Ten SpaceX Starlink updates

Starlink availability map (Source)

Starlink now has nearly 500,000 users and is available in 32 countries and nine languages. It is either available, wait-listed, or coming soon in every nation except Afghanistan, Belarus, Cuba, China, Iran, North Korea, Russia, Syria, and Venezuela.

There are now 15,000 Starlink terminals in Ukraine with service throughout the nation through connections to ground stations in Poland, Lithuania, and Turkey and they have made a significant contribution in the war with Russia. For better or worse, they have demonstrated the value of low-Earth orbit satellites in combat.

A Chinese research paper called for the development of systems to track, monitor, and disable Starlink satellites. Quotes include "The country needs to be able to disable or destroy SpaceX’s Starlink satellites if they threaten national security" and "A combination of soft and hard kill methods should be adopted to make some Starlink satellites lose their functions and destroy the constellation’s operating system.” The paper in Chinese is here and an English translation is here.

SpaceX has begun launching version 1.5 satellites with optical terminals and Elon Musk says they will be operational by the end of the year. They are about to run a test using laser links to provide connectivity in Polar regions (above 53 degrees latitude) where there are no ground stations. It will take some time for them to connect all the satellites in the constellation in an orthogonal mesh, but this simulation predicts a ~2x latency improvement over long links when the mesh is complete. 

Astronomers worry about reflections from larger Version 2
satellites. (source)

The version 1.5 satellites do not have visors that reduce reflection which interferes with astronomical observation because they were incompatible with the laser terminals. Principal engineer David Goldstein says they're working on technology that will make version 2 satellites ten times dimmer than Vantablack paint.

SpaceX tested Starlink roaming in the US and later in Ukraine and they have now rolled out two new commercial services, Starlink for RVs, a low-priority service that can be paused, and Residential Starlink + Portability which provides priority service at your registered residence, but low-priority service when away. Elon Musk reported that they had 30,000 Starlink-for-RV orders within three weeks of its availability.

Service is not yet offered for moving vehicles, but it has worked well in the tests mentioned above and in Ukraine. Since the technology works, they may be delaying availability until they have a capacity allocation/pricing scheme, or they might be working on a new terminal. How long will it be before they have a terminal for Tesla and other cars and trucks? (They will eventually be competing with Geely for automobile connectivity).

Starlink's new business service is intended for locations with up to twenty users. It includes a $2,500 terminal with a high-gain antenna and promises download speeds of 150-500 Mbps and latency of 20-40ms. It includes 24/7, prioritized support and a publicly routable IPv4 address, but does not offer on-site support or a guaranteed service-level agreement.

Starlink began by serving the consumer market, but they have begun selling to air and ship lines. Starlink has contracted to offer service on JSX and Hawaiian Airlines and has applied to serve Royal Caribbean cruise ships. The airlines will use a new high-gain antenna that can deliver 500 Mbps down and40-50 up with 400w peak power consumption. I'll be curious to see what sort of terminal they will have for cruise ships with thousands of passengers but little nearby congestion when at sea.

SpaceX published a detailed description of its approach to space sustainability and safety in a February 22 post on their Starlink Updates blog. The post reports that Starlink satellites made 3,300 autonomous maneuvers to avoid collision during the second half of 2021 and describes the policies and technology Starlink uses to avoid collisions, including "ducking" -- retracting the solar panel and orienting their attitude to have the smallest possible cross-section in the direction of conjunction. Unfortunately, SpaceX cannot solve the space debris problem unilaterally and all debris objects cannot be tracked from Earth, though improved tracking from satellites might help and at least one company, Privateer, is exploring that possibility.

(Click here for a Starship update).

Update 6/29/2022

Royal Caribbean has run an informal test of Starlink aboard one of their cruise ships. Starlink significantly outperformed geostationary satellite service with a Ping time as low as 38ms and up and download times as high as 77/03 and 15.72 mb/s. That is fast enough to stream video or teleconference but does not shed much light on what to expect when connectivity is distributed throughout a ship with 5,000 passengers and a crew. It will be interesting to see how LEO service on ships ends up competing with O3b Mpower MEO connectivity

Update 6/30/2022

The Minister of Digital Transformation of Ukraine, Mikhail Fedorov, said that in the future, Internet from Starlink will appear on Ukrainian trains and it has been demonstrated in a successful pilot test in which Starlink technicians were surprised that it worked at high speeds. Ukraine’s Special Communication and Information Protection State Service chief Yurii Shchyhol said Starlink will be available in all Ukraine Railway trains by the end of 2022.

It looks like SpaceX is using Ukraine as a testbed for connectivity in moving vehicles and as a war tool in general.

Update 7/1/2022

The FCC has authorized Starlink connectivity to RVs, large trucks, ships, and planes. They expect to be on planes "very shortly" according to Jonathan Hofeller, Starlink's commercial sales chief. Elon Musk has said he didn't see "connecting Tesla cars to Starlink, as our terminal is much too big." That's true for today's terminals, but I wouldn't rule it out in the future -- by SpaceX or, as mentioned above, Geely.

Update 7/7/2022

Starlink Maritime is available now near coasts where they can reach terrestrial ground stations. Coverage will expand in the fourth quarter of this year and expand to global coverage in the first quarter of next year as more satellites are equipped with inter-satellite laser terminals.

With a price of $5,000 per month and a one-time hardware cost of $10,000 for two high-performance, ruggedized terminals for a download speed of "up to" 350 Mbps, they are marketing to merchant vessels, oil companies and Russian oligarchs with Yachts, not the few thousand customers on a Royal Caribbean cruise ship. Latency will on average be greater than terrestrial Starlink due to hops through inter-satellite links. Customers can pause and resume service at any time, but when they turn it on, they will have to pay for the month.

For those who might be worried about performance in severe weather and rough seas, SpaceX points out that Starlink is currently being used to get high-quality video of rocket landings at sea, providing continuous coverage next to engines capable of generating up to 190,000 pounds of force.

Update 7/15/2022

The first phase of Starlink launches consists of 4,408 satellites in five orbital shells. Over half of those satellites are in orbit already, but just last week they launched the first forty-six of the 520 operational V 1.5 satellites that will be in near-polar orbits with inclinations of 97.6 degrees.

Phase 1 orbital shells -- 4408 satellites

These polar orbit satellites will enable them to offer global coverage on land and sea as described above.

Update 8/31/2022

Following on the trial described above, Royal Caribbean has agreed to provide Starlink connectivity on their entire fleet of 63 ships. They will begin outfitting ships immediately and expect to have them all online by the end of the first quarter of next year. 

Other cruise lines connect through geostationary satellites, but Royal Caribbean has been using SES middle Earth orbit satellites on two of their ships. Starlink LEO satellites will provide better performance and offer full global coverage, which the others cannot.

Can SpaceX launch version 2 Starlink satellites this year?

A Starship spacecraft being lifted onto a Super Heavy booster

The answer to the question in the title depends on the availability of SpaceX’s new Starship spacecraft and Super Heavy booster rocket, collectively referred to as Starship. Elon Musk says he is highly confident about getting Starship to orbit this year. He also says, “At Space X we specialize in converting things from impossible to late.” 

Starship is critical to Starlink because the version 2 satellites are seven meters long and weigh about 1.25 tons and the current Falcon 9 rockets have neither the cargo volume nor the mass-to-orbit capability to economically launch them. As Musk put it, they "need Starship to launch and fly frequently or Starlink version 2 will be stuck on the ground." 

Version 2 satellites will be ejected
using a "Pez dispenser" mechanism
modeled after industrial pallet
stackers.

Musk has said the performance of the version 2 satellites will be nearly an order of magnitude better than that of the current satellites and they will include inter-satellite laser links. He did not elaborate on his order of magnitude estimate and it will be interesting to see how they handle a mixed-satellite constellation, but all the analysts predicting the profitability and capacity of Starlink will have to go back to work. The projected improvement also reminds us of the advantage of short satellite life.

Today's terminals will work with the version 2 satellites, but performance will be better with a new terminal that is being developed. Perhaps Starlink will allow current terminal owners to trade them in on Version 2 terminals.

But can SpaceX launch the remaining 9,500 approved satellites or the 30,000 that are awaiting approval in a timely manner? They can if they achieve rapid reuse of Starships. The total world mass-to-orbit to date is 16,000 tons. Musk estimates that a single Starship launching three times a day could put 109,000 tons in orbit in a year.

Today they are doing about one launch per week and the current average turnaround time for boosters is 21 days -- can they reuse a Starship three times a day? Boosters fly for about six minutes before returning to Earth and SpaceX plans to have them land back on the launch pad ready for refueling, installation of a spaceship, and re-launch. (They will need more spaceships than boosters). 

During an interview/tour of the Starbase manufacturing and launch site in Boca Chica, Texas, Musk spent time at a launch pad discussing the 10-second booster landing and capture sequence and the design decisions it entailed with Andrew Krebs, Director, Starship Launch Engineering. It was apparent that various tradeoffs are still being evaluated and change is constant. Musk said there is a good chance they will fail to capture the booster on the upcoming orbital flight. (Recall the "unscheduled disassembles" tape of attempts to safely land Falcon boosters). 

They spent some time on the tour at an assembly bay talking about Starship design and manufacturing. Since they need Starship to launch and fly frequently, they expect to eventually have ten or twelve manufacturing bays and Musk expects to manufacture a new spaceship and booster every month. (He estimates that they will eventually have 1,000 Starships for building a city on Mars).

Raptor 1 vs Raptor 2 (source SpaceX)

The greater mass of the Starship and satellites requires more powerful rocket engines than SpaceX is currently flying so Starship will use thirty-nine Raptor 2 engines -- 33 for the booster and six for the spaceship. The Raptor 2 is more powerful than its predecessor, has a much simpler design, and costs about half as much. They had trouble manufacturing them at first but now expect to make seven or more per week.

If all goes well, Musk estimated the variable cost per orbital flight as being a few million, "maybe" as low as one million, dollars. 

In addition to engineering and manufacturing, there are regulatory hurdles. The Federal Aviation Administration (FAA) requires an environmental review before they can do the first orbital test launch and they are currently being required to make over seventy-five environmental adjustments to do an orbital launch in Texas. I have no idea how long these adjustments will take, and Musk has considered moving the first orbital flight to Florida. You can see the latest FAA report here.

There are a lot of unknowns and precisely landing and catching a 68-meter long, 9-meter diameter rocket might turn out to be impossible, but SpaceX specializes in converting things from impossible to late.

---

This post is based on a Starship update talk Musk gave at Starbase in Boca, his recent all-hands company presentation, and videos of interviews with Tim Dodd, the "Everyday astronaut," conducted while touring Starbase. Those videos are here, here, and here. I've only selected some Starlink-related material from these sources. Check them out to learn a lot more about Elon Musk and rockets.

(Click here for a Starlink update).

Update 6/17/2022

The FAA has concluded that continued launches "would not significantly affect the quality of the human environment" and Elon Musk, who has said there would be no more "hops," expects to attempt a Starship orbital launch in July. Musk thinks they will be able to satisfy the FAA quickly and get on with the orbital launch test. He tweeted “There will probably be several launch countdowns before we pass all the abort triggers, but hopefully the first countdown is this month.”

Update 8/4/2022

Alternative plan -- booster returns to launch pad

The answer to the question in the title is "no," but they are pushing hard and next year is a possibility. Version 2 satellites are already at Starbase, and the next Starship launch will try to reach orbit. It would be surprising if the first attempt reached orbit, but Elon Musk is confident that they can do it within twelve months. Whenever they do it, if all goes well, they will try to land the booster on the launch pad rather than allowing it to fall into the ocean. 

Musk previously predicted that the first orbital test would be as early as last January, so he is nearly a year behind schedule, but that prediction did not include an attempt at landing on the launch pad.