The Spirit briefing is about to begin. I'll post notes as soon as it's concluded (If I can keep up enough to take a second to post, and there's anything really good or really bad to report, I'll get something up sooner).
Really rough preliminary note - excuse the spelling and grammar :-) You can read the "official" press release here.
Pete Theisinger: We have been able to command SPirit and have gotten limited response. Flight software not behaving normally. When I left you yesterday I said we had recieved a response from the craft and that was correct. We attempted to command the craft to send us telemetry. That command did not work because it was too late in the day. This morning we sent an early beep and did not get a response. When we were about to send a second, it talked to us. We got 20 minutes of transmission with a single frame of engineering data. We received another 15 minutes of data. Spacecraft attempted a short comm at end of day. It's in a processor reset loop. Processor wakes up, loads the software and waits for 15 minutes twice and resets. The indications we have are that the things that causes the reset is not always the same. We know that the sequence that began wed mroning to calibrate Mini-TES motor, that didn't finish. Craft is in an x-band fault condition could be failure to move high-gain antenna. 1 oclock wed craft did not believe itself to be in a fault condition. Team is taking data it has collected and moving forward and analyzing what they no to prepare a plan of action. We believe that we can sustain the state of the craft from a health point of view for a long time. No indication of a power or thermal point of view. Indication that we're not going to sleep at night. An anomaly team has been formed separate from Opportunity team. They will be working on schedule from 500 mars time to 1500 mars time. They'll be coming in tonight at midnight. First, examination of data and go forward plan for the day, then on consoles for the day, the post planning. I expect this to go on for several days, talking to thecraft, gathering data, making theories, testing theories. Many days perhaps a couple of weeks before we restore spirit. Opportunity coming up. We've made appropriate personell adjustments. It is likely that we may not continue the opportunity impact to egress at the same pace as we made with Spirit. That's a decision the project will make in consultation with management.
Wayne Lee: Landing Opportunity on Saturday night. This is going to be extremely challenging. Highest altitude landing NASA has ever attempted. Working around the clock reconstructing what happened with Sprit EDL to figure out if there is anything different this time around. Reason for that is that what happened 3 weeks ago was the first time this actually happened with EDL. All systems were analyzed and behaved exactly as expected. Only one area it didn't, in the descent rate limiter. As you know, a safe landing depends on interaction with atmosphere. Surprises in Martian atmosphere, two gusts of wind. One at 40,000 feet at mach6. Clear air turbulence caused the capsule to wobble more than expected. Also had a big gust of wind at the bottom and had we not countered that we could have been driven into the ground at 50 mph. Also atmosphere thinner than we had hoped. Global dust storms. Bottom line is what are we going to do different. Almost nothing. One change. Deploy parachute 4500 feet higher than Spirit. Reasons are three-fold. Higher altitude landing. Atmosphere variability. We meant to guard against that variability so system will have more time to fire retro rockets. Third reason the decent rate limiter. Confident but respectful that this is a risky endeavor.
Miguel San Martin: I'm going to tell you the story about the second gust of wind. The MER EDL is based on Mars Pathfinder design. That design had a sensitivity to horizontal wind. If you encounter a gust, the system starts to oscillate. If we fired rockets while at an angle, we could add horizontal velocity. That combined with the normal velocity from the wind, might add up and airbags don't like large horizontal velocity. So we added a system that consists of a set of sensors and steering rockets in the back shell. We assumed the worst about wind and implemented this system. I'm going to show Spirit landing data. The answer was that we did get a gust of wind, up to 50 mph in the last 1,000 meters, the system did activate and compensated successfully. In this image we show path of spacecraft superimposed on DIME image. The blue arrow is the velocity of the system due to wind pushing the system to the east. The purple arrow is the velocity we would have had if we hadn't fired the rockets. The yellow arrow pointing down is the velocity that we actually got. If you look at the size of the large arrow, we went from a predicted 52 mph to 25 mph. We are very pleased with that. In the next image we can see the reconstructed trajectory. Next is an animation using the telemetry from the craft.
Adam Steltzner: Another animation, high resolution video. Almost at the surface of Mars at realtime, real speeds. You can see the rover pushed to an angle and then corrected by the rockets and coming to a stop 23 feet from the surface of Mars. We're trying to learn everything we can from Spirit. One difference was the performance of what's known as the descent rate limiter. We had anticipated a deployment time of about 6 seconds and telemetry shows 11 seconds. Forced us to examine telemetry and that subsystem and we've concluded that the most probable cause is an increased breaking friction in the breaking device attributed to the 7 month journey to Mars and the loss of volatiles inside the breakpad material making it perform better than we had anticipated. Improved breaking slowed the separation. We've folded that in to our understanding of the Opportunity EDL process and we expect Opportunity to display a similar behavior. That, atmospheric data and altitude all folded in to the decision to release the parachute earlier. We feel comfortable with our understanding of that. We've done this reconstruction effort in an attempt to understand exactly how the craft performed and Dr. Malin has the photos that let us know how we performed.
Mike Malin: All the data that these guys are looking at were relayed through MGS. MGS obtained lock before landing and retained lock throughout the rollout period so we had telemetry coming in. Also, as principal engineer for MOC, I've been working on getting a picture of the landing site at high resolution. This is the DIMES view. Next slide shows the changes. See bright dot in lower portion of image and two dots in the upper left and a dark streak. Lower dot is the lander, upper two dots are the backshell and the parachute and the dark streak is the heatshield impact and slide down crater wall. We got all the bounces too. Then there's a rollout streak. Tim Parker was within 5 meters when he told you the location of the site. We have a positive location to within half a meter. We can see orientation.
Pete: I've been in this business for a long time and this is amazing stuff.
Mike: We'll get the rover in future pictures.
Q. Pete, we're all interested in what data you got.
Pete: Helps us a great deal. We get a set of error messages for software, we get engineering on subsystems, gives us pointers on what information to ask for tomorrow. It tells us that the software is in x-band fault mode. We surmise it got there because of HGA pointing. It gives us a little bit of a tell tale for what's going on with the processor now. Flight software not operating normally. We do not have assurance that the next time we go ask it that we'll get either one to the two different behaviors we've gotten. How come it's behaving that way. We've got quite a bit of information, obviously not as much as we'd like.
Q. Pete, translate that to a ground effort to duplicate what you've seen. If spirit was in the hospital what's its condition.
Pete: We are still critical. We do not know to what extent we can restore functionality because we don't know what's broken. I think it's difficult at this stage to assume we did not have some type of hardware event so we don't know to what extent we can work around that. We have been able to establish that we can command it, that the power system is OK and the that thermal system is OK and that's good. I don't expect us to get a big sea change in knowledge or theory in the next several days. Until we get a root cause with hardware it's very hard to duplicate this on Earth. There's two kinds of things that can get you in trouble. a hardware fault that impresses itself on software. second is software with a lot of modules talking to each other. We need a starting place to be able to set up the testbed.
Q. If it's mechanical... Can you say "what have we got that's good" and get at that through software.
Pete: I don't know. We do know we were doing a motor operation that did not run to completion. We don't know exactly what is good. We can get at engineering info that tells us that we're safe and stable but we haven't established that it's "good". We need to get software to get behaved so we can get more direct access to the functionality of the system.
Q. Pete, you mentioned earlier that Spirit wasn't shutting down at night. Depleting batteries?
Pete: Yes. But not a problem that will end the life of the rover because we can wake up on the solar arrays only. We'd like not to get into a low-power situation. We attempted at the end of the day to shut down the vehicle to conserve battery and keep thermal under control. It's not a duration threat. Batteries can go to zero and be recharged.
Q. how many lines of code in software. Is it rebooting and how many times trying.
Pete: I don't know. It has attempted to reset since Wed over 60 times.
Q. Is there a lot of science data on the rover?
Pete: I don't have a good answer to that question. We've gotten a lot of stuff back. There were some science products on the rover but I don't know how much.
Q. what motor.
Pete: motor inside for mini-TES. At cold temp (night) you need to increase voltage. We were going through a set of increasing current and moving motor. A sequence of events that the last command didn't complete.
Q. For pete, you said you did get engineering data back? Any reason engineering and not science data
Pete: good data. We do not expect to get MGS or Odyssey passes trying to shut down. Data tomorrow will be around 10 or 11 Mars solar or 4:30 in the morning here. Engineering data because we got it broke and Steve looses out here.
Q. Play the parent. How concerned are you.
Pete: I don't view the situation that way. We have a serious problem. That it's stable will give us time to diagnose that problem. It's giving limited but good information. Engineers have data to chew on and they're off chewing. I expect to get functionality out of this rover. Chances that it will return to perfect are low. Chances that it won't work at all are also low.
Q. If it's not going to sleep...
Pete: it will kill things that we don't want anyway. It takes one or both batteries offline and that may be something we'd like to undo. Better for us if it doesn't go low power but we can cope if it does.
Q. Can you explain what you feel about Opportunity.
Pete: the team always has to be mindful that this may be systemic but it looks like a singular occurrence to me. Unknown if this has anything to do with the HGA.
Q. Pete, there are OSes that allow you to go back to a date when everything was working correctly.
Pete: We can attempt to reboot out of a better image but we don't know what we've got here. If we've got a hardware issue then this really isn't a software fault. We are resetting the computer. This is just like pushing the reset button. This is very close to pulling the socket out of the wall.
Q. You suspect hardware, is that the same as mechanical?
Pete: something outside the software, not necessarily mechanical thing.
Q. Predictions that it would ever be perfect or not working at all. What about the chances of obtaining science data.
Pete: science team working hard but until we know the state of the vehicle, forecasting isn't very fruitful. We have not shot the last arrow in our quiver. I'm cautioning patience. We need time.
Q. Given the clarity of these images do you intend to go hunting for other crafts.
Mike: I've had discussions with the Beagle people. Very surprised how bright this is. Not covered with dust. Good illuminations. We are going to try to take some images of the Beagle site. We did try for Polar Lander and the lighting conditions were much more poor. We are using a new 50cm technique. It would take about 60 years to cover the Polar Lander ellipse. We have imaged Pathfinder and VL1 and those pictures are on our site but they don't look anywhere as good as this but you can see the airbags on Pathfinder.
Q. are you focusing on the motor's electronics box or the entire rover.
Pete: symptoms look like it was triggered outside of the software domain. The fact that motor operation didn't complete leads you to look there first but we're not limited to that.
Q. Can you take a picture of the craft from it's cameras.
Pete: we don't' think it would be visible from external. Inside the system.
Q. Could one of you describe the differences in altitude of the two sites. Can you show us again where that motor is?
Wayne: Meridiani 1.3 KM below the reference surface. 4000 feet below. Gusev was 1.6km so 1000 feet lower than where we're going on Saturday
Pete: (points out the area on the model rover)
Q. How has what's happened with Spirit heightened anxiety.
Pete: I haven't asked the team that question. Personally it does not change my anxiety at all. We want very much to have two successful missions here. We're committed to do the best we can. If we get a root cause that leads us to believe it is systemic, that might change our sequence on Opportunity.
Q. How are resources being diverted.
Pete: a couple of people are moving. Anomalies to engineers have a gravitational force. We're making sure that the people are where they need to be.
(Schedule for tomorrow will begin at noon PST. At 2PM Mars program overview. At 7:30 PST live coverage for 9:05 PST landing.)
Q. Could you put the trouble you're having with Spirit in context with Bush's call for NASA to put humans on Mars.
Pete: No. I'm happy to talk about robots on deep space missions because I've been doing that all my life and that's my area that's my area of expertise but that's as far as I'm willing to go.
*End of briefing*
Posted by asa at January 23, 2004 09:58 AM