Author Topic: SpaceX FCC filing for a 4425 satellite constellation providing Internet service  (Read 62820 times)

Online Robotbeat

  • Senior Member
  • *****
  • Posts: 26624
  • Minnesota
  • Liked: 6493
  • Likes Given: 4705
It's not just bandwidth that determines bitrate but also the signal to noise ratio. So with higher gain, you can get higher bitrate.
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 MikeAtkinson

  • Full Member
  • ****
  • Posts: 1621
  • Bracknell, England
  • Liked: 391
  • Likes Given: 50
It is ~20GBit/s per satellite so with 25 satellites over the USA then that is 162,000,000 users on 100Gbyte/month plans. Say at $10/month that is almost $20B/year.

I know that is almost certainly not how it will be priced, and there are inefficiencies that mean that 162M users are not likely to be accommodated, but this shows that a vast amount of revenue is possible.
« Last Edit: 11/17/2016 04:52 PM by MikeAtkinson »

Offline SWGlassPit

  • I break space hardware
  • Full Member
  • ****
  • Posts: 432
  • Liked: 248
  • Likes Given: 35
Reposting this great data dump:

Also twitter from @malderi

"Musk: Talking about around 4000 satellites. 4025 exactly in current design."
- that is probably 100s of launches, maybe 50/year.

"Musk: Talking mostly around 1100km level. Space debris not much of a problem there."
- with that number of satellites, they will have to take great care to dispose of them at end of life.

Bolding is mine

Lots of features from two years ago bang on.

What?!  No.  Elon is flat out wrong here.  See the graphs on page 17 and 18 (pdf pages 18 and 19) of this presentation.

1100 isn't as bad as 800 km or 1500 km, but it's still bad.  Adding 4,400 satellites to that altitude range in 83 different orbital planes will make it worse.  Furthermore, any debris generated there will be there ... well ... basically forever.
« Last Edit: 11/17/2016 04:56 PM by SWGlassPit »

Offline GreenShrike

  • Full Member
  • *
  • Posts: 128
  • Liked: 108
  • Likes Given: 391
The current F9 fairing has a 4.6m internal diameter, that starts to taper off at 6.7m with a 70deg angle.
If the satellites are 4m x 1.8m x 1.2m at launch, they are lucky if they can put four in a ring. At 4m long, they can't put two rings in the current fairing. Since the solar panels are 6m x 2m each, I would tend to think that the first dimensions are main body.

That 4 x 1.8 x 1.2 m set of dimensions is for calculating orbital decay characteristics. Since this is a communications satellite, it will have antennas that fold out, which are included in the "Satellite Body" dimensions for the orbital decay model. So the actual satellite bus in a stowed away launch configuration should be significantly smaller than 4 m in that one dimension, bringing the launch dimensions a lot closer to 1.5 x 1.8 x 1.2 m or something like that (we aren't given the completely folded up dimensions).

So if SpaceX can put 4 in a ring, in 6.7m they can fit four 1.5m rings, plus spacing between rings, for 16 sats. Maybe one or two more in the inclined section to bring the total to the 17-18 sats an F9 can lift to 1150km and RTLS.

FH uses the same fairing, and if the fairing is already full increasing your lift isn't helpful.

Somehow I doubt SpaceX wont' be using the FH. ;-)

If they doubled the fairing length, they would fit 32-36 sats. But is doubling the fairing's length reasonable?

Also, if fairing volume limits mean FH needs two launches to fill a plane anyway, 26 sats is almost bang-on 10t. Would that leave sufficient performance to recover S2? If F9 RTLS can lift 8.5t to 1150km, FH should double that, which would leave 6t or so of performance -- possibly enough for recovery equipement and a deorbit burn?

80 orbital planes to fill at two launches per is 160 flights to get the constellation completely filled. If a Falcon second stage is $10M, that's $1.6B just in S2s. And that's assuming F9 isn't used at all.

I think the prospect of saving some fraction of $1.6B on launching their own comsats may justify developing a reusable S2, even if current GTO launches do not.

And if the current fairing is a few million, how much would a super-long fairing cost? At ~$6M each, that's another ~$1B in expenses.

So $2B+ in expendable S2s and fairings -- is reducing that possibly enough to justify a recoverable orbital stage with an integrated fairing/dispenser? Say, a 5m Raptor-powered one?


Considering the magnitude of the number of launches, it's easy to see why SpaceX is serious about the need to drive the costs of launch down.  At say 200 launches to fill out and replenish the constellation for a few years, the difference between $10M per launch and $40M per launch is $6B.

It's easy to say SpaceX will have its hands full developing ITS and that the Falcon series should just be flown out as it currently is, but that would leave a hell of a lot of money on the table -- money which SpaceX will need to develop ITS. They need to go fully-reusable to get this constellation into space; replacing the current Falcon S2 seems the quickest and cheapest way of doing so.

I wonder if that 12m composite test tank has a 5m cousin in the room next door.
TriOptimum Corporation
Military - Science - Consumer

Offline guckyfan

  • Senior Member
  • *****
  • Posts: 6249
  • Germany
  • Liked: 1547
  • Likes Given: 1329
There will be active deorbit according to the application. Sats will be deorbited in one year after they go out of service. The deorbiting function is fully redundant, so it would not be affected by one failure.

Quite a satellite rain coming. 880 up and 880 down a year, assuming a 5 year life.

Offline oldAtlas_Eguy

  • Senior Member
  • *****
  • Posts: 2922
  • Florida
  • Liked: 1247
  • Likes Given: 85
It is ~20GBit/s per satellite so with 25 satellites over the USA then that is 162,000,000 users on 100Gbyte/month plans. Say at $10/month that is almost $20B/year.

I know that is almost certainly not how it will be priced, and there are inefficiencies that mean that 162M users are not likely to be accommodated, but this shows that a vast amount of revenue is possible.
90% to backbone and only 10% to direct to household (DTH). But even 10% DTH is by your figures $2B/yr just in the US. For the Globe much more + the revenue from the backbone links.

The probable deployment looks to be 12 sats per F9 at ~20 per year (out of just SLC-40) for initial 800 would take just less than 4 years to populate to operational. Launches start in 2019 and operational in 2023 or earlier. Cost is ~$5B for first 800 mostly launch costs.

Later launches possibly using an ITS that could deploy ~104 sats/launch at $.6-1M per sat would cost $12B to replace the first 800 and to also fully populate the constellation at 50-80 launches/yr would take ~ the lifetime of the sats meaning launches would continue at that rate forever. So a continuous expenditure rate for the sat launch, manufacture, etc of ~$2-3B/yr. Full constellation existence in ~2028.

For Mars the loaded ITS goes to Mars and lands. Refueled goes to LMO, deploys the sats and then lands again to be refueled for its return trip to Earth. The ITS that does this and all its adaptations is exactly the same as the one used to deploy the sats at Earth.

Mars missions would be just at first a small ripple in that launch rate.

Offline M.E.T.

  • Full Member
  • ****
  • Posts: 403
  • Liked: 151
  • Likes Given: 12
Reposting this great data dump:

Also twitter from @malderi

"Musk: Talking about around 4000 satellites. 4025 exactly in current design."
- that is probably 100s of launches, maybe 50/year.

"Musk: Talking mostly around 1100km level. Space debris not much of a problem there."
- with that number of satellites, they will have to take great care to dispose of them at end of life.

Bolding is mine

Lots of features from two years ago bang on.

What?!  No.  Elon is flat out wrong here.  See the graphs on page 17 and 18 (pdf pages 18 and 19) of this presentation.

1100 isn't as bad as 800 km or 1500 km, but it's still bad.  Adding 4,400 satellites to that altitude range in 83 different orbital planes will make it worse.  Furthermore, any debris generated there will be there ... well ... basically forever.

I'm wondering whether SpaceX, through this venture, will not be creating another market for themselves - namely orbital debris clearance. Something for which demand is only going to be growing as space grows ever more crowded.

Online TrevorMonty

Besides high speed between end users, there is high security. The only weak points are ease dropping on microwave link downlink at customers end terminals. Then you only get half conversation. Optical links between satellites are good as it gets for security. Add rolling encryption codes sent via different route eg internet, to make it even more secure.

Offline SWGlassPit

  • I break space hardware
  • Full Member
  • ****
  • Posts: 432
  • Liked: 248
  • Likes Given: 35
Reposting this great data dump:

Also twitter from @malderi

"Musk: Talking about around 4000 satellites. 4025 exactly in current design."
- that is probably 100s of launches, maybe 50/year.

"Musk: Talking mostly around 1100km level. Space debris not much of a problem there."
- with that number of satellites, they will have to take great care to dispose of them at end of life.

Bolding is mine

Lots of features from two years ago bang on.

What?!  No.  Elon is flat out wrong here.  See the graphs on page 17 and 18 (pdf pages 18 and 19) of this presentation.

1100 isn't as bad as 800 km or 1500 km, but it's still bad.  Adding 4,400 satellites to that altitude range in 83 different orbital planes will make it worse.  Furthermore, any debris generated there will be there ... well ... basically forever.

I'm wondering whether SpaceX, through this venture, will not be creating another market for themselves - namely orbital debris clearance. Something for which demand is only going to be growing as space grows ever more crowded.

At typical encounter velocities upwards of 10 km/s, there's a technology problem.  Furthermore, the overwhelming majority of debris is far too small to track or even observe until you get hit by it.  Folks talk about orbital debris removal as though the problem is all big chunks of derelict satellites.  It's not.  It's paint flakes, bolt heads, coolant droplets, and generally small bits of detritus.  For every piece of debris you can track, there are thousands you can't even see, and every collision creates more tiny debris.

Offline SWGlassPit

  • I break space hardware
  • Full Member
  • ****
  • Posts: 432
  • Liked: 248
  • Likes Given: 35
There will be active deorbit according to the application. Sats will be deorbited in one year after they go out of service. The deorbiting function is fully redundant, so it would not be affected by one failure.

Quite a satellite rain coming. 880 up and 880 down a year, assuming a 5 year life.

They will have their perigee brought low enough that they will "self-deorbit" within a year.  That's a year of dipping in and out of other populated orbits in a more or less uncontrolled fashion.

Offline M.E.T.

  • Full Member
  • ****
  • Posts: 403
  • Liked: 151
  • Likes Given: 12
Reposting this great data dump:

Also twitter from @malderi

"Musk: Talking about around 4000 satellites. 4025 exactly in current design."
- that is probably 100s of launches, maybe 50/year.

"Musk: Talking mostly around 1100km level. Space debris not much of a problem there."
- with that number of satellites, they will have to take great care to dispose of them at end of life.

Bolding is mine

Lots of features from two years ago bang on.

What?!  No.  Elon is flat out wrong here.  See the graphs on page 17 and 18 (pdf pages 18 and 19) of this presentation.

1100 isn't as bad as 800 km or 1500 km, but it's still bad.  Adding 4,400 satellites to that altitude range in 83 different orbital planes will make it worse.  Furthermore, any debris generated there will be there ... well ... basically forever.

I'm wondering whether SpaceX, through this venture, will not be creating another market for themselves - namely orbital debris clearance. Something for which demand is only going to be growing as space grows ever more crowded.

At typical encounter velocities upwards of 10 km/s, there's a technology problem.  Furthermore, the overwhelming majority of debris is far too small to track or even observe until you get hit by it.  Folks talk about orbital debris removal as though the problem is all big chunks of derelict satellites.  It's not.  It's paint flakes, bolt heads, coolant droplets, and generally small bits of detritus.  For every piece of debris you can track, there are thousands you can't even see, and every collision creates more tiny debris.

That's kind of what I was alluding to. Some kind of technological, space based innovation to sweep areas of space clean of tiny pieces of dangerous debris.

Offline cscott

  • Senior Member
  • *****
  • Posts: 2455
  • Liked: 1682
  • Likes Given: 656
Besides high speed between end users, there is high security. The only weak points are ease dropping on microwave link downlink at customers end terminals. Then you only get half conversation. Optical links between satellites are good as it gets for security. Add rolling encryption codes sent via different route eg internet, to make it even more secure.
There are sigint satellites already in orbit that sit next to comm sats and eavesdrop on stray emissions.  Optical comms aren't a panacea: the beam size at the distances involved could well be big enough to accommodate an extra listener close up, and that's before you start thinking about stray specular reflections or even deliberately introduced partial reflectors.

I agree this would increase the sigint challenge significantly compared to GSO sats, but I wouldn't jump to the conclusion that highly motivated state actors won't be able to compromise the flow, at least in parts.

The biggest issue is that Elon is pushing the state of the art so hard that state agencies would have to run to keep up.  What would the launch rate for Intel sats be like if the commercial market starts launching 880 sats/year?
« Last Edit: 11/17/2016 05:59 PM by cscott »

Offline SWGlassPit

  • I break space hardware
  • Full Member
  • ****
  • Posts: 432
  • Liked: 248
  • Likes Given: 35

At typical encounter velocities upwards of 10 km/s, there's a technology problem.  Furthermore, the overwhelming majority of debris is far too small to track or even observe until you get hit by it.  Folks talk about orbital debris removal as though the problem is all big chunks of derelict satellites.  It's not.  It's paint flakes, bolt heads, coolant droplets, and generally small bits of detritus.  For every piece of debris you can track, there are thousands you can't even see, and every collision creates more tiny debris.

That's kind of what I was alluding to. Some kind of technological, space based innovation to sweep areas of space clean of tiny pieces of dangerous debris.


If I may analogize: imagine a nineteenth-century era battlefield -- a largely empty field, with bullets zinging across in multiple directions.

You're looking for an innovation that would allow someone to somehow make this random, omnidirectional hail of bullets vanish into thin air -- more importantly, making them vanish without fragmenting into even more bullets.  All without being able to even see them coming in.

To the extent that it's even technologically possible, it's an absolutely ludicrous business proposition until space gets to the point that it is literally unusable.

Offline oldAtlas_Eguy

  • Senior Member
  • *****
  • Posts: 2922
  • Florida
  • Liked: 1247
  • Likes Given: 85
Besides high speed between end users, there is high security. The only weak points are ease dropping on microwave link downlink at customers end terminals. Then you only get half conversation. Optical links between satellites are good as it gets for security. Add rolling encryption codes sent via different route eg internet, to make it even more secure.
There are sigint satellites already in orbit that sit next to comm sats and eavesdrop on stray emissions.  Optical comms aren't a panacea: the beam size at the distances involved could well be big enough to accommodate an extra listener close up, and that's before you start thinking about stray specular reflections or even deliberately introduced partial reflectors.

I agree this would increase the sigint challenge significantly compared to GSO sats, but I wouldn't jump to the conclusion that highly motivated state actors won't be able to compromise the flow, at least in parts.

The biggest issue is that Elon is pushing the state of the art so hard that state agencies would have to run to keep up.  What would the launch rate for Intel sats be like if the commercial market starts launching 880 sats/year?

The pure raw data rate is so high that in-orbit listeners would need to be nearly the equivalent of the comm system itself. Else they would need very large computational/packet sifting capabilities to just intercept the probable packets of interest. But if the data is also encrypted with a 1024 or even higher level of encryption the on-orbit sat would only have the source and destination IP addresses to trigger off of. So data flows of target data in the multi Tbps would require a significant comm system and a radiation footprint that would identify all of the listeners positions. A smart security routing protocol could literally rout around the listeners.

« Last Edit: 11/17/2016 06:43 PM by oldAtlas_Eguy »

Offline MikeAtkinson

  • Full Member
  • ****
  • Posts: 1621
  • Bracknell, England
  • Liked: 391
  • Likes Given: 50
It is ~20GBit/s per satellite so with 25 satellites over the USA then that is 162,000,000 users on 100Gbyte/month plans. Say at $10/month that is almost $20B/year.

I know that is almost certainly not how it will be priced, and there are inefficiencies that mean that 162M users are not likely to be accommodated, but this shows that a vast amount of revenue is possible.
90% to backbone and only 10% to direct to household (DTH). But even 10% DTH is by your figures $2B/yr just in the US. For the Globe much more + the revenue from the backbone links.
Good point.


The probable deployment looks to be 12 sats per F9 at ~20 per year (out of just SLC-40) for initial 800 would take just less than 4 years to populate to operational. Launches start in 2019 and operational in 2023 or earlier. Cost is ~$5B for first 800 mostly launch costs.

Given 52 satellites per plane obvious numbers are 13, 18 and 26. It looks like F9 may be able to accommodate 26 sats within its fairing, the satellites are probably folded down to something like 1.5 x 1.8 x 1.2. Mass-wise 10.4 tonnes probably enables a RTLS.
 
Later launches possibly using an ITS that could deploy ~104 sats/launch at $.6-1M per sat would cost $12B to replace the first 800 and to also fully populate the constellation at 50-80 launches/yr would take ~ the lifetime of the sats meaning launches would continue at that rate forever. So a continuous expenditure rate for the sat launch, manufacture, etc of ~$2-3B/yr. Full constellation existence in ~2028.

ITS can probably launch about 500 satellites at once at 400 kg/satellite, however I expect the 2nd generation satellites to be larger as they will need a larger antenna to reduce the spot size. So perhaps 250 satellites per launch. If ITS costs are to be believed entire constellation could then be launched for $500M or $100M/year, if so launch costs would be in the noise. The savings would pay for ITS, at least its booster and cargo elements, even if not the full Mars missions.

For Mars the loaded ITS goes to Mars and lands. Refueled goes to LMO, deploys the sats and then lands again to be refueled for its return trip to Earth. The ITS that does this and all its adaptations is exactly the same as the one used to deploy the sats at Earth.

Mars missions would be just at first a small ripple in that launch rate.

Offline oldAtlas_Eguy

  • Senior Member
  • *****
  • Posts: 2922
  • Florida
  • Liked: 1247
  • Likes Given: 85
Back to the discussion of the filing.

By doing this filling SpaceX's has shown that the business case and technology investigations has led to a "PDR" for the sats and system. They are moving forward with a high likelyhood of actually implementing the constellation. Or at least the first 800 sat version. That would cost about $4B. $2M per sat + $3M to launch at 12 per F9 launch[$32M/launch at spaceX's cost].

Offline MikeAtkinson

  • Full Member
  • ****
  • Posts: 1621
  • Bracknell, England
  • Liked: 391
  • Likes Given: 50
Capital requirements for the initial system are high, they will likely need further financing at a level which would dilute Elon's equity too much. Suggest that CommX will be a joint venture, possibly with google or a consumer electronics manufacturer for the user terminal.

They also need to have a worldwide sales and marketing operation, to both consumers, businesses and perhaps telecoms companies.

Offline dror

  • Full Member
  • ****
  • Posts: 535
  • Israel
  • Liked: 115
  • Likes Given: 328

At typical encounter velocities upwards of 10 km/s, there's a technology problem.  Furthermore, the overwhelming majority of debris is far too small to track or even observe until you get hit by it.  Folks talk about orbital debris removal as though the problem is all big chunks of derelict satellites.  It's not.  It's paint flakes, bolt heads, coolant droplets, and generally small bits of detritus.  For every piece of debris you can track, there are thousands you can't even see, and every collision creates more tiny debris.

That's kind of what I was alluding to. Some kind of technological, space based innovation to sweep areas of space clean of tiny pieces of dangerous debris.


If I may analogize: imagine a nineteenth-century era battlefield -- a largely empty field, with bullets zinging across in multiple directions.

You're looking for an innovation that would allow someone to somehow make this random, omnidirectional hail of bullets vanish into thin air -- more importantly, making them vanish without fragmenting into even more bullets.  All without being able to even see them coming in.

To the extent that it's even technologically possible, it's an absolutely ludicrous business proposition until space gets to the point that it is literally unusable.

10 km/s is practically standing compared to C, Droplets and paintflakes can be vaporised with low kw, and can be seen by radar from a few tens of km.
So to vanish these tiny standing droplets into the very thin air could become possible if need arises, using very similar equipment to that on their sats.
Bolt heads and bigger debries are a different story.
"If we crave some cosmic purpose, then let us find ourselves a worthy goal. "
Carl Sagan, Pale Blue Dot

Offline cscott

  • Senior Member
  • *****
  • Posts: 2455
  • Liked: 1682
  • Likes Given: 656
The pure raw data rate is so high that in-orbit listeners would need to be nearly the equivalent of the comm system itself. Else they would need very large computational/packet sifting capabilities to just intercept the probable packets of interest. But if the data is also encrypted with a 1024 or even higher level of encryption the on-orbit sat would only have the source and destination IP addresses to trigger off of. So data flows of target data in the multi Tbps would require a significant comm system and a radiation footprint that would identify all of the listeners positions. A smart security routing protocol could literally rout around the listeners.
From what we have learned from Snowden et al, pervasive monitoring is usually done with the backbone's (at least limited) cooperation. Eg AT&T sets aside a room for you to put your equipment in, and routes fiber through it.  It would be plausible that SpaceX would be convinced, for patriotic (and cost) reasons, not to try too hard to route around listeners.  The state agency might also find itself in possession of some relevant encryption keys, whether SpaceX likes it or not.

But your larger point is right on: the high bandwidth, constellation sizes, and deployment rates would require equally-aggressive technology development on the black op side.  The result could be a boon for launch service providers and others.
« Last Edit: 11/17/2016 09:54 PM by cscott »

Online gongora

  • Global Moderator
  • Full Member
  • *****
  • Posts: 1641
  • US
  • Liked: 1082
  • Likes Given: 862
Found the FCC deadline, it was in a filing about the OneWeb application:
Quote
Additional applications. We invite additional applications and petitions for declaratory ruling for
NGSO-like satellite operation in the 10.7-12.7 GHz, 14.0-14.5 GHz, 17.8-18.6 GHz, 18.8-19.3 GHz,
27.5-28.35 GHz, 28.35-29.1 GHz, and 29.5-30.0 GHz frequency bands. Applications and petitions filed
by November 15, 2016, will be considered together with the OneWeb Petition. Requests filed after this
date may not be entitled to shared use of this spectrum
with respect to any grant of applications or
petitions filed prior to the cut-off date. Applicants and petitioners that file by the cut-off date will be
afforded an opportunity to amend their requests, if necessary, to conform to any requirements or policies
that may be subsequently adopted concerning NGSO-like satellite operation in these bands.

The document is attached.

Tags: