Q&A: Pegasus Designer Dr. Antonio Elias

• #460 by HMXHMX on 21 Feb, 2008 04:09
• antonioe - 4/2/2008  9:54 AM
  
  "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. "
  
  
  The limit for expander cycle goes higher, probably at least 250K and maybe to 600K or more according to P&W.  See Dick Mulready's book "Advanced engine Development at Pratt & Whitney"
  
  We see eye to eye on that vehicle propulsion option, by the way.  Nothing beats expander-cycle turbomachinery, if one has to use pumps.
  
  Gary C Hudson
• #461 by yinzer on 21 Feb, 2008 04:15
• Quote

  HMXHMX - 20/2/2008  9:09 PM
  We see eye to eye on that vehicle propulsion option, by the way.  Nothing beats expander-cycle turbomachinery, if one has to use pumps.
  

  
  I'm interested to see what XCOR's expander-like hydrocarbon cycle is once their patent issues.  Anyone have any ideas that they can share?
• #462 by antonioe on 21 Feb, 2008 04:20

• 250K? 600K?  What units are you referring to, Gary?  Also, is the practical limit is higher for engines that start at sea level temperature than cold-soaked?

  "When Gary Hudson speaks, Antonio Elias listens!(TM)"

• #463 by HMXHMX on 21 Feb, 2008 22:28
• Quote

  antonioe - 20/2/2008  11:20 PM
  
  

  250K? 600K?  What units are you referring to, Gary?  Also, is the practical limit is higher for engines that start at sea level temperature than cold-soaked?

  "When Gary Hudson speaks, Antonio Elias listens!(TM)"

  
  I am out learning to be a blacksmith at our Mojave test stands and don't have my references with me.  I'll send you something on the weekend.
  
  "If Antonio is listening, I'll be more care what I say!"
• #464 by HMXHMX on 24 Feb, 2008 03:55
• Antonio,
  
  As promised, here is what I have at hand concerning higher thrust expander cycle engines.  Quotes are taken from the SAE-published book: Advanced Engine Development at Pratt & Whitney (2001) by Dick Mulready (ISBN 0-7680-0664-3).  There are AIAA papers or RPL reports dealing with higher thrust expanders, but all those I have are boxed away awaiting the time to PDF them, someday.
  
  In Jan., 1960, NASA asked for a 200K thrust engine proposal (which Rocketdyne would win, i.e., the J-2).  P&W submitted the idea for the RL200.
  
  p 76-77:  "The engine was named the RL200 (Fig 4.9), and it employed a shunt expander cycle similar to the RL10 but with the addition of separate gear-driven inducer stages for both propellants...  It is a shame that this engine design was slated not to be developed.  It is likely that a  big RL10 would have had a major impact on the future of space."  
  
  [D'oh.  You think?]
  
  p 59:  "Another excuse used to discredit our proposal for the 200K engine...was that the expander cycle could be applied only to small engines.  This was grossly untrue, and in the early 1990s, a design was submitted for a 600,000 pound thrust split expander cycle engine, with a turbine inlet approximately at room temperature.  This design employed a thrust chamber fabricated from copper tubes."
  
  While at Rotary, we looked at a variety of expander cycles (using LOX!) and concluded this was feasible up to Pc of about 1200 or so, as I recall.  I have no memory of what thrust levels.
  
  Of course, another cycle is the ox-rich NK33-type cycle which can be tuned to deliver close to equivalent results as the tried and true expander, in respect to turbine inlet temps.
  
  Gary C Hudson
  
  "If Antonio is listening, I'll be more careful what I say!"
  
• #465 by tnphysics on 27 Feb, 2008 10:55
• How much will the NK33s for the Taurus II cost?
• #466 by antonioe on 28 Feb, 2008 02:15

• Quote

  tnphysics - 27/2/2008 5:55 AM How much will the NK33s for the Taurus II cost?

  I'd have to ask Aerojet permission to give a numeric answer to that question, but let me just say that we are buying a lot more from Aerojet than the raw NK33s: new gimbal, new TVC (we are supplying them with the engine control computer, but they will have to program it) feed line slip joints, etc. etc.  As far as Aerojet is concerned, the engines are essentially free (to them, that is!)

  Buying/developing/building a rocket engine is a little bit like my story about buying a launch vehicle ("oh, so you want tires/air conditioning/nav/sunroof/CD player with that car?"); the basic rocket is one thing, but after you end up getting all the "accessories' the cost has almost doubled!

  Unfortunately, the mods that Aerojet did many years ago to make them useable in the K1 application (reuseable, air-restartable) do not apply to Taurus II (expendable), hence there will be some additional development cost, too, but a minuscule fraction of what developing a new motor would require, especially a motor with the NK33's performance and test data.

• #467 by Flometrics on 28 Feb, 2008 06:05
• Do you have any idea what the reliability of the NK-33 engine will be?
  For example, the SSME is expected to cause a loss of vehicle once every 500 or so flights. Reference: AIAA 2002-3758  (SSME)OptionsForTheFutureShuttle.pdf
  
  The issue being, what is more important, the performance of high pressure staged combustion engines, or the reliability of lower pressure engines. I am guessing the performance was more important in your studies. Or was there no other engine available to trade against?
  
  
  Steve
  
  
• #468 by Jim on 28 Feb, 2008 11:21
• Quote

  Flometrics - 28/2/2008  2:05 AM
  Or was there no other engine available to trade against?
  
  

  
  Name an engine to trade against that exists, has good performance and is "cheap"
• #469 by antonioe on 28 Feb, 2008 12:37

• Quote

  Jim - 28/2/2008 6:21 AM

  Quote

  Flometrics - 28/2/2008 2:05 AM Or was there no other engine available to trade against?

  Name an engine to trade against that exists, has good performance and is "cheap"

  Or that has the test time that the NK-33's have...

• #470 by tnphysics on 29 Feb, 2008 21:57
• RD-180. But that has almost 50% more thrust than 2 NK-33s.
• #471 by Lampyridae on 29 Feb, 2008 22:59
• Quote

  antonioe - 5/2/2008 12:54 AM
  
  

  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.

   The Japanese are aiming to cut launch costs by 2/3 as well. They're taking the labour issue head on and trying to work in as much automation and streamlining as possible. I don't think they'll quite get there, but automation and lean design are two areas where they are world leaders. They're starting out simple with solids and then moving the process on to liquid boosters and so on.

  I think the next thing that has to be developed is containerisation of payloads - especially all stuff that can be launched into a common orbit, and maybe processing at the other side, for example at a space station. Otherwise, or maybe together, perhaps some kind of space tug service to take payloads to wherever they need to go.

  The neat thing about COTS is that it is encouraging standardised flights and payloads, as well as payloads other than satellites. If you follow this logic, then eventually a manned space station becomes a lot more cost-effective than several small purpose-designed satellites.

• #472 by Jim on 29 Feb, 2008 23:16
• Quote

  Lampyridae - 29/2/2008  6:59 PM
  If you follow this logic, then eventually a manned space station becomes a lot more cost-effective than several small purpose-designed satellites.

  
  That is not where the logic takes you.  Manned station will never be more cost effective than unmannned spacecraft
• #473 by Lampyridae on 01 Mar, 2008 02:34
• Quote

  Jim - 1/3/2008  9:16 AM
  
  

  Quote

  Lampyridae - 29/2/2008  6:59 PM
  If you follow this logic, then eventually a manned space station becomes a lot more cost-effective than several small purpose-designed satellites.

  
  That is not where the logic takes you.  Manned station will never be more cost effective than unmannned spacecraft

  
  The logic is that of using a station with available resources (RMS, space tug, pallet mounts, humans etc) to minimise what you need design in a satellite and minimise all the custom analyses, launch profiles etc. In effect, the station becomes one big satellite bus for a lot of satellites. (Admittedly the ISS is not a good example of this). This also saves with having to dispose of satellites at the end of their lives.
  
  You can't replace satellites with stations, perhaps I should have been clearer with that. But whatever doesn't need a specific orbit and environment can be mounted on a station. It's a facility that you can use and leverage off of.
  
  The station needn't be manned, either. Packages can be installed on one big satellite bus by COTS-type "supply" missions.
• #474 by Solarsail on 01 Mar, 2008 04:39
• Antonio,
  
        Has Orbital studied using the AJ-10 engine to power a larger third stage on the Taurus II than you've mentioned for LEO missions?  I'm not certain, but using a bi-prop engine may help the payload capacity for GTO and Earth-escape missions, due to the slightly better ISP.  Or are you guys working mostly on the design of the first two stages?
  
  
  
  Thank you in advance
  -Solarsail
• #475 by antonioe on 01 Mar, 2008 11:20

• Quote

  Solarsail - 29/2/2008 11:39 PM Antonio, Has Orbital studied using the AJ-10 engine to power a larger third stage on the Taurus II than you've mentioned for LEO missions? I'm not certain, but using a bi-prop engine may help the payload capacity for GTO and Earth-escape missions, due to the slightly better ISP. Or are you guys working mostly on the design of the first two stages? Thank you in advance -Solarsail

  Michael:

  Yes and yes.  We've been working with P&W for the past two years or so on a possible high-energy upper stage for Taurus II.  At this time, however, we are concentrating our resources on the easier, lower development cost and lower risk (but also lower-performance) solid upper stage version.

  Earlier in the design trades, we had both a 3 meter diameter and a 3.9 meter diameter core.  There were several factors in favor of the larger diameter, one of them being the ease of adapting a LOX/LH2 or a LOX/Methane upper stage.

• #476 by antonioe on 01 Mar, 2008 11:38

• Quote

  Lampyridae - 29/2/2008 9:34 PM In effect, the station becomes one big satellite bus for a lot of satellites.

  Troy:

  People ask that question very often, and wonder why it isn't done.  Believe me, it is not because people don't think about it!  There are very good reasons:

  The energy required to reach orbital velocity is so high (launching 1 mT to low Earth Orbit requires, for a space of about 15 minutes, enough power to light the entire country of Uganda) that the margins you have to play with in designing a space application are very, very slim.  Therefore it pays to build satellites in a very custom way, as opposed to "general purpose" busses, which by necessity eat up your margins: if your payload only needs 500W Orbit Average Power, why pay for a 5 KW electric power system?  If your ocean surface altimeter only needs about a one degree pointing precision, why carry the mass associated with a 50 milliradian accuracy ACS?  If your mission life is only three years, why pay for redundancy?

  Quote

  But whatever doesn't need a specific orbit and environment can be mounted on a station.

  Every space application needs "a specific orbit", be it in terms of altitude, eccentricity, inclination or, in some cases, RAAN.  That includes extremely low-cost, very requirements-insensitive applications such as amateur radio satellites.  And even if there are a lot of satellites that need the same "class" of orbit, such as commercial geocoms (they are all  in "Geostationary Orbit") or GPS (they are all at the same altitude, eccentricity and inclination) it turns out that have to be in separate positions or separate orbital planes (or both), so you can't co-locate them.

  That said, we are helping Thomas Jefferson HS in Fairfax County to build a satellite (this may be the first satellite built by High-Schoolers, unless anybody knows of another example), and we are also trying to get them a ride to orbit.  they can tolerate almost any orbit but, even then, not any orbit...

• #477 by tnphysics on 01 Mar, 2008 12:01
• Which ones can't they tolerate?
• #478 by antonioe on 01 Mar, 2008 12:29

• Quote

  tnphysics - 1/3/2008 7:01 AM Which ones can't they tolerate?

  A very important one, because there are usually many free - or "almost free" - rides there: Geo Transfer Orbit (GTO).  AMSAT, the radio amateur satellite organization (by the way, I'm KA1LLM) took advantage of a free Ariane 5 piggyback to launch its OSCAR-40 spacecraft.  GTO is a popular "free ride" because of the frequent Ariane and other GTO launches with large satellites that allow small piggybacks (as long as they don't threaten the mission success of the big, paying payload).

  GTO has some very interesting orbital properties for amateur radio - think of it as a "poor man's Molnya" in that the argument of perigee wanders instead of remaining fixed (if you don't know what that means, I apologize - other readers may appreciate it; if you want an explanation of what that means, please ask and I'll explain).  But in order to communicate from those altitudes, the power and/or antenna directivity of the spacecraft must be just so.

  The Thomas Jefferson cubesat (a bit down that Wikipedia article it claims that "[ I]ts completion will mark the first CubeSat to be successfully launched by highschool students into space"), on the other hand, does not have that power or antenna directivity, so it must remain in LEO.

• #479 by antonioe on 01 Mar, 2008 12:34

• By the way, I'm having some unusual trouble with Private Messages: I can receive and read them, but I can't either reply or send a new one.  I tried both my regular browser (IE on Windows) as well as Mozilla Firefox on both Windoes and Linux (Ubuntu).  Same problem on all three configurations.  Maybe I've exceeded some quota, maybe I've managed to really tick Chris off :laugh:

  So if you've PM'd me and I have not answered, that's why.  Sorry.

  Feel free to e-mail me at my home address which is on my public profile.