Thanks for the info–interesting. It sounds like they got “lucky” with regard to axial loads, in that the ET was so overdesigned to handle the bending loads that it did not need additional reinforcement. Of course, this implies that the STS center tank may still be overdesigned as compared to a clean-sheet design, which is still a problem.
It is interesting but not surprising that the thrust structure would be a larger challenge. The F9 octoweb is the central load structure for that vehicle; the rest of the design is effectively built around that. The idea of going backwards, starting with the tank and retrofitting a thrust structure on that is clearly an immense engineering challenge.
As you know, the reuse of ground support facilities for SLS didn’t exactly go as planned. NASA has spent ~$1B retrofitting the “mobile launcher” tower for the SLS. It was supposed to be a savings, but almost certainly wasn’t. Also, it leans to one side. It will only be used one or perhaps zero times.
It isn’t that the External Tank is overdesigned per se; it is just that the axial thrust load does not dominate the overall shape or wall thickness. This is unsurprising since the peak thrust load doesn’t come from the aft-mounted engines but the SRBs which are mounted in essentially the same way for the Jupiter as the were on the Shuttle. The mass of the CSM and the upper stage is such a small portion of the overall mass that it only impacts the design of the interface structure, and since the forward ogive is being eliminated anyway that can be redesigned into whatever structure is necessary. Bending and local buckling tend to dominate failure modes and dictate wall thickness.
The problem with the SLS isn’t any particular design choice or feature; it is that it is a vehicle without a purpose. The system isn’t meant to serve as a launcher for larger interplanetary explorers (too expensive, flight rate and schedule too restrictive), and as a crewed launcher it is overpowered for LEO and yet still underpowered for any kind of serious interplanetary crewed missions or even a serious effort at a Lunar outpost for whatever value someone might imagine in that. Like the Shuttle, it is the result of a political process rather than an end goal and I’ll be surprised if it flies more than a handful of times before being retired.
I’m not sure I fully agree with that. Single-point design decisions can have such a cascading effect as to effectively ruin a design. As you have noted many times, rockets live close to the margins. Spending that slim margin on politics rather than solid engineering choices is a mistake.
Although it’s dead now, the choice of an SRB for the first stage of Ares I had serious downstream effects. It caused too much vibration, so mass budget had to be spent on a vibration damping system. It meant also that there were no abort modes between T+30s and T+60s: the burning propellant would have destroyed the capsule parachutes. To achieve a particular safety rating was probably still possible, but that meant spending more on qualification or enhancing the survivability of other phases of flights. And so on.
The physics must dictate the design, not politics. With tremendous effort, almost anything can be made to fly, but you will inevitably come up with a worse product.
The same is true of Orion–perhaps even more so. It is obviously useless for Mars, etc. It is also overkill for LEO. And it doesn’t make sense for any kind of intermediary mission, like transfer/return from a larger orbiting craft.
If money were no object, SLS could be used to build a large trans-LEO craft, say for a mission to an asteroid. But you would still only want a small craft to bring the passengers up to and down from LEO. We (will) have Dragon 2 and the CST-100 for that. As a component of the main spacecraft, it doesn’t make sense.
Looks like a good launch for TESS! This one is a very exciting payload! (Also, the Falcon 9 first stage looks to have made a nice landing on the droneship.)
Another second-stage burn in about half an hour or so.
Interesting that they chose a droneship landing for such a relatively small craft–but perhaps it was the difference between a single-engine and triple-engine burn, and they decided the single-engine burn would be more gentle. It did look like it had quite a long landing burn.
TESS is definitely a cool mission. Can’t wait to see the results from that.
I’ve been looking forward to TESS for a long time. As an amateur astronomer, I’m very excited that it will almost certainly discover planetary transits close enough to us that we amateurs can actually monitor transits and contribute to the science.
And, discovery of planets around our closest stars will give us awesome targets for the James Webb and the new generation of giant scopes. We will likely be able to measure their atmospheres in detail, even directly image them and see continent-sized features one day, etc.
I was terrified that something might go wrong today, but SpaceX came through again. And beautiful landing on the barge! It looked like pinpoint accuracy - important if SpaceX wants to achieve the 24 hour turnaround they promised to have with block 5.
I should clarify when I said we might be able to image continent-sized features ‘one day’. By that I didn’t mean with anybod the current or proposed scopes. It would take a giant interferometry array in space to do that, and the one mission designed to maybe do that was cancelled. Bur we can and will be able to measure atmospheric content and a few other characteristics which will greatly improve our understanding of what’s out there.
Space is the right place to do it, but it doesn’t even have to be that big. A 100 km “diameter” scope could image (in visible light) 1000 km diameter features on planets up to 20 light-years out.
In principle, a pair of interferometer scopes at the Earth-Sun L4 and L5 Lagrange points could do 30-meter imaging up to 1000 light-years out.
I know what you mean–this was one launch where I wasn’t really all that interested in the first-stage landing, however cool that always is. Very glad the launch went well, and very much hope everything continues to go well with the mission.
I guess the next “heart-in-our-throats” launch will be JWST–not just the launch on that one, but also that whole “unfold the origami telescope a million miles out in space” part, too. Definitely won’t be any helpful astronauts coming around to fix that one if some stupid little thing goes wrong.
Here is an explanation of TESS’s nifty orbital insertion.
I’ve used the same trick a number of times in Kerbal Space Program for inclination changes. I’m a little curious why they didn’t just start with a polar orbit, but this must have ended up with less delta V.
I think you’ll be more limited by photon count more than resolution, though. Maybe not for the 100km mirror, since you have a large collecting area, but the giant interferometry array. At some point, there just aren’t enough photons coming from smaller areas to be able to resolve them without an equally gigantic collecting surface.
This is really the ultimate vindication for Musk, since he started SpaceX after getting dicked around by Russians when trying to buy some launches. Also, Rogozin was the one that made the stupid “trampoline” comments with regard to sending astronauts to the ISS, so for him to go all sour grapes on the launch market must be very satisfying.
I’m already feeling nervous about that one. I wonder if a BFR flight could launch a couple of astronauts out there and back? I have no idea what the delta-v requirement would be for that.
I know the thing is not designed to be serviced in space, but you could imagine a failure mode where a fix might be possible, and maybe even easy if someone went there. There would be a lot of pressure to try it, if it could be done at all.
Yeah–this definitely becomes the limiting factor for the big array. At 1000 l-y, you get 4x10[sup]-18[/sup] photons/sec per square meter of emission per square meter of collection, assuming 1000 watts of visible light emission per square meter. That said, if we want a 1 km resolution, and the scopes have roughly a 10x10 km collection area, then we get about a photon every 1000 s. That’s not a lot, but hey, we can go for a very long exposure. That’s about 100 photons a day per “pixel”. Usable.
Of course, a 10 km diam scope isn’t a trivial undertaking, but it’s not totally unthinkable.
Don’t forget the planet would be rotating, so you are also limited by how long features are visible. You could probably correct for rotation in software - I have a program that does that now for Jupiter when you are image stacking, as Jupiter rotates pretty fast. But there are limits to the technique, and it falls apart for anything moving on the surface. So you can probably build up a pretty good picture of continents and bodies of water, but anything much below that is probably pushing it. But who knows? By the time we do such a mission we may have an entirely different way of doing this.
Mmm, right. Though it should be fairly easy to determine the rotational period of the planet using temporal correlation, and then place the incoming photons on the right position on the surface.
Really, I’m hoping at some point that we are just tracking each photon as it comes in with a direction, frequency, and timestamp (within quantum limits, of course). These get correlated between the two detectors and placed in the right spot on the virtual planet’s surface. It might be just doable given how few photons we’re talking about.
They do for about another year or two. Both Boeing and SpaceX have man–rated capsules that should be flying within a year or so. And Blue Origin might not be far behind.