..or will that drag stop the mole from progressing?

They are obviously aware of the risks. But it seems to be the risk of damaging the teather with the scoop:

In a slightly different context, they say that the force should be about 7 N:

I presume that this is considerable more than the friction between a layer of a few centimeters of sand and a smooth teather as long as there are no perpendicular forces in the same order of magnitude. The friction force induced between sand and teather by pushing the scoop will push the teather downward, too.
So, the upward motion of the mole might be the result of the elastic reactio of the phase boundary between loose sand and the suggested "duricrust" on hammering, not necessarily just the recoil of the hammering mechanism.

I'm not sure what you mean by elastic reaction, but the thinking seems to be that the recoil happens because of lack of friction due to too stiff duricrust material. Hence adding sand to try increase the friction with the mole. See eg the July 7th 2020 log entry here.

About the cable, I guess that was designed to handle drilling metres into the soil so should tolerate the current configuration.

In the simplest case, even with friction, if you are hammering on a solid elastic surface, your hammer will bounce back from the surface by Newton's 3rd law, the "reactio" of your hammering, or stated differently by elastic collision of the mole against the duricrust or whatever elastic material around.
The assumption of an inelastic collision doesn't require to be a given. But my impression is that an inelastic collision of the mole with its environment is assumed at least approximately, and only the recoil due to the acceleration of the hammer within the mole is considered appropriately. An elastic collision can cause much larger forces than the seven Newton the scoop balances. The force is depending on the hardness of the (elastic) surface you hammer on.

Well, according to the diagram in that July 7th post, they believe the tip of the hammer is well beneath the duricrust, so it doesn't seem like a collision against the duricrust is likely.

If we believe that assumption, then we have two plausible scenarios:
- Either there is just a pebble at the tip of the mole, or
- the mole is hitting bedrock.

In both cases, the mole needs to find a way to return to partially inelastic collisions to make any significant progress.
In the case of a pebble, it may either be possible to destroy it or to work around it. The loss of surface material might hint towards a pebble sinking into other regolith and leaving a void.
In the case of bedrock there may exist two options:
- Find a fracture or a soft fracture fill, or
- find a way to reduce elastic bouncing when hitting hard bedrock,
-- either by a larger force of the scoop pushing the mole, the risk of which needs to be assessed thoroughly by the mission engineers, or
-- by fixing the mole to a larger mass, which is looking even more challenging to me, since sand probably won't do it.

By the images, I'm inclined to presume that they are trying to work with a larger force of the scoop, but very cautiously at the same time in order to reduce the risk of damage.

The distinction between elastic and inelastic collisions is not present in the mole team's description of its intended operation.

• The design: "for loose soil to flow around it, providing friction that keeps the mole from bouncing backwards with recoil"
• The problem: "a lack of friction in the soil"; "The soil around InSight provides much less friction than what we've seen before on Mars."

The expectation is that the collision will be elastic, drive soil out of the mole's path, the friction from soil lateral to the mole to prevent significant recoil, and for the mole to thus fall a bit into the cavity thus created, and experience a net downward motion each time. The problem isn't that the mole is failing to sticking to the soil in front of (below) it.

https://mars.nasa.gov/news/8444/common-ques...e/?site=insight

Thanks! So, the mole in the soil seems to be considered to behave similar to the way it would behave in some kind of non-Newtonian fluid, quicksand for instance, with low (inelastic) resistance against low velocities during recoil and high (elastic) resistance against high velocity when the hammer is striking.
That way, the mole will counterintuitively move in the wrong direction.
Something for my backhead to think about.

It's amazing how little we still know about Mars. Years of research and tests are done on equipment, such as the mole, and the soil is not what they expected, so it doesn't work. Or even the unexpected wheel damage on Curiosity. Years of testing means nothing when the planet is holding the cards so close to its chest. But these setbacks only enrich our learning and experiences.

https://twitter.com/NASAInSight/status/1310978911002484736

More scooping and hole-filling to come!

Phil

Blink GIF of processed / cropped IDC images from sols 651 and sol 654. A little movement on the science tether. Wind, settlement in the filled pit, or thermal variations in the cable?
Click to view attachment

Blink GIF of processed / cropped IDC images from sols 654 and sol 656, more movement of the science tether!
Click to view attachment

It feels like reading tea leaves here, because we don't see what might have happened between frames, but the tether seems to buckle upwards in response to a tiny pull downwards. The scoop does not move between the frames, so that would seem to indicate that the mole moved a small distance downwards, which would be a positive sign.

However, the dynamics of a flexible object can be hard to interpret and it's not really going to be a good sign until we see the tether pull at least a couple of cm in the right direction.

The only fairly plausible idea my backhead came up with in the meanwhile to explain the strange behaviour of a soil poor of pebbles was triboelectricity between very dry grains of different composition within a very well-mixed fine dust phase. I discarded van-der-Waals forces and magnetism as much less plausible primary causes. Triboelectricity should be strongest where hammering forces are strongest and might make dust behaving temporarily like an elastic solid.
I think that it's fairly difficult, but not impossible, to produce such a kind of Mars analog in a lab and to use it in a test bed to develop a hammering strategy. A thorough mixing would probably require the mixing and settling of aerosols of the different mineral species rather than mixing powders directly. That's making the whole experiment less easy.

That said, thanks Paul for keeping us up to date!
I'll have to return to the next planet further out with my active participation.

I couldn't help to dig a little deeper in literature, and found that, in contrast to my first expectations, mere van-der-Waals forces can already be sufficient to cause temporary jamming (transition into a solid) of fine powders on high stress (Jamming Threshold of Dry Fine Powders, by J. M. Valverde, DOI: 10.1103/PhysRevLett.92.258303, see especially Fig. 1, investigated in the context of toners).
More tricky electrorheological properties of the dust phase in the Martian soil would still remain possible.

Great- so it is basically "electric oobleck"?

I started with an electrostatically stabilized suspension (but a mineral powder replacing the liquid electrolyte), considered a rheopectic fluid model induced by tribolectricity, and got stuck somewhere between a shear-thickening Herschel-Bulkley fluid (related to oobleck, but with a positive yield shear stress) and the Casson model.
It appears that you'll need a Mars analog testbed with a sufficiently large dust phase (the fluid) of a realistic composition to find out what's really going on.

Arm / Scoop activity during Sol 659: Simple blinking GIF, using just 2 images from the Instrument Context Camera (ICC)
Appears to show the robotic arm pulling up from the mole and folding back the scoop.
I guess there will be a few more images in the pipeline, but they may be setting up to scoop more material on top of the mole at a future date
Click to view attachment

InSight has now been on Mars for a full Martian year.

Phil

More soil scraping over the pit on sol 673.

Click to view attachment

Phil

This is how it was predicticted in the DLR blog post link
Click to view attachment

So, things are on a two-week cadence:

My three indoor cats approve.

Would it still count as "unmanned" spaceflight if we sent a cat to Mars? We sent a mole.

A cat on Mars was already done! This website shows how Steve the Cat flew on Phoenix http://www.stevethecat.com. Not sure how the website creator got those photos since many of them are unreleased ones from our build, but he did. On Phoenix we followed Steve's exploits with a mix of humor and confusion.

Nice story!

Phil

Low power generation is affecting surface operations:

Dust on the solar arrays has been steadily accumulating since landing, that dust has reduced the electricity generated and delivered to the landers batteries. Seasonal storms have have recently lofted dust into the atmosphere, filtering sunlight, and reducing the available power even further.

Mission data like electrical charge rates and atmospheric opacity (tau) is made available to the public via NASA's planetary data system (PDS). The most recent PDS mission data is about 3 months old, and the next update is not scheduled until January next year.

Mission Manager Reports (MMR) in the PDS document the electrical charge rates between landing and about 3 months ago. They record a general decline in charge rates, with a sharp drop in the most recent report.

As a result of the lower rates of energy generation, the team elected to disable the survival heaters in the robotic arm to save approximately 300 Watt hours (Whr) per sol. However at the end of June, power levels continued to drop so the team elected to place HP3 and some of the environmental instruments (APSS) to be placed into safe-mode. They also adjusted the trip levels for low power which would have placed the entire craft into safe mode.

Since then we've seen some of those instruments returned to service. We don't yet know if they were returned to service as a result of an increase in power levels, or if power was diverted from other services, but we did learn in the recent DLR HP3 blog that the HP3 radiometer (RAD) was not fully available due to power issues.

A selection of power levels reported in earlier MMRs is detailed below, but the power dropped from a peak at landing of >3000 Whr/Sol to less than 1300 Whr/Sol by the end of June. The sudden drop at the end was a result of a rapid increase in atmospheric opacity (tau) caused by dust storms. In the last report tau was ~1.3

* MMR for Sol 1: >3000 Whr/sol

* MMR for Sol 103: ~2800 Whr/sol

* MMR for Sol 225-232: ~1950 Whr/sol

* MMR for Sol 301-308: ~1900 Whr/sol

* MMR for Sol 376-402: ~2100 Whr/sol

* MMR for Sol 478-484: ~2100 Whr/sol

* MMR for Sol 519-525: ~1975 Whr/sol

* MMR for Sol 539-545: <1750 Whr/sol

* MMR for Sol 560-566: ~1300 Whr/sol

Click to view attachment

The image used here features one of the landers solar arrays.

* Sol 10: charge rate ~3000 Whr/sol.
* Sol 227: charge rate ~1950 Whr/sol.
* Sol 578: charge rate ~1300 Whr/sol.

The decline in power that Paul posts is similar in degree to that of Opportunity, and they are located at about the same latitude. Hopefully, Insight will experience similar good fortune with cleaning events that boost the solar panel production again. Insight, however, lacks the ability to pick a winter parking location that orients its panels at a favorable angle.

https://www.nature.com/articles/s41598-018-35946-8

Does anyone know whether after 684 SOLS the radio science experiment has been running for long enough to gain useful data? I remember that it was said that Opportunity would have needed to stay parked for more than two Earth years to gain more useful radio science data than it had already collected over its three months parking.

You could do the analysis at any stage but the results will be better the longer you wait.

Here:

https://ui.adsabs.harvard.edu/abs/2020EGUGA....5324B/abstract

is a description from May 2020 of the work needed to get good results. it includes lots of other modelling and incorporates atmospheric data, the orbit of Phobos (which is constantly being refined) and other things.

In other news, the scoop just pressed down on the soil pile made 2 weeks ago. Here is a visual summary of the last 150 sols of activity with the mole.

Click to view attachment

Phil

Re: RISE results:

here:

https://ui.adsabs.harvard.edu/abs/2019AGUFM...B0026F/abstract


The authors reported nearly a year ago that they already had better results than before.

Phil

Re: RISE results:

here:

https://ui.adsabs.harvard.edu/abs/2019AGUFM...B0026F/abstract


The authors reported nearly a year ago that they already had better results than before.

And here:
http://www.ursi.org/proceedings/procGA20/p...ctLeMaistre.pdf

a month ago they are reporting results - except these links are to abstracts without numerical results rather than full papers. But you can see that the work is ongoing and will soon be out.

Phil

Finally! Good news.

P

Sol 700 arm activity animated GIF (8 IDC frames).

It looks like there are gaps in the timestamps, so more images are likely to follow

(reduced to 800x800 to fit upload limit)

Click to view attachment

I missed the NASA article from 1 month ago that forecasts no hammering until 2021
https://www.jpl.nasa.gov/news/news.php?feature=7765

If they do two rounds of scrape-position-tamp at the 2 week arm-movement cadence, yup, that will take awhile.

Return to a weekly cadence to speed things up? On sol 700 soil was scraped over the hole again. On sol 707 the arm was placed over the mole just above the surface, probably to press down and compact the soil again.

Phil

No, no change in the 2 week operations. On sol 720 the scoop has just been pushed down to compress the sol 700 soil pile.

Phil

processed GIF using the 3 available IDC images
Click to view attachment

'End Game for HP3'

In the last paragraph of this Nature.com article dated December 15..


Click to view attachment

Bummer.

Yes, indeed.

Phil

Well, the "no digging till 2021" is now combined with "if that doesn't work-- that's it."

Here's hoping it works.

As a baseball fan, I'm familiar with the state of affairs where you're behind but suddenly win on the last play of the game. I'll hold out hope until that final attempt concludes.

If it does fail, I hope that a heat probe can be flown on a subsequent mission in the foreseeable future.

The rheology of the Martian soil appears to differ significantly from expectations.
Wouldn't it be worth to redefine the science objective of the mole (and of the scoop) to learn more about the mechanical properties of Martian soil?
Those data might turn out relevant for future heat probes as well as for rover design and operation.
Who knows, maybe the results could even be used to resume the primary science goal, provided, of course, the lander will stay operational for long enough.

Some of the speculation from the team included the possibility that the soil at the landing site may have differed, crucially, from that at other locations. Obviously, there's no way to know this with any detail or certainty, but it could be that simply landing somewhere else might have led to success of the mole. Certainly, the properties of Earth's soil different enormously from one location to another, and there's no doubt that this is also true on Mars, but we can't know conclusively that the same failure wouldn't have happened at any other location.

As I mentioned far upthread, the first effort to place a probe in the lunar surface was unexpectedly difficult, and even after a redesign, the second effort was also unexpectedly difficult, and this was true even when there were astronauts in the loop to modify the approach on the spot.

Also, to be precise, the Deep Space 2 probes were in a sense one or two earlier attempts to get a probe into the martian subsurface, but we didn't learn the cause of failure in that case.

The mole was a rather idiosyncratic architecture – perhaps a wise one given the knowledge we had at the time – but there are other, radically different options. Perhaps the best path forward is not to refine the Insight mole so much as to try a different tack, just as the unfortunate happenstance of the Galileo probe entry site was solved not with another entry probe but a completely different approach (Juno).

Shouldn't we try to squeeze out as much science as reasonably possible with the assets that are already available on Mars, even if this might require to redefine the science objectives?
One reliable data point would be better than speculation.
Soil properties may vary elsewhere, but deposits of settled dust should be quite ubiquitious on Mars.
Think at future applications, for example, at kind of a hammering mechanism for a rover to self-liberate when getting stuck in such a dust deposit, or at a method to cross dunes with kind of a hopping mechanism rather than roving. Any such considerations would require data. Now is the opportunity to collect some of them.

But first let them perform their planned attempt to address the initial science objective...

What are you suggesting they do that they haven't done already?

In case the next hammering attempt won't work in the intended way, and the mole would otherwise be given up, I'd suggest to further narrow down the root cause of the presumed bouncing.

One activity would be an attempt to verify with the scoop that there actually isn't a pebble, bedrock, or another crust layer, i.e. perform some digging. The goal is to ensure that the behaviour of the mole is unambigously caused by the property of the fine-grained soil phases. A first digging campaign would be performed in a safe distance from the mole to study the general layering structure of the soil. If this doesn't find obvious obstacles, digging next to the mole may be considered. This latter location would also be the only option to dig, if the upper duricrust turns out that it can't be penetrated by the scoop alone.

Another conceivable explanation to be ruled out is whether the density of the mole is less than the density of the soil, which would cause bouyancy.

Other experiments would repeat (cautious) hammering with different inclinations of the mole in order to get responses within a more homogenious soil environment. The bouncing might be caused by the vertical gradient of some property of the soil. If such a gradient exists, it would be smaller for a less inclined mole, with a possibly favorable change of its behaviour, but at least with additional data about inclination-dependent soil behaviour. A less inclined mole would also be able to discriminate between bouyancy (parallel to the gravity field lines) and bouncing (parallel to the hammering vector).

Since I'm not an expert in rheology, but this is a broad field of ongoing research, a student or post-doc in rheology would certainly give more advice than I could.
As usual, any experiments should be sorted by increasing risk. And I'm of course aware that things are much easier said than done.

I'll speculate the termination of further investigation is primarily a matter personnel scheduling and budget.
Implementing these ideas would require a non-trivial amount of engineering time. And if "the system" is working
well, those engineers are already scheduled to be working on and receiving their funding from other projects in 2021.
I imagine that some amount of the project's reserves budget has already been consumed by mole engineering work done in 2020.
And reducing a projects reserves budget is another form of risk.

Yeah, soil investigations would be an XM sort of thing. But it might not pencil out.

Primary mission was to be 709 sols and we're past 730? Wonder what's up there.

Different topic.
In addition to the Nature article with the "final try" comment, I notice there's a JPL release from a couple days ago, which doesn't say much about the mole:
https://www.jpl.nasa.gov/news/news.php?feature=7802