Q&A: Pegasus Designer Dr. Antonio Elias

• #440 by deiter on 12 Dec, 2007 02:33
• This is a carry over from the COTS thread, has there been any buzz about an Orbital COTS2 submission?
• #441 by madscientist197 on 12 Dec, 2007 07:48
• Hold on - what's the Campbell's tomato soup doing?!
  
  Very cool models though.
• #442 by gospacex on 13 Dec, 2007 05:48
• Quote

  kevin-rf - 11/12/2007  6:56 AM
  Actually, somewhere in the thread Antonio discussed why they did not go with a supersonic carrier aircraft... Have fun rereading the thread :-)

  
  I read it. Basically, there are no big supersonic planes which are even remotely affordable. However, it may change.
  
  Tupolev_Tu-160 bomber has 15 km ceiling, Mach 2 speed, can take 40 ton payload, has 10 000 km range, and has two 8+ meter weapon bays. Sooner or later it will become obsolete, and then?...
  
  Theoretically, by how much this kind of carrier plane will help?
• #443 by madscientist197 on 13 Dec, 2007 08:10
• ...and then you can't find spare parts.
• #444 by pippin on 13 Dec, 2007 09:22
• It's more like later...  Isn't Russia building new Tu-160s? And regarding spare parts: More of them have been built so far as of the L1011 and that one has been out of (regular) service for a while, too, hasn't it?
• #445 by Space101 on 14 Dec, 2007 09:49
• And now AndrewsSpace claim to be proposing, in their own words a "cargo logistics system to service NASA and emerging commercial providers, including a low cost launch system to replace the Delta II launch vehicle." How does this affect Orbital's plans? unexpected rival - or to be expected?
• #446 by jimvela on 14 Dec, 2007 15:55
• Quote

  Space101 - 14/12/2007  3:49 AM
  
  And now AndrewsSpace claim to be proposing, in their own words a "cargo logistics system to service NASA and emerging commercial providers, including a low cost launch system to replace the Delta II launch vehicle." How does this affect Orbital's plans? unexpected rival - or to be expected?

  
  They're only a rival if they actually fly.  
  
  I have zero doubt that Orbital will fly D-II class vehicles successfully if the customers show up to buy them.  I have great doubts about any other newcomer except SpaceX successfully flying any launch vehicles anytime soon.
  
  I doubt Orbital is loosing much sleep over Andrews...
• #447 by vt_hokie on 20 Dec, 2007 21:26
• Quote

  antonioe - 30/10/2007  9:55 PM
  
  Yeah... exactly what we are wondering... we're keeping an eye on Globalstar and even good 'ol Orbcomm, but it's hard to compete with the Ukranians... not to mention the up-and-coming Indians...

  Indeed, I'm willing to bet a chocolate milkshake (G. David Low's favorite bet) that India will emerge as the leading low-cost international supplier of commercial launches.  They are technologically sophisticated, and their labor costs are low.  Can't beat that!

  
  I'm curious about how you see that shaping the industry's future.  Could U.S. based launch providers find it impossible to compete in the global marketplace?
• #448 by tnphysics on 04 Feb, 2008 00:40
• antonioe,
  
  Why do you believe that SpaceX can charge only $35 million for a Falcon 9 and still make a profit?
• #449 by antonioe on 04 Feb, 2008 07:16

• Quote

  tnphysics - 3/2/2008 7:40 PM antonioe, Why do you believe that SpaceX can charge only $35 million for a Falcon 9 and still make a profit?

  Unfortunately, I don´t believe they can do that.  SpaceX has not made public any logic for that claim that I can feel comfortable with.  It would be nice if it happened (it would actually help Orbital´s growing spacecraft niche).  But hoping it will come true doesn´t help in this case.

  BTW, thanks for bumping the thread.  I had not checked it for a while (busy with COTS and LAS) and there are a few questions I should answer).

• #450 by antonioe on 04 Feb, 2008 07:25

• Quote

  simonbp - 10/12/2007 6:18 PM So, here's the real question: Have you built your own paper model Pegasus and Stargazer?

  I can do a lot of things: design (real) rockets, fly (real) airplanes, and so forth.  There are also many things I CAN´T DO.  I can´t touch plants: they die upon contact with my skin.  I can´t water a pot without what´s in it dying.  That´s why, much as I would like one, I don´t have a nice plant in my otherwise nice office. 

  Another thing I can´t do is decent airplane models, paper, plastic or otherwise.  And I´ve tried, since I was 8 or so.  They look like (censored).  Curiously, another thing I simply cannot do is fly RC models.  I have several thousand dollars in broken aircraft, helicopters and dust-covered RC gear to prove it :frown:

• #451 by antonioe on 04 Feb, 2008 07:41

• Quote

  vt_hokie - 20/12/2007 4:26 PM

  Quote

  antonioe - 30/10/2007 9:55 PM  hard to compete with the Ukranians... not to mention the up-and-coming Indians...  technologically sophisticated, and their labor costs are low.

   I'm curious about how you see that shaping the industry's future. Could U.S. based launch providers find it impossible to compete in the global marketplace?

  If things continue the way they are now, I don´think any US supplier has much of a chance to compete internationally on a pure price basis (policy considerations aside).  On the other hand, conditions, such as relative labor costs or unforseen events could change all that in a jiffy.

  For example, three years ago, if somebody had told me that Delta II was going to disappear, I would refer said individual to Alcoholics Anonymous.  Yet, a "perfect storm" of increasing labor costs, aging infrastructure, dwindling supply chain, corporate mergers and evolution, EELV flight rates and USAF Corporate Policy blended almost perfectly to kill an excelent LV.  Any of these causes, by itself, would not have killed Delta II.  Why, even any THREE of them probably would not!  But six...

• #452 by JIS on 04 Feb, 2008 13:33
• Quote

  antonioe - 4/2/2008  8:16 AM
  
  

  Quote

  tnphysics - 3/2/2008 7:40 PM antonioe, Why do you believe that SpaceX can charge only $35 million for a Falcon 9 and still make a profit?

  Unfortunately, I don´t believe they can do that.  SpaceX has not made public any logic for that claim that I can feel comfortable with. 

  
  What about using plain sheet with welded-in rods instead of isogrid tank structures? (If I understand correctly what they are doing) Have you considered something similar for TII or is the 1st stage isogrid?  
  Also the gas generator cycle Merlin manufactured in bigger series could help. I'm sure that AeroJet and ATK have high overhead costs with their engines and motors.
• #453 by aero313 on 04 Feb, 2008 14:18
• Quote

  JIS - 4/2/2008  9:33 AM
  What about using plain sheet with welded-in rods instead of isogrid tank structures?

  
  One of the not-so-obvious things you learn in the aerospace industry is that big machined parts cost less than assembly of lots of little parts.  Once you program the NC mill to make the isogrid, it chugs along fat, dumb, and happy with little human interaction.  On the other hand, assembly of parts requires lots of touch labor, and humans often make mistakes, requiring expensive rework.
• #454 by antonioe on 04 Feb, 2008 14:54

• Quote

  JIS - 4/2/2008 8:33 AM  What about using plain sheet

  plain sheet of what?  Aluminum?  Stainless steel?  Titanium?

  Quote

   with welded-in rods instead of isogrid tank structures?

  Again, what material rods?  What welding process?  Arc?  Tig?  Stir-friction?  Manual?  Robotic?  Using rigid jigs or tack-welded?  Do the welds deform the sheets?  What happens to the welds when you roll the panels into shape?  Remember, in a word of high labor costs, isogrid is a cost-saving process compared with welding.  You DO waste a lot of aluminum, granted.

  Quote

  Have you considered something similar for TII or is the 1st stage isogrid? Also the gas generator cycle Merlin manufactured in bigger series could help. I'm sure that AeroJet and ATK have high overhead costs with their engines and motors.

  IMHO there is a rather wide technological "bathtub" where launch vehicle structural and propulsive efficiencies and hardware manufacturing costs are approximate the same.  You can waste a lot of money if you wander "outside" this bathtub, but within it, costs and efficiency are approximately the same.  From a technology and manufacturing standpoint, Falcon 1/9 and T II are both within this "bathtub": not in the same point, but within the flat range.  The ATK vehicle is a different beast and lives in a different bathroom.  SO major cost differences between F9 and T II have to come from other sources, such as lowering the development, labor and fixed costs.  Spreading costs with similar but separate product lines helps a bit.

  If memory serves me right, the last major LV cost improvement due purely to technology was the change from metallic (usually steel) SOLID MOTOR cases to carbon composites.  Interestingly enough, this improvement was "indirect": the composite case is more expensive than the steel case, but the ensuing increase in payload more than makes up for the increased cost.  Also, the same carbon composites do not help liquid stages (different structural problem).

  Also IMHO, the "high overhead costs of ATK and AeroJet" have to be examined in the right perspective:  If ALL these suppliers produced were the items you buy from them (i.e. they don´t build anything else) then you probably would be better off manufacturing the items yourself.  On the other hand, if these suppliers can use their people, buildings, machinery, etc. for other products and customers, you save big.  It´s the good ´ol "make or buy" decisionmaking which has to be applied wisely (sometimes companies - not just SpaceX! - bias their make/buy decisions in favor of "make" for reasons that have nothing to do with good management; a large aerospace prime recently lost a multi-hundreds of millions contract just because of this.)

  I worry about SpaceX´s "vertical integration" approach:  under different circumstances, it could be the right thing to do; but in their case (expected rates of flight, labor costs, cost of capital), I´m afraid it could be their undoing.  Same goes for how much of their vehicles are "from scratch" designs.  I believe in balance.

  Having said that, let me point out that SpaceX has smartly avoided some of the fundamental mistakes that others - especially the advocates of the "big dumb boosters" - have made, such as pressure-fed sea-level engines (any attempt to use pressure-fed sea-level stages is, alas, doomed to economic failure even it it is acheived technically).  Also, they wisely avoided the need for SRB´s (at least initially).  I still have not made up my mind about clustering more than, say, four or five engines.  It may or may not be a good idea (I can see advantages and disadvantages).  I hope the SpaceX experience will shed some light.

  And, speaking  of engine cycles, if somebody gave me $500M to develop a Delta II class first stage from scratch (i.e., development money were no object) I would probably develop the largest possible expander-cycle LOX-LH2 that can be made (there is a maximum practical size for an expander-cycle engine: probably around 40-50 MT.  Jim French, if you are reading this, can you comment?).

  But my main reply to your question (boy, I´m long-winded, aren´t I?) is that you seem to espouse the premise that the principal cost of launch is materials, manufacturing techniques or some "overhead" of one type or another.  As I said previously, tyhe main cost of launches is labor, labor, labor. What good is to spend $500M reduce the Stage 1 hardware costs from, say $16M to $8M when the rest of the vehicle, launch ops, amortization of R&D, mission-unique analysis, ground infrastruture maintenance, etc. costs another $40M?  Oh, and don´t forget a decent profit, as in 10%.

  India.  We should outsource launch services to India.  They have what it takes to make and operate cheap LV´s.

• #455 by simonbp on 04 Feb, 2008 15:39
• Well, I guess you answered your own question: SpaceX expects to lower costs though a leaner workforce and smaller overhead from their vertical integration. Having Musk as a sugar-daddy doesn't hurt, either...
  
  Simon
• #456 by JIS on 04 Feb, 2008 16:34
• I based my assumptions on following: The cost of LV is in the engines and tanks.
  Falcon 9 has 10 common engines made in bigger series and two stages using the same (or very similar) manufacturing process. There is also Falcon 1 rocket which shares a lot of expenses.
  Taurus II has more complex engines built in Russia many years ago, but provided by AeroJet who charge some extra cost for maintaining and updating them. I don't know the cost but those two engines could end up being more expensive than 10 Merlins. There are also (two?) upper stages using solid rocket motors which needs to be supplied by vendor with it's own overhead (probably shared whit other projects).
  
  Now, I agree that the principal cost is a labour. But, using two stages with same diameter, same heads, same technology, just different length could be actually the magic behind the cost. I think they are using the same aluminium friction stir welding for all tanks. Just different numbers of stiffeners. Is it labour effective or not? I don't know. We should ask Mr. Musk.  
  
  I also agree that lowering the LV manufacturing cost is not always enough to make successful and affordable vehicle. The reliability suffers as can be seen on Falcon 1 failures. However, the loss of cheap test vehicle (what I consider Falcon 1) is not a tragedy.
  The real workhorse is Falcon 9. They still have a chance to succeed with Falcon 9 if they tune-up Falcon 1 to reliable rocket. I think that Falcon 1 was a really smart move which saved them millions on development and a lot of problems with Falcon 9.
  Orbital doesn't have any cheap test bed for TII. On the other side they have the experience in house (and they have to pay for it) or they can hire the experience from contractor (and they have to pay for it).
  
  Still, I'm not saying Falcon 9 can be much cheaper than TII at the and of the day. But, Musk invested his $100m and NASA's few $100m and he actually doesn't need to get these money back through the launch contracts. If he wants his investment back he can sell SpaceX. All he needs is just reliable Falcon/Dragon and COTS 2 contract to make a personal profit by selling SpaceX. Until then he can charge costumers whatever he considers sufficient.
  
• #457 by JIS on 04 Feb, 2008 16:49
• Quote

  JIS - 4/2/2008  5:34 PM
  Now, I agree that the principal cost is a labour. But, using two stages with same diameter, same heads, same technology, just different length could be actually the magic behind the cost. I think they are using the same aluminium friction stir welding for all tanks. Just different numbers of stiffeners. Is it labour effective or not? I don't know. We should ask Mr. Musk.  
  

  
  I know that your point was that Orbital can buy the tanks from a vendor who build similar tanks all the time. But consider how many lawyers, business men, accountants and managers are involved over the skilled labour actually manufacturing that tank. Musk just needs to buy the welding machine, hire some skilled labour and dump few Falcon 1 rockets into the ocean.
• #458 by antonioe on 04 Feb, 2008 19:02
• You all make good points.  We´ll see.
• #459 by jeff.findley on 20 Feb, 2008 22:01
• Quote

  antonioe - 27/8/2007  1:24 AM
  
  As soon as it is prudent, I will share with you all the vehicle and operational details that ITAR would allow me to post in this forum, including sketches. No yellow ruled paper, though - we use IDEAS.

  
  I used to work on I-DEAS CAE software, now I work on NX CAE software.
  
  Jeff