Author Topic: DARPA Phoenix Satellite Service and Salvage program  (Read 24688 times)

Offline GuessWho

  • Member
  • Posts: 94
  • Liked: 8
  • Likes Given: 0
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #20 on: 11/17/2013 12:22 AM »

1.  Ground testing does not qualify "demonstrated" for flight operations

2. Business arrangements that were never carried out do not qualify as proof.

Response to 1)  Ground tests is vacuum certainly qualify the hardware as a micro-g environment has no impact on that part of the operations.  Zero-g fluid transfer has been demonstrated on orbit numerous times since Shuttle started flying.  There was even a demonstration of a Coke machine transferring carbonated fluids so the physics and processes are understood.  No tech dev there.  Orbital Express demonstrated sat-to-sat fluid transfer.  ISS flight ops by GSFC has demonstrated end-to-end valve connection, fluid transfer, valve close-out.  TRL 9 Bub.  Learn before you post.

Response to 2)  Whether a business case closes was not the topic.  You stated that "It is magnitudes greater." in response to a question about "the cost difference between refueling a sat and moving a sat to a graveyard orbit is likely not that great ...".  Intelsat signed a contract with MDA for $280M to provide fuel to 5 existing satellites to extend their operational life by 5 years each.  That contract was a legally binding agreement.  Whether it was ultimately executed doesn't alter the value of the service being offered.  That $280M contract was for 25 years of cumulative operation for an average of $11.2M per year/per sat.  Intelsat generates between $50M and $100M in revenue per year per satellite or between $4M and $8M in revenue per month.  Executing a graveyard maneuver consumes between 2-3 months of station-keeping fuel equating representing $8M - $24M in revenue potential.  Thus refueling is not "orders of magnitude" greater cost than simply executing a GEO graveyard maneuver.  Learn before you post.  Think before you respond.

Besides, the challenge is not making the physical con

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 30391
  • Cape Canaveral Spaceport
  • Liked: 8683
  • Likes Given: 284
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #21 on: 11/17/2013 03:17 AM »

Response to 2)  Whether a business case closes was not the topic.  You stated that "It is magnitudes greater." in response to a question about "the cost difference between refueling a sat and moving a sat to a graveyard orbit is likely not that great ...".  Intelsat signed a contract with MDA for $280M to provide fuel to 5 existing satellites to extend their operational life by 5 years each.  That contract was a legally binding agreement.  Whether it was ultimately executed doesn't alter the value of the service being offered.  That $280M contract was for 25 years of cumulative operation for an average of $11.2M per year/per sat.  Intelsat generates between $50M and $100M in revenue per year per satellite or between $4M and $8M in revenue per month.  Executing a graveyard maneuver consumes between 2-3 months of station-keeping fuel equating representing $8M - $24M in revenue potential.  Thus refueling is not "orders of magnitude" greater cost than simply executing a GEO graveyard maneuver. 


Quite wrong.  A executed contract and closed business case are required to back up your assertions, otherwise they are just meaningless numbers.   The $280M was not validated.

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 30391
  • Cape Canaveral Spaceport
  • Liked: 8683
  • Likes Given: 284
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #22 on: 11/17/2013 03:20 AM »
  There was even a demonstration of a Coke machine transferring carbonated fluids


that didn't really work.  Know something before you post.  Anyways, the issue isn't fluid transfer. 
« Last Edit: 11/17/2013 03:40 AM by Jim »

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 30391
  • Cape Canaveral Spaceport
  • Liked: 8683
  • Likes Given: 284
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #23 on: 11/17/2013 03:37 AM »

1.  Yes, the technologies have been ground tested.  At MDA and at NASA.  NASA is flight testing those technologies on-orbit at ISS.

2.  No, it is not "magnitudes" greater.  Graveyarding costs 2-3 months of effective satellite revenue thus on the order of $10-$12M in lost revenue.  Prior business arrangement have shown that refueling is worth $10M-$12M per year of added life per satellite to an owner operator.

1.  Ground testing does not qualify "demonstrated" for flight operations

Jim... haven't you seen the Robotic Refueling Mission videos? They've been testing it on orbit on ISS on simulated task boards. That may not be TRL 9 yet, but that should count for something, even to a skeptic, shouldn't it?

Quote
2. Business arrangements that were never carried out do not qualify as proof.

While it is true that nobody has demonstrated the business case closes...isn't that true of even of successful business models before someone has made them work the first time? I agree there's room for *some* skepticism, particularly about timing and busines model closure at this point in time. But there's plenty of examples of technologies or business models that people thought of long before all the pieces were there to make them happen. Cars and heavier than air flight come to mind for instance.

~Jon

The issue I have is doing the refueling on spacecraft that were not design for it.  Look at all the times NASA serviced a spacecraft that was not design for it, there were always dimensional problems and the crew had to do workarounds.   

SMM - MMU/TPAD would not engage.  Ended up using RMS to grab spacecraft
HS 376 retrieval - Handling frame would not engage, crew had to hold on to spacecraft by omni antenna
Intelsat VI reboost - capture frame would not engage, 3 EVA crewmembers had to grab  spacecraft.

Offline a_langwich

  • Full Member
  • ****
  • Posts: 735
  • Liked: 211
  • Likes Given: 48
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #24 on: 11/17/2013 06:27 AM »

Response to 2)  Whether a business case closes was not the topic.  You stated that "It is magnitudes greater." in response to a question about "the cost difference between refueling a sat and moving a sat to a graveyard orbit is likely not that great ...".  Intelsat signed a contract with MDA for $280M to provide fuel to 5 existing satellites to extend their operational life by 5 years each.  That contract was a legally binding agreement.  Whether it was ultimately executed doesn't alter the value of the service being offered.  That $280M contract was for 25 years of cumulative operation for an average of $11.2M per year/per sat.  Intelsat generates between $50M and $100M in revenue per year per satellite or between $4M and $8M in revenue per month.  Executing a graveyard maneuver consumes between 2-3 months of station-keeping fuel equating representing $8M - $24M in revenue potential.  Thus refueling is not "orders of magnitude" greater cost than simply executing a GEO graveyard maneuver.  Learn before you post.  Think before you respond.

Wait, what?  What, in those numbers, actually represents the cost of refueling a satellite?  None of it.  That's what Intelsat was willing to pay, which of course is driven by their revenue numbers and alternative costs.   

But that's not the cost to refuel a spacecraft, that's the price point you have to meet, that's what you WISH you could accomplish with margin for a profit.  The cost to you to refuel a spacecraft is what it costs to design and build the refueling satellite, plus the cost to launch the equipment, including your refueling device, the amount of fuel it will need to reach the graveyard orbit, and the amount of fuel it plans to load onto the disabled satellite.  And what Jim is saying, as I understand it,  is since nobody has actually done that, those costs are still hypothetical.

I'm not arguing for orders of magnitude difference in price, but in order for it to close right now, I suspect you'd need at least a half-dozen EOL satellites, with not too much deltaV separation between orbits, that can be refueled from one launch of your equipment.  And it better happen reasonably quickly, too, because your debt costs money and your personnel can't wait a decade before a payout.  And if Jim is right, and it turns out your vehicle was a flight test vehicle that gave you lots of lessons learned but failed to actually refuel a satellite, you and your investors will have to eat that  hundred million or two or three. 

And why compare against the cost to graveyard a satellite?  There is zero chance within the next fifty years that graveyarding could be avoided.  It seems foolish to risk debris exactly coincident with the orbit of your on-service location.  Maybe you could choose a different "hospital" orbit instead of a graveyard one, but there will still be significant fuel cost to reach it--and the size of that cost is proportional to the size of the safety buffer you've given your operational orbit.  Therefore, it seems to me that the end-of-life orbit maneuver cost happens whether you planned to service it or not. 

Comparison with replacement, as opposed to the cost of an end-of-life safing maneuver, makes more sense to me.

All of this discussion is barely relevant to DARPA Phoenix, which although "Service" is in the title seems to be more about "Salvage."  That is, re-using a dead satellite's parts to pimp your new satellite ride.

Offline baldusi

  • Senior Member
  • *****
  • Posts: 7417
  • Buenos Aires, Argentina
  • Liked: 1413
  • Likes Given: 4390
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #25 on: 11/17/2013 02:13 PM »
I need some help with the refueling system. In particular how do you handle pressurization? AIUI, the pressurization system is a High pressure bottle of He, a regulator and a membrane on the propellant tanks. The regulator and the membrane keeps the propellant at a given pressure. Now come the part I see as a problem.
If you try to pump fuel/oxidizer back into the tank, you'd push back the membrane and, since the regulator is usually also a one way valve, you'd compress the He behind the membrane, increasin dangerously the pressure on the whole system. Since the attitude and main engines are usually pressure fed, this brings lots and lots of issues.
You might then be able to purge the whole He system, fill the tanks, refill the He system. But then you'd have to do not only another fluid, but you'd have to somehow have a compressor (to 3000psi may be?). And this is assuming that they somehow can actually access and fill the He bottle.

Offline GuessWho

  • Member
  • Posts: 94
  • Liked: 8
  • Likes Given: 0
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #26 on: 11/17/2013 02:22 PM »

SMM - MMU/TPAD would not engage.  Ended up using RMS to grab spacecraft
HS 376 retrieval - Handling frame would not engage, crew had to hold on to spacecraft by omni antenna
Intelsat VI reboost - capture frame would not engage, 3 EVA crewmembers had to grab  spacecraft.


SMM-MMU:  Fair statement, not every satellite is a candidate for servicing.  But putting a man in the middle of that effort introduces other considerations and constraints that without his presence, may have allowed a different approach to be more successful.  Wouldn't want to speculate though.

HS 376:  The actual capture went extremely well and showed that a capture mechanism designed to engage via the apogee engine works quite well.  Why you think failure of a storage device designed to hold the sat in the shuttle bay for return to earth constitutes a failure of on-orbit capture and servicing is beyond me.

Intelsat VI reboost:  And the lesson learned is not to design an overly complex solution to a simple problem.  The whole cross-bar solution was a stupid one.   Note that current approaches (DARPA OE, PHOENIX, MDA SIS) typically use one or more robotic arms with capture "hands" to capture the interface ring where an upper stage attaches.  Surprisingly, exactly how the STS-49 crew ultimately captured Intelsat VI.  Hardly a damning indictment of on-orbit servicing.  More a case of applying lessons learned.


Offline GuessWho

  • Member
  • Posts: 94
  • Liked: 8
  • Likes Given: 0
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #27 on: 11/17/2013 02:44 PM »

Wait, what?  What, in those numbers, actually represents the cost of refueling a satellite?  None of it.  That's what Intelsat was willing to pay, which of course is driven by their revenue numbers and alternative costs.   


And what MDA set the price at which they were willing to contractually sell the service for.  That price of course is driven by their business case which includes all of the costs you mentioned; i.e., what it costs to design and build the refueling satellite, the cost to launch the equipment, the amount of fuel it needs to reach client satellite, the fuel it uses to perform ARPO and capture,  the amount of fuel it plans to load onto the disabled satellite, the ground infrastructure costs to execute the mission, the insurance costs, taxes, interest, etc. such that at the end of the day, they actually realize a profit.

Quote

I'm not arguing for orders of magnitude difference in price, but in order for it to close right now, I suspect you'd need at least a half-dozen EOL satellites, with not too much deltaV separation between orbits, that can be refueled from one launch of your equipment.  And it better happen reasonably quickly, too, because your debt costs money and your personnel can't wait a decade before a payout.


Jim did argue for "orders of magnitude difference" and seems unwilling to retract that statement.  But that is not your point so let me respond.  Intelsat and MDA signed a contract for 5 satellites (plus a sixth to be used as the initial demonstration of the refuel capability) so yes, you need a half-dozen or so clients signed up to even consider executing a business case.  Rephasing from one sat to the next is not that delta-V intensive.  It depends on how far apart they are on the belt and how quickly you want to perform the phasing maneuver.  Given MDA was using an EP prop system to perform the phasing maneuvers, I suspect the timing was slow and the fuel use minimal.  Also, if you read up on the MDA approach, they were designed to refuel twice the number of sats that Intelsat actually settled for.  So presumably, MDA could refuel 10 satellites with a single servicer.  And who is to say that they couldn't then refuel themselves and continue operating for several years?

And from what I have gathered from several MDA presentations, they were expecting to launch within 3 years of signing the contract.  Problem is, they couldn't sell the other 5 refuel missions they had capacity for so didn't execute the contract.  Still doesn't alter the price that both the seller and the buyer were willing to pay for a refuel service which still isn't "orders of magnitude" greater than simply graveyarding a satellite at EOL.

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 30391
  • Cape Canaveral Spaceport
  • Liked: 8683
  • Likes Given: 284
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #28 on: 11/17/2013 02:59 PM »

HS 376:  The actual capture went extremely well and showed that a capture mechanism designed to engage via the apogee engine works quite well.  Why you think failure of a storage device designed to hold the sat in the shuttle bay for return to earth constitutes a failure of on-orbit capture and servicing is beyond me.

Intelsat VI reboost:  And the lesson learned is not to design an overly complex solution to a simple problem.  The whole cross-bar solution was a stupid one.   Note that current approaches (DARPA OE, PHOENIX, MDA SIS) typically use one or more robotic arms with capture "hands" to capture the interface ring where an upper stage attaches.  Surprisingly, exactly how the STS-49 crew ultimately captured Intelsat VI.  Hardly a damning indictment of on-orbit servicing.  More a case of applying lessons learned.


The issue isn't that on-orbit servicing or capture isn't feasible.  It is that interfaces (i.e prop and pressurant fill and drain valves) not designed for onorbit servicing will be problematic.

Offline GuessWho

  • Member
  • Posts: 94
  • Liked: 8
  • Likes Given: 0
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #29 on: 11/18/2013 04:48 AM »

The issue isn't that on-orbit servicing or capture isn't feasible.  It is that interfaces (i.e prop and pressurant fill and drain valves) not designed for onorbit servicing will be problematic.
 

The folks at Goddard would likely disagree with that statement given the ISS testing they have performed.

Online Robotbeat

  • Senior Member
  • *****
  • Posts: 25927
  • Minnesota
  • Liked: 5964
  • Likes Given: 4417
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #30 on: 11/18/2013 04:56 AM »
Problematic doesn't mean there isn't a solution, GuessWho.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline GuessWho

  • Member
  • Posts: 94
  • Liked: 8
  • Likes Given: 0
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #31 on: 11/20/2013 12:43 PM »
Problematic doesn't mean there isn't a solution, GuessWho.

Agreed.  Your point?

Offline jongoff

  • Recovering Rocket Plumber/Space Entrepreneur
  • Senior Member
  • *****
  • Posts: 5989
  • Lafayette/Broomfield, CO
  • Liked: 1938
  • Likes Given: 667
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #32 on: 11/20/2013 02:58 PM »
The issue I have is doing the refueling on spacecraft that were not design for it.  Look at all the times NASA serviced a spacecraft that was not design for it, there were always dimensional problems and the crew had to do workarounds.   

SMM - MMU/TPAD would not engage.  Ended up using RMS to grab spacecraft
HS 376 retrieval - Handling frame would not engage, crew had to hold on to spacecraft by omni antenna
Intelsat VI reboost - capture frame would not engage, 3 EVA crewmembers had to grab  spacecraft.

Agreed, but in the case of refueling, most of the US-made satellites have valves that come from one or two manufacturers (VACCO is one of them), and they know *exactly* what valve is on what craft. And Goddard has demonstrated tools that can accommodate differences. And you're talking to a guy who's working explicitly in the non-cooperative capture world (though we're nowhere near the experts yet). From what I've seen from work DARPA/NRL, GSFC, ATK, and MDA are doing, I think they've seen the same past problems you have, and have been working to make sure they aren't repeated.

~Jon

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 30391
  • Cape Canaveral Spaceport
  • Liked: 8683
  • Likes Given: 284
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #33 on: 11/20/2013 03:14 PM »
I have seen access problems during loading on the ground.

Offline a_langwich

  • Full Member
  • ****
  • Posts: 735
  • Liked: 211
  • Likes Given: 48
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #34 on: 11/21/2013 06:25 AM »

Wait, what?  What, in those numbers, actually represents the cost of refueling a satellite?  None of it.  That's what Intelsat was willing to pay, which of course is driven by their revenue numbers and alternative costs.   


And what MDA set the price at which they were willing to contractually sell the service for.  That price of course is driven by their business case which includes all of the costs you mentioned; i.e., what it costs to design and build the refueling satellite, the cost to launch the equipment, the amount of fuel it needs to reach client satellite, the fuel it uses to perform ARPO and capture,  the amount of fuel it plans to load onto the disabled satellite, the ground infrastructure costs to execute the mission, the insurance costs, taxes, interest, etc. such that at the end of the day, they actually realize a profit.


This gets back to what Jim said about not being validated.  Unless you happened to work the numbers at MDA for the proposal, you can't assume that price included a profit or break-even.  And it sounds like, even there, they HAD to have five other servicings to make the business case close.  Do you know how much revenue they depended on the second five to bring in?  Maybe they gave an extra-super-sweetheart deal to Intelsat, hoping a reputable anchor customer would bring in others.  Maybe they were willing to eat the development cost or even some of the build and launch costs, if the market could be shown to be large enough to recoup costs later.  See, for example, SpaceX's initial launches on Falcon 1, 9, and 9 v1.1.

You keep bringing up this firm contract but it must have had some large holes that allowed them to walk away without executing it. 

This raises again the question of deltaV and/or market size--they found five in a plane, maybe, or energetically nearby, but didn't find ten.  You say, "See:  it was easy.  They found five."  I say, "They didn't find ten, which apparently they had to have to purchase a launch."  Maybe they exhausted all the possibilities within reach?  Or maybe they realized they weren't going to get anywhere near their cost target for the vehicle?  Or, maybe they realized the market demand was not enough to support not just their pricing but any reasonable price, at the present.  Bottom line, you can't say it works at this price, when something clearly did NOT work.

Quote from: a_langwich
I'm not arguing for orders of magnitude difference in price, but in order for it to close right now, I suspect you'd need at least a half-dozen EOL satellites, with not too much deltaV separation between orbits, that can be refueled from one launch of your equipment.  And it better happen reasonably quickly, too, because your debt costs money and your personnel can't wait a decade before a payout.


Jim did argue for "orders of magnitude difference" and seems unwilling to retract that statement.  But that is not your point so let me respond.  Intelsat and MDA signed a contract for 5 satellites (plus a sixth to be used as the initial demonstration of the refuel capability) so yes, you need a half-dozen or so clients signed up to even consider executing a business case.  Rephasing from one sat to the next is not that delta-V intensive.  It depends on how far apart they are on the belt and how quickly you want to perform the phasing maneuver.  Given MDA was using an EP prop system to perform the phasing maneuvers, I suspect the timing was slow and the fuel use minimal.  Also, if you read up on the MDA approach, they were designed to refuel twice the number of sats that Intelsat actually settled for.  So presumably, MDA could refuel 10 satellites with a single servicer.  And who is to say that they couldn't then refuel themselves and continue operating for several years?

And from what I have gathered from several MDA presentations, they were expecting to launch within 3 years of signing the contract.  Problem is, they couldn't sell the other 5 refuel missions they had capacity for so didn't execute the contract.  Still doesn't alter the price that both the seller and the buyer were willing to pay for a refuel service which still isn't "orders of magnitude" greater than simply graveyarding a satellite at EOL.

If they didn't execute the contract, then they clearly couldn't make it on just the 5+1 sold to Intelsat.  So it sounds like 10 was the magic number.

And yes, of course they could refuel and keep going.  For the cost of another launch.  And yes, they could launch as a secondary.  For the cost of designing another vehicle that has the deltaV and guidance to get the fuel from wherever they get dropped as a secondary to their refueling spacecraft's location.  And both vehicles have to support a rendezvous and fuel transfer.  Refueling is not going to be cheap and easy, it is going to be a very substantial fraction of the cost of the original mission, where crunching the actual numbers would tell you whether it is worthwhile.  (What costs of the original refueling mission are avoided by a refueling mission to the refueler?)

All of this, oddly, ignores the risk of not being able to get it working on the first go, which is substantial I think.  That is, for most "next greatest thing" ideas, IF you stipulate that you can get it to work, the capability is a guaranteed money-maker.  For this idea, we've been stipulating that it works, and the business case is STILL arguable.  That ain't good.

Wandering off-topic...the Iridium satellites, and I suppose other relay satellite systems, have a fair amount of satellite-to-satellite communication built in.  If you beefed this up, I wonder if you could build multiple satellites that could coordinate spot beams and operate in the same orbital slot?  They would need not only to communicate with each other, but have the capacity to maintain position relative to each other.  Talk about redundancy.  EMC might be an issue, though.

Offline GuessWho

  • Member
  • Posts: 94
  • Liked: 8
  • Likes Given: 0
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #35 on: 11/22/2013 02:38 PM »
First, let me say these are good questions.  I will attempt to respond but quite frankly we are quickly reaching the point where this is not the right forum to discuss as the subject matter is very wide and very deep and one or two paragraph statements can't begin to address the issues involved.


This gets back to what Jim said about not being validated.  Unless you happened to work the numbers at MDA for the proposal, you can't assume that price included a profit or break-even.  And it sounds like, even there, they HAD to have five other servicings to make the business case close.  Do you know how much revenue they depended on the second five to bring in?


Clearly only someone from MDA could answer the specifics of their business case and I doubt they would do that in a public forum.  First off, I am not associated with MDA so I can't speak to their business case.  That said, I have worked a similar business case for a client in the not so distant past.  That business case did close with a respectable profit after an acceptable period of time given the client's overall risk posture.  I wouldn't assume that MDA HAD to have the other five missions to close their business case.  It may have broke even at 5, yielded a slight profit after 7 but not enough to convince their Board that it was worth the risk, and was very attractive at 10 missions.  It comes down to their investment risk posture, aggressive for income growth or more conservative to ensure steady, more predictable returns to their shareholders.

Quote

You keep bringing up this firm contract but it must have had some large holes that allowed them to walk away without executing it. 

This raises again the question of deltaV and/or market size--they found five in a plane, maybe, or energetically nearby, but didn't find ten.  You say, "See:  it was easy.  They found five."  I say, "They didn't find ten, which apparently they had to have to purchase a launch." 


Intelsat would pay upon services rendered so there is likely no penalty if MDA had to cancel early in the project.  Once Intelsat started to make financial commitments (i.e. cap-ex related) based upon a reasonable expectation of services being rendered, there would be penalties put in place.  But that is all standard contract stuff that would have been negotiated upfront so I don't see that as an issue.

Intelsat has 50+ satellites at GEO and they are only one of several comsat owner/operators.  So there are a significant number of potential clients in GEO.  And energetically, it is only a few m/s to a few 10's of m/s to get to just about any GEO satellite up there unless they are significantly inclined so delta-V is not the issue at GEO.  What is most significant about the MDA/Intelsat deal is that for the first time a buyer and a seller agreed upon what the financial value of a refuel mission was.  Prior to that, anyone contemplating a on-orbit service offer could only speculate about what a potential buyer might pay and base their business case on that assumption.  The fact that Intelsat stepped up and put a value to a kg of fuel or an additional year of operation is huge.  It set a millstone about which all other offers could assess their market assumptions.  For MDA, apparently a $11-$12M/year of added life was sufficient to close their business case.  From my own work, I actually had that type of service valued about 25% lower given the level of conservatism I injected while closing the business case for my client.  That likely means my other assumptions across the board are also one the low side relative to market value and thus makes the business case more attractive.  Conversely, if someone is out there thinking the value of that service was closer to $30M/yr, they need to go back and rethink their approach as it is unlikely that the market will bear their price point.

Quote
Maybe they exhausted all the possibilities within reach?  Or maybe they realized they weren't going to get anywhere near their cost target for the vehicle?  Or, maybe they realized the market demand was not enough to support not just their pricing but any reasonable price, at the present.  Bottom line, you can't say it works at this price, when something clearly did NOT work.


And you can't sit back and speculate about the viability of someone else's business case without having taken the time to actually run one to ground.  That is my issue with Jim's off the cuff and frankly ill-informed comment.  I performed due diligence for my client for two years, building the business case, surveying the market to understand size, value, timing, risks, legal and policy requirements and hurdles, etc.  All these factors go into the business case assessment.  That is weighed against the financial goals of the client, their risk posture, their business climate, their shareholder's expectations, etc.  For my client, the business case was attractive.  It was not the financial side of the equation, but the complete lack of coherent US policy regarding operations within space and the regulatory risks that would flow from that led to the decision not to execute the business plan.

Quote
All of this, oddly, ignores the risk of not being able to get it working on the first go, which is substantial I think.


That is what insurance is for.  You insure the loss of vehicle and you insure the loss of potential revenue, up to a point.  It all comes down to what you can negotiate with your insurance underwriters (at a premium structure that works for both parties) who in turn have to be convinced about the viability of your overall business case.  MDA had a signed insurance group on board so they must have convinced them that their business case had merit.  My client did as well.  Bigger issue is third-party liability.  There is absolutely no US or international law that even begins to address that subject when it comes to space servicing or even on-orbit debris mitigation.  That "unknown" dwarfs all other considerations.

Offline watermod

  • Full Member
  • ***
  • Posts: 335
  • Liked: 83
  • Likes Given: 94
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #36 on: 11/23/2013 03:21 AM »
At the bottom of page 2 of the "Economics of Repurposing in situ retired spacecraft components" paper assembly in space to build very large apertures for RF and optical systems is mentioned. This relates to something I had been wondering about lately. Taken to the extreme, might three huge antennas with appropriate electronics, built in GEO, be able to replace cell phone towers by allowing 2 way comm with existing cellphones directly?
 That would be what - a trillion dollar business?

CELLULAR:
Xerox Parc put it best in the late 80s and early 90s

     bandwidth / the population of earth = bits/sec per person

There are some 7 billion people on this world so that Tesla style solution doesn't work.
    with your 3 antennas canted to service 3 spherical triangles from on high and maybe covering 1/3 of the earth the best you get is:
    1 cell size of 3 (antennas or sectors) * 3 sats to cover the earth *bandwidth/7 billion people

You want lots of cells... micro-cells in urban areas larger narrow long cells on transit routes (so you don't hand off so much - esp with things like high speed trains racing through cells... so you have long narrow cells on the tracks enabled with leaky coaxial cable the full length of the tracks)  You want multi-spectrum, you want robustness in handoffs (hence the CDMA scheme with its soft and softer handoffs).   

Since Xerox Parc didn't happen but was schematically the cleanest concept we can discuss their drivers without ruining functional IP in use today.  Xerox Parc wanted to do it with a staged system with cheap cellphones.  Theirs were not RF but IR.   They envisioned fiber going street light to street light with IR LEDS and to/from IR cells and lasers or tight beam IR to the house with IR base stations all over inside your home or business.  Your car would have had IR to your phone connected to a radio based cellphone in the car.  The reason for that is if you had to hand off to each IR LED base station every 30 feet or so at auto speeds you would spend all your bandwidth handing off so you need to drop down in frequency to radio waves and take advantage of their greater cell size.  This would allow using cells for rural and transit routes.  Note how cheap this made cellphones.   A basic one only needed the LED/Photo Transistor in your TV remote and electronic of similar complexity.   They envisioned early 1990's costs of 50 cents a phone.     Obviously we went a much more complex direction but the underlying system is trying to enable the same solution to the formula by making cell size very small

   (time multiplicative factor {not allowing each user total time} * maximized cell count *  maximize sectors per cells * max frequency per cell )/ (Population of earth  * devices per individual )

It becomes obvious you don't want to go up to a few satellites except, possibly, for inter-regional backhaul in rural or remote areas.
 
The next level up from doing it Xerox's way and retro fit what we had would have been to have micro cell sites in every cable tv box or DSL modem.   That would have made the governments (all levels) in the USA very unhappy as they earn huge revenues on the large tariffs on backbone lines like T1s, E1s and higher speed currently required to service cell sites.    The IR light is unlicensed spectrum which would deprive governments of the billions of dollars received selling licensed spectrum.   

The sneaker way to do it would be to hand off to WiFi when available.   The Telcos and Government have revenue reasons not to make this a feature. 

Cellular is the way it is currently for political reasons and at the beginning for the desire of manufacturers to sell Big-Iron multi-billion dollar infrastructure to Telcos.   

Iridium and it's ilk have their niche as they have a multi-digit N sats and sectorization and time slots and large spectrum but it can never be a cellular system for the masses just one to serve remote areas and elites.





   

Offline ChrisWilson68

  • Senior Member
  • *****
  • Posts: 3353
  • Sunnyvale, CA
  • Liked: 1938
  • Likes Given: 2207
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #37 on: 11/23/2013 04:42 AM »
At the bottom of page 2 of the "Economics of Repurposing in situ retired spacecraft components" paper assembly in space to build very large apertures for RF and optical systems is mentioned. This relates to something I had been wondering about lately. Taken to the extreme, might three huge antennas with appropriate electronics, built in GEO, be able to replace cell phone towers by allowing 2 way comm with existing cellphones directly?
 That would be what - a trillion dollar business?

CELLULAR:
Xerox Parc put it best in the late 80s and early 90s

     bandwidth / the population of earth = bits/sec per person

There are some 7 billion people on this world so that Tesla style solution doesn't work.
    with your 3 antennas canted to service 3 spherical triangles from on high and maybe covering 1/3 of the earth the best you get is:
    1 cell size of 3 (antennas or sectors) * 3 sats to cover the earth *bandwidth/7 billion people

You want lots of cells... micro-cells in urban areas larger narrow long cells on transit routes (so you don't hand off so much - esp with things like high speed trains racing through cells... so you have long narrow cells on the tracks enabled with leaky coaxial cable the full length of the tracks)  You want multi-spectrum, you want robustness in handoffs (hence the CDMA scheme with its soft and softer handoffs).

I agree with you on this part.  Lots of cells means lower transmit power (longer battery life) and less interference between devices.

Since Xerox Parc didn't happen but was schematically the cleanest concept we can discuss their drivers without ruining functional IP in use today.  Xerox Parc wanted to do it with a staged system with cheap cellphones.  Theirs were not RF but IR.   They envisioned fiber going street light to street light with IR LEDS and to/from IR cells and lasers or tight beam IR to the house with IR base stations all over inside your home or business.  Your car would have had IR to your phone connected to a radio based cellphone in the car.  The reason for that is if you had to hand off to each IR LED base station every 30 feet or so at auto speeds you would spend all your bandwidth handing off so you need to drop down in frequency to radio waves and take advantage of their greater cell size.  This would allow using cells for rural and transit routes.  Note how cheap this made cellphones.   A basic one only needed the LED/Photo Transistor in your TV remote and electronic of similar complexity.   They envisioned early 1990's costs of 50 cents a phone.     Obviously we went a much more complex direction but the underlying system is trying to enable the same solution to the formula by making cell size very small

Thank goodness we didn't go with this IR system!  There are many ways in which it's inferior to the system we have.  IR is line-of-sight, so you need base stations everywhere indoors to make it work.  The phones would be dirt cheap, but the infrastructure astronomically expensive to cover every part of every room.  And then what happens if you put your phone in your pocket?  You can't receive a call (or a text or an e-mail message).  The same thing happens if the room is crowded and someone happens to be between you and each of the base stations in the room.

   (time multiplicative factor {not allowing each user total time} * maximized cell count *  maximize sectors per cells * max frequency per cell )/ (Population of earth  * devices per individual )

It becomes obvious you don't want to go up to a few satellites except, possibly, for inter-regional backhaul in rural or remote areas.
 
The next level up from doing it Xerox's way and retro fit what we had would have been to have micro cell sites in every cable tv box or DSL modem.   That would have made the governments (all levels) in the USA very unhappy as they earn huge revenues on the large tariffs on backbone lines like T1s, E1s and higher speed currently required to service cell sites.

Sorry, but that's nonsense.  Voice and data traffic through cable modems and DSL modems goes through exactly the same backbone lines.

Anyway, there's no reason to believe the people making the telecom regulations care at all about how much revenue the government gets from it.  It doesn't affect how much money their organizations get, it goes to general Treasury accounts.  If Congress cared, they could as easily assess fees on IR systems, cable modems, whatever they wanted to.

The carriers are the only ones who would care, and only to the extent it derives them of revenue.  So the carriers could conceivably exert political pressure to get the government to tilt the rules against femptocells.  But, there's no evidence they did.  The government wasn't really involved one way or another.  The carriers were slow to adopt it, and it wouldn't work on people's cell phones unless the carriers agreed, so the carriers made it more difficult.  But recently some carriers have been moving in the direction of supporting femptocells.

The IR light is unlicensed spectrum which would deprive governments of the billions of dollars received selling licensed spectrum.

First of all, the government didn't do anything to discourage IR systems.  Anyone who wanted to could have built them.  If anything, the government favored IR systems by charging carriers lots for radio spectrum but leaving IR unlicensed.

Secondly, the government getting revenue from the RF spectrum is not a motivation for the government to promote RF over IR because the government could simply decide to start licensing the IR systems and charging the IR system carriers however much the government wanted.

The sneaker way to do it would be to hand off to WiFi when available.   The Telcos and Government have revenue reasons not to make this a feature. 

Again, only the carriers (telco companies) have any incentive not to enable WiFi hand-off, not the government.  And the carriers, while initially opposing WiFi hand-off have started moving in that direction recently, since they've realized most of their customers are on unlimited plans these days, so they don't get more money per minute of celluar use, so handing them off to WiFi doesn't decrease revenue at all but instead reduces costs because fewer calls go through the carrier's cellular network.

Cellular is the way it is currently for political reasons and at the beginning for the desire of manufacturers to sell Big-Iron multi-billion dollar infrastructure to Telcos.

I think it's more accurate to say cellular is the way it is largely because it's more efficient than most of the alternatives, though, as you correctly point out, more use of WiFi hand-off and femptocells would improve efficiency even more -- and the industry has been moving in that direction recently.

Iridium and it's ilk have their niche as they have a multi-digit N sats and sectorization and time slots and large spectrum but it can never be a cellular system for the masses just one to serve remote areas and elites.

Offline ChrisWilson68

  • Senior Member
  • *****
  • Posts: 3353
  • Sunnyvale, CA
  • Liked: 1938
  • Likes Given: 2207
Re: DARPA Phoenix Satellite Service and Salvage program
« Reply #38 on: 11/23/2013 04:57 AM »
The next level up from doing it Xerox's way and retro fit what we had would have been to have micro cell sites in every cable tv box or DSL modem.   That would have made the governments (all levels) in the USA very unhappy as they earn huge revenues on the large tariffs on backbone lines like T1s, E1s and higher speed currently required to service cell sites.

Sorry, but that's nonsense.  Voice and data traffic through cable modems and DSL modems goes through exactly the same backbone lines.

Anyway, there's no reason to believe the people making the telecom regulations care at all about how much revenue the government gets from it.  It doesn't affect how much money their organizations get, it goes to general Treasury accounts.  If Congress cared, they could as easily assess fees on IR systems, cable modems, whatever they wanted to.

The carriers are the only ones who would care, and only to the extent it derives them of revenue.  So the carriers could conceivably exert political pressure to get the government to tilt the rules against femptocells.  But, there's no evidence they did.  The government wasn't really involved one way or another.  The carriers were slow to adopt it, and it wouldn't work on people's cell phones unless the carriers agreed, so the carriers made it more difficult.  But recently some carriers have been moving in the direction of supporting femptocells.



The sneaker way to do it would be to hand off to WiFi when available.   The Telcos and Government have revenue reasons not to make this a feature. 

Again, only the carriers (telco companies) have any incentive not to enable WiFi hand-off, not the government.  And the carriers, while initially opposing WiFi hand-off have started moving in that direction recently, since they've realized most of their customers are on unlimited plans these days, so they don't get more money per minute of celluar use, so handing them off to WiFi doesn't decrease revenue at all but instead reduces costs because fewer calls go through the carrier's cellular network.

By the way, it's worth noting that both WiFi and femptocells have huge technical challenges in terms of quality of service.  If you've used both Skype over WiFi and cell phones, you will have noticed Skype isn't nearly as reliable as cell phones -- sometimes the audio cuts out and sometimes there's a dropped call.  In years past, Skype and similar home VoIP services were much worse.  That's all because cell phone networks were built from the ground up to have hard real-time constraints on data transmission through every part of the network.  Home routers, WiFi, cable modems, etc. were not originally built for that, so packets can be delayed or dropped, particularly when there's other traffic.  In recent years, more of those other components have started supporting prioritization of certain traffic, and overall bandwidth and latency have improved enough that most of the time voice calling is fine.  But cellular companies have had a legitimate reason to worry about customer complaints about poor service because of things like old DSL modems and cheap or mis-configured home WiFi routers that are out of the control of the carriers.

I have a friend with a PhD in electrical engineering who tried to use a femptocell provided by his cellular carrier on his home network.  He was only able to get it to work erratically, in spite of a long time on the phone with the carrier's tech support and a lot of knowledge of networking.

Offline a_langwich

  • Full Member
  • ****
  • Posts: 735
  • Liked: 211
  • Likes Given: 48

Tags: