ApolloHoax.net
Off Topic => General Discussion => Topic started by: scooter on December 07, 2014, 10:56:54 PM
-
Enjoyed watching the Delta IV/Orion launch, and have a question. or 2...
The craft makes a very slow 90 degree roll during the climb...is there any particular reason for this?
Also, as it leaves the pad, I noticed a LOT of small wires/lines being broken/pulled out as it climbed past the gantry...telemetry/data lines or something?
It was also interesting being able to visually see the throttling of the center engine vs the outboards...great show, and hopefully the beginning of a strong (funded) program.
-
Enjoyed watching the Delta IV/Orion launch...
Same here.
...and have a question. or 2...
The craft makes a very slow 90 degree roll during the climb...is there any particular reason for this?
I'm not sure, but I assume it's the rocket rolling from its launch facing to the direction it needs to face to climb to its orbit, given its orbital inclination. Presumably the rocket was aligned N-S or E-W on the launch pad, while its orbit headed off to the south-east. For both Apollo and the Space Shuttle (and I assume Mercury and Gemini) the spacecraft climbed to orbit with the astronauts in a head-down position. That way the astronauts could see the horizon when looking out the window, and it presumably helped blood flow to the head rather than away from it.
Also, as it leaves the pad, I noticed a LOT of small wires/lines being broken/pulled out as it climbed past the gantry...telemetry/data lines or something?
Some of the lines would have been fuel lines, topping up the cryogenic propellants as they boiled off prior to launch. Other lines would have been providing power from the ground, so that if the launch was aborted Mission Control could control the spacecraft.
However, in each of these cases I'm happy for the experts to correct or add to my answer.
It was also interesting being able to visually see the throttling of the center engine vs the outboards...
Yes, I noticed that too, although I wasn't sure whether that was actually what I was seeing.
...great show, and hopefully the beginning of a strong (funded) program.
Agreed.
I have to say I'm ambivalent about the whole thing - what can Orion do that Dragon and the Boeing spacecraft can't? (Apart from keeping a lot of NASA engineers happily employed...?) And I've read a fair bit of criticism of the Space Launch System, although I'll watch it launch as avidly as any Saturn V, Shuttle, Falcon Heavy or Delta IV Heavy launch!
-
For ease of operation, the Saturn V and shuttle had fixed axis labels and always launched with the same orientation to the ground. So to generalize a bit speculatively, I'd guess that the same thing is true here. Since the launch direction of any rocket is set to make a specific orbit for the mission needs, there is no one orientation on the pad that will get it into the proper orbit for all missions. So you need a roll program. So why not just build the pad in the most efficient configuration for the pad and let the rocket take care of the flight orientation with the roll program.
-
Adding to what other have said... The reason the rocket must roll to its intended flight heading before pitching down is to get the pitch and yaw axes properly aligned. We want the pitch axis to be parallel to the ground and perpendicular to the heading the rocket will fly. This way when the rocket transitions to horizontal flight it is rotating strictly in pitch, which simplifies the control.
For example, let's say a rocket sits on the launch pad with its pitch axis aligned north-south and its yaw axis aligned east-west. We want the rocket to fly southeast along a heading of 135 degrees. After the rocket lifts off we must rotate it 45 degrees to the south (clockwise looking down on it). This aligns the rocket so that when we tilt the nose down it is rotating around the pitch axis. If we didn't execute the roll, bringing the nose down would require movements in both pitch and yaw, which is way more complicated than it needs to be.
-
Every STS launch I have seen had an obvious roll program. Here's a fine example; roll at 1:48 in the video
-
This of course raises the question of what governs how a space vehicle is oriented on the pad.
-
This of course raises the question of what governs how a space vehicle is oriented on the pad.
The pad, mostly.
-
The reason the rocket must roll to its intended flight heading before pitching down is to get the pitch and yaw axes properly aligned.
With the symmetric thrust of the Saturn V, I would guess the axes could be arbitrarily defined, although would be very practical reasons for the specific orientation that was used. With a side staged rocket like the Delta IV would this make difference?
-
The question I have about the Delta IV is why is the steam exhaust so bright? The STS main engine exhaust always seemed almost non-existent, aside from the shock diamonds. Was that only by comparison to the very bright exhaust of the solid boosters or is there something in the hydrogen fueled engines that makes them different?
-
It's a zillion times easier to write the guidance programs when there are designated pitch and yaw axes and roll is constrained -- even with symmetrical vehicles.
The civil engineering of the pad and launch site dictates where the hardpoints go, and that dictates the orientation of the vehicle on the pad. It is impossible to build a pad that doesn't require some roll at launch, since launch azimuths vary from mission to mission, and within the launch window for each individual mission. Hence launch vehicle orientation is not a consideration when siting launch facility structures.
Von Braun programmed a roll into each of his vehicles' launches specifically to exercise all the flight controls within a few seconds of liftoff, as a safety check. Pitch and yaw are exercised anyway as part of structure avoidance maneuvers and regular guidance.
Launch Complex 39 oriented the STS stack so that the vertical stabilizer points south. This was mostly a holdover from Apollo, but it had to do with the way the causeways were built on Merrit Island. The causeways for the crawler-transporter end with a northbound route, and the orientation of the stack(s) on the MLP was dictated in large part by orienting the most stable of pitch or yaw axes for ground transportation.
The engineering of umbilicals and their associated disconnection strategies is a field unto itself.
-
The question I have about the Delta IV is why is the steam exhaust so bright? The STS main engine exhaust always seemed almost non-existent, aside from the shock diamonds. Was that only by comparison to the very bright exhaust of the solid boosters or is there something in the hydrogen fueled engines that makes them different?
The RS-68 engines of the Delta IV are ablatively cooled, so maybe that's the reason. The ablative lining burns away during operation to keep the nozzle cool. Perhaps it's this material in the exhaust that gives it a different appearance.
-
Good discussion...
I'm familiar with the "azimuth roll" shortly after launch, but what I noticed in the following minutes is a roughly 90 degree roll that takes the booster from a horizontal alignment (strap on boosters on "either side") to a vertical alignment (with the strap ons above and below the core). Very slow, takes a couple/few minutes, but it's just something interesting I noticed.
I'm familiar with all the propellant feed disconnects pulling away at launch, but there were a lot of smaller "wires", very small, maybe attached further down...perhaps small telemetry feeds or something. Perhaps it's a Delta thing, just never noticed them on other launches.
Boy, that fireball on launch sure is attention getting!!
-
The civil engineering of the pad and launch site dictates where the hardpoints go, and that dictates the orientation of the vehicle on the pad. It is impossible to build a pad that doesn't require some roll at launch, since launch azimuths vary from mission to mission, and within the launch window for each individual mission. Hence launch vehicle orientation is not a consideration when siting launch facility structures.
The Scout launch vehicle used a rotatable launch pad to set the azimuth, but this was a much smaller vehicle that those we've been discussing.
-
When a modern launcher performs a roll after liftoff, it is almost always to point its various antennas in the correct directions.
Launcher antennas include those for the range safety (destruct) receivers, various radar transponders, telemetry transmitters and often a TDRSS antenna. I think this last one accounts for most of the roll maneuvers on US launches as there still aren't that many TDRSS satellites and you usually want to remain on one for as long as you can, at least during critical mission events, to avoid the interruption that occurs when handing off from one satellite to the next.
It's not just a matter of where the antennas are mounted; sometimes plume obstruction is an issue too.
The shuttle used to remain heads-down for its entire ride into orbit, but later missions often rolled into a heads-up position specifically for better TDRSS coverage.
Range safety reception usually isn't a serious constraint as there are multiple antennas (and receivers) arrayed around the vehicle specifically so the destruct command will be reliably received no matter what the rocket happens to be doing.
-
I tried to start a thread specifically on the question of the appearance of the Delta IV's plume, but I couldn't find the new topic button. I've since seen LO's note about accidentally disabling it.
I thought to bring it up not only for my own curiosity, but because I expect the hoaxers will eventually seize on it as "proof" the LM ascent plumes should also have been visible.
An ablative nozzle lining occurred to me too, especially since the onboard camera shows quite a few 'sparks' being thrown off. But that orange color is also the color of hydrogen burning in air, so I didn't want to jump to that conclusion. OTOH, all rocket engines run rich, including the LH2-burning SSMEs, but I never saw them produce orange plumes. As reusable engines, I don't think they used ablative coatings.
-
I also noticed those fine lines pulling away at liftoff. I can't imagine they were any kind of propellant lines, as that's what the gantry arms are for. They don't detach until liftoff.
All I can think of is that they pulled away some sort of protective insulation or cover at liftoff. The early versions of the Ariane launcher dropped a layer of insulation around their second stages at liftoff. The second stage burned hypergolic (storable) propellants, but they were cooled to cram more into the tanks in the tropical French Guiana heat. I always thought it was bad form for your launcher to drop bits of itself onto the pad at liftoff.
The tip of the Q-ball at the top of the Apollo stack also had a protective cover, but I think it was pulled away by a line a few minutes before launch.
And yeah, that hydrogen fireball around the base of the Delta IV has always bothered me. The engines are started with a fuel lead, and the excess hydrogen burns in air as it rises. In previous launches it actually scorched the insulation on the three sections. To minimize that effect, the three engines were stagger-started on the Orion launch for the first time and it seems to have worked; I noticed very little if any scorching of the insulation this time.
-
The first time I noticed such a thing was the maiden launch of the Titan 3 in 1965. This famously had a camera that took a series of pictures looking down the length of the rocket and documenting the jettison of the solid boosters. The line appeared to lead to the position that this camera had to be in and I would guess that it switched the camera on as the vehicle left the pad.
-
The first time I noticed such a thing was the maiden launch of the Titan 3 in 1965. This famously had a camera that took a series of pictures looking down the length of the rocket and documenting the jettison of the solid boosters. The line appeared to lead to the position that this camera had to be in and I would guess that it switched the camera on as the vehicle left the pad.
Link, please? Thanks.
-
An ablative nozzle lining occurred to me too, especially since the onboard camera shows quite a few 'sparks' being thrown off. But that orange color is also the color of hydrogen burning in air, so I didn't want to jump to that conclusion. OTOH, all rocket engines run rich, including the LH2-burning SSMEs, but I never saw them produce orange plumes. As reusable engines, I don't think they used ablative coatings.
The SSME used regenerative cooling. The reason the RS-68 uses ablative cooling is cost reduction. The trade off for lower cost was higher mass. The RS-68 has a thrust-to-weight ratio of 51:1 while the SSME is 67:1 (based on vacuum thrust).
-
Link, please? Thanks.
The line is visible about 15 to 20 s into this video:
I can't find any on-line pictures from the camera, but they were in the press at the time and also in this article:
http://adsabs.harvard.edu/abs/1993JBIS...46..122R
-
It's a gazillion times cheaper to build an ablatively-cooled or radiatively-cooled nozzle than a regeneratively-cooled one. Regeneratively-cooled thrust assemblies are monumentally difficult, monumentally expensive, and have a high rejection rate during manufacturing.
-
Okay, I finally got a good look at them and I think they're almost certainly safing lanyards. Signals and power go through the hardpoint umbilicals -- the big ones that get pulled away by mechanical arms and are properly deadfaced at liftoff. But certain range safety and ordnance systems require either 2- or 3-factor inhibitors for safety reasons (depending on the consequences of premature activation). Looking at where the lanyards attach to the vehicle, I'm thinking at least half of them are range-safety destruct inhibitors. After the rocket has climbed high enough, they pull loose and operate by severing breakwire inhibitors as one of the safing factors. That's confirmation to the simple-as-dirt range safety systems that the rocket is moving and has climbed high enough to allow a commanded destruction.
One of the lanyards goes to the middle of the second stage, where the instrument unit is on this vehicle. That would an additional breakwire trip that confirms vehicle motion to the guidance system. Basically you program the rocket to expect a breakwire event within a few seconds after commencing the launch. That would be the second confirmation factor of vehicle motion, whereas IMU reports would be the first. It can also be a dirt-simple confirmation of vehicle altitude to allow pitch and yaw manuevers that would otherwise be restricted when the vehicle is near the umbilical tower.
Incidentally I went over the flight dynamics specifically for the Delta IV H and was reminded that because of its unique profile it rolls in order present its most drag-insensitive aspect to the wind. This minimizes wind dispersion which, frankly, isn't that much of an issue for smaller vehicles. It fixes wind dispersions after staging and consequently after wind is no longer a factor.
-
So what about antennas? If the vehicle has to roll to minimize wind forces, which vary unpredictably with the weather, it would need extra antennas to communicate regardless of the roll.
I know several range safety antennas and receivers are always arrayed around the vehicle to ensure a rock-crushing signal regardless of attitude, but I don't know about telemetry transmitters, radar transponders, TDRSS terminals, TV transmitting antennas, etc.
You're right about those range safety systems being as simple as dirt. The commands consist of simple tone sequences, not much different from telephone Touch Tones. Last I heard, the frequency was somewhere around 410 MHz, with ground transmitters generating some tens of kilowatts -- there's no hiding them. (I'm told that listening for them was a good way to tell when a secret launch was coming up, before they lightened up and began to announce them a few hours before the event.) The range safety received signal strengths are telemetered to the ground, and anything less than full-scale at any time on any receiver is considered cause for concern.
-
I wonder what the range safety profile is for the Delta IV. The Saturn V had a rather complicated one that inhibited engine shutdown for the first 30 seconds so that the vehicle would not fall back on the pad. I'm not sure how this would have worked given that a single engine failure at liftoff left a total remaining thrust less than the vehicle weight. But the S-IC gobbles its propellants pretty quickly, so it probably didn't take long for this particular vulnerability to go away.
They seemed as concerned about engine gimbal hardovers as premature shutdown, and some of the failure scenarios looked pretty bad no matter what you did. Still, they took some precautions such as canting the four outboard engines slightly outward. This was said to make it easier in case of an engine hardover, but I'm not sure how.
-
So what about antennas?
Patches around the circumference. It's literally impossible for there not to be a range-safety antenna available.
but I don't know about telemetry transmitters, radar transponders, TDRSS terminals, TV transmitting antennas, etc.
The Delta guidance system and ground-support has a real-time ongoing telemetry margin monitor. If you invade the link margin, that becomes a guidance input. Naturally it's prioritized among wind and ordinary trajectory management, but Boeing really thought this through. That's why it was a pleasure to work with both their launch-vehicle team and their spacecraft engineering team.
Last I heard, the frequency was somewhere around 410 MHz, with ground transmitters generating some tens of kilowatts -- there's no hiding them.
ULA has put practically everything in the C- and S-bands these days, and there's an ongoing effort to uniformalize with the Atlas. Within then next 10 years they will fly the same guidance software and the same telemetry and up-/downlink systems.
-
I wonder what the range safety profile is for the Delta IV.
Very hard to find these days. I had to go back to my old Boeing documentation for Delta II and III to get the designs for ordnance safeties, and I can't find any detailed information online. The Delta IV User Guide describes the range safety procedures at a very high level, but that material is aimed at payload designers and doesn't really go into detail about launch vehicle design policies.
They seemed as concerned about engine gimbal hardovers as premature shutdown, and some of the failure scenarios looked pretty bad no matter what you did.
Bending moments for launch vehicles are heinous and unforgiving. You don't hardover an engine on a large launch vehicle unless several important people are going to die otherwise.
Still, they took some precautions such as canting the four outboard engines slightly outward. This was said to make it easier in case of an engine hardover, but I'm not sure how.
The ratio of hardovered engine to the opposing moment in the same plane would be reduced. But I agree with you; it still seems puzzling.
-
But doesn't the Air Force dictate the range safety requirements, and aren't they more than a little conservative when it comes to technology?
I've heard people complain about the ancient design and high cost of these receivers, but given the consequences of their not working when required I can sort of understand the Air Force's concern. The US has a perfect record of protecting uninvolved persons and private property on the ground. Several other spacefaring countries can't make this claim.
-
But doesn't the Air Force dictate the range safety requirements...
Yes. But a fair amount of that isn't static.
...and aren't they more than a little conservative when it comes to technology?
Yes. And for every iota of effort toward, "The vehicle must explode completely when told to," there are two iotas of, "Ordnance on your rocket may not explode prematurely or pose a hazard to integration and ground crews."
Both of those imperatives practically dictate a dirt-simple method of handling ordnance, both on the ground and in flight. So yes, three-factor safety where one of the factors is a physical pin pulled by a physical lanyard when the vehicle clears the tower is not at all out of keeping with the Air Force philosophy.
Some of the components we build for rocketry do indeed achieve or exceed "five nines" of reliability. We do not lightly set aside those designs, no matter who archaic or simplistic they may seem to the layman.
-
...a physical pin pulled by a physical lanyard when the vehicle clears the tower...
Whenever there's a question like, "How can we best get these two things working together?" there's a certain nobility and excellence in the best answer being, "Piece of string."
-
But doesn't the Air Force dictate the range safety requirements, and aren't they more than a little conservative when it comes to technology?
I've heard people complain about the ancient design and high cost of these receivers, but given the consequences of their not working when required I can sort of understand the Air Force's concern. The US has a perfect record of protecting uninvolved persons and private property on the ground. Several other spacefaring countries can't make this claim.
I'm not so sure about the private property part. On Jan. 17 1997 a Delta II carrying a GPS satellite "auto destructed" 13 seconds into flight after a SRB casing failed. Debris fell on a parking lot near the blockhouse complex, and damaged or destroyed 20 cars parked there. https://en.wikipedia.org/wiki/GPS_IIR-1#Debris
-
You're quite right, I'd forgotten about that parking lot. In my defense, I can argue that those cars belonged to NASA/Air Force/ULA/whoever employees, not the general public (for the most part).
I assume they were all compensated?
But I think they still have a perfect record in protecting human lives.
The possibility of a failure immediately after liftoff is what keeps launch vehicle and range safety people up at night. The recent Antares failure is a perfect example; I'm actually surprised the pad damage wasn't more severe. It seems to have fallen just slightly southeast of the pad. I quickly noticed that before the launch there were four lightning protection masts; just after, only two.
Thank our lucky invisible daytime lunar stars that we never had a S-IC failure in, what, 13 flights?
-
Yes. And for every iota of effort toward, "The vehicle must explode completely when told to," there are two iotas of, "Ordnance on your rocket may not explode prematurely or pose a hazard to integration and ground crews."
I'm no ordnance expert, but I'm rather amazed at its reliability on spacecraft and launch vehicles. They're almost boring: they fire when you want them to, and they don't fire when you don't want them to.
I know they go out of their way to ensure this with redundant initiators, firing units, det cords, dedicated batteries, isolated wiring and the like, but it's still quite an achievement. I think Curiosity fired over 70 pyros in its landing sequence, and a failure of just one could have ended it all right there.
When I first heard that explosives are often used just to open a valve, I almost wondered if my leg was being pulled. But many valves and the like only have to open once, and since long experience has shown pyros to be very reliable, they're still used despite the obvious hazards, costs and testing difficulties.
I know there have been failures, e.g., during the launch of the Skylab space station when second-plane separation failed after S-IC/S-II staging. But that seems to have been more of a systems-level slipup than an actual failure of the ordnance itself.
-
You're quite right, I'd forgotten about that parking lot. In my defense, I can argue that those cars belonged to NASA/Air Force/ULA/whoever employees, not the general public (for the most part).
I'll admit to nitpicking. :)
I assume they were all compensated?
I'd have to assume that, too.
But I think they still have a perfect record in protecting human lives.
Indeed.
The possibility of a failure immediately after liftoff is what keeps launch vehicle and range safety people up at night. The recent Antares failure is a perfect example; I'm actually surprised the pad damage wasn't more severe. It seems to have fallen just slightly southeast of the pad. I quickly noticed that before the launch there were four lightning protection masts; just after, only two.
Thank the invisible daytime lunar stars that we never had a S-IC failure in, what, 13 flights?
The Soviet N1 launch failure on July 3, 1969 gives us some idea what an early-in-flight S-1C failure might have looked like, so I'm with you on thanking those stars.
-
The Soviet N1 launch failure on July 3, 1969 gives us some idea what an early-in-flight S-1C failure might have looked like, so I'm with you on thanking those stars.
Not a pretty sight for rocket enthusiasts
-
...during the launch of the Skylab space station when second-plane separation failed after S-IC/S-II staging. But that seems to have been more of a systems-level slipup than an actual failure of the ordnance itself.
Could you discuss this event in a bit more detail, please.
That is, I already know (though I don't remember where I found out) that the S-IC/S-II interstage failed to separate for the launch of Skylab, and I know (from the Apollo Flight Journal) that normally an interstage separation failure meant an automatic immediate abort. I also know that the abort was overridden for Skylab, rather than being inhibited ahead of time.
But I'd be curious to know any background information about exactly what happened and why, and how Mission Control got around it.
Thanks very much.
-
...during the launch of the Skylab space station when second-plane separation failed after S-IC/S-II staging. But that seems to have been more of a systems-level slipup than an actual failure of the ordnance itself.
Could you discuss this event in a bit more detail, please.
That is, I already know (though I don't remember where I found out) that the S-IC/S-II interstage failed to separate for the launch of Skylab, and I know (from the Apollo Flight Journal) that normally an interstage separation failure meant an automatic immediate abort. I also know that the abort was overridden for Skylab, rather than being inhibited ahead of time.
But I'd be curious to know any background information about exactly what happened and why, and how Mission Control got around it.
Thanks very much.
A long and detailed report here
http://klabs.org/richcontent/Reports/Failure_Reports/Skylab/Skylab_Report.htm
CHAPTER X
SIGNIFICANT FINDINGS AND CORRECTIVE ACTIONS
Significant Findings
The launch anomaly that occurred at approximately 63 seconds after lift-off was a failure of the meteoroid shield of the OWS.
The SAS-2 wing tie downs were broken by the action of the meteoroid shield at 63 seconds. Subsequent loss of the SAS-2 wing was caused by retro-rocket plume impingement on the partially deployed wing at 593 seconds.
The failure of the S-II interstage adapter to separate in flight was probably due to damage to the ordnance separation device by falling debris from the meteoroid shield.
The most probable cause of the failure of the meteoroid shield was internal pressurization of its auxiliary tunnel. This internal pressurization acted to force the forward end of the tunnel and meteoroid shield away from the OWS and into the supersonic air stream. The resulting forces tore the meteoroid shield from the OWS.
The pressurization of the auxiliary tunnel resulted from the admission of high pressure air into the tunnel through several openings in the aft end. These openings were: (1) an Imperfect fit of the tunnel with the aft fairing; (2) an open boot seal between the tunnel and the tank surface; and (3) open stringers on the aft skirt under the tunnel.
The venting analysis for the tunnel was predicated on a completely sealed aft end. The openings in the aft end of the tunnel thus resulted from a failure to communicate this critical design feature among aerodynamics, structural design, and manufacturing personnel.
Other marginal aspects of the design of the meteoroid shield which, when taken together, could also result in failure during launch are:
a. The proximity of the MS forward reinforcing angle to the air stream
b. The existence of gaps between the OWS and the forward ends of the MS
c. The light spring force of the auxiliary tunnel frames
d. The aerodynamic crushing loads on the auxiliary tunnel frames in flight
e. The action of the torsion-bar actuated swing links applying an outward radial force to the MS
f The inherent longitudinal flexibility of the shield assembly
g. The non-uniform expansion of the OWS tank when pressurized
h. The inherent difficulty in rigging for flight and associated uncertain tension loads in the shield.
The failure to recognize many of these marginal design features through six years of analysis, design and test was due, in part, to a presumption that the meteoroid shield would be "tight to the tank" and "structurally integral with the S-IVB tank" as set forth in the design criteria.
Organizationally, the meteoroid shield was treated as a structural subsystem. The absence of a designated "project engineer" for the shield contributed to the lack of effective integration of the various structural, aerodynamic, aeroelastic, test, fabrication, and assembly aspects of the MS system.
The overall management system used for Skylab was essentially the same as that developed in the Apollo program. This system was fully operational for Skylab; no conflicts or inconsistencies were found in the records of the management reviews. Nonetheless, the significance of the aerodynamic loads on the MS during launch was not revealed by the extensive review process.
No evidence was found to indicate that the design, development and testing of the meteoroid shield were compromised by limitations of funds or time. The quality of workmanship applied to the MS was adequate for its intended purpose.
Given the basic view. that the meteoroid shield was to be completely in contact with and perform as structurally integral with the S-IVB tank, the testing emphasis m ordnance performance and shield deployment was appropriate.
Engineering and management personnel on Skylab, on the part of both contractor and government, were available from the prior Saturn development and were highly experienced and adequate in number.
The failure to recognize these design deficiencies of the meteoroid shield, as well as to communicate within the project the critical nature of its proper venting, must therefore be attributed to an absence of sound engineering judgment and alert engineering leadership concerning this particular system over a considerable period of time.
Corrective Actions
If the back-up OWS or a similar spacecraft is to be flown in the future, a possible course of action is to omit the meteorold shield, suitably coat the OWS for thermal control, and accept the meteoroid protection afforded by the OWS tank walls. if, on the other hand, additional protection should be necessary, the Board is attracted to the concept of a, fixed, nondeployable shield.
To reduce the probability of separation failures such as occurred at the S-II interstage Second Separation Plane, both linear shaped charges should be detonated simultaneously from both ends. In addition, all other similar ordnance applications should be reviewed for a similar failure mode.
"Structural" systems that have to move or deploy, or that involve other mechanisms, equipment or components for their operation, should not be considered solely as a piece of structure nor be the exclusive responsibility of a structures organization.
Complex, multi-disciplinary systems such as the meteoroid shield should have a designated project engineer who is responsible for all aspects of analysis, design, fabrication, test and assembly.
Observations on the Management System
The Board found no evidence that the design deficiencies of the meteoroid shield were the result of, or were masked by, the content and processes of the management system that were used for Skylab. On the contrary, the rigor, detail, and thoroughness of the system are doubtless necessary for a program of this magnitude. At the same time. as a cautionary note for the future, it is emphasized that management must always be alert to the potential hazards of its systems and take care that an attention to rigor, detail and thoroughness does not inject an undue emphasis on formalism, documentation, and visibility in detail. Such an emphasis can submerge the concerned individual and depress the role of the intuitive engineer or analyst. It will always be of importance to achieve a cross-fertilization and broadened experience of engineers in analysis. design, test or operations. Positive steps must always be taken to assure that engineers become familiar with actual hardware, develop an intuitive understanding of computer-developed results, and make productive use of flight data in this learning process. The experienced "chief engineer," who can spend most of his time in the subtle integration of all elements of the system under his purview, free of administrative and managerial duties, can also be a major asset to an engineering organization.
It seems that NASA didn't learn from this, hence, 12 years later, the Challenger accident.
-
Smartcooky beat me to it with the facts. NTRS is still massively broken after the decommissioning of CASI and the subsequent breakage of many links to technical reports.
*cracks knuckles*
This is my special area of (somewhat dated) expertise. My work on the Delta family of launch vehicles (chiefly the Delta III but also backward toward the Delta II and forward toward the Delta IV) was the payload-bearing structures and their relevant coupled-load analysis and separation sequence analysis. The half-dozen or so payload-attach fittings (PAF) for those vehicles make special use of pyrotechnical and ingenious mechanical solutions.
Was the staging anomaly on Skylab 1 the failure of a single pyrotechnic device? Yes. Normally we design these assemblies with redundant pyros so that the failure of any one device doesn't fail the system. Here they didn't. A single device was meant to sever all the straps holding the structures together. Any mechanical or thermal damage to that single device meant that the detonation front would be interrupted at the point of damage, which is what occurred on that flight. Systemic redundancy was limited to multiple detonation initiations of the single device, implemented electronically but not mechanically. Well, sort of. The electrical connectors that pass the "fire" command (both primary and redundant) to the device separate after less than a centimeter of separation has occurred between the interstage and the second stage, and only at the point where the connectors are mounted on the skirt. This means the interstage can partially separate while defeating the electrical redundancy. This is considered a component-level design factor, not a system-level design factor.
In the sense that this was dependent on electrical properties of the stage design and staging sequence design, this was a system-level design failure. The designers should have foreseen that mechanical damage to the single pyro -- which had a massive mechanical extent in the overall vehicle design -- could result in partial separation of the interstage, which could defeat the electrical-only redundancy program. A better design using the existing hardware would have initiated the circumferential pyro detonation simultaneously at both ends. This would have meant that a single discontinuity would have been perfectly sustainable; the interstage joint would have to be hit two or more times to result in partial pyro detonation.
The system failure argument incorporating the micrometeoroid shield is valid but weak. Clearly there was a failure to communicate design constraints between design teams, and that's what I did for the Delta LV team back in the day. It's weak because specific causes of debris damage aren't important to LV stage designers. You design launch vehicles (especially those with cryogenic propellants) to tolerate impact damage from any source, including from unknown sources. Literally anything forward of the point in question can break off and slam into the launch vehicle with supersonic force.
And yes, Challenger. The STS design was fully head-in-sand about this. By the early 1970s it was already known that launch vehicles were likely to suffer minor impact damage during boost. Modern launch vehicles had already been adapted to this fact. The Saturn V, by the early 1970s, was no longer a modern launch vehicle in this respect. It could be made resilient in the face of this particular failure by the modification I outlined above. But as this was the last flight of the Saturn V, that wasn't really a consideration. STS vacillated about this, with the knowledge that the orbiter TPS was especially vulnerable to impact damage and the naive expectation that debris shedding could be controlled with additional engineering.
Nope.
Payload attachment strategies today typically use clampband designs. The payload mates to the circular adapter structure in such a way that the shared circular "lip" is a V-shaped arrangement that accepts the V-shaped concave channel of a strong band going around the circumference. At 0-degree and 180-degree radials, the band is held in place and tensioned by studs that are severed by pyro cutters. Only one of the two cutters has to succeed in order for the clampband to release, and since they are at diametrically opposed points on the vehicle, the chances of them both being damaged by impacts is minimized.
Full-race linear shaped charges these days are permitted only in a few places, such as payload fairing separation. Since the payload fairing is the most forward structure, it isn't likely to be damaged mechanically. Gemini's "angry alligator" showed the problems with clampband methods for separating payload fairings, but a few designs still use(d) it (e.g., SpaceX Falcon 1).
But back to stage joinery. Rocket stage joints have to accept bending-moment loads on the order of 80,000 lbf-in. This requires them to be especially robust in tension around the perimeter of stage. This further requires separation devices also to be located near the vehicle skin and be susceptible to debris impact damage. Newer interstage joint designs make more extensive use of fuze pins and other simply-frangible structures to ensure both sufficient strength to withstand flight loads and reliability at separation.
-
...normally an interstage separation failure meant an automatic immediate abort.
For a manned mission, yes. Because the structural and thermal effects of carrying the interstage could mean an eventual catastrophic failure from which a safe abort would be problematic. Aborting during nominal flight was considered more survivable and the safe thing to do in this contingency.
For an unmanned mission you don't mind a catastrophic failure so much during second-stage flight. Only property would be destroyed. Hence you can fly more risky safety margins.
But I'd be curious to know any background information about exactly what happened and why, and how Mission Control got around it.
To summarize the voluminous information above, debris from Skylab's disintegrating shield damaged the separation pyro. That was traced to insufficient integration engineering.
Structurally there's little problem with leaving the interstage attached. The resulting vehicle is heavier and requires longer to reach the target velocity, but that's still way within the second stage's capacity. Ostensibly a hardover on the J-2 engines might have forced contact between the nozzle and the interstage, but that would have created other problems anyway. The worst effect was that the interstage collected and concentrated exhaust gases that flow up around the nozzle. That heated up the elements of the stage that were just behind the aft shield, but not to a dangerous extent.
-
When I first heard that explosives are often used just to open a valve, I almost wondered if my leg was being pulled. But many valves and the like only have to open once, and since long experience has shown pyros to be very reliable, they're still used despite the obvious hazards, costs and testing difficulties.
Consider also that some of the valves in question require considerable actuating force. Some are meant to hold back pressurized LH2 or LOX -- rather small molecules that require considerably tight fits for the valve components. Permissible leak rates could be very small. The combination of very tight mechanical tolerances and very high upstream pressures and very high eventual flow rates means it could require an enormous torque to open or close the valves. You can either try to apply one of the various fluid-power systems such as pneumatics and hydraulics. Or, depending on location and reliability constraints, you could us a pyro device. It requires only wiring, which is more robust than fluid-power piping.
The testing is still a problem. No ground-test machine exists to faithfully simulate the shock effects of pyro detonations on a payload or other structure, which can be prodigious. So the only way to do it is to fasten the payload to a test fixture that closely resembles the actual payload fitting, install actual pyros, back way the heck up, and set them off. Then you see if your satellite boots up afterward. And yes, I have seen spacecraft appendages fall to the test floor when the "simulated" sep pyros fire.
But pyros are dangerous. Even with today's containment structures, you don't want to be anywhere near them when they go off, which is why installing and arming them on launch vehicles and/or payloads involves a bunch of grim-faced Air Force guys watching closely every move you make. The buttload of paperwork required to clear the pyro designs is phenomenal.
-
Your reference to Challenger - I suppose you mean Columbia?
Edit: And aborting an unmanned mission versus a vehicle failure - the payload would be lost anyway, I suppose. So an abort with vehicle destruction would only be necessary if a major course deviation occured. Any orbit, I suppose, is better than self-destruct.
-
No, Challenger, which was brought down by debris striking the orbiter during initial boost. Happily the crew completed their mission before the catastrophe, but the tragedy is that we've known literally for decades that you can't rely on controlling the debris streaming back from a launch vehicle.
-
No, Challenger, which was brought down by debris striking the orbiter during initial boost. Happily the crew completed their mission before the catastrophe, but the tragedy is that we've known literally for decades that you can't rely on controlling the debris streaming back from a launch vehicle.
Challenger exploded during take-off due to faulty o-rings in the booster. Columbia disintegrated during re-entry due to debris strike during take-off which compromised the thermal protection.
-
Er, yeah, Columbia. ;D
-
We need another T-shirt here.
-
Indeed, I was having an entire day where I was constantly thinking one thing and saying/writing another. I finally just went home to pet the cat. I mean the dog.
-
Indeed, I was having an entire day where I was constantly thinking one thing and saying/writing another. I finally just went home to pet the cat. I mean the dog.
It's almost a shame those t-shirts are mythical.
With their rarity, a mercenary or desperate winner could probably sell their authentic one to the Blunder or Adrian etc. , for a nice premium price.;)
-
Indeed, I was having an entire day where I was constantly thinking one thing and saying/writing another. I finally just went home to pet the cat. I mean the dog.
or the macaw...
-
This is considered a component-level design factor, not a system-level design factor.
In the sense that this was dependent on electrical properties of the stage design and staging sequence design, this was a system-level design failure.
I guess this is a matter of semantics. If a given pyro fails to fire despite being installed correctly and undamaged at the moment of firing, that is clearly a component failure. This seems extremely rare in modern spacecraft and launch vehicle pyros.
But if the pyro fails to fire because it's damaged by something else, it's kind of hard to ascribe that to a pyro component failure. Sure, a separation system might have been designed to withstand a specific kind of damage, e.g., by initiating a linear shaped charge at both ends simultaneously, but it wouldn't have failed without that external damage. Any component can fail if sufficiently damaged, and it's hard to blame that strictly on the component. So I'd classify this as a system problem.
Regarding the effect of the second-plane separation failure on Skylab, I had understood it did not trigger an abort because there was so much mass margin that the vehicle had no trouble making it to orbit. My understanding is that there was so much less mass margin on a lunar mission that just the mass of the interstage would have prevented a successful TLI. You seem to be saying this is not true, is that right?
-
I guess this is a matter of semantics.
Yes, or more specifically a matter of scope. That's why I wrote two paragraphs with a different perspective and a different answer. You can consider each pyro initiator a component along with the ordnance load, or the whole installed pyro assembly together a component. It really matters only when you want to have a specific kind of engineering discussion to exclude the other paradigm.
As I said, I believe there's a system-level argument to be had based on how the separation sequence was designed to proceed. There are holes in that design (no pun intended) that would properly be system-level design questions. And as you note, there's a system-level design argument to be had at a broader scope that incorporates not only general hazards from the environment and the rest of the vehicle but also the specific hazards from the payload. If you want to launch on a Delta, you spend a lot of time in the payload integration and integration testing phase.
You seem to be saying this is not true, is that right?
No, I'm not making the mass-budget argument, although I know that the Skylab launch did indeed have a pretty huge margin. The mission report says there were no guidance anomalies aside from a slightly longer S-II burn. But I'm not sure about the margin for the typical Apollo stack. My impression, talking to the ordnance engineers from Boeing, was that it was strictly for safety -- the abort for a manned mission. An immediate abort wouldn't be a mission rule if it were just a performance issue.
The concerns I've always heard from Boeing are (1) the aft skirt heating -- which happened on Skylab 1 -- and (2) the various aerodynamic, structural dynamic, and J-2 interference issues from a partially-separated interstage. Ostensibly you'd have a few minutes of flight following an indication of S-II interstage separation failure to decide on the abort, but the gist of what I heard was that unless it separated cleanly, things could conceivably go very bad very quickly. Better to abort when you can command a clean S-II shutdown and not give the LES too much to do, instead of trying to separate from a possibly tumbling, possibly damaged and burning vehicle.
-
Here's the joint in question. Or rather, on a test fixture. The charge is the white band being fitted to the joints. The tension straps are the short pieces of metal that connect the stringers (the ridges that stick out). For flight, the charge was covered with an aerodynamic fairing that can be seen as two narrow white bands near the top of the S-1C, but that's essentially it -- no significant protection from debris.
(http://www.clavius.org/img/sat-v-lsc-test-pre.jpg)
And here you can see the tension straps severed after the ordnance has fired.
(http://www.clavius.org/img/sat-v-lsc-test-post.jpg)
-
Is that a shaped charge? Looks more like a piece of quite heavy det-cord.
-
I agree it looks like det-cord, but I'm told it's the linear shaped charge by the person who supplied the photos, who was on the Boeing ordnance team that designed and installed it. As I'm sure you're aware, ordinarily the shaped charge for this application would have a V-shaped cross section with the concave side facing the straps. However, the encapsulation of the charge need not have the same cross section. There are rectangular and beveled encapsulations -- plastic, epoxy, etc.
You're used to the thin copper casings used in building demolition. Keep in mind the encapsulation doesn't have to contain pressure or contribute to the brisance of the ordnance. It just keeps the plastic explosives in the proper shape and protects it from light damage and contamination until it's needed. If you build the linear shaped charge (LSC) as a copper-clad, form-fitting component, it would have to be a circle in the same circumference as the S-1C and the S-II interstage. If you install it as an epoxy- or elastomer-clad unit, it has the same effect but it's easier to handle and install. In the top photo you can see the spool from which the ordinance package is being unwound.
-
Is there a marking on it which shows the cutting direction?
-
For rectangular cross sections, yes. For beveled cross section, the side away from the cutting direction bears the bevels. The cutting side remains flat.
-
I know this picture is from way back in 1963, but doesn't it look just a little strange to see a stereotypical engineer in white shirt, black tie and pocket protector handling explosives?
Well yeah, the guy to his left is actually holding the spool, and we can't really see what he's wearing...
Oh, and they're still working while the rest of the country was watching JFK's funeral...
-
Oh, and they're still working while the rest of the country was watching JFK's funeral...
I hadn't noticed, but you're right.
-
Oh, and they're still working while the rest of the country was watching JFK's funeral...
That has a nice sort of symmetry to it; engineers pressing ahead on the project that JFK started even on the day of his funeral.
-
They seemed as concerned about engine gimbal hardovers as premature shutdown, and some of the failure scenarios looked pretty bad no matter what you did. Still, they took some precautions such as canting the four outboard engines slightly outward. This was said to make it easier in case of an engine hardover, but I'm not sure how.
Are you sure that was for the case of a hardover and not an engine shutdown? Pointing the direction of thrust of each engine closer to the center of mass rather than parallel reduces torques if one underperforms or unexpectedly shuts down.
-
I'd have to go back and find the actual discussion, but I know they were concerned about both hardovers and shutdowns. The IU would quickly try to compensate, but the gimbals on the remaining engines probably couldn't move fast enough to avoid some momentary but catastrophic bending moments on the vehicle structure.
Not being a mechanical or aerospace engineer, it wasn't until I read some of the Saturn V flight reports that I realized just how critical those bending stresses in the vehicle structure really were. Imagine a car with an engine so powerful that merely losing traction on a tire or turning the steering wheel just a little too quickly or too far would rip your car apart even before it had a chance to roll over and crash.
Then again, judging from their subjective reports of what it felt like on top of a Saturn V as it lifted off the pad, the astronauts probably had a very keen awareness of these possibilities...
-
Not being a mechanical or aerospace engineer, it wasn't until I read some of the Saturn V flight reports that I realized just how critical those bending stresses in the vehicle structure really were.
A bending moment of 80,000 lb-ft was typical in a Saturn V flight. When you look at the construction, the miracle is how something that looks so flimsy can be so frakking strong.
-
A bending moment of 80,000 lb-ft was typical in a Saturn V flight. When you look at the construction, the miracle is how something that looks so flimsy can be so frakking strong.
Am I correct in my understanding that the thickness of metal varied along the length of each stage which caused unique welding problems, which made it even more of a miracle that it stayed together?
-
Along the length and around the circumference, with resonances and varying elasticity in nearly every dimension in which vibration could occur. This is why aerospace is the in high priesthood of both engineering and manufacturing.
-
Along the length and around the circumference, with resonances and varying elasticity in nearly every dimension in which vibration could occur. This is why aerospace is the in high priesthood of both engineering and manufacturing.
...and all designed without modern computers in the 1960s. I bet that was easy :o
Would tensor analysis featured heavily in the design?
-
It featured heavier in the guidance math than in the vehicle design. But yes, in the mid-1960s tensor analysis was one of the sexy mathematical techniques in high-end engineering.
-
It featured heavier in the guidance math than in the vehicle design.
I'm guessing tensors would be used in guidance math to store and process the parameters that describe postion and motion. I have seen them used in rotational dynamics. Would they also have been convenient from the perspective of making efficient computing algorithms given how powerful tensors are? While notoriously horrendous, I would imagine that their functionality and the operations that can be carried out on them make them convenient and very powerful for computing.
But yes, in the mid-1960s tensor analysis was one of the sexy mathematical techniques in high-end engineering.
Sexy??? Sexy??? :o Have you momentarily lost your mind? Having ploughed through a course of general relativity, the words sexy and tensor were never used in the same sentence... ever!!! ;)
-
While notoriously horrendous, I would imagine that their functionality and the operations that can be carried out on them make them convenient and very powerful for computing.
That's the thing. Do the horrendous math on the chalk board in order to devise correct and simple control laws for the computer.
Sexy??? Sexy??? :o Have you momentarily lost your mind?
I'm channeling Dr. Elaine Cohen, one of my advanced math professors in graduate school. That was her term for any technique or branch of mathematics that was especially powerful in what it could attain for practical purposes.
-
I'm channeling Dr. Elaine Cohen, one of my advanced math professors in graduate school. That was her term for any technique or branch of mathematics that was especially powerful in what it could attain for practical purposes.
Yes. Quite often ideas that are tricky to begin with prove to be extremely powerful and easier to use in the longer term. It is worth persevering. I found Lagrangian mechanics to be quite tricky when I first met it, but I put that down to the awful lecturing where we spent 3 lectures running through some esoteric mathematics. By that point I was switched off, but now understand the beauty and power of the Lagrangian. Anyway, this is running well off topic.
-
What an absolutely incredible discussion.
Where else would one find something like this....thanks to all.