GAO Sees Improvement With Webb
James Webb Space Telescope: Actions Needed to Improve Cost Estimate and Oversight of Test and Integration
“The National Aeronautics and Space Administration (NASA) has provided significantly more time and money to the James Webb Space Telescope (JWST) than previously planned and expressed high confidence in the project’s new baselines. Its current cost estimate reflects some features of best practices for developing reliable and credible estimates.”
Launch isn’t scheduled until 2018,when its finally lofted it’ll take nearly a week to be inserted into its orbit , full deployment then its lifespan is projected for 5 years.Is this a worthy investment?
JWST won’t start to operate until it has thermally equilibrated at Earth-Sun L2. It’ll deploy just outside of LEO, but will take several months after launch to get to L2. Once it is there, it will take many more months to do full system checkout. Working launch date is October 2018, so we won’t see any results from it until well into 2019.
The worth of JWST is going to be measured by what it accomplishes, not how long it operates. Scientifically, you could simplistically say it is hugely economical compared to suborbital science missions! Note that HST was in the development stage for longer than JWST is going to be. In fact, HST was only really functional several years after launch. FWIW, JWST probably has enough consumables (mostly stationkeeping propellant) to last for ten years, though it is designed for five.
It’s another question about whether what JWST is likely to accomplish is going to be worth $8B.
Of course Hubble cost $3B in older dollars, *plus* the cost of the shuttle servicing missions. Was that a worthy investment?
Then Hubble sat in storage for years awaiting a shuttle ride. Then when they found out the scope was nearsighted, the critics went off the edge of the cliff predicting the end of NASA Big Science. So we will see. The only problem I see is that if we have to fix the thing (JWST) it’s going to be a long, long ways away to do it. It is expected that the JWST will provide some answers concerning the Big bang theory and could also make a paradigm shift in Physics. Or not.
JWST could be returned to Earth-Moon L1 or L2 without too much trouble or propellant. That’s well understood. So the “long, long ways away” is not necessarily as “long” as you might assume. That being the case, we’re still not likely to send astronauts or telerobotic surrogates to fix it. Why? Because the damned thing wasn’t designed for servicing. There is a gratuitous docking adapter, but docking is the easy part.
Now, if it were a deployment problem, a swift kick might do the trick. But anything else is going to be hideously hard to fix.
The decision not to design JWST to be serviceable is hopefully one that won’t come back to haunt us. Of course, a serviceable JWST would have been far more expensive, and a servicing mission would be fabulously pricey as well.
As to NASA BIg Science ending, that’s arguably already happened. As a result of huge and unanticipated JWST cost burdens.
Of course, a serviceable JWST would have been far more expensive, and a servicing mission would be fabulously pricey as well.
In all likelyhood a serviceable JWST would be cheaper than the $8 billion.
In all likelyhood a serviceable JWST would be cheaper than the $8 billion.
Why do you think that is likely?
Gonzo
Sorry gotta answer up here. We did a study for DARPA on this in 2004 and Roger Angel of the Steward Observatory at the University of Arizona did as well. An on orbit assembled telescope that then left the station and transported to any other orbit would have been considerably cheaper than JWST.
Our DARPA study, which was bolstered by a LOT of work by NASA Langley, showed that a 14 meter telescope (one meter segments) with a diffraction limit well into the visible (lamda over 10 at 550 nm) was quite doable and a lot of the assembly task timelining as well as in pool work had already been done in the early 90’s.
This telescope could be built for no more than a billion dollars.
That’s hard to believe.
Serviceability layers a mission with many new design requirements — subsystem accessibility, connection compatability, and disconnect survivability. Also human-rating. That’s all very expensive. The only way you save money is by backing off on reliability, which is never a good strategy.
As to the DARPA study, I am reminded that Rogel Angel was part of the team that originally proposed JWST, back when it was a $1B mission. With all due respect, and not to lay the present budget fiasco entirely on his back, why should I believe an early cost estimate by him on a serviceable JWST? Was that cost estimate even formally vetted with standard cost models?
What serviceability (whether by humans or telerobots) WOULD add, as it did for HST, is potential capability. That’s a scientific decision, as to whether upgrading this particular existing telescope has more value than building a new one.
Helen
Roger Angel’s ten meter telescope that would have been assembled at ISS had a cost of about $400 million. The biggest problem with JWST has been to get the segmented mirrors to remain aligned after launch, something that is trivial for an on orbit assembled system. A senior GSFC scientist told me that it cost $500m just for the risk reduction exercise on JWST.
I also know a LOT of the details about what has gone on inside of the program at Northrup and it is not pretty.
JWST will be the last ground assembled large telescope.
Most of the servicing done on Hubble was done in the areas that were not designed for servicing, yet Grundsfeld and company made it work.
We have to quit living in the past when designing space systems. I was at DARPA this past week having a nice discussion on why spacecraft morphology has not advanced in 40 years. That must change.
“JWST will be the last ground assembled large telescope.”
I think we’d all like to believe that, and large telescopes should eventually be assembled in space (not necessarily by astronauts, and not necessarily in LEO), but we’re pretty far from that capability right now. In fact, the next large telescope is much more likely to be one assembled on the ground, and stuffed in an SLS, making use of the spacious 8-10m diameter launch shroud.
Understand that assembly and alignment of JWST at ISS would not necessarily have been optimal. JWST will be fully deployed AFTER ejection out of LEO. Why? Because the g-stresses from that propulsive maneuver would not be tolerated by the design. So you can tweak the alignment in LEO until you’re blue in the face, but you still have to kick it off to L2 and cross your fingers that the alignment will be maintained. Yes, you could make it a lot stiffer and heavier, and yes, you could send it to L2 more slowly, perhaps with an SEP. But that’s not JWST.
Again, the issue is whether science is advanced by having a upgradable telescope. In the case of HST, that was certainly the case. Not entirely clear about JWST.
$500M for risk reduction in an $8B telescope is money well spent. In fact, recent NRC cost-control recommendations for SMD missions are that investment in risk reduction should be expected to come to 5% of the target cost. That’s pretty close to what we see in JWST.
Yes, you could make it a lot stiffer and heavier, and yes, you could send it to L2 more slowly, perhaps with an SEP. But that’s not JWST.
With a thrust of no more than a few newtons, a SEP is the means whereby a very large telescope could be moved from LEO to wherever. The NASA Langley built structure from the early 1990’s was easily able to handle this scale of loads and that was a 14 meter structure. This telescope structure was built and timelined in the pool at MSFC along with test segments.
Also, I will bet you a large coffee that SLS will never fly an operational mission.
Much of the cost of JWST comes from designing, building, and qualifying the instruments and other items for the first time. The cost of building and launching a copy would be significantly cheaper and would make more sense than trying to perform a dubious servicing mission if necessary.
And as for the politically mandated docking adapter: that thing is long gone.
Copies are only inexpensive if built at the same time. There is no such thing as “build to print” in space. Inevitably, there are components that no longer exist and need replacement, idiosyncratic knowledge that didn’t make into redline drawings, etc.
Not to mention that “new science” (i.e. collected with new instruments) is usually more compelling than “incremental science” (more of the same data). The possible exception is where you are trying to capture a long term data set (over decades), because the value is in the length of the data series, not in the individual measurements.
There is a theory that the universe is infinite and that we really see very little of it. And, IIRC from the Astronomy magazine article that I recently read, it could also mean that the Big Bang could have been a relatively “local event” in an infinite universe. So JWST *could* help provide some guidance or corrections on the Big Bang theory as well other phenomenom that are puzzling astronomers and physicists. Such as where is the universe getting its energy to continually expand? After all, it takes energy to create more space. So yes, I think $8 billion is worth it. I operate under the theory that ignorance is not bliss. And it could, somewhere down the line, possibly, maybe, improve future generations’ lives here on earth and in space.
Would anyone familiar with JWST please tell me if there are any full scale thermal vacuum tests of the final assembly? My understanding is NASA is not using the SPF chamber at their Plum Brook Station; and that SPF is the only thermal vacuum chamber large enough to test the entire final assembly. Thanks.
The integrated optics and instruments will be tested at JSC’s Space Simulation Chamber.
The SPF at Plum Brook Station was investigated early in the program but ruled out for several reasons.
why would SPF be ruled out? Testing the optics & instruments packages is a good start, but shouldn’t the entire assembly be tested if possible?
Bobby,
That question is always asked too soon. We can only ever know after the fact whether something was a “worthy investment.” That’s how science is, right? If we knew the answers beforehand, we wouldn’t have to do the experiments. Only afterward can either the science or the experiment be evaluated. And any “value” assessed is subjective. To some JWST findings will turn out to be wonderful; to most people they probably won’t be of any interest at all. So, how do you measure it?
Steve
This seems to me one of the few really worthy projects NASA is currently working on. I hope, for the sake of the science and also for NASA’s morale, that they get it done on time and on budget. Not holding my breath, though.
Huh? They already went overtime and overbudget long, long time ago. Moving goalposts will not change anything.
I know; I meant the new deadline and new budget. Of course I agree that it would have been nice had none of that been necessary. But given the circumstances, I’d still like to see this project salvaged. This would be of major scientific value — unlike a lot of the flimsy climate stuff NASA seems to have gotten involved in somehow. Just my two cents.