Philae landing on the nucleus of Comet 67P C-G Options

[Gerald]

Considering two grains, one on top of the other, each charged 1e, and find the ratio of electrostatic force to gravity:

Fe/Fg = 6*k*(e^2)/(pi*rho*g*(d^5))
...

Great, thanks a lot!

For the potential energies (how high would a grain jump?) I got a slightly different result:
h = 3 k e² / (8 pi r^4 rho g)
by using m g h, with h the height, as the potential energy in a homogenious field of gravity, and k e² / 2r for the electric potential energy between point charges e of distance 2r, leading to a (dynamic) equilibrium at about 4.1 mm height for a 1 um grain (r = 0.5 um), and 41 m height for a 100 nm grain (r = 50 nm).
Within an order-of-magnitude estimate enough to move fine dust even in the absence of sublimation.


Details about actual ionic charge states of solar wind, see e.g. this paper.

[Weywot]

In a short article in a german newspaper from January 2nd, Holger Sierks the PI of the OSIRIS camera system says that Philae could not be found in the images taken in the 20 km orbit.

Leider konnten wir nichts von dem Lander sehen. Das wäre sonst eine schöne Bescherung gewesen.

"Unfortunately, we could not see anything from the lander. Otherwise it would be a nice present for christmas eve."

[flug]

Does the angular size of the CIVA image projected onto the comet model match the expected FOV of the camera?

According to one seemingly reliable source, the CIVA cameras have a 70 degree field of view. (Most sources say 6 cameras X 60 degrees field of view to cover 360 degree panorama - but I find it hard to believe they don't have any overlap at all? The source is MCSE, which claims to have developed the CIVA cameras, so I assume they know what they are talking about.)

Attached is the deconvoluted CIVA image, plus a screen shot of the scene visualized in STK with a 70 degree field of view.

Now, it turns out that the original CIVA image has a slightly wider field of view than the deconvoluted image--something got trimmed in the deconvolution process. The deconvoluted imaged shows maybe 60 degrees FOV, assuming it started with 70 degrees. But all that means is that Philae must have been a bit higher than illustrated here--or perhaps our 3D model is a bit off, or perhaps the CIVA camera's geometry is a bit different the STK. We don't have Philae's height above the surface constrained very tightly at this point (at least, I don't).

Point is, it is very close and well within the constraints we have at hand--all of which have a fair bit of play in them.

Attached thumbnail(s)

 

[Brian Lynch]

For the potential energies (how high would a grain jump?) I got a slightly different result:
h = 3 k e² / (8 pi r^4 rho g)

The difference between these values and those in the previous plot is expected since this is a different (and more sensible) analysis compared to my previous post. Here we are looking at the peak height for a particle sent on a trajectory where the initial velocity is defined by the electrostatic potential energy. Previously, I was looking at the height at which the electrostatic force and gravity forces are equal -- both analyses suffer from an assumed constant value for g, but nevertheless it looks like this erosion process would be process.

[Gerald]

Ah, ok I see, where the discrepancy comes from:
0.001 mm = 1 um = 1,000 nm = 1,000,000 pm.
1e-10 m = 100 pm = 0.1 nm.

[Brian Lynch]

Ah, ok I see, where the discrepancy comes from:
0.001 mm = 1 um = 1,000 nm = 1,000,000 pm.
1e-10 m = 100 pm = 0.1 nm.

Argh, stupid mistake on my part! 1 nm = 1e-9 m...

[Brian Lynch]

As I mention above, I'm even closer to 100% certainty on this now, after examining it even more carefully . . .

You've put a lot of consideration into the comparison, and I am not saying it is incorrect (your gifs are very convincing and I could easily believe this is the correct location). However, a lot of assumptions have gone into the analysis and stating things like "exactly where we know Philae was at 5 minutes after 1st touchdown" and "I am 99.8% certain, perhaps a little more than 99.8%" makes me cringe! We don't know exactly where Philae was anytime after touchdown -- our best guess about the trajectory is based on aligning Rosetta and Sun vectors using the images at 15:35 and 15:43.

If you are relying on Malmer's model and a presumed trajectory that simply connects waypoints with spline trajectories then things get even more uncertain. Plus, you are not transforming the image at all but instead assuming it is only scaled and rotated, meanwhile the (also unknown) direction of the camera will likely not be directly downwards.

Note this is just constructive criticism, I really hate to sound like I'm bursting your bubble! After all, you might be spot on -- just wouldn't say 99.8% sure...

[Deepnet]

In ESA Blog: How & Where is Philae there is an OSIRIS montage. At the bottom of the image is a possible Philae (green arrow).

Roughly scaling the Philae Landing & Bounce OSIRIS montage to the same size by eye from the terrain and then comparing it's Philae at 15:43 (green box) the size seems OK.


To narrow down when it was taken, compare the shadows (left half ) with the OSIRIS Bounce montage (right half), they seem to be longer and at a bit of a different angle, also Rosetta has moved - so this means later than 15:43.

[Brian Lynch]

At the bottom of the image is a possible Philae (green arrow).

Nice catch, it would be great if we could figure out when that image was taken! Here is a rough attempt at projecting the mosaic onto a NavCam image without so many shadows. There are a couple of rocks jutting up that are most likely the bright spots seen nearby the potential Philae location -- I can't pick out anything that could be mistaken for the lander after assuming the obvious rocks are the other bright spots.

Attached thumbnail(s)

 

[Deepnet]

@Brian Lynch , your overlay is well placed and there does not seem to be anything there that could eliminate the Philae candidate.

The same area in cross eye stereo from NAVCAM 20141030


On the OSIRIS bounce montage, is it possible to estimate the height of 15:43 Post Bounce Philae by size ( relative to the other known heights at 15:14, 19, & 23 ) ?

[Hungry4info]

I hate to be that guy, but doesn't that assume it bounced in the opposite direction to what has been established already?

[Brian Lynch]

I hate to be that guy, but doesn't that assume it bounced in the opposite direction to what has been established already?

Not really, it is easy to look at the OSIRIS landing sequence and assume that Philae is flying above the locations in the image, but in fact if you were to drop lines directly down to the ground you would find it is not even close. Attached is a figure I posted before but with the details emphasized a bit more:

- the blue line is Philae's trajectory (from SPICE data) up until the first touchdown
- magenta lines are drawn from Rosetta to each of the locations where Philae is seen in the OSIRIS images (red dots)
- the intersections of the magenta lines and the blue line are the actual positions of Philae at each of those times (blue dots)
- post-touchdown locations are treated similarly but with green lines and green dots (no SPICE data so it requires more analysis to determine the actual Philae positions, but this has been discussed previously in this thread based on Sun vectors and the shadow in the image at 15:35:32)

With the scene oriented so we're looking straight down at the image, the blue dots are the locations on the terrain over which Philae is really flying at each of the times in the OSIRIS sequence -- the huge difference is due to the position of Rosetta, who is looking down at the landing from a significant offset angle. As Philae climbs during the bounce trajectory, we should expect to see it appear lower and lower in the image -- although the comet's rotation and Rosetta's motion makes it hard to guess if this location makes sense.

If we figure out the time that this OSIRIS image was taken then I can look at the vector from Rosetta to the potential Philae location and see if it aligns with the expected bounce trajectory.

@Brian Lynch , your overlay is well placed and there does not seem to be anything there that could eliminate the Philae candidate.
On the OSIRIS bounce montage, is it possible to estimate the height of 15:43 Post Bounce Philae by size ( relative to the other known heights at 15:14, 19, & 23 ) ?

This might not be too hard, although it is a bit involved since you have to account for what I said above (ie. Philae's size will be related to its distance to Rosetta, not the ground).

Attached thumbnail(s)

 

[scalbers]

According to one seemingly reliable source, the CIVA cameras have a 70 degree field of view. (Most sources say 6 cameras X 60 degrees field of view to cover 360 degree panorama - but I find it hard to believe they don't have any overlap at all? The source is MCSE, which claims to have developed the CIVA cameras, so I assume they know what they are talking about.)

On the other hand, this document suggests the side of the square is 60 degrees, and the 70 degree measurement is partway to the corners. There are some diagrams starting on page 8. To some extent there would be overlap since the camera field centers are pointing a bit below the horizontal (ranging from 15 to 25 degrees).

[Brian Lynch]

4th rock from the sun posted a great diagram of the CIVA FOVs here.

[fredk]

In ESA Blog: How & Where is Philae there is an OSIRIS montage.

Where did you find the OSIRIS montage you posted with the green arrow? The blog post you linked to has only a tiny version, and the version on the image site has only about half the resolution of the one you posted. Your version really has more resolution, it's not just upsampled.