Unexpected Wheel-Test Results

http://www.jpl.nasa.gov/freespirit/index.cfm

This topographic map of Scamander Crater suggests that the RF wheel is already on solid ground.

Looking at that map with my untrained eye, it looks like the entire right side is on solid ground, but we know that is not the case. What's different where the RF wheel is?

When I wrote that "Hmm, maybe they should try to send some commands to Oppy's Joint-1 azimuth motor now... " it was meant as a humourous remark, but I guess it didn't work too well...

"The plan for Spirit on Sol 2116 (Dec. 15) is to command a drive. This drive will further investigate functionality of the right-front and right-rear wheels. The results are expected Wednesday. "

Sounds like they may tell the right wheel to do the same thing as the other working wheels, and see what happens?

If the RF wheel is working again, it seems Spirit will be living up to its name, just as Opportunity has done. Curious thing, names!

This is Incredible!

How about a test of Opportunity's RF steering actuator? It has not been used since Sol-433

Maybe a pebble got stuck in the wheel in 2006 and got dislodged recently
if so the same thing probably happened to the RR wheel

This could be a possible cause

Whether or not a rock caused it to stall
5 out of 6 wheels is definitly better than 4

You know what they say
"Always expect the unexpected"

The most interesting tidbit in the new update is:

Normal resistance. Wow. That's something the pictures couldn't show. That sounds darn encouraging, but they also say:

This is making me wonder if there's some connection between the RF and RR wheels. One breaks and the other is perhaps restored. If the RF is working, though, who knows how long it's been working.

About the crust or not on the right side, there's no really "solid ground" over most of the land Spirit has visited, including the current right side. But the left side is into Scamander, which is filled with a very loose powder under a crust. So the wheels can dig in on both left and right (we can see that in the rear hazcams), but when they dig in on the left the traction is much worse than when they dig in on the right. So my guess is if the RF is working, and even if it does dig in, it would still have better traction than the left wheels.

At the time they detected an open circuit in the RF wheel. Presumably that meant a wire or connection had broken.

I think the link should be posted, even though most know where to look by now: http://www.jpl.nasa.gov/news/news.cfm?release=2009-197



Just had to quote that again. This is certainly... suspenseful!

Fredk - I was thinking the same thing about RR wheel failure, and now normal resistance in the RF wheel, and a possible connection of the two. It seams pretty ironic - but then again it's probably jumping the gun abit, and as you said there is no way of knowing what the actual condition of the RF wheel was before the RR wheel failure.

I find it pretty extraordinary that they haven't tried turning the wheel for about 3 years.

Presumably the risk of trying it was deemed too high, but with the RR possibly dead, the risk was now deemed worth the potential reward.
The irony is that if they had tested the RF a lot earlier (and found that it still worked) then we wouldn't even be stuck here at Troy trying to skirt around home plate, ie Spirit would have been able to backup onto home plate from her winter haven. (or we would already be at Van Braun and never had to retreat to the north edge of home plate in the first place). This just shows that there is sometimes risk from being overly cautious and not taking risks. This is all 20/20 hindsight however...

It's wise to remember that the RF has not yet been proven to be truly fixed. Spontaneous malfunctions happen MUCH more readily then spontaneous repairs; entropy always wins in the end.

A few more details in the figure caption:

One full motor rotation is more than I would've guessed.

WAY more, actually. Certainly not an open winding lead (unless that's what got re-connected.)

Oh, this last bit of info is really good! It's absolutely not the same thing to have a contact for a fraction of a degree then for a complete rotation.

Motion...I have this stupid-giocondesque smile on my face...
Guess we'll make it to Land of Salvation sooner than expected hey?

Mars rover driver Scott Maxwell said in a recent Free Spirit update that being stuck at Troy was "...very much like your car breaking down right next to Disney Land".

I think that everything that the scientists had hoped to find in the wall of Goddard crater has been found at Troy. I think that Spirit will eventually get to Goddard. However, I do not think that anyone wishes that Spirit had not carried out science measurements at Troy.

How big are those wheel motors? 1,500 x gearing to my mind says that they're really small....

Oh, found some info on the motors:

http://trs-new.jpl.nasa.gov/dspace/bitstre...9/1/01-2004.pdf

Page 8 and figure 9 are of particular interest.

These plans may have been delayed (one day?) since I see only a "navcam movie" and a few pancam shots on tosol's downlink.

Thanks on that info! Very nice..

For those of us (me) that may not know what a harmonic drive gearbox is take a look at this page, look at the two videos. The animated one will immediately bring you up to speed (no pun intended).

http://www.harmonicdrive.net/reference/operatingprinciples/

and the gearbox catalog containing more info about the Silk Hat series.
http://www.harmonicdrive.net/media/support...shg-catalog.pdf

Yes, it really is a beautiful gearing principle! The outer ring having a different number of teeth than the inner ring, thus creating an *everlasting* gear box, because so many teeth are acting at the same time and the materials are hardly stressed.

Sublime engineering like this has given us such long MER missions....

"marsroverdriver

Unless this changes at the last minute, we'll be including the RF wheel (a.k.a. "Lazarus") in our next extrication attempt."

Then:

"marsroverdriver

Sequencing the next #FreeSpirit extrication drive (A-2117) now. Diagnostics, then relatively aggressive extrication attempt."

If the wheel works, "Lazarus" would be an excellent name, indeed.

Wow, Keatah, thanks for those links! For anyone still getting their mind around the relative motions of the harmonic gear components after watching the animations in the first link, check out the beautifully detailed explanation of everything on pp 5 & 6 of the pdf catalog in the second link. One or the other ought to do it -- and I must say it is totally worth the effort for non-engineers (like me) to appreciate such a marvel.

Are folks aware of any other particular tech that helps account for the rovers amazing longevity?

Of course a lot of it is in the painstaking engineering and testing of how the systems fit together and cope with the demands of the Mars environment, as seen in the paper that Oersted linked to. The Martian winds have also helped a lot (didn't some scientists expect the rovers to die within the first year as their solar panels clouded up with dust?). The operations team, over the years, has skillfully safeguarded and upgraded the initial investment.

But what about other elegant or unusual or just unsung technologies that enable the rovers to do what they do? Anyone have some pointers to info on that?

Yeah, I was puzzled by the lack of front hazcams too. But Maxwell wrote this just minutes ago when asked how much distance was commanded for the RF on 2116:

I would've guessed that they'd want to actually see the wheel move (and not just get telemetry), considering how unbelievable this is!

My oh my, I can't believe I'm reading the words "including", "RF", and "extrication attempt" in the same sentence! And the word "aggressive" on top of that!

Exciting times for Spirit indeed!

Lot's of good newsy bits here that I haven't seen as of yet.

"Engineers are currently analyzing an unusual voltage in the mobility system that could be the cause of the wheel troubles on Spirit.

"There's a voltage we measure between what's called the spacecraft chassis and the spacecraft electrical ground, and that voltage is normally zero and it has been zero for most of the mission," Callas said. "We're now seeing a voltage show up on that that seems to be correlated with the drives." "

http://www.spaceflightnow.com/news/n0912/16spirit/

And this clarification from Maxwell on the 2116 vs 2117 question:

Re: "...the A-2116... drive was pulled at more or less the last minute..."

Hmmmm. Why.

Any one of a thousand possible reasons. Have a perusal through 12 months (nearly) of Scott's 5 year retrospective blog and you'll see the complexity involved at every step.

Just curious as to which one.

While we are waiting for the next suspenseful update from Mars ( I cant believe we have been on this fantastic ride for 2100+ Sols!)... Here is an animation of the RR buried wheel (2072 -- 2095). Shadows are reduced -etc.
Click to view attachment
Cheers

In the words of John Lennon... "So this is Christmas"... Seems we MIGHT be getting an unexpected present under the tree on the 24th! ...Or in the sock on the 25th...

Uh-oh. That usually means a deteriorated chassis ground. Those are really good at causing weird, intermittent buggy failures. Possible that might also relate to the memory anomalies as well.

Thanks Nick, I was wondering about the significance of that. I'm now wondering
(1) what could cause a 'deteriorated chassis ground' (still a broken wire, but not in the wheel actuator?) and
(2) implications for system reliability? Could we be gaining a wheel (or half a wheel) but losing volatile storage? If the latter, ISTR the workaround is to uplink each day's new data on the same day, which as long as MRO's out of safe mode would seem like not such a problem (?)

Wish I knew enough about MER detailed design to comment with any sort of authority, man, but here are some general thoughts.

The basic idea of a frame ground is to establish a single reference point for all voltages. For example, it's important that 5 VDC "highs" for electronics/logic functions are at the same level from box to box, since signals will be transmitted between them via a data bus; if received signals show a 3 VDC differential or something like that (since the receiving transceiver will be powered through a different circuit & interpret incoming signals with respect to its ground voltage level), the electronics will get, uh, confused.

To achieve this uniform reference, all components of a system are normally grounded in some way to the overall vehicle structure. Some are literally bolted right onto the frame; some use "ground straps", which are usually heavier-gauge wire attached to the frame. This protocol also applies within boxes, of course, which is a good visual aid for the concept of "systems of systems".

So, in answer to your question, it's very hard to predict system behavior in this situation. What's particularly difficult is distinguishing secondary effects from direct symptoms of a loose ground; unless the faulty chassis ground is associated with a component with a very limited function set (output), the damn thing can cause some really bizarre system behavior in direct proportion to the number of direct & secondary interfaces (uses/consequences of the output). It's a function of both complexity & level of integration.


(Jeez, I hope that made some sort of sense! Sometimes it's hard to describe things like this.)

EDIT: Oops, forgot to answer your question about root causes, sorry! Interestingly enough, the Martian environment precludes the most common terrestrial cause: corrosion. Takes liquid water for that, or high ambient water vapor (plus salt ions really accelerate the process; you can literally WATCH grounds corrode on ships in tropical seas!) For Spirit, things like a connection working loose or perhaps even dust intrusion into a sealed connection that might've become less sealed due to thermal cycles would be more likely. The main idea is to make a formerly highly conductive interface less conductive.

Some tracking data for the sol 2117 drive is available now.

About 5mm forward, but about 7mm down if you believe the numbers.

Can't wait to see some images...

Wow!



Woah!

Pictures are required

Isn't down bad?

Pics up!

Definitely some spinning of the RF! But not much movement.

http://qt.exploratorium.edu/mars/spirit/fo...86P1214R0M1.JPG

Pictures are already available and the RF wheel is moving!
I was expecting to see it heavily eroded but it's not.

Look at that RF wheel spin!
http://qt.exploratorium.edu/mars/spirit/fo...cam/2009-12-17/

An animated GIF.
Click to view attachment (1.6MB!)

Edited: Made it a little bit slower (5 fps instead of 10 fps).

A Xmas miracle...it has to be...
Go Spirit GO! GO! GO!

Thanks for that, Tesheiner, you can really see Lazarus spin in that animation!

The RF clearly dug in somewhat when it started spinning. But looking at the frames, it looks like the digging in was almost entirely in the first 2117 frame. Then there's no further visible digging in between the first and the last frame. So it may be that the wheel was just establishing a new equilibrium in its driving mode. Perhaps further driving won't see excessive sinkage.

Regardless, driving with 6 wheels minus RR is going to be different than driving with 6 minus LF. I would guess we'll see some brief testbed runs to determine the best strategy. Brief since winter approaches.

Keep gunning it !!!!

My recollection is that there's a fuse between chassis ground and the main system ground and that this fuse ended up being blown before launch (this was described in ROVING MARS, as I recall.) So this sounds like some kind of triboelectric charging thing, perhaps caused by the wheels spinning, and not a recent deterioration. But I could be misremembering.

I was just thinking of the pre launch pyro hunt and the concerns about shorting out resistors... would it make any sense that maybe these were weakened and maybe have shorted leading to these strange voltages?

I suppose it's probably more likely that our girl is just getting old. I guess it's something we may never know.