Launching Starlink with Starship

[Robotbeat]

I suspect it may be a lot fewer Starlink satellites on Starship, at least at first. Maybe like 200 or so. Preserves more propellant margin. Plus early Starships might be really heavy.

If you stick only to the cylindrical part and scale 2*30 satellites per 6.7m stack to 6*8m stacks, you get 214.  You could make the stacks somewhat higher and stay inside the static envelope in the ogive, but I think you're right that, between performance uncertainty and the dogleg, somewhere between 50t and 70t is about right.

Also, I suspect that SpaceX will make Starlink satellites a little heavier when they have the performance and lower cost per kg of Starship, plus the need for laser links and more bandwidth.

[envy887]

I suspect it may be a lot fewer Starlink satellites on Starship, at least at first. Maybe like 200 or so. Preserves more propellant margin. Plus early Starships might be really heavy.

If you stick only to the cylindrical part and scale 2*30 satellites per 6.7m stack to 6*8m stacks, you get 214.  You could make the stacks somewhat higher and stay inside the static envelope in the ogive, but I think you're right that, between performance uncertainty and the dogleg, somewhere between 50t and 70t is about right.

Also, I suspect that SpaceX will make Starlink satellites a little heavier when they have the performance and lower cost per kg of Starship, plus the need for laser links and more bandwidth.

Is there a practical limit to the size of satellite that would still be useful? It seems that they could keep adding antennas to add more spot beams, and also increase the antenna diameter to make the spot beams smaller.

[oldAtlas_Eguy]

The primary purpose to grow sat size and mass is to increase the comm data throughput per individual sat. It does not take a 2X heavier sat to have a 2X increase in throughput. Electronics as in CPU's throughput via increase in cores per chip increase every 18 months. So digital electronics portion to handle a 2X throughput increase on the sat may decrease in mass. Analog electronics if any increase is required other than the receiver preamp and transmitter final would not be a significant mass or possibly a net zero increase in mass and volume for all electronics except batteries in the sat for 2X data throughput. The major mass increase is in Solar array and batteries. More transmitters needs more power. A practical 2X increase for transmit power budgeted on the sat. But that is not the only consumer of electrical power so the Solar array and battery increase would not be 2X increase but some fraction of that. Also because solar and battery do not increase by 2X the sat total mass is likely to be a 50% or less total increase to accomplish a 2X increase in data throughput. But there are some practical limitations on spot size and incidence angles so at some point the max practical data throughput would be reached for an individual sat until significant UT improvements in number of sats able to be tracked, shorter duration per individual connection to an individual sat, smoother transitions between sats/spots and more often, higher incidence angles by the phased arrays...

Interesting you can probably get more by increasing number of sats than by increasing sat data throughput after a point. Because some UT needed improvements needed to increase individual sat throughput would increase the cost per UT. Making it easier and more cost effective to just add more sats. But even with minor UT changes, mostly software and faster/more cpu core CPU's, there is likely to be some 2X/4X improvements in throughput per sat in the Starship V2.0 version.

[Twark_Main]

If you're worried about debris, they're deployed so low (well under 300km, even though that's what I used above), that the tensioning rods, which have low ballistic coefficients, will deorbit in less than a month.

Last I checked only the tensioning rods from the v0.9 launch have deorbited, so they have a demonstrated orbital lifespan of many months.

The v0.9 tension rods are still going to be up there for a while (still around 430km).  After that, the v1.0 flight 9 tension rods will take a little while (at about 400x380km), and the rest of the flights have either deorbited or will shortly for the last couple flights.

Ahh, thanks for the correction. I thought the v0.9 rods had already deorbited.

Do you have the COSPAR id(s) handy?

[Robotbeat]

If early Starship has a usable capacity of 50 tons to Starlink’s altitude and inclination, and if SpaceX wants 30k+12k=42k total satellites each weighing at least 260kg, then that’s 10,920 tonnes. Or 219 early Starship launches.
“Hundreds” of launches.

[oldAtlas_Eguy]

If early Starship has a usable capacity of 50 tons to Starlink’s altitude and inclination, and if SpaceX wants 30k+12k=42k total satellites each weighing at least 260kg, then that’s 10,920 tonnes. Or 219 early Starship launches.
“Hundreds” of launches.

A good possibility is that after a few launches with a modified V1.0 sat they start launching an up to 50% heavier V2.0 sat for a total of some 200 to 300 launches by EOY 2026. Even though that on an early Starship with a 50t capability to the desired Starlink orbit would launch just 180 V1.0 sats and then when capability grows to 75t they start launching 180 V 2.0 sats. The end result is even with a growing tonnage capability is still some 200+ launches over a period of every 5 to 7 years. Or an average of 30 to 40 Starship (the full capacity version of 100t+) launches every year until the Starlink program stops providing service.

Also after the V2.0 sats have been launching for 5 years expect an even larger more capable data throughput sat whose mass is likely to increase another 50%. Such that by 2027/28 even more launches will take place to launch the same number of sats or 45 to 75 every year (7 year replacement or 5 year replacement respectively). One a week. If each SH does 20 launches/landings and each SS does 10 launches.landings. spaceX would need to produce 3 -4 SH and 5-8 SS each year just to launch Starlink sats. At some point doing a refurbishment which hopefully is cheaper than a brand new vehicle and results in just as safe a vehicle as a new vehicle. But if the production rates and manufacturing cost continue to drop doing significant refurbishment may not be cost effective until vehicle improvements make refurbishment cheaper as well as the number of flights before such refurbishment is needed.

Starlink again like it has for the F9 life extension (number of flights) development, would do the same for Starship tacking the higher risk until the risk has been retired and them moving on to the next risk followed by all the other customers on a higher quality vehicle.

[wannamoonbase]

If early Starship has a usable capacity of 50 tons to Starlink’s altitude and inclination, and if SpaceX wants 30k+12k=42k total satellites each weighing at least 260kg, then that’s 10,920 tonnes. Or 219 early Starship launches.
“Hundreds” of launches.

A good possibility is that after a few launches with a modified V1.0 sat they start launching an up to 50% heavier V2.0 sat for a total of some 200 to 300 launches by EOY 2026. Even though that on an early Starship with a 50t capability to the desired Starlink orbit would launch just 180 V1.0 sats and then when capability grows to 75t they start launching 180 V 2.0 sats. The end result is even with a growing tonnage capability is still some 200+ launches over a period of every 5 to 7 years. Or an average of 30 to 40 Starship (the full capacity version of 100t+) launches every year until the Starlink program stops providing service.

Also after the V2.0 sats have been launching for 5 years expect an even larger more capable data throughput sat whose mass is likely to increase another 50%. Such that by 2027/28 even more launches will take place to launch the same number of sats or 45 to 75 every year (7 year replacement or 5 year replacement respectively). One a week. If each SH does 20 launches/landings and each SS does 10 launches.landings. spaceX would need to produce 3 -4 SH and 5-8 SS each year just to launch Starlink sats. At some point doing a refurbishment which hopefully is cheaper than a brand new vehicle and results in just as safe a vehicle as a new vehicle. But if the production rates and manufacturing cost continue to drop doing significant refurbishment may not be cost effective until vehicle improvements make refurbishment cheaper as well as the number of flights before such refurbishment is needed.

Starlink again like it has for the F9 life extension (number of flights) development, would do the same for Starship tacking the higher risk until the risk has been retired and them moving on to the next risk followed by all the other customers on a higher quality vehicle.

The picture you paint is that SS/SH could be successful if it just served Starlink. 

A few commercial launches and a little service to the moon and we're all set!!

[Robotbeat]

If early Starship has a usable capacity of 50 tons to Starlink’s altitude and inclination, and if SpaceX wants 30k+12k=42k total satellites each weighing at least 260kg, then that’s 10,920 tonnes. Or 219 early Starship launches.
“Hundreds” of launches.

A good possibility is that after a few launches with a modified V1.0 sat they start launching an up to 50% heavier V2.0 sat for a total of some 200 to 300 launches by EOY 2026. Even though that on an early Starship with a 50t capability to the desired Starlink orbit would launch just 180 V1.0 sats and then when capability grows to 75t they start launching 180 V 2.0 sats. The end result is even with a growing tonnage capability is still some 200+ launches over a period of every 5 to 7 years. Or an average of 30 to 40 Starship (the full capacity version of 100t+) launches every year until the Starlink program stops providing service.

Also after the V2.0 sats have been launching for 5 years expect an even larger more capable data throughput sat whose mass is likely to increase another 50%. Such that by 2027/28 even more launches will take place to launch the same number of sats or 45 to 75 every year (7 year replacement or 5 year replacement respectively). One a week. If each SH does 20 launches/landings and each SS does 10 launches.landings. spaceX would need to produce 3 -4 SH and 5-8 SS each year just to launch Starlink sats. At some point doing a refurbishment which hopefully is cheaper than a brand new vehicle and results in just as safe a vehicle as a new vehicle. But if the production rates and manufacturing cost continue to drop doing significant refurbishment may not be cost effective until vehicle improvements make refurbishment cheaper as well as the number of flights before such refurbishment is needed.

Starlink again like it has for the F9 life extension (number of flights) development, would do the same for Starship tacking the higher risk until the risk has been retired and them moving on to the next risk followed by all the other customers on a higher quality vehicle.

The picture you paint is that SS/SH could be successful if it just served Starlink. 

A few commercial launches and a little service to the moon and we're all set!!

That's right. I think the Starlink constellation is huge in part because it provides sufficient demand to justify Starship...

[TheRadicalModerate]

If early Starship has a usable capacity of 50 tons to Starlink’s altitude and inclination, and if SpaceX wants 30k+12k=42k total satellites each weighing at least 260kg, then that’s 10,920 tonnes. Or 219 early Starship launches.
“Hundreds” of launches.

"219 early Starship launches" is an oxymoron.  The moment they've got something up and running in Florida, you can likely get close to 90t (that's as many as I could fit with the existing form factor), and I'd be surprised if the additional 30K of birds wasn't Starlink v2.0, with a form factor more conducive to filling the space up to 120t or so.

That would take the 30K birds down to 65 launches.  Of course, by then you'll have regular replacement flights going as well...

[gongora]

Do you have the COSPAR id(s) handy?

I added the launch designations on the Starlink Index thread to make it easier to look up launches

[Robotbeat]

If early Starship has a usable capacity of 50 tons to Starlink’s altitude and inclination, and if SpaceX wants 30k+12k=42k total satellites each weighing at least 260kg, then that’s 10,920 tonnes. Or 219 early Starship launches.
“Hundreds” of launches.

"219 early Starship launches" is an oxymoron.  The moment they've got something up and running in Florida, you can likely get close to 90t (that's as many as I could fit with the existing form factor), and I'd be surprised if the additional 30K of birds wasn't Starlink v2.0, with a form factor more conducive to filling the space up to 120t or so.

That would take the 30K birds down to 65 launches.  Of course, by then you'll have regular replacement flights going as well...

90 tons to a 200km, 26 degrees orbit with low margin, absolutely. (Maybe even 100 tons.) But 400km and 53 or even 98 degrees (i.e. roughly polar) latitude? With perhaps a (reusable) dispenser and extra margin? Easily could be just 50 tons of useful Starlink payload.

Shuttle got 27.5t to LEO (200km, 28 deg latitude) but just 12.7t to polar and 16t to ISS orbits. ISS is fairly close to Starlink's orbit. Shuttle is probably closest to these early, heavier Starships than any other vehicles. So this is a pretty reasonable comparison and illustrative of what kind of knockdowns they might have for the earlier, heavier vehicles to different orbits without refueling and with extra margin for landing.

[wannamoonbase]

If early Starship has a usable capacity of 50 tons to Starlink’s altitude and inclination, and if SpaceX wants 30k+12k=42k total satellites each weighing at least 260kg, then that’s 10,920 tonnes. Or 219 early Starship launches.
“Hundreds” of launches.

"219 early Starship launches" is an oxymoron.  The moment they've got something up and running in Florida, you can likely get close to 90t (that's as many as I could fit with the existing form factor), and I'd be surprised if the additional 30K of birds wasn't Starlink v2.0, with a form factor more conducive to filling the space up to 120t or so.

That would take the 30K birds down to 65 launches.  Of course, by then you'll have regular replacement flights going as well...

Elon talks a big long term goal on flight rate.  But look at how hard it is to move these monsters around.  Even 65 flights, that is going to take years to get up to that.  Maybe 4-6 years

[envy887]

As a 2sto Starship has a much higher staging velocity and better upper stage mass fractions than 1.5 stage systems, so it isn't nearly as sensitive to minor changes in mission delta-v as Shuttle was.

SpaceX is positioning Starship as capable of 21 t to GTO. even with all kinds of derating, it will easily exceed 50 t to low polar orbits or ISS. They might fly it with "only" 50 t for various reasons, but lack of performance would be pretty far down the list.

[TheRadicalModerate]

If early Starship has a usable capacity of 50 tons to Starlink’s altitude and inclination, and if SpaceX wants 30k+12k=42k total satellites each weighing at least 260kg, then that’s 10,920 tonnes. Or 219 early Starship launches.
“Hundreds” of launches.

"219 early Starship launches" is an oxymoron.  The moment they've got something up and running in Florida, you can likely get close to 90t (that's as many as I could fit with the existing form factor), and I'd be surprised if the additional 30K of birds wasn't Starlink v2.0, with a form factor more conducive to filling the space up to 120t or so.

That would take the 30K birds down to 65 launches.  Of course, by then you'll have regular replacement flights going as well...

90 tons to a 200km, 26 degrees orbit with low margin, absolutely. (Maybe even 100 tons.) But 400km and 53 or even 98 degrees (i.e. roughly polar) latitude? With perhaps a (reusable) dispenser and extra margin? Easily could be just 50 tons of useful Starlink payload.

Shuttle got 27.5t to LEO (200km, 28 deg latitude) but just 12.7t to polar and 16t to ISS orbits. ISS is fairly close to Starlink's orbit. Shuttle is probably closest to these early, heavier Starships than any other vehicles. So this is a pretty reasonable comparison and illustrative of what kind of knockdowns they might have for the earlier, heavier vehicles to different orbits without refueling and with extra margin for landing.

I arm-waved the 120t and it was a little optimistic, but I think you're way too pessimistic.

In the "non-early" timeframe, the addition of RaptorVacs, thrust improvements, and dry mass reduction to 105t should get Starship to 140t payload to 200 x 200 x 28.5º in the "non-early" timeframe (i.e., when Florida ops are up and running).  If that's the case, then going to 300 x 400 x 53.8º costs an extra 350 m/s, which will take you down to about 110t.

[Robotbeat]

Yeah, but launching lots of Starships isn't a bad thing, either. Higher flight rate does more to prove rapid reuse than large payloads. And it proves it out for launching crew.

[TheRadicalModerate]

Yeah, but launching lots of Starships isn't a bad thing, either. Higher flight rate does more to prove rapid reuse than large payloads. And it proves it out for launching crew.

Higher flight rate than you need is simply wasting money.  Testing while you fly live payloads makes all kinds of sense.  Testing sub-optimal payloads doesn't--at least not when you're talking about 70 flights more than you need.

[Robotbeat]

Yeah, but launching lots of Starships isn't a bad thing, either. Higher flight rate does more to prove rapid reuse than large payloads. And it proves it out for launching crew.

Higher flight rate than you need is simply wasting money....

Larger payload than you need is also wasting money. Less margins and higher fuel loads. More wear and tear requiring more maintenance. Have to wait longer in between each flight, and it takes longer for each flight's payload to reach operations.

[TheRadicalModerate]

Larger payload than you need is also wasting money. Less margins and higher fuel loads. More wear and tear requiring more maintenance. Have to wait longer in between each flight, and it takes longer for each flight's payload to reach operations.

We're not talking about larger payload than is needed.  As you pointed out up-thread, we're talking about needing to put more than 10,000t into orbit.

Let's take your points one at a time:

1) Less margins and higher fuel loads?  That might be true when fuel is >50% of the amortized cost of a flight, but that's not happening for a while, if for no other reason than early versions of Starship will be retired to keep the fleet relatively homogeneous, and to drive the costs down on the more stable versions.  That drives the amortized flight cost up. Even if that weren't the case, I'm skeptical that the first 50-100 flights have short turnaround cycles, if for no other reason than SpaceX needs the data to improve the product.

2) More wear and tear?  It's not a minvan with too heavy a trailer on the back.  The only thing that minimizes wear and tear is fewer engine minutes.  It makes no sense not to fill SuperHeavy full, so you're dealing with varying prop loads on Starship.  So... you reduce the burn time on 6 engines by maybe 30%, but it stays the same on 28 engines?  That's a 5% reduction in engine minutes.

3) Have to wait longer between flights?  This is a little bit like the old joke: "We lose a little bit of money on each thing we sell, but we make it up in volume."  Increasing cadence without increasing payload is more expensive than reducing cadence to match demand, even if the increased cadence makes you slightly more efficient.  It's certainly not going to make you 50% more efficient.

4) Takes longer for each flight's payload to reach operations.  Eh?  The only time that would be true for Starlink is if there were suddenly a burning need for a half a load, and you couldn't wait for the second half to be manufactured.  Do you really see that being an issue any time soon?  Even if you got to the steady-state maintenance phase, where you were only launching spares in anticipation of upcoming retirements or failures, if you've planned things so that you have a burning need for 200 right now instead 400 two weeks from now, you've done something horribly wrong.

[Alvian@IDN]

Each Starlink launches with less than maximum capacity would means faster time to high launches, which in turns they could get this vehicle to carry a human faster

We know that Gywnne had said if they didn't meet 3 years from first orbital flight at its longest to the first crewed mission, it would be considered as company failure

If they want to meet or at least not terribly delayed the first crewed mission and esp Dear Moon (still within a middle of this decade), they have to meet the 100th uncrewed launches fast, and that would be more difficult if they fill each Starlink launches to the max instead of spreading it to more launches

[envy887]

...
4) Takes longer for each flight's payload to reach operations.  Eh?  The only time that would be true for Starlink is if there were suddenly a burning need for a half a load, and you couldn't wait for the second half to be manufactured.  Do you really see that being an issue any time soon?  Even if you got to the steady-state maintenance phase, where you were only launching spares in anticipation of upcoming retirements or failures, if you've planned things so that you have a burning need for 200 right now instead 400 two weeks from now, you've done something horribly wrong.

It's at least plausible that more frequent launches with less payload could be worthwhile. It would take some modeling of the vehicle costs and constellation design and revenue to find the optimal point, and I don't think we have good enough data to tell for sure.

SpaceX can save a few months of precession drifting by launching more times into more planes. Since equal satellite distribution is necessary to provide equal bandwidth, and customers won't like wildly varying bandwidth as more or less densely populated constellation planes come overhead, they can grow the Starlink customer base faster with more frequent launches. Faster customer growth means more revenue.

This is a similar effect to all-electric GEO commsats paying more to get inserted closer to the operational orbit. A month of lost revenue using ion thrusters to get into service is a lot of money.