Why The Delta III Rocket Exploded On Its First Flight - Why Rockets Fail
Вставка
- Опубліковано 29 лис 2019
- In 1998 Boeing debuted the new Delta III rocket with a payload demonstrating the larger capacity of the latest evolution of the Delta series rockets. However 72 seconds into the launch the vehicle was lost. The reason for the failure boiled down to a guidance system which used up the limited guidance capability of the strap on solid rocket boosters and ultimately lost control.
Lots of data was taken from Boeing's Delta III Payload guide
web.archive.org/web/200111191...
Also Ed Kyle's history of the Delta III provides lots of good links:
www.spacelaunchreport.com/tho... - Наука та технологія
I worked on Delta III. I was part of the team that developed the composite components: The 4 meter fairing, The 4 meter 1666 Payload Attach fitting and the 4 meter composite interstage where the new was upperstage hung inside.
It was heart breaking to watch that 1st launch. When I put many hours and heart and soul into developing a new launch vehicle then watch it blow up almost brought me to tears.
I hope further, later launches was successful and that somehow the problems of this maiden launch was overcome.
Sorry I am intrigued by the name/reference 1666 Payload, unless a classified code, was it related to lb's or kg's, or from a historical date ? ..excluding the Great Fire of London as a too obvious and too British an event for the Americans.
@@razor1uk610 1666 references the diameter in centimeters of the payload interface. It's a common interface diameter for commercial satellites, usually Boeing commercial satellites..
As it does to us all.
Can anyone please please give me something like 66million dollars to give humanity absolutely NOTHING!!!?
Dat hurt
My favorite part of this video is at 5:16, where they have the "Laws of Nature" as a component of their control system
It's not just a suggestion, it's the *law*
I mean, that is a good thing to take into account.
Why waste money on programming for the control surfaces if mother nature volunteers to do it for you?
That part is in there only as a placeholder for "all the uncontrollable and not fully knowable stuff doing things to our vehicle that we then have to try to counter".
Lol, I write control systems though and this is common practice. Just makes for clutter if you include stuff that is static anyway
Many of my KSP creations have been destroyed because of overzealous gimbaling.
Mine too. The damping factor is not controllable in the stability assist, so if the gimballing is too good, the rocket oscillates.
@@adamrak7560 The vanilla SAS is not nearly as bad as MechJeb though. I love the ability to point to an angle with MechJeb, but the damping is terrible. It's probably configurable, but it needs better defaults anyway.
Seems like putting amount of hours played in KSP on a CV may not be a terrible idea.
Thats why I keep wanting KSP to implement user defined control programming. They have done an absolutely awful job with the automatic control stability.
Same, just about to comment the same thing
I worked on the RS-27A abd RS-68 engines. Specifically, I designed and built automated test systems for their engine control units (ECU). So your video brings back memories for me.
The RS-27A had a dome-shaped ECU that was sometimes called the "salad bowl". It's logic circuitry used diodes and electromechanical relays. The guy who designed the ECU (Rudy something) was a young engineer when he did it, but when I joined to project (to replace the ancient ECU tester with a modern one in the late '90s) Rudy was an old fart and I went to his retirement party. The old tester was so poorly designed that it didn't work whenever it was raining outside. Seriously.
The RS-68 ECU used microcontrollers and FPGAs. The guy who ran that team frequently called himself "a damn fine systems engineer" but in reality he had NO IDEA what he was doing; he just wanted a really big team so he could feel important. For example, there were about 20 software engineers on the team to design firmware that had to fit in a 2kB ROM. Once, someone raised the question of software reliability, and this guy laughed like it was the funniest thing he'd ever heard, and said "software doesn't break!" with attitude in his voice, like he was talking to an idiot. I think a few months after that an Ariane rocket blew up because of a firmware bug in it's ECU. I could fill a book with all the stupid, ignorant things that guy did.
You might be the only one to archive the stories. I for one love to read about the real goings-on in other spheres.
And at least 63% of the dorks here, myself included, would love to buy that book.
There are buffoons at every level. I once watched a crew supervisor in the "shake & bake" labs at Martin Marietta (1980s, Waterton CO Space Systems campus) balance a launch vehicle's primary rate gyro package on his head and spin it. It was flight hardware. We were standing in a clean room too!
They promoted him so he wouldn't be touching flight hardware.
At the same joint, I watched a test technician attempt to roll a full-height rack stuffed with maybe half a million USD$ of high-end HP (then, now Agilent) RF test gear down a ramp alone. Caster caught on a carpet edge. (yes, the ramp had carpet on it - the place was ghetto AF) Rack stopped rolling. The tech pushed harder, and over it went, like a felled tree, landing on its side, BOOM! It was heard throughout the building.
They promoted her so she would no longer be touching hardware.
This proved to be a pattern there. Given that their Titans went from being the most reliable launch vehicle at the time to the least reliable in the space of 36 months, it was my impression that the whole organization was a dying dinosaur from top to bottom which would not survive the end of the cold war. If they hadn't merged with Lockheed, they'd have vanished.
How come a Delta IV hasn't blown up because of him?
@@railgap god, I was in charge of our electronics lab at my institute, and you bringing up memories of Agilent VNAs, Cascade Microtech probe stations just caused me pain. I've had to replace cables, probes etc so many times, because people would crossthread the connectors ("I didn't get a good signal so I tightened it more, now it doesn't work"->after being told not to, of course)...mind you, this is at one of the top universities in the world. Or my pet peeve: We have 4 probe stations. We have several DC measurement instruments. All of them can communicate with the host PCs. One of them is designed specifically to interface with 2-3 other 100k$+ pieces of equipment and has software written for it to make automated, wafer scale DC+RF measurements and analysis. It can also run standalone DC, which all the other units can too. But of course, they're too lazy to read the manual for 5-10 minutes, so they always wheel the 200k$ rack around, so we can only use 1/4 of the equipment at a time, and then they complain about lack of availability. Can't wait to get out...
As my engineering prof was fond of saying "Then bad things happen in rapid succession!"
Lesson learned: Pull that control authority slider on your gimbals down, people!
System instability: exists.
LOL .. if only 1960's rocket designers had access to the KSP as a design simulator!!!
The rocket knows where it is at all times. It knows this because it knows where it isn't. By subtracting where it is from where it isn't, or where it isn't from where it is (whichever is greater), it obtains a difference, or deviation. The guidance subsystem uses deviations to generate corrective commands to drive the rocket from a position where it is to a position where it isn't, and arriving at a position where it wasn't, it now is. Consequently, the position where it is, is now the position that it wasn't, and it follows that the position that it was, is now the position that it isn't.
In the event that the position that it is in is not the position that it wasn't, the system has acquired a variation, the variation being the difference between where the rocket is, and where it wasn't. If variation is considered to be a significant factor, it too may be corrected by the GEA. However, the rocket must also know where it was.
The rocket guidance computer scenario works as follows. Because a variation has modified some of the information the rocket has obtained, it is not sure just where it is. However, it is sure where it isn't, within reason, and it knows where it was. It now subtracts where it should be from where it wasn't, or vice-versa, and by differentiating this from the algebraic sum of where it shouldn't be, and where it was, it is able to obtain the deviation and its variation, which is called error.
Trying to read that with a hang-over from hell. I'm just gonna go back to bed. That made my head hurt
@@zuestoots5176 ua-cam.com/video/bZe5J8SVCYQ/v-deo.html
use as lullaby
But where am I?
ua-cam.com/video/bZe5J8SVCYQ/v-deo.html
Sounds a lot like cricket.
When you said “the third Delta III” I half expected you to follow up with “burned down, fell over, and _then_ sank into the swamp” 😂
😆😆😆😆😆
But the fourth one stayed up.
🎶"You will not go to space today..."🎶
When your Delta Three rocket is oversteered...
came to say this. "if your booster runs out of booster fuel, you will not go to space today" - god, now I have to watch it again. I wish Scott's daughter wasn't so young, I get vaguely creepy vibes from repeatedly watching the daughter of "some guy on the internet" sing a silly song, but dammit, it's funny!
PIO - pilot induced oscillation. When you start over compensating for your over controlling.
See it on highways all the time, especially idiots in 4WD vehicles driving too fast for conditions, thinking that 4WD is a magic spell which gives them a hall pass through the laws of physics.
Definitely haven't done that before...
I love this series! They're like detective stories...with rockets. This one is a lot easier to understand than that 1950's one from a few months back.
I saw a Delta II launch once in the early 70s or 80s. We were taking a tour of the cape and just at the end of it, the tour guide asked us if we'd like to see a rocket launch. Naturally, everyone on the bus was willing to stay a little longer...
The bleachers we went to were very close to the pad. I don't know how far, but I bet it wasn't more than a couple of miles. (Safety wasn't as big of a concern in those days.) Anyway, the launch was at sunset and the combination of the low light, close distance, and the solid rocket boosters made it the most spectacular launch I've ever seen, and I have seen a Saturn V, a couple of space shuttles and a Falcon 9.
Truly a great treat!
cool
Congratulations in advance for reaching *1 million subscribers!*
It feels like it’s taken forever. Maybe that’s because I really want this channel to grow.
@@jerry3790 Well, the big KSP days are a bit over. Maybe a 2nd wave when KSP2 hits? A whole generation of KSP players got to learn how to play it by following Scott.
My father worked on the Delta II's and the Delta III's as a propulsion engineer. I remember watching the launches and failures from Cocoa Beach. He was also in the blockhouse when the Delta II blew up just off the pad in 1997...Even with those failures, my dad had a very long and successful career with McDonnell Douglas and Boeing. He retired around 2000. Spaceflight was in our family's blood, I was a second generation space worker as an aerospace technician on the Space Shuttle Orbiters. Always love your videos!
They say space is hard..
So you know that it's okay
That your rocket just crashed into the ground
and you didn't go to space today
I remember this
No! No, it's not ok😐
Takeaway: don't trust the new one just because the old one was great. Let some fanboys test out the new one, and then invest if they survive that test.
Pronto not a lesson Boeing took to heart.
The takeaway is Test fully before you light the fire.
:D thats my favorite programming problem! Changing control outputs to direct a system state that has control and sensing latency. It comes up eeeeverywhere, from simple dc voltage regulators to drone flight controls to massive power plant burner and turbine controls. Its one of the reasons not to buy a knock-off hoverboard: even if it had identical electronics (which it won't) it probably has garbage-tier control programming and will throw you off the first chance it gets.
Awesome video, great explanation! I can't imagine how much research you do for each video like this. Thanks!
Always appreciate your commentary and analysis. I learn a bit with every segment!
Always amazed by the people who create and program the flight and guidance computers. This rocket fought hard to save itself. Another is the shuttle Columbia. She fought like hell to keep flying until the airframe disintegrated.
This is my favorite style of video of yours. Great video
Scott,
Love these 'Why rockets fail' Vids, and the time you put into researching and making them 👍
More please 😊
Well done production Scott. Thank you!
Fly safe indeed - informative and brilliantly delivered as always.
I absolutely love your videos like this and learn about rockets and satellites! I’m a huge fan of the Delta II and I always wondered why it never flew! Thanks Scott keep it up!!
And yet another fantastic video!
4:07 - As soon as you said "Three of the solid rocket boosters had gimbaling nozzles", I instantly knew what went wrong -- but I watched the rest of the video anyway.
Great overview as always
Thanks for sharing :-)
I appreciate that the shuttle you used in this intro is Endeavour, as its final flight was on my birthday!
I have a question Scott. Do the lightbulbs on the launch tower break during every launch? Do a lot of launch platform systems need maintenance after a launch?
Man, those shuttle SRB's are fast reacting!
Sounds exactly like an sas malfunction in ksp. Reduce gimbal range or fin control authority.
If only main-stream science programs could be this informational. Thanks Scott.
Fred Derf I grew up on the Discovery Channel, and I used to lament how long-dead that is, but then made peace with it being over and done with as a great thing in the past. Now we have things like Scott Manley, Veritasium, Smarter Every day, etc. The whole paradigm for where to get quality edutainment is shifted. TV is dead but don’t cry over it, embrace what we have today instead that’s just as good as what we used to have.
ALMOST makes you think, doesn't it? Good thing you found Scott to think for you!
Arianne 5 and Delta 3: both inaugural launches lost because they reused computer hardware from earlier rockets and didn’t bother to update the code for the new one.
off topic- hearing the word "gimblling" puts a smile on my face every time - thank you Lewis Carroll
I love this series. "Space is hard" in action.
hello Mr. Manley, I was wondering if you could tell me were you got those star-ship models on the shelf behind you. love the video as well :D
It should be recalled the Delta III RS-27A rocket engine had an interesting lineage. The RS-27A was an improvement of the RS-27. RS-27 was a development of the H-1 engine of Saturn I and IB rockets. The H-1 was a follow on engine based on the Rocketdyne S-3D that powered both the PGM-19 Jupiter and PGM-17 Thor missiles and the Juno II rocket. The S-3D shared its lineage from the Rocketdyne LR-79. Rocketdyne developed the LR-79 engine in 1955-56 for the U.S. Army. The LR-79 was derived directly from the German V-2 rocket technology.
The Delta II was finally retired last year, and the "single stick" Delta IV retired this year. Note that of the "Delta" family only the Delta IV Heavy is still in use.
Great explanation of the rocket mate , 6.51 the engine under power gimbaling looks brutal , hard yakka for the hydraulics .
Very good video
AW YES! I love Why Rockets Fail episodes!
Nice video... I can tell Scott's got a soft spot for this interesting but unfortunate rocket.
He forgot to mention that the 4m DCSS developed for Delta III lived on until the final Delta IV medium launch on August 22 2019 of the GPS III-2 satellite, together with the final 2 GEM-60 SRBs that grew from the GEM-40 and GEM-46...
TFW you look at youtube and see a video from Scott Manley posted 21 seconds ago.
Incredible, don't forget to mention it to your mom when you have a chance!
Adam
I‘ll do it for him when i see her tonight
In addition to the spectacular explosion of the vehicle, the payload completely separated from the upper stage and continued a ballistic trajectory down range. A group of us sat at a local beach side bar discussing what we just saw when several minutes after the initial explosion there was a large fireball on the horizon. It was the satellite finally reaching the end of it's flight. I have witnessed most of the failures at the Rocket Ranch in the last 20 years and I must say seeing that satellite hit the water was the most unique explosion I have seen.
I worked this launch and even filmed it with my then girlfriends video camera from the left window of an HH-60 pave hawk helicopter. We had spent the previous 90 minutes out over the Atlantic clearing out boat traffic from the launch danger zone and at L-5 were at our mission support positions in a hover over the Banana River. As it climbed overhead I hung out of the helo trying to video it but stopped and came back inside just a few seconds before it blew up. We were ordered to break away by cape control and we immediately flew off to the west just in case any debris came toward us. Of course by that time the rocket was well over the Atlantic so no danger to us but we did get to spend the next ten minutes watching pieces fall into the ocean through our night vision goggles. In my career, I worked dozens of rocket and space shuttle launches and landings and a few of them stand out. This is one of those launches that Ill never forget.
God damn Scott Rocket scientists love your channel
Great job
Really awesome!
Hi Scott, What's with the unusual configuration of the strap-on boosters on the delta II as shown @1:50? Why aren't they placed symmetrically?
@DJSnM are there any good stories of rockets that had similar simulation/algorithm issues but just “barely” got to orbit?
This is very good.
I love this video, it has it all.
That in-flight computer must've been a beast to work on.
PID loops aren't actually that complicated
Scott, love your vids. Intelligent , insightful, ACCURATE, and I love the funny way you talk !!!
That's Scots, och!
It’s not a “funny” way of speaking - it’s a Scots accent.
@@F-Man , conveying humor with the written word is tough. I like the Scottish accent; Scott Manley, Ewan McGregor, Sean Connery, and of course Craig Ferguson to name a few.
Very interesting!
Hey Scott, can you go over launch termination systems? Is it explosive pyrotechnics or is it some other system?
I've heard details of only a few local designs. I'm sure there are quite a few solutions out there trying to satisfy the high requirements of failsafe that will cause extreme mission fail by erring either way.
Idk if its the same on the Delta rockets, but on the titan 4 and 3 they used strips of explosives so maybe its that
Thank you Scott.
Your Videos remind me, that there are people with same interest. For 10 Minutes I forget lovesickness and trouble in private life.
What was that transition at 6:42? That kinda hurt my eyes.
This sounds like my KSP rockets, with similar results. Okay it sounds just like my KSP rockets with THE SAME results!
please do a video on Direct Fusion Drives and Torchdrives
Mr. Manley, I think it would be interesting if you did a video on the balloon-borne ARCADE cosmic background radiation sensor. My understanding is that ARCADE helped show where CBR has a Doppler shift that probably indicates Earth’s speed and direction relative to the universe as a whole.
It would be nice a video about vibration mode, what it means how it’s managed.
Good one.
RIP delta III. You will always be my favorite rocket of all time
What's the shelf ship in the top left?
awesome tale :)
thnx
Were these modes determined from the mathematics, or through simulation?
Sounds a lot like a computerized version of PIO - Pilot Induced Oscillation.
Bit of trivia:
Alongside the I-5 freeway in Santa Ana, California there is a Delta 3 upper stage and payload fairing. It's in a remote parking lot outside the Discovery science museum.
I haven't been there in a long time to read the plaque, so I don't remember its history, if the plaque even details it. Perhaps it's one that didn't get rolled into the Delta 4 program for some detail?
Heyyy, is there a Rifter in the Back? ;)
Hi Scott--The cryo upper stage for the Delta 3 had its genesis as the 'Hydrogen Upper Stage', or 'HUS' that MDAC had wanted to develop in 1977. NASA wanted no part of it, as it would have been competing with the Shuttle.
Does it use PID loops for guidance?
6:41 - Why didn't it detect that it was overcorrecting and automatically decrease the gain in its flight-control system to compensate?
not designed that way
Hey Scott why do they face the nozzles of the boosters outwards, shouldn't they start vertical? 6:19
They point inwards towards the center of mass because if the thrust varies they don’t want extra torque
Does the flight control use a "PID" algorithm? Just curious because I tune PID temperature control loops as part of my job and wonder if similar control is employed for rockets.
Mr. Manley. No model of the Rocinante on your shelf... sad :(
Literally trying to calibrate my 3D printer to make a new one
I noticed that the design of Roscosmos Angara, spacex Falcon and Boeing Delta 4 are very similar. Apparently we have reached the limit of our technology, space rockets are simply unified tanks with engines and are very similar due to maximum optimization. I think nothing new will be seen in the coming years in this area. Only if breakthrough technology does not appear, for example, a new type of engine or fuel.
Great reporting Scott! I liked the student steering analogy. Thought for sure by now the PHD engineers had overcompensation dampening down to a science! Guess not. Seems to be an Applied Physics wake up call. To run out of compressed CO2 or whatever they used to push the hydraulics does seem also embarrassing. This double embarrassment needs to be shown to our middle and high school students as an incentive to be interested in and focus on math. If I were still teaching I'd show your video.
the Delta family was a fine example of the old adage: "If it ain't broke, don't fix it!" trying to just stretch it out a 'little' further was an example of 'fixing' a problem that would have better been addressed by designing a new vehicle.
Another fantastic video. Im a little bit smarter then when i woke up!! Thank you Scott!!..check six buddy!!
When you do this series you should have your daughter singing her song in the background while you explain what happen.
So close to that gold UA-cam button. Spread the word people.
The Delta IV should be called "The DINO" for Delta in name only.
If anyone wants to see something of a Delta III, there is a DCSS off a Delta III in Santa Ana California on Mainplace Dr and Broadway off the 5 Freeway
and also the front of Endeavour for some reason?
@@moosemaimer that was a mock up from a store that's now closed. Endeavour is further up the 5 at the California Science Center
I wonder if an alternate preemptive fix could have been to sense the hydraulic supply and command neutral position and hold if either member of a triad of boosters got low for any reason. Less control just before loss of control but a predictable and partial loss of control.
My goodness Scott, you are on a roll with RUDding rockets, aren't you? :)
When will they finally make a Delta V? The perfect rocket name
Sadly there isnt any plans for a new delta rocket as ULA will move to the vulcan. Delta heavy is probally the last delta launch vehichle
killed because someone screwed up the tuning parameters just like my senior year modeling project, though that was just water valves
Can you do a video on the failure of the first Titan IIIE launch? The one caused by a guy retiring and taking his rocket assembly tricks with him.
So it was a somewhat similar failure than with the Ariane 5: they made a more powerful hardware, but forget to update the software to take it into account.
WOW. That thing looks so kerbal, I would have been surprised if it DIDN'T fail.
I do like it whenever we get a good ol' "Fly Safe (This was How Not To)"
Too much play in the steering wheel. I get that because I also drove old classic cars with sloppy 'gimbals'....
It's a shame the rocket didn't go up obviously, but I won't shed a tear for the corporate board that was like "eh screw it we'll buy a ride on the first try."
But how much does the Delta V provide?
And a quite serious "Fly Safe" today
The reason for the explosion was actually because of a screw-up with the payload. It wasn't actually the Galaxy 10. It was the Galaxy Note 7.
never been this early before, nice vid
Thank you, as always. And your input is desperately needed on why Starship may fail as a lunar lander - regolith as shrapnel, damaging itself or orbiting craft. Could it throttle the Raptors to hit zero velocity at 15 meters altitude, instead of zero altitude? As 1/6g slowly draws it down, fire a cluster of hot gas thrusters, similar to its RCS thrusters. (Double as ullage motors?) Workable?
Since Starship is a private investment by SpaceX I think it'd be great for them to just send one up to the Moon and see what happens. NASA doesn't lose any of its budget and SpaceX/NASA can learn from what happened.
Isnt a lunar base intended to be buried anyway? Maybe SS can inadvertently help with that lol. That is a legitimate concern though. Infact it makes me think of this setup which IMO is superior to Starship. ua-cam.com/video/p_56U0RZId4/v-deo.html
I am intrigued. Did you come up with this scenario yourself or is there a place where this is getting discussed? I mean the acceleration of regolith to the degree that is poses a risk for orbiting spacecrafts or satellites? I want to know more :)
@@z33r0now3 Dr Robert Zubrin of the Mars Society brought this up. He's a prominent long-time advocate for Mars missions. Others have chimed in agreeing or disputing his claim that Starship will kick up rocks hard enough to send them on a ballistic trajectory to orbit. No atmosphere, so even rocks moving at a slight angle to the ground can reach orbit or orbital altitude if moving fast enough. I've followed this 2nd and 3rd hand, but the debate is extensive, there may be a concern.
Came up with the idea of shutting off the Raptors at a certain altitude and using smaller engines myself, but on my post about this on reddit someone suggested the hot gas thrusters based on the RCS design they'll use. Then I realized Starship will also have ullage engines, but they may be too small for what I'm proposing. It will be falling slowly, not much room to accelerate - but will have the inertia of its ~250t mass.