P2K: So, February 26. What's special about today?

Adam Steltzner: Well, one of the special things about today is that today is a Tuesday, and it's a Tuesday on a week that I'm actually here on Tuesday, and I was able to have a team meeting, which we don't get to do - they used to be weekly and now they're about every two or three weeks because of travel considerations. So, it's really nice to be home and relaxing to be here.

Technically, with respect to the project, what's special about today - let's see. We just... we just have done... We're in a cost controlling mode right now, where we are getting close to being done, we're looking at how much money we have, and how much money we have to spend and how much money we think we need to spend, and those two numbers aren't the same. And so we are in a cost containment mode, and we've just made a decision today to swap out a couple of our aeroshell units.

We produce... In the entry phase of the mission, we have an aeroshell that encases the spacecraft and takes it through the initial part of the atmosphere and protects it against the heat and plasma of gas that we penetrate through as we come in, very, very hot and very fast.

And we've just made the decision to take our flight "spare" aeroshell and not cover it in TPS, which is the "thermal protection system," not cover it in this ablative material, and defer that for the potential emergency of needing to use it. And at that time we would put the TPS on. Typically we would have put a TPS on the flight spare so the flight spare is exactly like the flight units and just ready to go. And in this case we've decided to defer that final stage of preparation in which we coat the aeroshell in this cork-like material for thermal protection, and we're deferring that. In doing that we're saving quite a bit of money.

P2K: How meaningful is the fact that yesterday was the official start of ATLO, officially, by the books?

Adam Steltzner: Not terribly meaningful. (laughs) It is always the case - or at least in my experience, and I don't have... decades of experience - but in my experience projects are always behind schedule. The fact that ATLO started yesterday is a passing milestone that you don't have much time to acknowledge because you're busy getting the hardware that is late into ATLO into ATLO, getting the stuff built. So we're kind of nose to the grindstone and staying focused on the job at hand. So yesterday came and passed, and today has come and passed... and the milestone doesn't really affect... where we are in the project. We are moving to put that hardware into ATLO as soon as we can get it there. Then those are the milestones. When you deliver hardware, that's a great day, (smiles) and we're very close to delivering - well, we have essentially delivered the cruise stage. Pam Hoffman, the Cog-E (ed. Cognizant-Engineer) for the cruise stage, is a very happy woman, and she's walking around two or three inches taller than she was because she has that burden off of her shoulders.

P2K: What is DIMES? Where did the idea for DIMES come from and was that there in the EDL plan from the beginning? What does DIMES stand for?

Adam Steltzner: DIMES stands for - let me see if I can remember this-descent, image.... Dual image... Okay, it stands for Descent Image Measurement Estimator, but I don't know where the V went because horizontal velocity is supposed to be there. So I'm actually not sure. We refer to it as the "descent camera." DIMES, or the descent camera, is a system that came about for two reasons.

One, the airbags that we are using, the Pathfinder design, essentially, that we're using for this mission, have proven to fall a little bit short of our hopes in terms of their performance. We were hoping to see solid twenty-four meter a second impact velocity capability, and we are a little bit short of that, like twenty to twenty-two meters a second. That's one thing.

The second is as the project has progressed, we have been doing work, the atmospheric scientists have been working hard to try and develop better wind models than we used for Pathfinder to define the winds at our landing site. Pathfinder used a "winds aloft" model developed for the Kennedy Space Center, and it is the world's best "winds aloft" model, but it happens to be designed and built for use in Florida, not on Mars.

And so the atmospheric scientists have been working hard to develop a better wind model. The preliminary work, it's called "mesoscale modeling," the technique that they're using, and their preliminary work suggest that, in fact, it is definitely possible that we will (get) sustained winds, or constant wind, at any of these landing sites that we're going to. And that constant wind was not present in the wind models that were used when we built Mars Pathfinder, or the wind models that we used when we were initially building up our system. And in the presence of a constant wind, a steady state wind, or a steady wind, we have severe performance problems. (Grimaces) We have a hard time making our requirements. So what the DIMES system allows us to do is take two pictures of the surface as we are descending and, using information on the IMU, the inertial measurement unit, on the spacecraft, correct for the swinging and the pitching motion that the spacecraft's going through as it's on parachute, it's making its way down, and adjust those two photographs of the surface and lay them on top of each other as it thinks it should be. But in reality they aren't perfectly aligned, and the difference between what the IMU says the alignment should be, and what the actual image alignment is is attributed to horizontal velocity - velocity parallel to the surface of the planet - that we can't measure because it's associated with the steady state wind. We measure the attitude of the spacecraft in its terminal descent configuration, but we cannot measure it's horizontal velocity without some reference to the surface, and that's what DIMES does for us.

P2K: Is DIMES a brand new invention that was put onto the system as things progressed? Was it planned from the beginning?

Adam Steltzner: It was not planned from the beginning. I think you'll be speaking later on with Miguel San Martin ...I think it may have been Miguel's idea in the beginning; many of the good ones are.

It came about over the last about four months. We really committed to it at the end of last year, at the end of 2001. It was an addition that evolved. We'd been aware that it would be nice to make this measurement of what our real, absolute horizontal velocity is.

But we looked at Doppler radar, which is a possible way of doing it, and found that that looked very difficult and very ponderous to get through that in the amount of schedule that we had, and we put that to bed.

But as time progressed, this problem arose, and the problem of the steady state winds became more and more obvious. We had to find a solution. So we determined that we could take one of these cameras, which we were already having produced for the spacecraft and which we weren't using, and put it on the outside of the lander and make these measurements and get this horizontal velocity measurement through this system. We kind of uncovered that solution over about the last four or five months.

P2K: Is it a done deal, or is it - Wayne talked about a review board coming up to talk about that.

Adam Steltzner: Depends on who you talk to whether it's a done deal. We need to work... At the working level we move forward with it as though it's a done deal, because you can't get anything done unless you move forward as if you're doing it, we're doing it. There is some concern within the review community that there's a lot of risk in adding more to our plate.

You know, adding this DIMES system to our "to do" list. We have a very large "To Do" list and not very much time, and adding something to it seems like an unwise decision. There's been some people who have been critical of it. However, when you sit down and tell the full story of where our performance is without DIMES, and where our performance can be with DIMES, it is a very compelling story. So, you know, short of a desperate "descope" measure, I believe we're going to go forward with this.

DIMES has another feature to it that is very nice in that the way it integrates, the way software - wise and algorithmically it integrates with the rest of the flight software, is kind of in a component fashion. So if we do - I'm involved in mechanical engineering - if we do the work to implement the mechanical solution and get the camera on the lander, then we can do the algorithmic work and the tests and the verification work on the performance of DIMES and it will... make the choice as to whether we're going to use it or not at a very late date, conceivably as we're on our way to Mars. You can either turn it on or turn it off when you're half-way to Mars, without a problem. So that allows us - it mitigates the threats to the development of the rest of the spacecraft. By saying that we're going to do the real work, the yeoman's task on verification of this subsystem, in the cruise phase of the mission.

P2K: Mayor Koch of New York City used to ask his constituents, "How are we doing?" As you look at things, at Plum Brook the tests there are going better, but everybody we talk to says this is a really tough schedule. It's a really tight situation.

From your perspective, looking at EDL, how are you doing?

Adam Steltzner: Well, I'll give you... In terms of EDL, technically, in terms of EDL, the greatest threats that are out there to us - and I'm going to take away the programmatics for a moment, I'm going to not talk about the budget issues; I'm a technical guy (and) I'm solving a technical problem, I'm doing my best, we're doing our best to contain costs but I can't speak to what the threats to us are in terms of the financial - but in EDL the thing that threatens us the greatest, the biggest bogeyman out there for us, are the environmental unknowns, and that is to say we do not know the winds that we're really going to fall through on the way to the surface of Mars, and we do not know the surface that we're going to land on in a particular location on Mars. We have some decent guesses about the surface from MOC images, from Mars Orbital Camera (MOC), from observations, orbital observations, but we don't have a really good idea about the winds (laughs) and small details in the landing surface.

So the landing images tell us that we think we've got somewhere between, let's say, eight and fifteen percent rock coverage - that is to say, between eight and fifteen percent of the surface is covered in rocks. And then we use some extrapolations to think about what that really means in terms of the rock fields and what they really look like. But variations in those can have a large impact on our success. And when the MOC images come back and we get an estimate of an average rock abundance over an area, that doesn't mean that where we land might have a lot higher rock concentrations. It's just smeared out over the image site.

So the greatest challenges that the EDL team at, large faces, are environmental uncertainties. Environmental uncertainties are best combated with two spacecraft. So, I'm a little downhearted, right now, with the prospect that, to save money, we may descope from the two spacecraft to one spacecraft. Because we can do the best job on the face of the Earth, but if we go to Mars and if Mars ends up being different than we think it is, we can't be successful. And so, I'm really hoping that we do not descope from the two spacecraft to a single spacecraft mission.

P2K: Is there some kind of design review coming up in which that issue will be addressed head on, or is it a rolling decision that could be made at any time?

Adam Steltzner: Well, we are in the process right now of doing budgetary efforts to try and reduce our expenses. When those get done and we see what our new picture of the future costs are, I think the rolling possibility - it is a rolling decision, it's made at a fairly high-level - I think... at that time that decision might be made.

Adam Steltzner: (Equipment) starts trickling in at the beginning of ATLO. In other words, for ATLO, it's not as though the doors to the spacecraft assembly facility open up, and outside the doors are every single piece of spaceflight hardware that all marches in and starts the process. They trickle in, first the big pieces, then pieces get added on. In fact, there's delivery dates to ATLO that are not all on the start of ATLO because the guys assembling the spacecraft don't need it all at once, so they only ask for it when they're going to need it. And so there is this trickling in in any project, and typically in any project that trickling kind of slips past the dates that were the initially planned dates. And so we're sitting here on the second day of ATLO and we have not a lot of hardware being assembled because, partially by design and partially by the realities of this mission.