In light of the "Dual Ares V Lite" options with crew on an Ares V Lite, I wonder how this analysis fits into it. Especially Ares V's extremely high LOM number due to "Loss of control during Frist Stage burn" is kind of worrisome.
Does side-mounting the engines almost completely remove any possible thrust imbalance issues? Because that's what these slides seem to imply. I would like to know more about why Ares V suffers from such large thrust imbalance issues. It is fairly clear why a 3 core, 3 engine EELV would have thrust imbalance issues if one engine failed, but it is not very obvious to me why a single core design like Ares V is worse than a 3 core ELLV!
Base heating doesn't have anything directly to do with controllability under engine out conditions. The only relevance that base heating might have to controllability under engine out conditions is if base heating is driving engine layout to a point where the engines are spaced so widely that the core becomes uncontrollable. Quite frankly, if that is true then the engine layout is absurd. Name another vehicle with a single core and >= 4 engines that becomes uncontrollable when one engine fails... I bet you can't!
I want to know *why* the In-Line SD-HLLV is missing from the analysis.The answer would likely be quite revealing.My guess is that they either wanted to try to pretend it doesn't exist (first they ignore you...) or that perhaps they were afraid to show the figures because it competes a little too well with Ares-I (...then the fight you...).I would like to know which is true.Ross.
There's gotta be more going on in these charts when it comes to "Loss of control during First Stage burn" that what's been discussed here so far.The really perplexing question is why does side mount with 3 SSME have such a lower risk of "Loss of control during First Stage burn" than in-line Ares V with 6 RS-68B? One would think that side-mount aerodynamic loads would be more challenging than in-line, but that doesn't seem to be the basis of this metric.Is the newer ablative-nozzle RS-68 with much less flight history so much riskier, or is the greater number of engines (and their arrangement) so much riskier? In either case, you'd kinda think that those metrics would be captured under the constrained and unconstrained engine failure risk components.
While we're waiting, did this chart affect your mood earlier?
Quote from: kraisee on 11/05/2009 05:42 pmI want to know *why* the In-Line SD-HLLV is missing from the analysis.The answer would likely be quite revealing.My guess is that they either wanted to try to pretend it doesn't exist (first they ignore you...) or that perhaps they were afraid to show the figures because it competes a little too well with Ares-I (...then the fight you...).I would like to know which is true.Ross.NASA has done everything in its power to discredit, ignore, bad-mouth, insult, and fight DIRECT since its inception.Only a true competitor would engender such enmity. If DIRECT was as bad as they say, it never would have been included in the Augustine analyses (such as it was).And those charts were generated by NASA, not Aerospace Corp. So of course they left DIRECT/inline/NLS off. "La la la, I can't hear you..."Mark S.
This isn't new information (I said it's from a July 29 presentation). I just wanted to highlight that NASA will have a rather big problem in justifying a switch from Ares I to something like Ares V Lite Dual launch in light of the analysis they have done just recently.I mean, their analysis says Ares V's LOC numbers are only about 2 to 3 times better than Shuttle. That's a farcry from where Cx aims to be. Having said that, unless they go back to their LOM/LOC analysis and say "oh well, it wasn't correct back then, everything is different now", they won't be able to back up their decision to move away from Ares I to a Ares V Dual Lite scenario.Whether their high "Loss of Control during First Stage failure" LOM and LOC numbers are due to RS-68B or not is another question, the odd thing is, side-mount doesn't have that kind of high LOM and LOC numbers, it has a big problem for "Contained engine failure stage 1" but that's probably more of an engineering challenge than an actual problem at the end.
Launch risk is a tiny percentage of total mission risk.
Quote from: MP99 on 11/07/2009 06:35 amLaunch risk is a tiny percentage of total mission risk.That's not entirely correct. A. It depends on the mission and B. it of course depends on the LOC numbers of the vehicle you are using.<snipped>If your vehicle has a risk of LOC on ascent of 1:3000 and you are doing a complicated mission (e.g. extended lunar stay or Mars mission), thn your vehicle ascent risk is probably only a fraction of the mission risk.
However, once you get down to the numbers NASA's analysis above quotes (LOM being about 1:150 for Ares V and LOC about 1:400 if not less)
Jupiter's LOC is better than 1:1000, which would not require a large improvement in other areas to maintain the same overall Sortie mission risk.
These are figures from DIRECT_Summary_v2.0.2.pdf (ie RS-68) and http://www.launchcomplexmodels.com/Direct/documents/DIRECT_ISDC_2009.pdf (DIRECT 3.0, ie SSME - 2nd & 3rd image).cheers, Martin
Quote from: MP99 on 11/07/2009 01:40 pmThese are figures from DIRECT_Summary_v2.0.2.pdf (ie RS-68) and http://www.launchcomplexmodels.com/Direct/documents/DIRECT_ISDC_2009.pdf (DIRECT 3.0, ie SSME - 2nd & 3rd image).cheers, MartinI cannot say anything about LOM and LOC numbers for Jupiter vehicles, but the numbers for Ares I and V as well as the Shuttle numbers look very wrong indeed. They are so completely different from the numbers NASA engineering teams have come up with after pretty long analysis, I just can't take them for real.Is there any more detailed data anywhere on how the DIRECT team arrived for the LOM and LOC figures for Ares I and V? Maybe a detailed probability risk approach as outlined in the first post to this thread?
Dude, you're kidding, right? Guess who came up with DIRECT's numbers and what methodology they used, and what other vehicles those folks do LOM/LOC numbers for as part of their day job in Alabama. Hint - it wasn't Ross or Chuck or the Metschans who came up with these numbers. You need to do some reading in the Direct threads to understand the history of this.
NASA did provide a detailed overview of how they derived their estimates
However at the same time there have been so many different LOC figures for Ares I from 1 in ~1200 to above 3000 that it's pretty clear that NASA has no idea how safe Ares I is; complaining that the LOC/LOM figures don't match NASA's current values isn't fair as NASA has changed methodologies.
I'm not sure why you are complaining about the Shuttle figures, I've heard everything from 1 in 64 to 1 in 240. Although ISS means that LOM => LOC doesn't necessarily hold (in practice I don't think this will actually improve figures very much though).
Quote from: madscientist197 on 11/08/2009 08:59 amHowever at the same time there have been so many different LOC figures for Ares I from 1 in ~1200 to above 3000 that it's pretty clear that NASA has no idea how safe Ares I is; complaining that the LOC/LOM figures don't match NASA's current values isn't fair as NASA has changed methodologies.I have never seen a NASA estimate for the current Ares I architecture of 1: 1200.
The only LOM/LOC numbers worth the paper they are written on are the ex post ones. Sorry for the guys who calculate these things.
Quote from: alexSA on 11/08/2009 09:20 amQuote from: madscientist197 on 11/08/2009 08:59 amHowever at the same time there have been so many different LOC figures for Ares I from 1 in ~1200 to above 3000 that it's pretty clear that NASA has no idea how safe Ares I is; complaining that the LOC/LOM figures don't match NASA's current values isn't fair as NASA has changed methodologies.I have never seen a NASA estimate for the current Ares I architecture of 1: 1200. I have & can back up his statement. It's on a presentation somewhere...
If you find it at some point, I would appreciate if you could post it here. Thanks.
Quote from: Analyst on 11/08/2009 01:10 pmThe only LOM/LOC numbers worth the paper they are written on are the ex post ones. Sorry for the guys who calculate these things.Even then, they a still only comparative (and suitable for where to focus upgrade efforts). They are never predictive or absolute. They only account for random hardware failures, not for process failures which dominate real-life.
Other issues I have is the assumption a passively stable reentry vehicle that must shed a SM is naturally going to be much safer then one with active guidance but under goes no configuration changes for reentry.Real world data Shuttle vs Soyuz during reentry seems to suggest the opposite.Guidance and control systems have been proven to be so reliable that they probably would not even contribute to 1% of LOC/LOM events.Every time you fly on an airliner you are trusting your life to the exact same type of technology to work and to continue working even with multiple failures.
Some dead Soyuz crewmembers,
Quote from: Analyst on 11/08/2009 05:42 pm Some dead Soyuz crewmembers, Which ones would that be.Komarov died because of a faulty pressure sensor. The main chute failure was just a consequence of the sensor failure.
I think I know where I had seen this...it was from a Direct thread.I can't copy & paste on this stupid thing, but do a search on here for:Ares 1:2000There are two related. One is on the Direct thread #2, and describes how NASA changed the rules, the other has a chart made by Direct.Again, it's all theoretical. A 1:2000 could have a failure at launch #3, and nothing for the remaining 1997 launches. Are we going to make 2000 launches with the same vehicle? Never.
Quote from: robertross on 11/08/2009 04:57 pmI think I know where I had seen this...it was from a Direct thread.I can't copy & paste on this stupid thing, but do a search on here for:Ares 1:2000There are two related. One is on the Direct thread #2, and describes how NASA changed the rules, the other has a chart made by Direct.Again, it's all theoretical. A 1:2000 could have a failure at launch #3, and nothing for the remaining 1997 launches. Are we going to make 2000 launches with the same vehicle? Never.Even if there were 4 launches per year, over 50 years you would have 200 launches. I really doubt that with a 100% turnover of all the engineering staff over that time (at least once), various supplier changes as companies come and go, etc, that you're really going to be able to avoid accidents caused by loss of institutional knowledge. And that's assuming that somehow the rookie NASA team designing Ares-I is going to be able to pull of delivering a vehicle that is 8x more reliable than any other ELV that has flown...~Jon
Quote from: jongoff on 11/10/2009 03:38 pmQuote from: robertross on 11/08/2009 04:57 pmI think I know where I had seen this...it was from a Direct thread.I can't copy & paste on this stupid thing, but do a search on here for:Ares 1:2000There are two related. One is on the Direct thread #2, and describes how NASA changed the rules, the other has a chart made by Direct.Again, it's all theoretical. A 1:2000 could have a failure at launch #3, and nothing for the remaining 1997 launches. Are we going to make 2000 launches with the same vehicle? Never.Even if there were 4 launches per year, over 50 years you would have 200 launches. I really doubt that with a 100% turnover of all the engineering staff over that time (at least once), various supplier changes as companies come and go, etc, that you're really going to be able to avoid accidents caused by loss of institutional knowledge. And that's assuming that somehow the rookie NASA team designing Ares-I is going to be able to pull of delivering a vehicle that is 8x more reliable than any other ELV that has flown...~JonIt might be interesting to speculate on how R-7 escaped that fate. There can't be too many original enginneers left on the team since the early 1950s.
As Downix suggested, high flight rate helps keep people trained. But as it is, they did have several near failures in recent years that could've turned fatal quite easily.
Quote from: jongoff on 11/10/2009 03:59 pmAs Downix suggested, high flight rate helps keep people trained. But as it is, they did have several near failures in recent years that could've turned fatal quite easily. These were problems with the spacecraft, not with the LV.
Quote from: William Barton on 11/10/2009 03:44 pmQuote from: jongoff on 11/10/2009 03:38 pmQuote from: robertross on 11/08/2009 04:57 pmI think I know where I had seen this...it was from a Direct thread.I can't copy & paste on this stupid thing, but do a search on here for:Ares 1:2000There are two related. One is on the Direct thread #2, and describes how NASA changed the rules, the other has a chart made by Direct.Again, it's all theoretical. A 1:2000 could have a failure at launch #3, and nothing for the remaining 1997 launches. Are we going to make 2000 launches with the same vehicle? Never.Even if there were 4 launches per year, over 50 years you would have 200 launches. I really doubt that with a 100% turnover of all the engineering staff over that time (at least once), various supplier changes as companies come and go, etc, that you're really going to be able to avoid accidents caused by loss of institutional knowledge. And that's assuming that somehow the rookie NASA team designing Ares-I is going to be able to pull of delivering a vehicle that is 8x more reliable than any other ELV that has flown...~JonIt might be interesting to speculate on how R-7 escaped that fate. There can't be too many original enginneers left on the team since the early 1950s.Constant use, constant training, constant flights.Same reason why you can still get a car built today when Fords original workers all are dead or retired.
Quote from: alexSA on 11/10/2009 04:22 pmQuote from: jongoff on 11/10/2009 03:59 pmAs Downix suggested, high flight rate helps keep people trained. But as it is, they did have several near failures in recent years that could've turned fatal quite easily. These were problems with the spacecraft, not with the LV. No, there have been booster problems in the last ten years, see Foton.