Well, duh! The design did not work well at all, in hindsight, when it came to getting sample into the instrument.

That's the point Mike was making. It was probably the best stab we could have without knowing the exact properties of the soil in advance.

Doug

Let me know when you guys develop the ability to predict the future, because it would be a great help in mission planning.

Honestly, I think the jury is still out on how well TEGA will end up meeting its mission goals, and whether the design and fabrication were as good as they could have been given reasonable expectations of soil properties. I doubt if the TEGA team will be pulling any punches in the self-criticism department when they write their final result papers.

It would be easier to be equanimous about TEGA if it weren't for the development history of the instrument. I seem to recall that the sample delivery system went through several design iterations.

First, IIRC, there was just an open path for soil to travel to the ovens. Then the concern was raised that larger particles might block the soil chutes, so the screens were added to block all but the particles of appropriate size to successfully reach the ovens. Then it was determined that the desired soil grains wouldn't just fall through the screen, the larger grains would block up the screens, so the vibration motors were added. Then someone noticed that the ice that could be scraped by the rim of the RA scoop wouldn't pass through the screens, so the rasp was added to the scoop to break ice down to particle sizes small enough to pass through the screens (which also served to increase the surface area of the ice enough that it measurably increased the sublimation rate, making it that much more difficult to get it from the ground into the ovens before it all sublimated away).

See where I'm headed with this? The design concept itself started out rather short of being functional, even based on estimates of soil properties that we could easily determine pre-flight, and the instrument design was tweaked several times to try and make it work. Rather than starting with an instrument that was designed from inception to be able to handle everything we could imagine with a good amount of performance margin, we started with an instrument that failed to handle anticipated conditions and was tweaked several times to, with its best possible operation, push performance so that it could achieve its desired function -- with very little margin for error or unanticipated soil conditions.

I'm not finding fault here so much as I'm pointing out that, with extremely tight mass and monetary budgets, it is absolutely essential that you design your systems from the get-go with as much margin for error as possible if you're going to assure successful operation. If the last phases of design and development are spent pushing the system just far enough that its best possible operation just barely covers the requirements for success, you're courting failure. I mean, just look at the number of single-point-failure systems (particularly pyro events) that could have transformed either or both MERs from the incredible successes they have been into short-lived, frustratingly unproductive stationary lander missions. That's a good example of a design that required best operation for success -- and it was the element of the mission that probably caused the greatest intake of antacids amongst the MER team members prior to having all 12 wheels in the dirt.

-the other Doug

To the question, "Is TEGA in 20-20 hindsight poorly designed?", I just wondered how you could say "I have no idea". Of course, in hindsight it didn't work well where sample delivery is concerned. I'm not saying they should have known in advance.

Because I don't even know how successful or unsuccessful TEGA has been, much less how much of that is due to poor design and how much to random variation in circumstances.

Of course, maybe you know a lot more about the instrument design and all of the tradeoffs and how it's behaved in flight than I do, which is entirely possible. If not, then I'd say you are jumping to conclusions if you think that TEGA was certainly "poorly designed."

Exactly! Tradeoffs! If we had thrown money at all the soil adhesion worst-case-scenarios, would the instrument that launched have been substantially different, or was the end of the design cycle looming? There is a difference between anticipating every eventuality and designing for them. As dvandorn said, a robust design tolerates more eventualities, but when you're looking at an instrument with eight separate ovens rather than one superoven, it may have been reasonably assumed that there was error tolerance in redundancy.

Winston Churchill stuck to a very clear policy during the thirties, in the years leading up to the calamity of WWII. When looking back, he only lambasted the policies that he had criticised BEFORE they were enacted.

I am still trying to figure out what kind of ex-ante design would have been sufficiently robust, in line with otherdoug's contention that it should have been able to deal with a wider range of unknown sample properties. MSL will be able to zap rocks with its laser and sniff the vapors, eliminating all those transportation and deposition headaches. This could probably have worked with ice too, but the power needs may exceed what solar alone can supply. Or the deliveries could have initially been made into to a "coffee grinder" to pulverize and homogenize the contents before passing them on to the ovens. Then how to clean up the grinder to avoid cross-contamination of samples? Would one of these solutions, or the many other alternatives that have been proposed stand out enough to have been an obvious choice prior to the mission? I doubt it. But use of the word "failure" just refers to the fact that at the end of the mission we probably won't have answers about the isotopic composition of the water ice and what is dissolved in it. It does not require pinning blame on anyone, and should not be taken that way, (OK - except for the case of the doors not opening properly). Most aspects of the mission have been hugely successful.

Oven 6 got the multiple samples named Rosy Red, the latest Rosy Red N and finally Rosy Red N+1. I believe they came from the Snow White neighborhood, Burn Alive Trench.

Emily Lackdawalla has a good trench map and sample list: Planetary Society Weblog: Catching up with Phoenix to sol 133 [...]



I've followed closely and haven't seen success with OFB sample to oven #2, though I suppose they could be keeping it a secret. My scorecard at the September 29 briefing was:

#4, #5, #0 and #7: used.
#1 has had two delivery attempts of icy soil: one missed and one clumped on screen.
#2 appears to have since been targeted once for the Organic-Free Blank delivery, but that did not succeed.
#6 is barely open. (Now, successfully delivered.)
#3 is unopened.

An auger that went through the sample into the oven, maybe. I'm not an engineer.

If the ice table is interacting everyday with the atmosphere, is there much else expected to be found in the ice than in the soil on it?

The big hope was to analyze the water molecules themselves (and the soil contains little or no water), specifically the Deuterium/Hydrogen ratio. By the way, I see in this paper that TEGA was also to measure D/H values in the Martian atmosphere but don't recall hearing anything about this as far as the actual mission is concerned.

D/H FRACTIONATION IN THE ATMOSPHERE-GROUND ICE SYSTEMON MARS
"The main objective for this work
is to investigate the solid-vapor fractionation
processes of Deuterium/Hydrogen (D/H) in the
ground ice- atmosphere system on Mars....

...Investigating the stable isotope ratios in water
vapor and ice (e.g. in the Greenland ice cores) have
proved to be a valuable method for understanding
the past climate on Earth....

...With its mass spectrometer, the TEGA instrument
will measure D/H values in the atmosphere
and in the Martian subsurface."

It didn't before winter, but now there's frost over it. If they could analyze a good bit of frost, the ratios could be meaningful if that's the same water they are scraping from the ice table. They won't need to use the rasp with the frost, so the scoop won't get warm.

I think the D/H ratio from the frost would be the close to the same as for the atmospheric water vapor.

(OK, to be overly anal, there should be a slight theoretical increase in the D/H ratio for frost due to the heavier molecules fractionating out on deposition....)

-Mike

The whole point of the D/H measurement for both the ice table and the atmosphere is to see if they are different, because that would indicate whether the ice table is ancient (possibly from an earlier ocean) or recent (from H2O frost or snow). It would not be worthwhile to substitute frost for the ice chips because that would defeat the whole purpose of the measurement.