Emre Kelly has the story on the current status of Artemis 1.
33 thoughts on “Going Ahead With The Launch”
Peak winds hit 82 mph at 60 feet, but lightning tower sensors recorded gusts as high as 100 mph 457 feet above ground. That means NASA stayed within its limits at the 60-foot mark.
There’s a table you can find with the answers to (NSF pissing contests have a lot of said info easily found). Jim Free’s statement is, limits were nowhere exceeded at any point, and they had a 25% margin besides. This who thing is just typical interdweeb dopiness. Plus we have an on-site expert who has proclaimed certain failure. If it doesn’t fail, we can pick on him for ages! An if it does fail, his gloating privileges will be insurmountable!
Well, those hard limits are really more guidelines than rules…
Gee, my management hat must have shrunk in the wash! It’s fitting really, really, tightly right now, for a management XXXL.
Sorry I meant use metric, that’s an engineering XS.
LOL
Well speaking of unit conversion issues this is interesting
Though was originally going to state that 85 MPH limit may be at 60 feet it probably much higher at 457 feet up. But seeing the Feet per second makes my eye twitch.
85 mph is ~125fps
Unfueled limits @ 60′ up is 83.3 fps, full/partial fueled limit is 52.5 fps
85 mph is ~125fps
Unfueled limits @ 60′ up is 83.3 fps, full/partial fueled limit is 52.5 fps
Yea I noticed that. I keep thinking there is something I am missing. That NASA wouldn’t be dumb enough to have the unit conversion issue again but we will see.
But we also a bunch of people at someone who Wrote a book called “Safe is not option” With a few complaining pointing out that NASA not following a safety guidelines on a unmanned rocket that probably has a safety factor of 200%.
HAD. Had a safety factor of 200%.
What was the safety factor on Challenger?
Doesn’t matter, because the safety was compromised by expedience.
And when they light the candle and it blows, “mistakes were made”. Lets have Boeing build another….for 3x the original cost, to be finished in 12 years, and in the mean time NASA is still in the business.
HAD. Had a safety factor of 200%.
What was the safety factor on Challenger?
Doesn’t matter, because the safety was compromised by expedience.
And when they light the candle and it blows, “mistakes were made”. Lets have Boeing build another….for 3x the original cost, to be finished in 12 years, and in the mean time NASA is still in the business.
There was no safety factor for how cold it was. Some engineer has thought it would caused issues and ball parked a 53F since it was the coldest they launched the shuttle at.
But they had to launch the teacher…..
Try a 1.4 FoS, which is pretty common for structural design. 200%? Another sign you are not an engineer. In case you are confusing risk of loss of vehicle and thinking NASA actually over-designs things, keep in mind shuttle ended with LoC/LoV estimated at 1/90. BTW, for Challenger, the LoC/LoV risk likelihood was calibrated, based on the loss, to 1/37. Design ultimate FoS was still 1.4 for Challenger, and the softgood O-Ring should have been FoS of 4.0 against loss of life or crew, or again 1.4 for individual component failure (to get to 4.0 was why they had two, then 3 O-Rings later).
Try a 1.4 FoS, which is pretty common for structural design. 200%? Another sign you are not an engineer
Was part of it exaggerating and never claimed I was structural expert. But you don’t pay attention to derating guidelines do you Leland. Now de-rating is another purpose of extending life and increase reliability. Along with sometimes protect a stupid engineer from themselves.
Airplanes are built with a 150% margin (FAR 23). So, if you’re talking about a plane certified for aerobatics, it’s design load will be a minimum of +6/-3 Gs. That means you can pull 6 positive Gs and return to normal flight without anything bending. With a 150% margin, thing can start breaking at 9Gs. However, when you exceed the design limit, things do start to bend, and being bent means it isn’t as strong as before. Once that happens, the plane has to be inspected and repaired. Sometimes, it may have to be retired even though the damage may not be readily apparent to the naked eye.
If the SLS experienced wind loads above the design limits, then it no longer has a 140% margin. It would require extensive examinations to determine if anything is bent or if the design limits were exceeded. That would require rolling it back to the VAB and would probably take months. Instead, they’re going to roll the dice and hope nothing breaks, assuming they can get to the point of ignition.
Free clearly stated there was a 25% margin between the wind limits and the 1.4 margin. What’s that, 1.15? Looks like the bathtub caulk underperformed.
Who made that decision, NASA managers or NASA engineers? In 1986, NASA engineers said it was too cold to launch Challenger. They were overruled by NASA managers. “Mistakes were made”, and the result was the deaths of 7 crew and the loss of a very expensive Shuttle. In 2003, NASA engineers were worried about possible damage to Columbia’s heat shield tiles following the impact of a big piece of foam during launch. They wanted national reconnaissance assets to image Columbia so they could inspect for damage. NASA managers weren’t worried about it. More “mistakes were made” and the result was the deaths of another 7 crew and loss of another orbiter. If the engineers say there is no significant damage to the SLS, I’d be much more likely to believe them than if the managers made that decision.
Problem here is that the engineer’s voices are muted – they’re not allowed to speak out.
Thanks! I no longer save links, but do collect pdfs when they turn up.
Damn, that link is old. Had you not picked up on it, I probably wouldn’t have to correct, but here goes. We have this one:
ntrs.nasa.gov/citations/20210024522
This one is dated October 27, 2021.
Tables 3.1-3.1 & 3.1-3.2 between the two appear unchanged.
Hey it’s not their money if they blow it up by not being thorough.
Some of the engineering principles posted here are, at best, only trivially true. Airplanes, rockets, etc. are not monolithic objects conjured up whole, they’re assemblages of pieces and parts, each with its own distinct properties. NASA’s fabled 1.4 “margin” is not applied throughout, it is, as I said, a sort of lower bound. In this instance, I guessed it applied to the orange insulation, and, in fact, what failed was what amounted to a strip of bathtub caulk on the ascent cover of the capsule, which is discarded post booster burnout. Since it would have been exposed to near-hypersonic wind during ascent, it would probably have blown off then. The caulk was about as thick as a strip of duct tape.
There’s an old story about Henry Ford sending his engineers out to junkyards to inspect junked Fords so as to discover which parts had not worn out. Those parts were too well made (i.e., had too much margin) and could be made more cheaply in the future. I can tell you this does not happen with large, engineered systems, by and large, for “reasons.” Unfortunately, we run into a wall here, because some of what I know is covered by the terms of my old security clearance, and most of the rest is covered by the proprietary NDAs of my former employers (many of them absurd, but that’s life). I realize that’s annoying. I can tell you that the public margins on systems I worked on were sandbagged.
Some of what’s being posted here implies that NASA associate admin Jim Free is either a liar or a fool. Since he’s a public employee, you’re pretty much free to say what you want about the way he does his job (my Dad was, at the end of his career, New England States Liaison for the Bureau of Mines, so it’s something I saw first hand). But Free probably does have more facts than random Internet commenters. (Or Berger’s anonymous “unbiased” industry sources, for that matter.)
All of that said, I have never been an SLS supporter, or even a supporter of the in-line SDV concept, which I long ago said would be a collosal waste of time and money (And guess what…). In Spaceflight Magazine, in 2004, so long before any of this came along and the first successful Falcon 1 launch was still years in the future, I published an article titled “Off the Shelf and On to Mars” that championed the 1977 Type 1 SDV concept, which took the ET and 4seg SRBs and swapped the Orbiter for a thrust frame with the Shuttle boat tail (engines and OMS) at the bottom. As an expedable, it could push around 90 tons to LEO. I suggested an upper stage carried in the thrust frame derived from the Ariane V cryo core (needing an air lightable engine), and an Apollo-size capsule above that. I added a small orbital module in the SLA equivalent so the crew could be expanded to a notional 5 like the Skylab rescue vehicle). It would require two launches to make a lunar landing, the second launch sending a lander. You’ll notice this is pretty much what the Chinese are proposing to fly in 2027 as the CZ-5G? I did propose a recoverable version that only expended the ET and thrust frame, by adding a heat shield, parachutes, and airbags to the boat tail, with a notional recovery in western Australia. That would have reduced the LEO capability to a little below 70 tons, and would have been usable as a crew/cargo transport to service ISS. My guess at the time was, the booster would be ready around 2010 and the capsule a bit later. Best of all, it would have been compatible with continued use of the STS Orbiter.
It seems to me that the simplest way to tell if the structure exceeded design limits during a wind event is just to look at the camera footage during the storm to see the maximum sway and displacements that occurred, from which the structure’s stress strain curves should allow a quick back-computation of what kind of stress it was under. If nothing flexed more than it’s allowed to, it should still be good.
The key would be the “allowed to number” and whether or not it was nonsense. Remember the body flap deflection on STS-1. Young said if he’d known, he would have ejected over the Atlantic (and probably died). But it was fine, because the allowed deflection was much less than it could take.
I’m betting NASA’s confidence comes from looking at data from the rocket and how much it deflected, rather than playing what-if games about anemometer readings. Of course they’ve been wrong before.
On a side note, would a small drone be a cheap yet viable way to visually inspect the inside of a vertical SRB, perhaps with a laser beneath the nozzle aimed straight up along the middle that it could use to stay locked in position along the center axis?
I think you’ve conflated two body flap issues on STS-1. At ignition of the SRB’s, the reflected pressure pulse – much larger than estimated – exerted such force on the body flap that the hydraulic actuators reported pressures that translated to flap structural loads exceeding its limit loads. Nothing about deflection. However, on reentry the flap deflection was set to a value designed to trim the orbiter to an angle of attack where it would fly stably – but it was the wrong flap deflection. Inexplicably, NASA had used perfect gas properties of air in the CFD analysis used to predict the right flap deflection. The air was really ionized, and its properties nowhere near perfect gas. The difference meant that the orbiter wasn’t trimmed at a stable angle of attack, and the crew almost lost control. That, along with loss of a number of TPS tiles on the windward side, made that entire first flight one big close call.
The Soviets flew their version, Buran, completely autonomously on its one and only flight. They didn’t consider the risk to a crew acceptable, and the fact that they were able to succeed with purely autonomous control was a remarkable feat for them. I once heard, though have never verified it, that reason Buran flew only once was not due to the cost. Rather, the TPS was inadequate, and the airframe was degraded by overheating to the point where it couldn’t be fly again, and was beyond repair.
I think you’re right. I was mainly just remembering a public statement from John Young in an interview. I have the material to look up, but didn’t bother. Another issue with Buran may have been the erratic behavior of the EDL software. I think I remember Oberg saying if they dared launch Buran, they’d never see it again. But it worked at least to that point.
Michael/William can you enlighten me a bit? Were the trim actuators on the shuttle unable to be adjusted once deep into the re-entry profile because of loading? Meaning essentially they were adjusted ahead of time in orbit and then locked into position? If that were the case, flying essentially with no trim wheel, then it’s dicey indeed.
I’m sorry I mean the body flap.
Read the Wikipedia STS1-1 article’s anomalies section. It does state that there was deflection of the body flap (as well as defining said flap) sufficient to fracture the hydraulics (there was a lot of other launch damage beisides). This is why Young said he would have ejected had he know. STS-1 staged at 174,000 feet, too high for ejection, so they would have been ejecting above the SRB exhaust plume. So he was choosing between certain death (EDL failure) and probably death (SRB exhaust). It also says the flight computers were able to correct in flight for the calculation errors on reentry. i haven’t checked the footnotes, and my own references are behind a temporarily immovable object.
I suddenly remembered Von Braun’s joke about “four Neins” (0.9999) reliability.
Peak winds hit 82 mph at 60 feet, but lightning tower sensors recorded gusts as high as 100 mph 457 feet above ground. That means NASA stayed within its limits at the 60-foot mark.
…and above that mark?
There’s a table you can find with the answers to (NSF pissing contests have a lot of said info easily found). Jim Free’s statement is, limits were nowhere exceeded at any point, and they had a 25% margin besides. This who thing is just typical interdweeb dopiness. Plus we have an on-site expert who has proclaimed certain failure. If it doesn’t fail, we can pick on him for ages! An if it does fail, his gloating privileges will be insurmountable!
Well, those hard limits are really more guidelines than rules…
Gee, my management hat must have shrunk in the wash! It’s fitting really, really, tightly right now, for a management XXXL.
Sorry I meant use metric, that’s an engineering XS.
LOL
Well speaking of unit conversion issues this is interesting
Though was originally going to state that 85 MPH limit may be at 60 feet it probably much higher at 457 feet up. But seeing the Feet per second makes my eye twitch.
85 mph is ~125fps
Unfueled limits @ 60′ up is 83.3 fps, full/partial fueled limit is 52.5 fps
Yea I noticed that. I keep thinking there is something I am missing. That NASA wouldn’t be dumb enough to have the unit conversion issue again but we will see.
But we also a bunch of people at someone who Wrote a book called “Safe is not option” With a few complaining pointing out that NASA not following a safety guidelines on a unmanned rocket that probably has a safety factor of 200%.
HAD. Had a safety factor of 200%.
What was the safety factor on Challenger?
Doesn’t matter, because the safety was compromised by expedience.
And when they light the candle and it blows, “mistakes were made”. Lets have Boeing build another….for 3x the original cost, to be finished in 12 years, and in the mean time NASA is still in the business.
Yea yea go buy the book.
There was no safety factor for how cold it was. Some engineer has thought it would caused issues and ball parked a 53F since it was the coldest they launched the shuttle at.
But they had to launch the teacher…..
Try a 1.4 FoS, which is pretty common for structural design. 200%? Another sign you are not an engineer. In case you are confusing risk of loss of vehicle and thinking NASA actually over-designs things, keep in mind shuttle ended with LoC/LoV estimated at 1/90. BTW, for Challenger, the LoC/LoV risk likelihood was calibrated, based on the loss, to 1/37. Design ultimate FoS was still 1.4 for Challenger, and the softgood O-Ring should have been FoS of 4.0 against loss of life or crew, or again 1.4 for individual component failure (to get to 4.0 was why they had two, then 3 O-Rings later).
Was part of it exaggerating and never claimed I was structural expert. But you don’t pay attention to derating guidelines do you Leland. Now de-rating is another purpose of extending life and increase reliability. Along with sometimes protect a stupid engineer from themselves.
Airplanes are built with a 150% margin (FAR 23). So, if you’re talking about a plane certified for aerobatics, it’s design load will be a minimum of +6/-3 Gs. That means you can pull 6 positive Gs and return to normal flight without anything bending. With a 150% margin, thing can start breaking at 9Gs. However, when you exceed the design limit, things do start to bend, and being bent means it isn’t as strong as before. Once that happens, the plane has to be inspected and repaired. Sometimes, it may have to be retired even though the damage may not be readily apparent to the naked eye.
If the SLS experienced wind loads above the design limits, then it no longer has a 140% margin. It would require extensive examinations to determine if anything is bent or if the design limits were exceeded. That would require rolling it back to the VAB and would probably take months. Instead, they’re going to roll the dice and hope nothing breaks, assuming they can get to the point of ignition.
Free clearly stated there was a 25% margin between the wind limits and the 1.4 margin. What’s that, 1.15? Looks like the bathtub caulk underperformed.
Who made that decision, NASA managers or NASA engineers? In 1986, NASA engineers said it was too cold to launch Challenger. They were overruled by NASA managers. “Mistakes were made”, and the result was the deaths of 7 crew and the loss of a very expensive Shuttle. In 2003, NASA engineers were worried about possible damage to Columbia’s heat shield tiles following the impact of a big piece of foam during launch. They wanted national reconnaissance assets to image Columbia so they could inspect for damage. NASA managers weren’t worried about it. More “mistakes were made” and the result was the deaths of another 7 crew and loss of another orbiter. If the engineers say there is no significant damage to the SLS, I’d be much more likely to believe them than if the managers made that decision.
Problem here is that the engineer’s voices are muted – they’re not allowed to speak out.
The full PDF file to SLS-SPEC-159 Revision D CROSS-PROGRAM DESIGN SPECIFICATION FOR NATURAL ENVIRONMENTS (DSNE)
Can be found here:
https://ntrs.nasa.gov/api/citations/20160004378/downloads/20160004378.pdf
Thanks! I no longer save links, but do collect pdfs when they turn up.
Damn, that link is old. Had you not picked up on it, I probably wouldn’t have to correct, but here goes. We have this one:
ntrs.nasa.gov/citations/20210024522
This one is dated October 27, 2021.
Tables 3.1-3.1 & 3.1-3.2 between the two appear unchanged.
Hey it’s not their money if they blow it up by not being thorough.
Some of the engineering principles posted here are, at best, only trivially true. Airplanes, rockets, etc. are not monolithic objects conjured up whole, they’re assemblages of pieces and parts, each with its own distinct properties. NASA’s fabled 1.4 “margin” is not applied throughout, it is, as I said, a sort of lower bound. In this instance, I guessed it applied to the orange insulation, and, in fact, what failed was what amounted to a strip of bathtub caulk on the ascent cover of the capsule, which is discarded post booster burnout. Since it would have been exposed to near-hypersonic wind during ascent, it would probably have blown off then. The caulk was about as thick as a strip of duct tape.
There’s an old story about Henry Ford sending his engineers out to junkyards to inspect junked Fords so as to discover which parts had not worn out. Those parts were too well made (i.e., had too much margin) and could be made more cheaply in the future. I can tell you this does not happen with large, engineered systems, by and large, for “reasons.” Unfortunately, we run into a wall here, because some of what I know is covered by the terms of my old security clearance, and most of the rest is covered by the proprietary NDAs of my former employers (many of them absurd, but that’s life). I realize that’s annoying. I can tell you that the public margins on systems I worked on were sandbagged.
Some of what’s being posted here implies that NASA associate admin Jim Free is either a liar or a fool. Since he’s a public employee, you’re pretty much free to say what you want about the way he does his job (my Dad was, at the end of his career, New England States Liaison for the Bureau of Mines, so it’s something I saw first hand). But Free probably does have more facts than random Internet commenters. (Or Berger’s anonymous “unbiased” industry sources, for that matter.)
All of that said, I have never been an SLS supporter, or even a supporter of the in-line SDV concept, which I long ago said would be a collosal waste of time and money (And guess what…). In Spaceflight Magazine, in 2004, so long before any of this came along and the first successful Falcon 1 launch was still years in the future, I published an article titled “Off the Shelf and On to Mars” that championed the 1977 Type 1 SDV concept, which took the ET and 4seg SRBs and swapped the Orbiter for a thrust frame with the Shuttle boat tail (engines and OMS) at the bottom. As an expedable, it could push around 90 tons to LEO. I suggested an upper stage carried in the thrust frame derived from the Ariane V cryo core (needing an air lightable engine), and an Apollo-size capsule above that. I added a small orbital module in the SLA equivalent so the crew could be expanded to a notional 5 like the Skylab rescue vehicle). It would require two launches to make a lunar landing, the second launch sending a lander. You’ll notice this is pretty much what the Chinese are proposing to fly in 2027 as the CZ-5G? I did propose a recoverable version that only expended the ET and thrust frame, by adding a heat shield, parachutes, and airbags to the boat tail, with a notional recovery in western Australia. That would have reduced the LEO capability to a little below 70 tons, and would have been usable as a crew/cargo transport to service ISS. My guess at the time was, the booster would be ready around 2010 and the capsule a bit later. Best of all, it would have been compatible with continued use of the STS Orbiter.
It seems to me that the simplest way to tell if the structure exceeded design limits during a wind event is just to look at the camera footage during the storm to see the maximum sway and displacements that occurred, from which the structure’s stress strain curves should allow a quick back-computation of what kind of stress it was under. If nothing flexed more than it’s allowed to, it should still be good.
The key would be the “allowed to number” and whether or not it was nonsense. Remember the body flap deflection on STS-1. Young said if he’d known, he would have ejected over the Atlantic (and probably died). But it was fine, because the allowed deflection was much less than it could take.
I’m betting NASA’s confidence comes from looking at data from the rocket and how much it deflected, rather than playing what-if games about anemometer readings. Of course they’ve been wrong before.
On a side note, would a small drone be a cheap yet viable way to visually inspect the inside of a vertical SRB, perhaps with a laser beneath the nozzle aimed straight up along the middle that it could use to stay locked in position along the center axis?
I think you’ve conflated two body flap issues on STS-1. At ignition of the SRB’s, the reflected pressure pulse – much larger than estimated – exerted such force on the body flap that the hydraulic actuators reported pressures that translated to flap structural loads exceeding its limit loads. Nothing about deflection. However, on reentry the flap deflection was set to a value designed to trim the orbiter to an angle of attack where it would fly stably – but it was the wrong flap deflection. Inexplicably, NASA had used perfect gas properties of air in the CFD analysis used to predict the right flap deflection. The air was really ionized, and its properties nowhere near perfect gas. The difference meant that the orbiter wasn’t trimmed at a stable angle of attack, and the crew almost lost control. That, along with loss of a number of TPS tiles on the windward side, made that entire first flight one big close call.
The Soviets flew their version, Buran, completely autonomously on its one and only flight. They didn’t consider the risk to a crew acceptable, and the fact that they were able to succeed with purely autonomous control was a remarkable feat for them. I once heard, though have never verified it, that reason Buran flew only once was not due to the cost. Rather, the TPS was inadequate, and the airframe was degraded by overheating to the point where it couldn’t be fly again, and was beyond repair.
I think you’re right. I was mainly just remembering a public statement from John Young in an interview. I have the material to look up, but didn’t bother. Another issue with Buran may have been the erratic behavior of the EDL software. I think I remember Oberg saying if they dared launch Buran, they’d never see it again. But it worked at least to that point.
Michael/William can you enlighten me a bit? Were the trim actuators on the shuttle unable to be adjusted once deep into the re-entry profile because of loading? Meaning essentially they were adjusted ahead of time in orbit and then locked into position? If that were the case, flying essentially with no trim wheel, then it’s dicey indeed.
I’m sorry I mean the body flap.
Read the Wikipedia STS1-1 article’s anomalies section. It does state that there was deflection of the body flap (as well as defining said flap) sufficient to fracture the hydraulics (there was a lot of other launch damage beisides). This is why Young said he would have ejected had he know. STS-1 staged at 174,000 feet, too high for ejection, so they would have been ejecting above the SRB exhaust plume. So he was choosing between certain death (EDL failure) and probably death (SRB exhaust). It also says the flight computers were able to correct in flight for the calculation errors on reentry. i haven’t checked the footnotes, and my own references are behind a temporarily immovable object.
I suddenly remembered Von Braun’s joke about “four Neins” (0.9999) reliability.