Author Topic: Red Dragon Discussion Thread (1)  (Read 559199 times)

Offline Prober

  • Senior Member
  • *****
  • Posts: 10348
  • Save the spin....I'm keeping you honest!
  • Nevada
  • Liked: 724
  • Likes Given: 729
Re: Red Dragon
« Reply #120 on: 08/08/2011 03:56 am »
Might be useful to think of it as similar to a car with modern unibody construction instead of using a box steel frame and heavy body panels  like a 1955 Buick.

The sheetmetal in a unibody is so thin you could easily bend it, sometimes by hand, but welded up it can protect you in a car crash while still being very light - much lighter than that Buick.

Make it out of lithium-aluminum and you're on the bleeding edge of light & strong.

Next stop: someone builds a booster out of composites.

High strength steel is the new bleeding edge.  The changeover is in your cars.  Might be worth a look see for future rocket dev.
2017 - Everything Old is New Again.
"I fear all we have done is to awaken a sleeping giant..." --Isoroku Yamamoto

Offline Dave G

  • Senior Member
  • *****
  • Posts: 3326
  • Liked: 2173
  • Likes Given: 2076
Re: Red Dragon
« Reply #121 on: 08/08/2011 12:26 pm »
NC machining is easy, set it and forget it.  Welding stringers and hoops takes more touch labor.

Remember that the stir welding process is now automated.

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 38157
  • Cape Canaveral Spaceport
  • Liked: 22638
  • Likes Given: 432
Re: Red Dragon
« Reply #122 on: 08/08/2011 12:31 pm »
NC machining is easy, set it and forget it.  Welding stringers and hoops takes more touch labor.

Remember that the stir welding process is now automated.

Still have to put the hoops into place and clamp them

Offline kevin-rf

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 8823
  • Overlooking the path Mary's little Lamb took..
  • Liked: 1318
  • Likes Given: 306
Re: Red Dragon
« Reply #123 on: 08/08/2011 12:51 pm »
NC machining is easy, set it and forget it.  Welding stringers and hoops takes more touch labor.

Remember that the stir welding process is now automated.

Still have to put the hoops into place and clamp them

You also have to X-Ray/Ultra Sounds/Dye Penetrate all the welds when done.

It is worth noting that composites are mostly used where rigid non pressure stabilized structures are needed. The weight savings must be worth the cost in the cases of inter-stages and fairings. It is worth noting somewhere in the threads is a quote from Antonio that the performance gain by switching from metal to composite solid casings is worth the extra cost.
If you're happy and you know it,
It's your med's!

Offline Dave G

  • Senior Member
  • *****
  • Posts: 3326
  • Liked: 2173
  • Likes Given: 2076
Re: Red Dragon
« Reply #124 on: 08/08/2011 12:57 pm »
NC machining is easy, set it and forget it.  Welding stringers and hoops takes more touch labor.

Remember that the stir welding process is now automated.

Still have to put the hoops into place and clamp them

I don't know.  Are you sure?  It's not hard to imagine automation methods that replace placing and clamping.

In any case, it's clear that SpaceX believes this method saves money.

Offline baldusi

  • Senior Member
  • *****
  • Posts: 8389
  • Buenos Aires, Argentina
  • Liked: 2593
  • Likes Given: 8476
Re: Red Dragon
« Reply #125 on: 08/08/2011 01:16 pm »
Given enough production rate, it is cheaper to automate anything. This is more a question of whether it's worth making an automated stringer positioning and clamping device or not. My guess is that for less than twenty cores per year it isn't. And probably the number is closer to one hundred. A custom machine like that would cost how much? 500k?
An operative should be able to do at least two segments per day. Let's assume that the falcon takes 30 segments, that's 15 days of work (for that specific task). Let's say that the working year is 240 days, that's 16 cores per year. Times five years (machine cost vs operative's wage) that's 80 cores. And I haven't added the financial cost. Nope, I don't see an economic reason to actually automate this task. Not with less than 20 cores per year.
If SpaceX has any sort of serious financial officer, he'll keep it as manual as possible until the supposed 20 cores per year are realized. Again, this assumes that said operative is only doing that task. Since it takes some time for the welding machine to do it's work, he could work on two things at a time.

Offline kevin-rf

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 8823
  • Overlooking the path Mary's little Lamb took..
  • Liked: 1318
  • Likes Given: 306
Re: Red Dragon
« Reply #126 on: 08/08/2011 01:37 pm »
From my college days:

The four cases when you manufacture products.

1. Low production rate, immature product.
2. Low production rate, mature product.
3. High production rate, immature product.
4. High production rate, mature product.

As far as significant labor savings and recovery of your tooling costs. Assembly automation only makes sense in case four.

It does not make sense in case one where your overhead is higher than your labor costs and you are constantly investing in new tooling.
It does not make sense in case two where your overhead is higher than your labor costs, and the savings will be in the noise compared to other costs.
It does not make sense where you are constantly investing in tooling and never recovering the tooling costs.

It is also worth noting, anything designed for easy automated assembly is actually easier to manually assemble than something that wasn't. Lesson two, always design for easy assembly ;)

Even at 100 cores a year, Falcon 9 production is not high enough to justify full automation. Automation tools that aid in assembly and save man hours do make sense. They have a nice hybrid system in place (automation where it makes sense, manual labor for setup and moving parts). There is not enough money anywhere to afford a program throws away hundreds of cores a year. If we reach that point, we will not be going into space on ELV's that require this much labor.
« Last Edit: 08/08/2011 01:37 pm by kevin-rf »
If you're happy and you know it,
It's your med's!

Offline mmeijeri

  • Senior Member
  • *****
  • Posts: 7772
  • Martijn Meijering
  • NL
  • Liked: 397
  • Likes Given: 824
Re: Red Dragon
« Reply #127 on: 08/08/2011 09:47 pm »
So how much lighter are optimised structures like those used on modern launch vehicles compared to a plain aluminium or steel cylinder of uniform thickness and sufficient strength / stiffness? Two times? Five times? More than that?
Pro-tip: you don't have to be a jerk if someone doesn't agree with your theories

Offline Prober

  • Senior Member
  • *****
  • Posts: 10348
  • Save the spin....I'm keeping you honest!
  • Nevada
  • Liked: 724
  • Likes Given: 729
Re: Red Dragon
« Reply #128 on: 08/09/2011 12:50 am »
So how much lighter are optimised structures like those used on modern launch vehicles compared to a plain aluminium or steel cylinder of uniform thickness and sufficient strength / stiffness? Two times? Five times? More than that?

Might be able to help you understand this by the examples of the ET. Check out the weight savings in the three versions.

http://en.wikipedia.org/wiki/Space_Shuttle_external_tank
2017 - Everything Old is New Again.
"I fear all we have done is to awaken a sleeping giant..." --Isoroku Yamamoto

Offline HMXHMX

  • Full Member
  • ****
  • Posts: 1730
  • Liked: 2269
  • Likes Given: 683
Re: Red Dragon
« Reply #129 on: 08/09/2011 04:53 am »

Next stop: someone builds a booster out of composites.

Delta IV payload payload fairing, interstage, second stage intertank, first stage intertank and aft aeroshell.  Composites for cryogens is some years away or not at all.
Falcon 9 also has composite intertanks and fairings. I would even guess the aeroshell in the back, too. What's more, even Proton-M block III, have those parts made out of composites. I know that the Polish Stako company does 300Bar SCUBA tank out of composite (not aluminum lining with composite reinforcement, pure composite).
Obviously a SCUBA tank and an RP-1 are orders of magnitude differente sizes. But they SCUBA tank had a proof pressure of 450bar, while the Falcon 9 has what, 3.3Bar? I mean, eventually we might see or not cryogenic composite tanks, but non cryo should be seriously studied.
The other thing that I would study (and I guess was already but nothing came out of this) would be fibre reinforced cryogenic tanks. With some kevlar mesh around, should be able to support quite a bit of extra pressure, which should save on the wall thickness. It would most certainly require to be kept slightly pressurized for structural stability. But something like 1.2Bar of air should do it.

Minimum gauge becomes a significant issue for small diameter, thin composite pressure vessels.  Leakage risk goes way up, along with cost.  Pretty much nothing can beat aluminum for expendable tanks.

Offline baldusi

  • Senior Member
  • *****
  • Posts: 8389
  • Buenos Aires, Argentina
  • Liked: 2593
  • Likes Given: 8476
Re: Red Dragon
« Reply #130 on: 08/09/2011 01:13 pm »
Minimum gauge becomes a significant issue for small diameter, thin composite pressure vessels.  Leakage risk goes way up, along with cost.  Pretty much nothing can beat aluminum for expendable tanks.

I don't quite follow what you wrote, I'm sorry. The first phrase is about small pressure vessels, and has nothing to do with expendable tanks, right?
In that case, I've happily used Stako's plastic composite paintball tanks (1.1L), at a working pressure of 300Bar with no problem. Regrettable, that's not ideal for the two main uses of those kind of sized tanks: paintball and firefighters breathing air. The first simply don't have any training and some do stupid things, like putting oil in the feeding valves, at 300Bar air (that's equivalent to 60Bar or 870psi Oxygen). The second group, has to fight it's way through fire a debris. But in general for any other application, those tanks work great. You do have the buoyancy issue for SCUBA.
I'm sure nothing beats Al for expendable tanks, for now. What I was saying was that it was well worth it to keep researching some mix of materials and techniques to beat it. But as we always find, if you had the launch rate to do a machine to build composite tanks more cheaply thank Al tanks, then you'd better research a fully reusable launch system, right?

Offline HMXHMX

  • Full Member
  • ****
  • Posts: 1730
  • Liked: 2269
  • Likes Given: 683
Re: Red Dragon
« Reply #131 on: 08/09/2011 04:26 pm »
Minimum gauge becomes a significant issue for small diameter, thin composite pressure vessels.  Leakage risk goes way up, along with cost.  Pretty much nothing can beat aluminum for expendable tanks.

I don't quite follow what you wrote, I'm sorry. The first phrase is about small pressure vessels, and has nothing to do with expendable tanks, right?
In that case, I've happily used Stako's plastic composite paintball tanks (1.1L), at a working pressure of 300Bar with no problem. Regrettable, that's not ideal for the two main uses of those kind of sized tanks: paintball and firefighters breathing air. The first simply don't have any training and some do stupid things, like putting oil in the feeding valves, at 300Bar air (that's equivalent to 60Bar or 870psi Oxygen). The second group, has to fight it's way through fire a debris. But in general for any other application, those tanks work great. You do have the buoyancy issue for SCUBA.
I'm sure nothing beats Al for expendable tanks, for now. What I was saying was that it was well worth it to keep researching some mix of materials and techniques to beat it. But as we always find, if you had the launch rate to do a machine to build composite tanks more cheaply thank Al tanks, then you'd better research a fully reusable launch system, right?

Sorry.  I was thinking about the only application where composites really make sense for tanks, large reusables.  Those have diameters in the 5-15 M range, so referencing F9, 3-4 M dia. is "small."

At 50 PSIA and 12 ft diameter, the wall thickness for aluminum is in the 0.05 - 0.10 range (inches).  Using composites, it goes down to a few plys or perhaps 0.02 - 0.04 inches.  Without a liner, you can't easily prevent leaks, plus the installed cost of the composite may be 10-20 times that of the aluminum.  Composites will require a higher safety factor than aluminum, as well.  And the thin walls are very fragile, and easily damaged.

Composites work very nicely in high pressure tanks that have relatively thick walls and can afford thin liners for leak prevention without much mass growth, not low pressure tanks with thin skins.

(I also believe this is off topic for this thread, so should move to general discussion.)

Offline go4mars

  • Senior Member
  • *****
  • Posts: 3748
  • Earth
  • Liked: 158
  • Likes Given: 3463
Re: Red Dragon
« Reply #132 on: 08/25/2011 02:32 am »
Is part of the reason for the propulsive landing engines on the side to keep debris from bouncing into and wrecking the heatshield (whether on Mars or Earth)? 
Elasmotherium; hurlyburly Doggerlandic Jentilak steeds insouciantly gallop in viridescent taiga, eluding deluginal Burckle's abyssal excavation.

Offline beancounter

  • Full Member
  • ****
  • Posts: 1249
  • Perth, Western Australia
  • Liked: 106
  • Likes Given: 172
Re: Red Dragon
« Reply #133 on: 08/25/2011 04:17 am »
Given enough production rate, it is cheaper to automate anything. This is more a question of whether it's worth making an automated stringer positioning and clamping device or not. My guess is that for less than twenty cores per year it isn't. And probably the number is closer to one hundred. A custom machine like that would cost how much? 500k?
An operative should be able to do at least two segments per day. Let's assume that the falcon takes 30 segments, that's 15 days of work (for that specific task). Let's say that the working year is 240 days, that's 16 cores per year. Times five years (machine cost vs operative's wage) that's 80 cores. And I haven't added the financial cost. Nope, I don't see an economic reason to actually automate this task. Not with less than 20 cores per year.
If SpaceX has any sort of serious financial officer, he'll keep it as manual as possible until the supposed 20 cores per year are realized. Again, this assumes that said operative is only doing that task. Since it takes some time for the welding machine to do it's work, he could work on two things at a time.

There's more to stir friction welding than simply cost.  It also improves quality by standardising the nature of the welding.  In addition, I believe SpaceX x-ray their welds immediately after completion - again automated.  Seem to recall that this was the case originally made to Musk who wasn't all that keen on it in the first place.
Beancounter from DownUnder

Offline simonbp

  • Science Guy
  • Senior Member
  • *****
  • Posts: 7138
  • Liked: 314
  • Likes Given: 183
Re: Red Dragon
« Reply #134 on: 08/25/2011 03:36 pm »
The four cases when you manufacture products.

1. Low production rate, immature product.
2. Low production rate, mature product.
3. High production rate, immature product.
4. High production rate, mature product.

As far as significant labor savings and recovery of your tooling costs. Assembly automation only makes sense in case four.

I suppose that depends on how you define automation. If it is a specifically-designed machine/robot for assembly, then yes. But a highly automated general-purpose device is advantageous in the other cases.

I say that because I know that SpaceX mills out the entire Merlin combustion chamber/nozzle assembly in one big CNC machine (I've seen it myself). It is highly automated, and much more efficient that milling by hand or with a jig. Only in your case four (producing 1000 Merlins a week) would a special-purpose milling machine make more sense. But it would be just as automated as the current CNC machine.

Offline simonbp

  • Science Guy
  • Senior Member
  • *****
  • Posts: 7138
  • Liked: 314
  • Likes Given: 183
Re: Red Dragon
« Reply #135 on: 08/25/2011 04:03 pm »
Is part of the reason for the propulsive landing engines on the side to keep debris from bouncing into and wrecking the heatshield (whether on Mars or Earth)? 

That might be beneficial, but the IMHO there are two main reasons for it:

1) For crew launch, they can begin firing immediately in the case of an abort, without waining for a separation event (like CST-100).

2) For landing on Mars, it allows them to avoid a supersonic parachute. Terminal velocity on Mars is supersonic (because the air is so thin), and it is really hard to fire a retrorocket into a supersonic (incompressible) flow. Viking and Phoenix got around this with a supersonic parachute that slowed the capsule to high subsonic before the rockets were ignited. Dragon gets around this by putting the exhausts in the stagnant flow behind the shock, thus allowing the engines to ignite while the vehicle is supersonic. This is quite clever if it works, and removes an entire level of complexity from landing on Mars.

Offline go4mars

  • Senior Member
  • *****
  • Posts: 3748
  • Earth
  • Liked: 158
  • Likes Given: 3463
Re: Red Dragon
« Reply #136 on: 08/25/2011 04:27 pm »
Is part of the reason for the propulsive landing engines on the side to keep debris from bouncing into and wrecking the heatshield (whether on Mars or Earth)? 

That might be beneficial, but the IMHO there are two main reasons for it:

1) For crew launch, they can begin firing immediately in the case of an abort, without waining for a separation event (like CST-100).

2) For landing on Mars, it allows them to avoid a supersonic parachute. Terminal velocity on Mars is supersonic (because the air is so thin), and it is really hard to fire a retrorocket into a supersonic (incompressible) flow. Viking and Phoenix got around this with a supersonic parachute that slowed the capsule to high subsonic before the rockets were ignited. Dragon gets around this by putting the exhausts in the stagnant flow behind the shock, thus allowing the engines to ignite while the vehicle is supersonic. This is quite clever if it works, and removes an entire level of complexity from landing on Mars.

Thank you very much.  Fascinating response! 

There appears to be at least 3 significant advantages (with the main disadvantages being complexity and cosine losses). 
Elasmotherium; hurlyburly Doggerlandic Jentilak steeds insouciantly gallop in viridescent taiga, eluding deluginal Burckle's abyssal excavation.

Offline starsilk

  • Full Member
  • ****
  • Posts: 687
  • Denver
  • Liked: 269
  • Likes Given: 117
Re: Red Dragon
« Reply #137 on: 08/25/2011 04:47 pm »
Is part of the reason for the propulsive landing engines on the side to keep debris from bouncing into and wrecking the heatshield (whether on Mars or Earth)? 

That might be beneficial, but the IMHO there are two main reasons for it:

1) For crew launch, they can begin firing immediately in the case of an abort, without waining for a separation event (like CST-100).

2) For landing on Mars, it allows them to avoid a supersonic parachute. Terminal velocity on Mars is supersonic (because the air is so thin), and it is really hard to fire a retrorocket into a supersonic (incompressible) flow. Viking and Phoenix got around this with a supersonic parachute that slowed the capsule to high subsonic before the rockets were ignited. Dragon gets around this by putting the exhausts in the stagnant flow behind the shock, thus allowing the engines to ignite while the vehicle is supersonic. This is quite clever if it works, and removes an entire level of complexity from landing on Mars.

Thank you very much.  Fascinating response! 

There appears to be at least 3 significant advantages (with the main disadvantages being complexity and cosine losses). 

yes, thanks simonbp! I don't recall reading anywhere about the 'avoid supersonic ignition' point. do you have a reference?

Offline simonbp

  • Science Guy
  • Senior Member
  • *****
  • Posts: 7138
  • Liked: 314
  • Likes Given: 183
Re: Red Dragon
« Reply #138 on: 08/25/2011 07:23 pm »
Not on the Dragon-specific speculation; that's simply based on their own images of it landing on Mars sans parachute.

On the general point, here's a paper on the issues: http://smartech.gatech.edu/jspui/bitstream/1853/8390/1/IEEEPaper06ID0076FINAL.pdf

Quote
An additional supersonic decelerator possibility is simply to use propulsion. While this appears straightforward, there is little experience firing larger thrusters directly into a high dynamic pressure supersonic flow. Flow stability, flow-control interaction and thermal protection are some of the design issues that surround use of this technology.

And I recall being told point-blank by the coauthor (Rob Manning, designer of the Pathfinder/MER EDL system) that he has no idea how you would do a supersonic retrorocket. Apparently, SpaceX thinks they know.

Offline Geron

  • Full Member
  • **
  • Posts: 229
  • Liked: 60
  • Likes Given: 7
Re: Red Dragon
« Reply #139 on: 08/26/2011 06:38 am »
If they do they should patent it and then get it out in the open for us all to gawk at and discuss openly! Maybe save some tax dollars by preventing others from spending 10 times as much landing a fraction of the payload on Mars.

Tags:
 

Advertisement NovaTech
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
0