#### LouScheffer

• Senior Member
• Posts: 2000
• Liked: 2791
• Likes Given: 299
##### Re: Antares General Discussion Thread
« Reply #760 on: 06/13/2018 02:08 pm »
Speaking of which, can anyone help decipher the Antares 23x+ characteristic energy versus payload chart?  Where would GTO be, for example?

- Ed Kyle
I've never done this for a bound orbit, but using the definition that c3 = 2x specific orbital energy, we get

c3 = 2 * (v^2/2- u/r)

For GTO, v = 1596 m/s at the top of the orbit (if a 250 km perigee), r = 42157 km, and u for Earth is 3.986e14, then plugging in we get -16.4 km^2/sec^2.
For a 250 km circular LEO, v=7760 m/s and r=6621 km, giving C3 = -64.9 km^2/sec^2.   This agrees with the range of the horizontal axis.

For GEO, v=3075 m/s and r=42157 km, giving C3 = -9.45 km^2/sec^2.  But you can't use the chart for this, since it needs at least two burn sequence, and an inclination change (which takes delta-V, but does not change C3).  The chart appears to be for a single impulse starting at LEO altitude.

#### edkyle99

• Expert
• Senior Member
• Posts: 13234
• Liked: 4542
• Likes Given: 823
##### Re: Antares General Discussion Thread
« Reply #761 on: 06/13/2018 02:32 pm »
Some additional pain from the higher latitude of the MARS launch site.  Maybe enough performance to try for a GPS launch without direct injection, though?
Maybe 2.8/3.2 tonnes to a 200 x 20,370 km type transfer orbit, enough for the old Delta 2 GPS missions, but neither the mass nor the orbit meet the EELV-2 requirements (4.082 tonnes to a 1,000 x 20,368 km x 55 deg orbit).  Thus Omega, but they must have been tempted to turn Antares into an EELV-2 contender at one point.  It can't handle the Heavy missions though.

- Ed Kyle
« Last Edit: 06/13/2018 02:36 pm by edkyle99 »

#### edkyle99

• Expert
• Senior Member
• Posts: 13234
• Liked: 4542
• Likes Given: 823
##### Re: Antares General Discussion Thread
« Reply #762 on: 06/13/2018 05:33 pm »
In terms of payload performance, this Antares 230+ series could have handled about half of the missions launched so far this year world-wide.  Most of those were non-U.S. launches however, the exceptions being OA-9, Paz, and probably Zuma, which would have been perhaps interesting under a Northrop Grumman banner.

- Ed Kyle
« Last Edit: 06/13/2018 05:33 pm by edkyle99 »

#### brickmack

• Full Member
• Posts: 557
• USA
• Liked: 253
• Likes Given: 19
##### Re: Antares General Discussion Thread
« Reply #763 on: 06/13/2018 05:59 pm »
but they must have been tempted to turn Antares into an EELV-2 contender at one point.

Nonviable for political reasons. Russian engines, Ukrainian tanks. Hypothetically though, I wonder what they could have done in this market (or, perhaps, what they will do in the civilian market in this class if OmegA is not selected for EELV. They did mention that OmegA would not continue development in that scenario, and would be replaced with a different rocket for civilian use which wouldn't meet EELV requirements). The core stage of course is pretty similar in thrust, wet and dry mass, and ISP to the Atlas V core. A liquid upper stage was proposed early in the development process but fizzled. If you put a hydrogen upper stage on top (derived from their work on OmegAs US probably), you'd probably end up with something awfully close to Atlas V 401 in performance (maybe a tad better, especially to LEO, if they stick with 2 RL10s). We know strapon boosters were considered early in Antares's development as well, but were ditched because the Zenit derived tanks couldn't handle it without extensive mods, and the same was true of the ground interfaces too. Presuming those are still true, and would remain true after the mods needed for a liquid upper stage, what sort of performance might they manage with a solid or hypergolic third stage on top, with no strapons? This probably wouldn't improve LEO performance much/at all, but for high energy orbits I bet you could match the larger Atlas variants. Direct GEO insertion could be a bit easier too, since theres no need for long-duration cryo storage. Still not nearly good enough for EELV, but that'd cover most of the commercial market, and should probably be cheaper than Atlas V if flown in similar volumes.

#### edkyle99

• Expert
• Senior Member
• Posts: 13234
• Liked: 4542
• Likes Given: 823
##### Re: Antares General Discussion Thread
« Reply #764 on: 06/14/2018 02:22 pm »
but they must have been tempted to turn Antares into an EELV-2 contender at one point.
Nonviable for political reasons. Russian engines, Ukrainian tanks.
Agreed, at least for the current time-frame.  I see Antares in a non-EELV niche that has potential.  Just this year, I count (on further reflection) four or five missions that Antares 230+ series could have handled:  Paz, TESS, InSight, OA-9 (of course), and probably Zuma.

The market among these was LEO/ISS, LEO/sun synchronous (commercial and government, but could be civil), and deep space civil.  Falcon 9 and Atlas 5 handled four of these missions, so Antares 230+ has to compete with these on cost.  I believe it can for these smaller payloads.

Think about it.  Antares should be able to compete with Atlas 5 simply due to the lower-cost of the rocket.  That it can compete with Falcon 9 is hinted at by its lower bid for the CRS-2 program, though in that case the payloads obviously accounted for part of the cost.

I think of it this way.  The Russo-Ukrainian first stage is essentially "free", certainly lower cost than it would be if built in the U.S.  That leaves the upper stage or stages, which should cost less than the Falcon 9 or Atlas 5 upper stages (which are also expended).

- Ed Kyle
« Last Edit: 06/14/2018 02:33 pm by edkyle99 »

#### envy887

• Senior Member
• Posts: 5610
• Liked: 3423
• Likes Given: 1667
##### Re: Antares General Discussion Thread
« Reply #765 on: 06/14/2018 02:39 pm »
but they must have been tempted to turn Antares into an EELV-2 contender at one point.
Nonviable for political reasons. Russian engines, Ukrainian tanks.
Agreed, at least for the current time-frame.  I see Antares in a non-EELV niche that has potential.  Just this year, I count four or five missions that Antares 230+ series could have handled:  Paz, TESS, InSight, OA-9 (of course), and probably Zuma.

The market among these was LEO/ISS, LEO/sun synchronous (commercial and government, but could be civil), and deep space civil.  Falcon 9 and Atlas 5 handled four of these missions, so Antares 230+ has to compete with these on cost.  I believe it can for these smaller payloads.

Think about it.  Antares should be able to compete with Atlas 5 simply due to the lower-cost of the rocket.  That it can compete with Falcon 9 is hinted at by its lower bid for the CRS-2 program, though in that case the payloads obviously accounted for part of the cost.

I think of it this way.  The Russo-Ukrainian first stage is essentially "free", certainly lower cost than it would be if built in the U.S.  That leaves the upper stage or stages, which should cost less than the Falcon 9 or Atlas 5 upper stages (which are also expended).

- Ed Kyle

The GAO puts the price of Antares at 80-85 million dollars, which is competitive with Atlas and Vulcan, but not with Falcon 9.
https://www.gao.gov/assets/690/686613.pdf

The booster is definitely not anything approaching "free", and the upper stage probably is not significantly cheaper than the F9 upper stage. And the low flight rate hurts it compared to F9 and even Atlas.

#### edkyle99

• Expert
• Senior Member
• Posts: 13234
• Liked: 4542
• Likes Given: 823
##### Re: Antares General Discussion Thread
« Reply #766 on: 06/14/2018 03:27 pm »
but they must have been tempted to turn Antares into an EELV-2 contender at one point.
Nonviable for political reasons. Russian engines, Ukrainian tanks.
Agreed, at least for the current time-frame.  I see Antares in a non-EELV niche that has potential.  Just this year, I count four or five missions that Antares 230+ series could have handled:  Paz, TESS, InSight, OA-9 (of course), and probably Zuma.

The market among these was LEO/ISS, LEO/sun synchronous (commercial and government, but could be civil), and deep space civil.  Falcon 9 and Atlas 5 handled four of these missions, so Antares 230+ has to compete with these on cost.  I believe it can for these smaller payloads.

Think about it.  Antares should be able to compete with Atlas 5 simply due to the lower-cost of the rocket.  That it can compete with Falcon 9 is hinted at by its lower bid for the CRS-2 program, though in that case the payloads obviously accounted for part of the cost.

I think of it this way.  The Russo-Ukrainian first stage is essentially "free", certainly lower cost than it would be if built in the U.S.  That leaves the upper stage or stages, which should cost less than the Falcon 9 or Atlas 5 upper stages (which are also expended).

- Ed Kyle

The GAO puts the price of Antares at 80-85 million dollars, which is competitive with Atlas and Vulcan, but not with Falcon 9.
https://www.gao.gov/assets/690/686613.pdf
But this well-known OIG report suggests that Falcon 9 prices may have, or will be, increasing.
https://oig.nasa.gov/docs/IG-18-016.pdf#bi

Either way, it is clear that Northrop Grumman will have to tightly control Antares costs to compete with Falcon 9.  I believe that the apparent decision not to stretch the first stage or to quickly build a West Coast launch pad are signs of such cost-control, while SpaceX continues to announce big new spending, like the KSC control center and the BFR project.  We'll have to see an actual launch contract win for a mission like those I listed earlier to see evidence of competitiveness.

- Ed Kyle

#### envy887

• Senior Member
• Posts: 5610
• Liked: 3423
• Likes Given: 1667
##### Re: Antares General Discussion Thread
« Reply #767 on: 06/14/2018 04:33 pm »
but they must have been tempted to turn Antares into an EELV-2 contender at one point.
Nonviable for political reasons. Russian engines, Ukrainian tanks.
Agreed, at least for the current time-frame.  I see Antares in a non-EELV niche that has potential.  Just this year, I count four or five missions that Antares 230+ series could have handled:  Paz, TESS, InSight, OA-9 (of course), and probably Zuma.

The market among these was LEO/ISS, LEO/sun synchronous (commercial and government, but could be civil), and deep space civil.  Falcon 9 and Atlas 5 handled four of these missions, so Antares 230+ has to compete with these on cost.  I believe it can for these smaller payloads.

Think about it.  Antares should be able to compete with Atlas 5 simply due to the lower-cost of the rocket.  That it can compete with Falcon 9 is hinted at by its lower bid for the CRS-2 program, though in that case the payloads obviously accounted for part of the cost.

I think of it this way.  The Russo-Ukrainian first stage is essentially "free", certainly lower cost than it would be if built in the U.S.  That leaves the upper stage or stages, which should cost less than the Falcon 9 or Atlas 5 upper stages (which are also expended).

- Ed Kyle

The GAO puts the price of Antares at 80-85 million dollars, which is competitive with Atlas and Vulcan, but not with Falcon 9.
https://www.gao.gov/assets/690/686613.pdf
But this well-known OIG report suggests that Falcon 9 prices may have, or will be, increasing.
https://oig.nasa.gov/docs/IG-18-016.pdf#bi

Either way, it is clear that Northrop Grumman will have to tightly control Antares costs to compete with Falcon 9.  I believe that the apparent decision not to stretch the first stage or to quickly build a West Coast launch pad are signs of such cost-control, while SpaceX continues to announce big new spending, like the KSC control center and the BFR project.  We'll have to see an actual launch contract win for a mission like those I listed earlier to see evidence of competitiveness.

- Ed Kyle

I don't see any evidence in the OIG report that F9 prices will increase; all the SpaceX increases for CRS-2 are attributed to Dragon 2, not to F9.

Orbital managed to reduce their prices CRS-2, though it's not clear if that is related to Antares or Cygnus of both.

Antares' lack of non-NASA commercial contracts is already good evidence for it's competitiveness, or rather, lack of competitiveness. Why would a customer pay 30% more for 1/2 the performance of F9R. Atlas V offers unique services, and so is able to sustain higher prices, but Antares doesn't offer anything customers can't get cheaper on F9R.

#### edkyle99

• Expert
• Senior Member
• Posts: 13234
• Liked: 4542
• Likes Given: 823
##### Re: Antares General Discussion Thread
« Reply #768 on: 06/14/2018 05:50 pm »
Antares' lack of non-NASA commercial contracts is already good evidence for it's competitiveness, or rather, lack of competitiveness. Why would a customer pay 30% more for 1/2 the performance of F9R. Atlas V offers unique services, and so is able to sustain higher prices, but Antares doesn't offer anything customers can't get cheaper on F9R.
Obviously, customers won't pay more, certainly not 30% more.   The key will be for Northrop Grumman to exploit the smaller size of this rocket, its largely outsourced first stage, its low-cost solid upper stage motors, its smaller launch site footprint, etc..  It should cost less than the much larger and more capable Falcon 9.  It is roughly half the weight and thrust.  It will have to cost less, ultimately, or it won't have work.

- Ed Kyle
« Last Edit: 06/14/2018 05:58 pm by edkyle99 »

#### RonM

• Senior Member
• Posts: 2723
• Atlanta, Georgia USA
• Liked: 1490
• Likes Given: 1129
##### Re: Antares General Discussion Thread
« Reply #769 on: 06/14/2018 06:00 pm »
Antares' lack of non-NASA commercial contracts is already good evidence for it's competitiveness, or rather, lack of competitiveness. Why would a customer pay 30% more for 1/2 the performance of F9R. Atlas V offers unique services, and so is able to sustain higher prices, but Antares doesn't offer anything customers can't get cheaper on F9R.
Obviously, customers won't pay more, certainly not 30% more.   The key will be for Northrop Grumman to exploit the smaller size of this rocket, its largely outsourced first stage, its low-cost solid upper stage motors, its smaller launch site footprint, etc..  It should cost less than the much larger and more capable Falcon 9.  It will have to cost less, ultimately, or it won't have work.

- Ed Kyle

The problem is the largely outsourced first stage. Part of SpaceX's price advantage is building most of the parts in-house. NG needs to look at reducing costs of the first stage so they can be competitive. Anyone have ideas on how NG can do that?

#### russianhalo117

• Senior Member
• Posts: 5044
• AR USA / Berlin, DE / Moscow, RF
• Liked: 1388
• Likes Given: 599
##### Re: Antares General Discussion Thread
« Reply #770 on: 06/14/2018 06:17 pm »
Antares' lack of non-NASA commercial contracts is already good evidence for it's competitiveness, or rather, lack of competitiveness. Why would a customer pay 30% more for 1/2 the performance of F9R. Atlas V offers unique services, and so is able to sustain higher prices, but Antares doesn't offer anything customers can't get cheaper on F9R.
Obviously, customers won't pay more, certainly not 30% more.   The key will be for Northrop Grumman to exploit the smaller size of this rocket, its largely outsourced first stage, its low-cost solid upper stage motors, its smaller launch site footprint, etc..  It should cost less than the much larger and more capable Falcon 9.  It will have to cost less, ultimately, or it won't have work.

- Ed Kyle

The problem is the largely outsourced first stage. Part of SpaceX's price advantage is building most of the parts in-house. NG needs to look at reducing costs of the first stage so they can be competitive. Anyone have ideas on how NG can do that?
Shift stage production to the US like Firefly Aerospace Inc. (FAI) is doing. FAI is having YUZHMASH build its ground test, qual and test flight articles whereas operational launcher production can be done in both the US and Ukraine. Doing the same would reduce costs and provide an increased launch rate. Switching the first stage structures to Carbon fibre would also help but I do not see that happening any time soon.

#### edkyle99

• Expert
• Senior Member
• Posts: 13234
• Liked: 4542
• Likes Given: 823
##### Re: Antares General Discussion Thread
« Reply #771 on: 06/14/2018 06:29 pm »
The problem is the largely outsourced first stage. Part of SpaceX's price advantage is building most of the parts in-house. NG needs to look at reducing costs of the first stage so they can be competitive. Anyone have ideas on how NG can do that?
The standard method used to control subcontractor pricing is competitive bidding

- Ed Kyle

#### envy887

• Senior Member
• Posts: 5610
• Liked: 3423
• Likes Given: 1667
##### Re: Antares General Discussion Thread
« Reply #772 on: 06/14/2018 06:57 pm »
The problem is the largely outsourced first stage. Part of SpaceX's price advantage is building most of the parts in-house. NG needs to look at reducing costs of the first stage so they can be competitive. Anyone have ideas on how NG can do that?
The standard method used to control subcontractor pricing is competitive bidding

- Ed Kyle

That doesn't work well for buying hardware when the design is proprietary and very user-specific. Now if you're buying services, and multiple contractors can deliver the same service, you can do that. But it harder for hardware. Especially low-volume single application hardware like rocket engines.

Antares' lack of non-NASA commercial contracts is already good evidence for it's competitiveness, or rather, lack of competitiveness. Why would a customer pay 30% more for 1/2 the performance of F9R. Atlas V offers unique services, and so is able to sustain higher prices, but Antares doesn't offer anything customers can't get cheaper on F9R.
Obviously, customers won't pay more, certainly not 30% more.   The key will be for Northrop Grumman to exploit the smaller size of this rocket, its largely outsourced first stage, its low-cost solid upper stage motors, its smaller launch site footprint, etc..  It should cost less than the much larger and more capable Falcon 9.  It is roughly half the weight and thrust.  It will have to cost less, ultimately, or it won't have work.

- Ed Kyle

The avionics probably aren't smaller,or cheaper than F9 since AIUI they don't use off-the-shelf hardware. Nor the booster airframe, as it's purchased rather than built. Nor the engines - ULA pays ~24 million dollars for a single RD-180, and I doubt a pair of RD-181's is much if any, cheaper.

Horizontal integration is a great way to avoid development and capitalization costs, but it's a poor way to reduce the per-unit price compared to a vertically integrated company.

#### ZachF

• Full Member
• Posts: 790
• NH, USA, Earth
• Liked: 953
• Likes Given: 183
##### Re: Antares General Discussion Thread
« Reply #773 on: 06/14/2018 08:23 pm »

The avionics probably aren't smaller,or cheaper than F9 since AIUI they don't use off-the-shelf hardware. Nor the booster airframe, as it's purchased rather than built. Nor the engines - ULA pays ~24 million dollars for a single RD-180, and I doubt a pair of RD-181's is much if any, cheaper.

The +/-\$24 million for the engines is probably not that far off the cost of an entire F9 lower stage. The lower stage is probably ~\$40 million for the Antares and ~\$30m for the F9 if I had to guess.
artist, so take opinions expressed above with a well-rendered grain of salt...
https://www.instagram.com/artzf/

#### TrevorMonty

##### Re: Antares General Discussion Thread
« Reply #774 on: 06/14/2018 09:21 pm »

The avionics probably aren't smaller,or cheaper than F9 since AIUI they don't use off-the-shelf hardware. Nor the booster airframe, as it's purchased rather than built. Nor the engines - ULA pays ~24 million dollars for a single RD-180, and I doubt a pair of RD-181's is much if any, cheaper.

The +/-\$24 million for the engines is probably not that far off the cost of an entire F9 lower stage. The lower stage is probably ~\$40 million for the Antares and ~\$30m for the F9 if I had to guess.
Antares is never going be competitive with F9R on price or performance and not likely to get high enough flight rate to match F9R current reliability.
The Antares will stay as long as there is Cygnus missions to LEO. If they are lucky sell the odd commercial launch.

#### rayleighscatter

• Full Member
• Posts: 1050
• Maryland
• Liked: 488
• Likes Given: 232
##### Re: Antares General Discussion Thread
« Reply #775 on: 06/14/2018 09:48 pm »

The avionics probably aren't smaller,or cheaper than F9 since AIUI they don't use off-the-shelf hardware. Nor the booster airframe, as it's purchased rather than built. Nor the engines - ULA pays ~24 million dollars for a single RD-180, and I doubt a pair of RD-181's is much if any, cheaper.

I could easily see RD-181 being cheaper for NG than RD-180 is for ULA. Aside from being a different and less complex engine NG also isn't a captive customer. They can always move on to another engine, another launcher, or exit the launch business altogether. ULA doesn't(didn't) really have that option with Atlas V. For national security it was the only game in town so any price could be asked.

It's like Soyuz. We've seen what a privately purchased Soyuz seat costs, and what a NASA purchased Soyuz seat costs. A private customer can always walk away, NASA is captive.

So it comes down to business. If Energomash wants to sell engines, they can't price themselves out of the market. Even Antares isn't a free lunch, NG has already shown they'll switch launchers on CRS.

#### russianhalo117

• Senior Member
• Posts: 5044
• AR USA / Berlin, DE / Moscow, RF
• Liked: 1388
• Likes Given: 599
##### Re: Antares General Discussion Thread
« Reply #776 on: 06/14/2018 10:09 pm »

The avionics probably aren't smaller,or cheaper than F9 since AIUI they don't use off-the-shelf hardware. Nor the booster airframe, as it's purchased rather than built. Nor the engines - ULA pays ~24 million dollars for a single RD-180, and I doubt a pair of RD-181's is much if any, cheaper.

I could easily see RD-181 being cheaper for NG than RD-180 is for ULA. Aside from being a different and less complex engine NG also isn't a captive customer. They can always move on to another engine, another launcher, or exit the launch business altogether. ULA doesn't(didn't) really have that option with Atlas V. For national security it was the only game in town so any price could be asked.

It's like Soyuz. We've seen what a privately purchased Soyuz seat costs, and what a NASA purchased Soyuz seat costs. A private customer can always walk away, NASA is captive.

So it comes down to business. If Energomash wants to sell engines, they can't price themselves out of the market. Even Antares isn't a free lunch, NG has already shown they'll switch launchers on CRS.
RD-180 sales have to go through a middle man which takes a cut of each engine purchase on top of the base price. That middle man is United Technology Corporation's Joint Venture RD Amross, LLC formed between subsidiary Pratt & Whitney and JSC NPO Energomash (fka NPO Energomash). Due to various reasons UTC's 50% share was not transferred to Aerojet General upon Rocketdyne's spinoff from PW and its merger with AG to become Aerojet Rocketdyne.

RD-181 is directly purchased from JSC NPO Energomash at export price.
« Last Edit: 06/14/2018 11:29 pm by russianhalo117 »

#### edkyle99

• Expert
• Senior Member
• Posts: 13234
• Liked: 4542
• Likes Given: 823
##### Re: Antares General Discussion Thread
« Reply #777 on: 06/14/2018 10:57 pm »
The +/-\$24 million for the engines is probably not that far off the cost of an entire F9 lower stage. The lower stage is probably ~\$40 million for the Antares and ~\$30m for the F9 if I had to guess.
Falcon 9 first stage almost certainly costs more then \$24 million.  If it only cost that much, SpaceX wouldn't bother trying to recover the stage!  Musk has said that the first stage accounts for 60-75% of the total Falcon 9 cost (varying amounts depending on when he was asked).  Gwynn Shotwell also once said that it cost less than half the cost of a new first stage to refurbish and refly a stage.  That is the number that Northrop Grumman needs to target.

- Ed Kyle

#### fthomassy

• Full Member
• Posts: 237
• Austin, Texas, Earth, Sol, Orion, Milky-Way, Virgo, Bang 42
• Liked: 158
• Likes Given: 2342
##### Re: Antares General Discussion Thread
« Reply #778 on: 06/14/2018 11:26 pm »
Falcon 9 first stage almost certainly costs more then \$24 million.  If it only cost that much, SpaceX wouldn't bother trying to recover the stage!  Musk has said that the first stage accounts for 60-75% of the total Falcon 9 cost (varying amounts depending on when he was asked).  Gwynn Shotwell also once said that it cost less than half the cost of a new first stage to refurbish and refly a stage.
I think your idea of value is different than most. A >50% additional margin per reuse seems worth the effort.
gyatm . . . Fern

#### envy887

• Senior Member
• Posts: 5610
• Liked: 3423
• Likes Given: 1667
##### Re: Antares General Discussion Thread
« Reply #779 on: 06/15/2018 01:26 am »

The avionics probably aren't smaller,or cheaper than F9 since AIUI they don't use off-the-shelf hardware. Nor the booster airframe, as it's purchased rather than built. Nor the engines - ULA pays ~24 million dollars for a single RD-180, and I doubt a pair of RD-181's is much if any, cheaper.

I could easily see RD-181 being cheaper for NG than RD-180 is for ULA. Aside from being a different and less complex engine NG also isn't a captive customer. They can always move on to another engine, another launcher, or exit the launch business altogether. ULA doesn't(didn't) really have that option with Atlas V. For national security it was the only game in town so any price could be asked.

It's like Soyuz. We've seen what a privately purchased Soyuz seat costs, and what a NASA purchased Soyuz seat costs. A private customer can always walk away, NASA is captive.

So it comes down to business. If Energomash wants to sell engines, they can't price themselves out of the market. Even Antares isn't a free lunch, NG has already shown they'll switch launchers on CRS.
How is 2x RD-181 less complex than a single RD-180? The pair have all the same components, plus an extra powerhead, compared to the single engine.

And Northrop can't exit the launch business and leave NASA hanging on CRS. Not if they want to keep building multiple billion dollar satellites for NASA.

Tags: