster didn’t work?
The manufacturer of those devices might know the answer if it was a hardware failure or if the used software that should initiate their function was faulty. The first test of the thruster prior to the release failed as well but nobody seamed to be concerned and with 1 out of 3 failed it might have been wise to change the command set a bit from its origin to take into consideration that the anchors mith have been affected to. This would have made it a bit easier to track and trace the lander. Although the rest of the scenario would have been the same landing on a totally new, never planned for, no name spot.
Nope. They were expecting to be steady and grapped on the commet at once. But the harpoons did not work well (i don’t know why).
The harpoon propulsion system contained 0.3 grams of nitrocellulose, which was shown by Copenhagen Suborbitals in 2013 to be unreliable in a vacuum
SOURCE
Djursing, Thomas. “ESA skrev til danske raketbyggere om eksplosiv-problem på Philae” English translation Ingeniøren, 13 November 2014. Video
The propellant was in a hermetically sealed capsule for sure and not in open vacuum. I would have used sodium azide instead as this is what is used in airbags due tho its high reliability and no need of a high explosive igniter. Another even safer way would be to use compressed Helium and a valve. This also would not pollute the environment.
Hi Mike,
Yes, mission planners and engineers did expect the rebound to occur as Philae was fitted with a cold Nitrogen thruster on top to repel any liftoff, as well as two harpoons and ice screws, none of which for reasons still unclear did not activate.
Andrew R Brown.
Just to be clear the single harpoon and single thruster did not activate. The ice screws fitted to each of the three feet are “powered” by the downward force of the legs so did not need activation. However, they never stood a chance as there is no surface ice whatsoever.
Of course they did. And not only them.
That’s why there are multiple systems onboard to dampen and counterwork bouncing, ie the landing legs configuration itself, a jet to press Philae against the comet, a screw, hooks to keep Philae onto the ground.
Unfortunately it was already known that the jet didn’t work, so they did decide to take the risk nonetheless, but without jet the screws weren’t applicable at the moment of touchdown, so apart from the landing gear configuration only the hooks were left but they haven’t been activated by Philae (no one knows eactly why but they are obviously in working condition) so in the end only the dampening effect by the landing gear came into effect.
On the other hand Philae did land eventually so most of the planned experiments could be executed which by itself hasn’t been for sure, so most of the attention lies on Rosetta whilst Philae has always been a bonus.
Keeping it alife for a longer period would have been a big extrabonus but many observations can be made by Rosetta itself. And hopefully Philae will come to life again (if it’s true that the momentary position will force it to sleep).
Given the size of the image and the presumed velocity of Philae as it rebounded from the surface, it seems safe to assume that the probe is imaged somewhere in this field of view?
I agree with John. We should be able to spot some trace of the lander in this shot; or is the lander size below the resolution capability of Rosetta’s imager?
The NavCam has a pixel size of about the same size as Philae has from a 12 km altitude. With the amount of pixel errors noted in those two frames its a blindfolded search. The Osiris would have made it with its far superior resolution and dynamic range. Hard to guess why it was not used.
Hi John McDonald,
Excellent post, but I am afraid not. Philae bounced well outside of that area, far over to the right / west I think.
Andrew R Brown.
If the probe already travelled about 33m when the second snapshot was taken, and based on fact that the boulder in top right corner was stated to be 5m in its size, the probe should be somewhere in 6x(size of the boulder) radius from the calculated landing spot.
“…based on fact that the boulder in top right corner was stated to be 5m in its size…”
Do you mean the boulder above and to the right of the original landing area? I don’t see anything boulder-like in the top right corner of the images. That region looks flat, featureless.
Yes, I actually meant that one. I should rather identify the place as “top right quadrant” instead..
But the second image is taken just 1 minute 26 seconds after the first touchdown. Even if Philae’s motion was entirely horizontal, it would have traveled just about 20 pixels from the touchdown spot.
Assumed velocity of the probe after first bound was stated to be about 38cm per second. I am not sure about dynamic evolution of the velocity from the very first moment to next bounce, but I would expect it to be constant. Given that, the probe already travelled about ((15:35:32 – 15:34:06) * 0.38)m = 86*0.38m = 32.68m.
No, per la scarsissima attrazione gravitazionale del corpo.
Comunque erano previsti mezzi (come dire: dei piccoli razzi?) per contrastare la forza eventuale del contraccolpo…
There’s a white smooth spot somewhat 10 pixels east-southeast of the boulder close to the circle (and a dark smooth spot few pixels below it – a shadow?), that could be something like 25m away from the contact point, a plausible distance after a 1’26” rebound.
Maybe just a random texture variation… or maybe…
Not exactly. I think the lander can be seen about 22 pixels to the east and 2 pixels up from where the lander actually touched down (center of the red circle). The shadow is visible 6 pixels south of the lander. I am pretty sure, because this would be consistent with other shadows being cast.
This information should help calculating the trajectory / path of the lander and thus locate the second touchdown.
Agree. The combination of 1 light pixel (philae) and 1 dark pixel (shadow, in an likely place) distance from impact etc. seems on the spot. Will be important to consider exposure time etc. Long exposure will turn moving spots into a pale smear. If indeed so, the distance between shadow and Philae, inclination, distance from impact, rebound speed estimate of 0,38 m/s, combined with nucleus shape-model, should predict impact no 2. quite accurately.
…and adding nucleus rotation.
…and a bit of luck, of which there seems to be plenty.
I have the same impression this could be the Philae moving above the surface .
It would be nice to konw from ESA in which direction the comet rotates in this picture .
If it is confirmed the comet rotates from right to left in this image the spot might indeed be Philae .
Seconded. (I came here to post the same guess…)
Furthermore, the shadow (if so) seems to match the directions of the other shadows.
Haring, I think you are right.
I marked the smudge you mention in this picture:
https://pbs.twimg.com/media/B2gbcTpCQAABKQi.png
Thanks for the image. It’s nice to be sure we’re talking about the same feature.
But I have some questions: Looking outside the predicted landing area, there’s a fair amount of variation between the frames. In particular, some craters (for lack of a better word) disappear, or suddenly appear. The triangle of “craters” in the top right corner is a good example. They appear to be single-pixel changes, but the pixel values seem to change by a significant amount.
So might these be reflect actual changes to the comet’s surface? If so, it doesn’t help the Philae shadow theory (although the back-of-the-envelope calculations are persuasive). Of course, watching the comet’s surface evolve is itself pretty awesome!
I understand that pixel values between frames are likely to vary, but does anyone have an idea of how much they’re likely to vary? I assume each pixel assumes a gray-scale value in the range [0-255].
Bravo Haring,
I think you were the first to indicate the location. Distance in your entry was just an quick estimate.
I second Homo stupidus,
It cannot be said enough, how much we are in awe of this success, Philae IS a HUGE SUCCESS, despite the turn out of events.
Also echoing Homo stupidus, is there a signal? I would go further and add, are there any new images from Philae?
One thing occurred to me, during the firth ‘bump’ in this remarkable NavCam image series, did Philae capture a partial panorama with the CIVA? The spacecraft was moving at no more than a slow walking pace. I hope we get to see all of the ROLIS imagery during the bounces as well as more NavCam and OSIRIS images of said areas too?
I think Philae may yet surprise us with her robustness.
I will be back 😀
Andrew R Brown.
Well, maybe the incident is itself an answer: Next Philae should be a ‘jumper’ 🙂
Way wait for the next lander, as this one was so from the very beginning, just a bit misinterpreted and misused.
a lot of clever drone robotics is being developed to let a lander decide on its own where to land. More technology = more failure modes. The drop bombing part was actually the part that worked to perfection 🙂
It should be a big harpoon.;)
From the ROLIS image the nearby boulder is ~5m across, I would say the green square is less than 2m from the actual landing spot. Thats some amazing shooting from 22.5Km away to a moving target 500 million Kilometres away. It can’t be more than 50m from the original landing ellipse centre either. Utterly staggering precision!
It is indeed, incredible precision indeed.
I hope ESA dig out the blueprints of Rosetta and Philae, rebuild both and send them to another comet, perhaps an Oort Cloud comet if there is enough lead time or an asteroid, or Phobos or Deimos 😀
Andrew R Brown.
Wow! What a great message!!!
Well, those screens seems to be Linux or Unix 🙂
I see the lander 20 pixels right of the touchdown point and its shadow is five pixels below that. If correct, that means it was moving 14 pxels per minute.
So this would mean 32,68m flown distance and around 18m on the ground. That gives a rebounce angle (towards a flat surface) of around 55 degrees, so it has almost 27 meters height at that point (again assuming flat terrain and if I calculated correctly). Considering Sun elevation and viewing angle, would the distance between white and dark spot fit to this height?
Correct calculation, wrong input … the surface distance is in fact about 27 meters, so the rebounce angle would have been about 35 degrees and the height at this point about 19 meters. For a 111 minute flight time this means more than 2 km horizontal hop distance. When taking the comet’s rotation into account (52 degrees during this time) would that end up in a place which fits to the CONSERT data? That’s just a rough calculation as it does not consider all effects and there is some uncertainty in the data but perhaps this really brilliant flight dynamics team should give it a try…
In the second image there’s brighter patch to the right of the impact point, with slightly darker patch below it. And they’re not pixel artifacts. I’d swear it is the lander.
Here I circled the two spots:
[IMG]https://i.imgur.com/KlBeC6Y.png[/IMG]
Are you kidding? Kudo’s to all involved for a VERY successful mission regardless of the final outcome. Even if we did not receive further data, photos and other materials, our ability to actually get it there, land it and speak to it is extraordinary! The rest is all bonus!
philae should be flying somewhere in this picture or maybe just a bit of a viewing angle. is it possible to spot moving object with enough resolution from source frames?
El sitio elegido era magnífico. Una pena, una verdadera pena el que no funcionasen los anclajes.
Sorte txarra.
Great job.
when Philae landed on the right spot it was not really an adventure.
I mean now we can hoping and watching and and and…
it is extra value for the same money. 3 landings!!
Has anyone noticed the small white object casting a shadow along the right side of the second photo?? It is just beyond half way up the right side of the photo.
It wasn’t multi-touchdown?
As the one below?
https://www.youtube.com/watch?v=6FbH4Y97ucc
Just reiterating on what Haring stated: Beside the dark spot at the assumed landing site there is a signifcant second difference between the two images: A tiny white spot exactly to the right of the landing site in a distance of 21 pixels (a bit less than twice the distance between landing site and the boulder). 5 pixel below there is a dark spot which is also not present on the image before touch down. Both spots seem not to be image artefacts. Could that be a trace of Philae?
I think you are totally right. correspoding to the released informations this is exactly the distance i would expect philae at this moment.
maybe the flight dynamics team should analyse this with more accurate calculations!
it could provide better information about the movement angles after the first bounce.
Assumed the spots are Philae and its shadow and assumed that Sun’s elevation angle is low (below 45 degrees) it looks like it gained considerably more distance than height on rebouncing.
The navcam apparently is not able to detect Philae, how about the Osiris and some image stacking. Will it be within technical reach to detect Philae at all?
Single pixel analyzing is not going to be of any value, only image processing with a lot of pictures is going to be fruitful. The Osiris NAC has a pixel size from a 30 km altitude of about 1/3 m2 and thats about 3 pixels for Philae. If image cropping/ stacking is used then the resolution can improve a factor ten. The bad news is that its in the shades for most of the time and to collect images where it is in the sun is going to take quite some time.
A geostationary orbit has about a 3280m or so radius and that is very close to the surface. A new concept of a lander would be that one part of the lander would stay a bit outside this radius and the other part anchored to the ground, both connected with a cable. The advantage would be that the orbiting unit will be pulled forward securing its stable position a few kilometer above the ground and most of the time in the sunlight. Also the communication link window would be a lot longer.
The disadvantage is that you have to relay on harpoons again and as two out of two failed, this concept is prone to some development. Possibly a jumping rabbit concept would be safer and this actually could have been accomplished with this lander as well if from the start if the comands would have been a bit more flexible and not just relayed on the function of the anchoring system. To start the experiments at the tuchdown signal was too optimistic. The safe way would have been to just keep the cameras running and have a command series that did not start any power consuming experiments untill an optical confirmation was received and possibly also made som decision to continue to make use of the active jump capacity of the lander. It would have taken a day or so to start the important experiments but then at an optimal and final landing site.
Well, congratulation to the team for an almost 100% success and hope fully the errors that occurred are analyzed to the full and this valuable information will be shared to whom it concerns. To learn from mistakes is the best way to develop. A repeated FMEA is to be done. With some luck the lander will wake up again and also located by then. Thereafter it might still be possible to jump into a better position to continue this mission to its full. The Rosetta does what it is supposed to do and if the lander looses a halve year of mission time is not very critical i suppose. The lander team can focus on, during this period, to create some plans that will save the situation for the future. Wish you luck, and please enlighten us with some part of your plans.
The decision regarding starting experiment on touchdown signal is driven by the fact that whatever happens, Every minute/second battery cells needs to be heated to operate … waiting any longuer would have mean less time for experiment …
They got clues about truster probably failing on touchdown before the release (where expecting it was a sensor error) and certainly knew that charge within the harppons could fail but the balance between gains vs risks leaded to a GO and it been a good decision regarding all result collected – sure with a little bit a luck it could have been even better but achievement has been fantastic.
Special reward to Andrea Accomazzo and his team – best pilots in the known universe so far – driving eyes closes on an absolutly unknown location with 511+ millions distant steering and 58mn light round trip delay, yet touching the target with such accuracy.
Even if the drilling results prove to be the last data that Philae will ever send, this enterprise will still be a resounding succes in my opinion. Great job!
In one of my previous Jobs I became familiar with principles of gas chromatography and mass spectrometry. What am amazing achievment to send portal versions of such highly sophisticated devices onto such a distant target! Congratulations for this breathtaking and successful endeavor.
Mastering such challenges provides the courage that we need to tackle the manyfold challenges on our own small planet!
Thank you!!!
One way you can tell any difference between the two photos is to over lay them exactly on a computer screen (e.g. open two images or webpages), and then ‘toggle’ back and forth real quickly. You’ll immediately see any small differences in the photos.
Has anyone noticed the small white object casting a shadow along the right side of the second photo?? It is just beyond half way up the right side of the photo.
I am loving the images n daily updates 🙂 for me the mission is already a success, all the science n learning n images are a besutiful tasty bonus…you landed a functional man made space probe on a comet, 510 million kms from the earth, travelling up to 135000kms/per hour….OMG!!! Bravo 🙂
Incredible precision! Important experience gained for eventual future missions to comets or asteroids.
Maybe a stupid question – but maybe no other was “brave” enough to ask….;o)…but:
if Philae is lying on its side, maybe a execution of one of the harpoons can pull himself out either with kinetic energy of the harpoons flying off an pulling, or while Philae tries to fasten the string of the harpoon and gets some grip on the ground with the “spreading” harpoons head…..??? Would this procedure cost much energy/battery?
I tried my best in matching the two images together and while it is not perfect I think it is a success. I also removed most of the two-pixel artifacts but left the bright/dark thing to the right of the impact since it is way bigger than two pixels.
https://i.imgur.com/j4dbMJV.gif
I hope very much that the mission succeed. They have done so much to reach there. Noble cause! I wish you all the best great people!!!
I have a sweet calculation here for you:
All this assumes that the image was taken roughly vertical (which is probably not the case) and that the terrain under the white spot is at roughly the same altitude as the touch-down spot (which is probably given.)
Distance touch-down to white spot in x direction of image
dx = 22px x 1,3 m/px = 28,6 m
Distance touch-down to white spot in y direction of image
dy = 3px x 1,3 m/px = 3,9 m
Total distance of white spot from touch-down point projection to ground
dp = sqrt(dsx^2 + dsy^2) = 28,86 m
Sun angle estimated from shadows is 60 – 75°, let’s say
a = 65°
Distance white spot (projected to ground) to its shadow
dsp = 5px x 1,3 m = 6,5 m
Height of white spot over ground
h = tan(a) x dsp = 13,9 m
Total distance of white spot from touch-down point
d = sqrt(dp^2 + h^2) = 32 m
Time difference touch-down to second image
t = 15:35:32 – 15:34:06 = 86 s
Speed of white dot
v = d/t = 0,37 m/s
Hey, does that last value ring a bell?!?
Come-on ESA, give us a comment! 😀
BTW:
With a distance of the white spot projected to ground of 28,86 m in 86 seconds the total distance after 1:50 hours (the time between the touch-downs 1 and 2) would be 2.2 km on a flat surface and significantly more on the curved surface of a comet that is only 3 to 5 km in size. It would very likely result in a landing site not being on the smaller lobe any more.
So I guess that the whole idea of that white spot being Philae is void after all (even if the numbers match so nicely).
@Jörg Wagner: Smart calculation! I think the important thing here is the comet’s own rotation. From the published data the first touchdown point seems to be near to the comet’s “equator”, so it moves pretty fast due to the comet rotating. A point 2 km from the rotation centre and on the comet’s “equator” proceeds about 1800m during the time Philae needed for its first hop.
Beside that I think flight dynamics could do a quick check to see whether the “bright spot hypothesis” could make sense: The flight direction after touchdown we assume and the rebounce angle (just 28 degrees according to your calculation) must fit to the direction and angle Philae had when first approaching the comet. If Philae did not come more or less from the “left” in the image we should forget about the bright spot. If it did approach from there, then…..?
ESA has confirmed per Twitter that the dots 22px to the right of the touchdown location are indeed Philae and its shadow! I am quite sure that Philae is now somewhere in the neck region or even slightly beyond it on the inner side of the larger lobe.
https://twitter.com/ESA_Rosetta/status/534007368708022272?s=09
I don’t think we see any dust caused by Philae, it is just the changing of the shadows. A bit to the right we can see the same thing happening. My thoughts about the whole mission is that we should send landers with more capabilities or just send multiple landers.
Is it possible that the harpoons just did not have an adequate surface to grab onto, perhaps breaking through a crusty veneer like crusted over snow, only to find sandy texture underneath with inadequate structure to hold down the upward motion of Philae? It looks to me like the landing area could easily have had such a soft composition.
Similarity of the NavCam image with the Rolis image at 40m is striking (on the low frequency features).
If you tilt by 22° the rolis image and low-pass it the H shadow pattern below the boulder is clearly visible on the rolis image.
I don’t know how to put an image in a post here, but I have the illustration of this (here: https://www.normalesup.org/~cantallo/rolis_rotated_lowpass_scaled.jpg ).
It is sad that the Rolis did not continue imaging during the bounces, because by matching these images with the Navcam images from orbiter, locating the lander would have been possible.
By the way, I think that the touchdown area could not appear darker because of dust deposition, because as the lander (and perhaps even more because the dust is not fitted with choc absorber on its landing gear) the dust projected could not have reached the surface before a few hours at least after landing. (in the vacuum, the lander and the lightest dust grain fall at the same rate…)
More interesting is that the material slightly BELOW the surface is indeed a bit darker than the surface (at the opposite of what occured on the moon: the LM touchdown area is slightly LIGHTER than the surrounding ground). That is an interesting observation to match with drilling results.
With all my wishes of luck to Philae for its hibernation…
and thanks again for the marvelous images.
Does the data obtained from Philae tell us anything about:
– why did the gas-release pushdown system fail to operate?
– why did the harpoons fail to deploy on landing?
– have we been able to collect any data on these things?
– what is the current surface temperature on the comet?
Do we know how long the batteries will remain usable in the current state?
As 67P gets nearer to the Sun, things will hot up. Outgassing will get more and more intense, and the surface of the comet may well start moving. In this environment, it seems to me not unlikely that Philae may get moved. Including getting pushed off the surface again, after which the weak gravity may cause her to make more “landings”! Almost any move from her present position under a cliff, it seems, may be preferable from the point of view of obtaining more sunlight. Hence my question about battery life.
How long have we got?
This seems to me to be a question which the Rosetta team should be able now to clearly answer.
Is anyone listening?
@Christian R : If it is true, that the lander gained considerably more distance than height on rebouncing, could this perhaps be due to the rotation of the comet itself? I’m just guessing here, ’cause I’m not an expert in physics/astronomy.
@Jerome Verhogen: The comet’s rotation has an influence. Depending on the exact distance from the comet’s rotational axis it is in the order of a few tens of centimetres per second. This is not so far away from Philae’s initial approach speed. But it depends most on the direction and angle the lander had when approaching the comet (which we do not know here in the discussion) and how sloping the terrain is. Flight dynamics have most of these data so for them it should take only minutes to find out whether the hypothesis of the bright spot being Philae is worth to be analysed further. Once someone tells them what we are discussing here.
@Christian: With a period of 12,5 hours @2000 m radius, roughly linear speed of comet under Philea is 28cm/s as you wrote. Somewhere in the blogs I had read that the flight directors equalized the horizantal speed of Philae with landing point, so as to be a vertical landing on a non rotating body. If I recall right.
from all the hypothesis presented here it seems that something is missing: the lander did end in a dark place, so it is most likely not in this image. Second point: maybe it is just because of difference in lighting angle, but there seems to be an elongated trace of dust next to the touchdown spot. Could this give an idea of the direction of the impact? One would expect the lander moved some dust in the direction where it bounced. Some of the dust would of touched ground before the lander. Looking at a larger view from https://www.jpl.nasa.gov/spaceimages/details.php?id=PIA18875 maybe a place to search are the dark places to the lower right part of the image? PS: Is it well established now if the lander bounced once or twice? Is there an intermediate touchdown spot?
My comment was made before the revised version was uploaded. I think the whole credibility is compromised. First we get an animation from two pictures and than suddenly the pictures seem to have been changed. Big question mark here!
Great great success for ESA!!
So proud of you guys-thank you for letting us witness history!
Xxxxxxxx
Freewheeling about the search for Philae: Is there a possibility to find the actual location by the incomming sunbeams on panel 2 and the other panels?
When you know the moment when the production of the elektricity was maximun on panel 2 and you combine this knowledge with the position of the comet towards the sun at this time. If you also know the maximum incomming sumpower on the other panel(s), also combining with the position of the comet at this time with the sun.. Then it might be possible to draw a strip over the small lobe where Philae may rest… No? If this is possible and you can combine this line with the radius between the first and last touchdonw… Just freewheeling.
Philip
Slightly better fit, this time without circles (using updated image from the post)
https://i.imgur.com/xncdrwZ.gif
On the second image, at the right of the darkened area (dust) one can see a brighter spot with a dark spot a few pixel below, both absent in the first image.
Could it be Phylae and it’s shadow?
Vladimir Dzhanibekov effect ?
Very likely that Philae hit 67P/Churyumov–Gerasimenko surface with her legs upward (that is why we can see only one spot in this photo and not three distinguished footprints).
This caused it bounce with significant fraction of initial speed (without rebound) back into space.
Quite likely the reason is that ESA “stabilises” its landers by rotation and the latter may cause Dzhanibekov effect
https://www.youtube.com/watch?v=L2o9eBl_Gzw
Therefore, in the key moments landers might occur to be upturned.
(It is quite possible that also Beagle-2 met Mars atmosphere upward-down, due to the same effect.)
Somebody from ESA should take a better look at this comment. I don’t know if the mentioned effect happened or not, but landing on its back would explain a lot… why the harpoons did not fire, why none of the landing systems deployed in fact for that matter… Having a soft landing on a dusty area, foam like, most probably saved the poor Philae from breaking apart… it makes some sense… I’m not a scientist, but what are the odds of 3 different systems failing (top jet, harpoons, drills) at the same event? Landing on its back would explain it all.
This proves
https://www.esa.int/spaceinimages/Images/2014/11/OSIRIS_spots_Philae_drifting_across_the_comet
that Philae landed on her four (I mean three :).
There was no Dzhanibekov effect involved here
(this time).
Exactly how big is Rosetta’s comet? Here it is compared to Los Angeles.
https://www.iflscience.com/space/graphic-shows-size-rosettas-comet
I don’t understand why the landing site of Philae should show a patch which is darker than the surrounding, which is already very dark. Shouldn’t the touch down area be brighter than the surrounding area? Maybe, it’s only Philae’s shadow what we see
@manyworlds: Philae is about one pixel big in these images, there’s no way to see three footprints of its legs at this resolution,
@dhd40: It’s hard to say but my guess is that in general dust on the surface is arranged by gases coming out of the hidden surface, while dust shaken by the impact will be more chaotic and might have different light scattering properties, that’s why it appears darker.
Regarding the white/black spot being Philae, I think it is likely. The second image was recorded a minute and half after touchdown and Philae rebounded at approximately 30 cm/s, so the distance it traveled in those ~90 seconds is about 30 meters, which is very close to the distance at which the spot appears to be from the impact zone (remember the boulder is about 5 meters in diameter).
I believe we will get more info on that when the team finds Philae’s final resting spot.
@Kasuha: “Philae is about one pixel big in these images, there’s no way to see three footprints of its legs at this resolution”
But this doesn’t necessarilly mean that there would be three footprints at higher resolution, isn’t it?
Very unexpected that the impact resulted in a darkened ‘plume’. Being so dark already, ejected dust should almost invariably produce a more reflective plume, maybe with an underlying shadow. What we see is a more or less uniform darkening. Is there a chance, that the lander made the whole area collapse? In order to explain the subsequent bounce, the landing site would have to consist of a several meters thick “fairy castle” on top of a “crust” strong enough to make Philae bounce, and rough enough to make the bounce other than perpendicular to surface, as it seems? Somehow I find this scenario to correspond well with all bits of information. New images will show if the changes seen are more or less permanent. Of course, changing back to the original albedo in the same lighting angle shoots this bit of speculation to pieces.
I will now revise that completely: no collapse, but no dark plume either! There is a plume, but less obvious, because the plume is almost indistinguishable in albedo from the original surface. This plume streches from the darkened area up towards the direction of the sun. The darkened portion is the shadow cast by this ‘invisible’ plume.
Will be important to follow every trace of Philae, before they are erased: knowing the location will help if it should later be attempted to make a ‘hop’ ( given batteries charged in some future)
Hi Jacob. I expected it: the difference between unstructured ‘dark gray/red’ ‘dust’ and structured ‘void full’ ‘stealth’ ‘Lakritz’.
My question is: didn’t ESA design, or couldn’t ESA have designed a set of ‘legs’ which would collapse in a calculated way so as to absorb the impact energy of Philae.
Springs wouldn’t work ’cause they’d simply release the absorbed/stored energy; but for example, something that would fold (‘crushed’ if you will) and remain static after absorbing any energy. Or even something like a cylinder that has a release valve that lets gas out of a small vent when the pressure increases on impact, much like when a stunt man lands on one of those large air-bags after jumping from a 30-story building.
It could have been relatively light and simple in design, I would guess.
It would seem fairly feasible to calculate and design something since the, gravity, velocity and mass of Philae are all relatively small; and the forces would likely not be too great so as to damage components, etc.
It would seem so much simpler than all the exploding anchors, etc. they used, and apparently did not work very well.
Or am I just being stupid or what?
You are not stupid, but they are not either. You miss something in your backward evaluation this is the “backward”.
In 1993 the comet model was about ice ball with massive gas exhaustion that can blow out a shuttle as viewed in those Hollyword’s block buster. Of course we now realise the model was not even close to reality …
The plan failed because of mechanical failure plus inapproprite model .
AFAIK landing system was not spring based but try to convert so little energy (1g x 1m/s does make a lot of energy to absorb – will be difficult to crunch something able to resist earth takeoff acceleration btw) in ice screw rotation – should have been helped by the failing truster
“…something that would fold (‘crushed’ if you will) and remain static after absorbing any energy ” Like soda cans?
You are talking of the moon lander’s shock absorbers, Neil. Amazing tech in its simplicity and reliability. Well known by the time Philae was designed. Surely there is a very interesting history behind implementation decision.
Don’t forget that you had the most ever powerful horse at that time. Weight restrictions were high, but not as astringent as with the Ariadne.
Maybe Philae was ‘discretely’ designed as a ‘jumper’ from the beginning, Neil. And budget constraints ‘crippled down’ the original intent. Just a ‘playful’ guessing.
I wondered whether Philae should be in range of the NAVCAM image taken 3:34 before touchdown. I looked into the ROLIS image and the fine work done by cantallo (see his avove post). Provided that the ROLIS camera is oriented such that it exactly points into the direction the lander moves (which is an assumption I have no proof for) the deviation of the actual touchdown point towards the centre of the ROLIS image should give a clue to the actual movement of the lander relative to the comet’s surface.
From that it looks like the lander moved from “north east” to “south west” on the NAVCAM picture.
The published altitudes for NAVCAM and ROLIS images were 40m vs. 250m (with all uncertainty of these figures). So if we assume constant movement the lander should be about 6.25 times farther away from the touchdown location as it was in the ROLIS image.
This extrapolated leads to an area just outside the ROLIS picture but well in the NAVCAM picture. It is almost exactly “north” of the boulder about two to three times the boulder’s size away.
There is in fact a very faint change of one pixel in that area. But it is a wild, wild guess and perhaps too near to seeing pink elephants 🙂
So, will it be possible to narrow the possible post-bounce locations from the position of Philae in these images?
doekia, You’re 1993 inappropriate model point is well taken; as to the other, you just connect a timer/microprocessor to prevent an “premature” triggering. I’m sure they’ll figure it out for the next go around. 4 billion miles is insane! It gives hope that there may be someone out there smart enough to solve some of the self-inflicted problems here on Earth.
ESA has confirmed per Twitter that the dots 22px to the right of the touchdown location are indeed Philae and its shadow! If this is true, then I am quite sure that Philae is now somewhere in the neck region or even slightly beyond it on the inner side of the larger lobe (see my calculations above, which were obviously fully correct).
https://twitter.com/ESA_Rosetta/status/534007368708022272?s=09
Bravo! You were spot on!!
Hey, how about without too much guessing here, lets just say its science and the most likely is that the hypothetical situation or even better, the initial analysis of comet missed something that never helped the situation with the lander? Could the temperatures, the consistence or substance of the comet have thrown up something unexpected?
Its a good lesson learned and the next mission will probably be better equipped to land and better still, include a better battery + maneuverability like the 2 guys running around Mars so that an attempted recovery would be much “easier”!
This may be crazy, but is there any way to reflect sunlight off Rosetta toward Philae, so it could power up? Just a thought… wacky though it may be.
Wow, realy nice pictures! I Hope you get a lot of interesting data from oft there!
Does anybody happen to know if the Rosetta has sufficient propellant and power left to reduce its orbital distance to the surface of 67P for better mapping the surface and looking for the lander without the danger of orbiting into the surface or compromising future goals?