SpaceX launches another 60 Starlink satellites while setting two reuse records
cnbc.com
cnbc.com
Traditional ISPs are in trouble if the Starlink latency is really in the sub 100ms due to the LEO distance advantage.
Global coverage with decent latency!
I wonder if that's why google dipped out on rolling out fiber infrastructure.
Physical infrastructure is about to become obsolete as more advancements in wireless tech and satellite internet become available to the masses.
Which isn't that great.
The system Starlink is trying to beat is BGAN and regional Ku/Ka providers. If you're expecting them to beat the pants off of your terrestrial wired service you're probably going to be disappointed, unless you're on the edge of the range of some old neglected AT&T DSL line or still trapped on dialup. Even then it's almost certainly going to be much more expensive but at least it will have a sizeable speed advantage.
I’m seeing figures of 1 million$ per satellite. Over 10 years that’s roughly 100k/month in mortgage or around 1k customers at 100$/mo.
My guess is Starlink is going to need FEC for Doppler effects and multipath in urban environments. Maybe they will be able to overcome multipath with modulation. But Doppler shift will be significant with LEO.
The latest 802.11n standards have been increasing physical layer frame sizes, not decreasing them. I don't know of any modern wireless standard that doesn't use FEC. Coding gain is significant in most link budgets I have seen.
The last satellite RF downlink I worked with was for JPSS-1 in LEO nearly 6 years ago. There was a (255, 223) Reed-Solomon code with an interleave depth of 4 frames. So you effectively had to receive 1024 bytes, decode, and you got 4x 223 bytes worth of information. While latency wasn't a major concern, this scheme seemed pretty standard to me. If you don't believe me, the spec is public information:
https://www.star.nesdis.noaa.gov/jpss/documents/CDFCB/GSFC_4...
I'm sure SpaceX is trying really hard to control latency - so it doesn't get out of hand! One set of RF processing here, another over there, each with their own filters and decoding algorithms and then you have an extra 100 ms of delay in the system.
https://mobile.twitter.com/elonmusk/status/11329039145865297...
Besides we’re talking about the latency of the satellite to satellite links, not the downlink (since that’s the difference between using a downlink close to the client vs close to the server).
I'm not talking satellite to satellite latency, I'm talking total latency end-to-end. I'm also not saying the system will be 100 ms, just that it takes significant engineering effort so latency doesn't get out of control. IMHO, it's a fundamentally much harder problem than existing transatlantic links, despite the greater signal speed in a vacuum. I'm skeptical of what the latency will be. You disagree and that's fine.
Maybe we'll see an actual measurement soon enough. The competition can do 60 ms round-trip to NYC to London.
I’m speculating they’re smart guys working hard to keep latency low, and that routing will be a more challenging latency issue, while serialization should be minor.
But we’ll see and I agree we’re both speculating
Starlink allows high bandwidth density on the ground, and at the satellites because every antenna in the system is an extremely capable steerable phased array. Every transceiver can fully distinguish between signals of equal power and frequency that differ whose sources differ at least 10 degrees.
Reflections from buildings are irrelevant when the antenna must already be capable of picking the correct transmission out of the 50 or so equally powerful beams on the same frequency that only differ in angle.
Also, since every terminal already needs to be constantly aware of the position of every satellite, the correct answer to doppler is not fancy error correction, it's modulating the transmission to eliminate it, or just expecting it at the receiver.
This was a big problem with Iridium. With only one ground station available your latency figures were all over the place thanks to the complicated routing that every packet had to go through.
I thought downlinks were going to be pizza box-sized antennae. Why would it be far away from you? That could live on your roof.
This suggests to me a low data cap and/or rate, perhaps comparable to DSL.
12000 * 20Gbps / 2 * 100 / 100Mbps / 3 = 40,000,000 customers
A service like that could sell for a lot in rural North America, but maybe not to 40m customers. Let's assume most of those customers are from lower income countries or from places with more competitive ISP environments. Maybe $20/mo average revenue per customer. That's $800m a month. That already sounds compelling to me, before considering the potential for more or higher bandwidth satellites, or for higher margin commercial or government contracts.
A few HFT customers will probably finance the whole thing, tbh.
Last I saw, they were reconfiguring deployment plans to go for higher density constellation in the lower latitudes before worldwide coverage, so I'm not sure if this is even knowable at this point.
150kg sat to launch at $2,719kg/leo costs $0.4m
At this point I'd just like to have a built-in wifi hotspot for my own usage in my car. My Chevy can do this. My fancy high-tech Tesla cannot.
Google stopped deploying fiber for several reasons:
a) they weren't able to get reasonable access to poles for deployment.
b) incumbents started large rollouts of gigabit capable services (FTTH from telcos and DOCSIS 3.1 from cablecos) in markets Google had announced, but hadn't built out
c) maybe? lots more LTE deployment
The Starlink receiver is the size of a pizza box (and cannot get smaller, since that is the size of the necessary antenna) - and doesn't work while moving.
So it's impractical to use on cars.
What Tesla probably could do is put a receiver at every Supercharger and broadcast wifi there. It would be especially useful at the more rural locations.
[0]: https://spacenews.com/spacex-plans-to-start-offering-starlin...
My Volt included unlimited data for 20$ a month, and it was fast enough for streaming video and worked at times where my phone didn't (presumably because of the better positioned antenna)
* 4th landing for this first stage booster. Estimated first stage service life is ten launches prior to inspection, one hundred flights total
* Fairings were reused from a previous launch, unable to be recovered due to rough seas this launch
* 360-400 satellites required in the constellation to transition from "bent pipe" to intra-constellation data routing
* 1200 satellites provides global coverage
* Starlink satellites have a five year service life
* SpaceX board recognized in 2012 that satellite comms has a much higher margin than satellite launch services
But they launched it anyway.
SpaceX is going for vast scale with this project. Failures up to and including the loss of a launch vehicle can't derail the rest of it.
My guess is that they did a cost benefit analysis of doing this, and realized it was worth more to them to launch a dead satellite and verify it deorbits as planned than it was to miss a launch deadline, take apart the stack, fix the bum satellite, reintegrate the stack, put it back on the rocket, and schedule a new launch date.
The satellite's cheap - that's literally the point. Verifying failure modes with NASA and the USAF? Now that's expensive. It's worth more as a science experiment than a time sink.
There could be some systematic failure that affects them all, and just this was the first to fail. If the rest fail over the next 2 weeks, that will be a big goof.
PS: besides, who said they don't know the reason?
https://twitter.com/SpaceX/status/1193687615528042496
And in this case, if dead, the satellite will be used as a science experiment
There is an annoying interstitial ad, but the URL really says it all. From March 21, 2019.
They say the first 75 have iron thruster parts and steel reaction wheels which can survive reentry, but that in subsequent builds no parts will come down with lethal energy.
We have better than half a century where space launch was so expensive, and so rare, that this had an impact on the payloads themselves. Perversely it caused the payloads to become so highly engineered (for a variety of reasons) that they also became ultra-expensive in-line with the launch vehicle and launch opportunity cost.
This also created a negative feedback loop that because the payload was expensive, the launch vehicle had to be extra reliable and thus cost more.
Think of the entire lifetime program expense (R&D, launch, repair, etc.) for the Hubble program, billions upon billions of dollars to basically put a large 1 MP sensor at the end of a tube with a digital radio for communications.
What would happen if this program cost went into hundred of launches, every month, for decades? We could launch not only an equivalent, but technologically improving payload every month? Why not make the device cheaper? Experiment a bit? Who cares if one fails? Maybe launch 2, 3, 4 at a time? Have constellations of Hubbles spread throughout the system? What if Hubble was 300 telescopes with massive 20MP sensors clustered into a distributed virtual telescope the size of the orbit of Mars?
When the launch costs, and cadence drop so low, it doesn't really matter if your satellite doesn't work because you didn't spend 15 years of R&D, and a billion dollars putting up as technologically "perfect" a device as possible, you fab another one, maybe with better parts, and put it up next month instead and write-off the couple hundred thousand as cost of doing business. It's all still cheaper than the alternative.
Smallsats were a kind of way to try to squeeze costs down by simply cutting capability, size, and power. But the cost of a smallsat in orbit was still dominated by the cost of the launch. What if you could launch a fully capable space platform (not a compromised small sat) for less than the cost of a shared-ride smallsat.
With the tonnage going up comes more capability as a by-product. The ISS took over 40 flights to get ultra expensive space hab parts into orbit. A SpaceX Starship might be able to life equivalent mass in 5 launches, at a fraction of the cost. What if, instead of one space station every 20 years, we get two per year?
We have no idea what's about to happen in the next decade.
That's such a gross over-simplification of what the Hubble platform is and what it is capable of that it annoys me.
First, there are _six_ scientific sensor packages on the Hubble craft. These have been swapped out and upgraded several times during its life with a process that allowed us to bring the failed instruments back to surface and study them.
Second, it's not just scientific instruments.. the Hubble is one of the few satellites to have a fine guidance system, that allows it to lock onto and then _track_ with extreme precision exceptionally distant objects. The newer JWST is the only other telescope with similar capabilities.
This system has also been upgraded during the several servicing missions that were a planned part of the entire mission, which is another reason the initial costs were so high. This is an exceptionally advanced program to launch one of the most advanced scientific instruments we created up to that point.
Finally, I think it's also worth pointing out that Hubble was launched into orbit over 29 years ago and it's still working.. which is incredibly impressive considering it has to transit between Earth's shadow and direct sunlight every 47 minutes.
This thing is as _far_ from "a 1MP sensor at the end of a can attached to a radio" as you can possibly get. By the way, we could have made it larger.. but congress wanted to keep the price under $100 million. (Or about $201 million in Today's dollars).
> We could launch not only an equivalent, but technologically improving payload every month?
If all you're doing is making phone calls and routing packets, sure. If you're trying to do legitimate astrophysical science it seems almost an entirely worthless attribute.
> We have no idea what's about to happen in the next decade.
I think your reasoning is exactly backwards. We only launched highly specific and purpose built scientific instruments because it was the only thing _worth_ doing with our previous limited and highly government subsidized launch activities.
There was simply no way to commercialize those launches, and now that we entering an era where you easily can see long term value in them, you're simply going to see typical commercial activities and value realization behavior when they utilize this destination. It's absurdly predictable.
We know today that the Hubble instrument cost about $4.7 billion, and lifetime costs as of 2010 are around $10 billion [1] which doesn't appear to have included launch costs (1 initial launch and I believe 5 servicing missions) which at around $450 million per launch is quite a bill of around $2.7 billion in total program launch costs.
One might try to make the argument that shuttle missions involved multiple mission objectives for each launch (and thus the total cost shouldn't be borne by the HST alone). However, HST servicing missions were designed around the shuttle, making it mandatory for the servicing mission and thus the entire cost is a minimum regardless of other shuttle duties.
For example, STS-61, the mission to repair the HST (plus a few other odds and ends) required one of the most complex mission profiles in history. I can't find solid figures on how much the mission cost, but I'm guessing it was on the order of around $1 billion dollars, launch costs included. I think it could be argued that such a complex mission was worth the cost in terms of maybe crew and mission planning experience, but my central argument is, in a cheap launch future, such a mission (and all servicing missions) would be entirely pointless -- which not only would be cheaper, but allow more advanced sensing packages to be put up over time with less of a risk to the lives of the astronaut crews.
Suppose Falcon-9 Heavy technology existed when the Hubble was launched. Here might be a take on an entirely different program:
1 - Separate satellites for each orthogonal sensing package. Say maybe, 1x FOC/FOS, 1x GHRS/HRS, 1x HSP, and 1x WFPC. So a constellation of 4 satellites to start with.
2 - Let's build all missions off of a single designed "things in a can" platform so we can realize cheaper costs over time. Let's assume an expected life-span of 4 years (the time between STS-61 and STS-82). So we can reduce the per device cost from $4.7 billion/per to probably something more in line with reality. Maybe $250 million/per so we can at least gold plate it a bit, but keep it within the original budget per instrument.
3 - Launch mass of the original HST was about 11.1 metric tonnes. Since we're getting rid of stuff, we can shave that down to probably <11 tonnes per satellite.
4 - A F9Heavy can shove 63,800 kg to LEO, which is about where the HST sits (with periodic reboosts during servicing missions). That means we can put the entire constellation up in one launch with room to spare.
5 - So no we're now in program year 5 and total cost (minus ground station and staff and such) is $1.09 billion. Even with a complete failure of one of the satellites, the replacement cost is $300 million ($250 for another satellite + $50 for an F9 launch). At 5 years in, with an equivalent optics failure as the HST, we're at $1.39 billion for a program of greater capability and one failed device.
6 - Over the years, the HST servicing missions added the ACS, COS, NICMOS, STIS, and the WFPC2 and WFC3 instruments. The COSTAR was never needed, and never needed to be replaced by the COS. That's two servicing missions that never had to happen at all! We've just shaved off almost another billion dollars from the program. For that cost, we can almost repeat the entire first launch and shove another 4 satellites of the the same capability in the air. So we do that and we're extend the life of the HSP and the WFPC missions beyond what actually happened and we're at a total cost of around $2.5 billion.
7 - We go ahead and do two more F9Heavy launches to clear out the mission backlog and now we have 14 satellites conducting mission, in orbit, for almost exactly the cost of just the original HST device, and no lives risked on servicing missions. With the remaining funds, we can deorbit failed satellites, engineer better/different sensing packages, and put another 14 satellites in orbit. Hell, there's enough spare launch capacity in the flight plan to put a couple JWSTs, CXOs, and roadsters up as well.
8 - We don't know what Starship launch costs might be, but they could significantly expand these kinds of mission profiles. Musk's crazy claim is $2 million per, and could shove maybe 6-8 HST-class satellites up in one go.
9 - For $2 million per launch, why the hell are we putting up $250 million dollar devices? Why not get crazy, set up an assembly line for "things in a can" space sensing devices and crank them out at $500,000 per can and radio setup with a "bring your own sensor" policy? Who the fuck cares if the STIS mission didn't pan out? Deorbit the sucker and put two more up next month!
At the scale of nation states and large corporations, it means space platforms become as disposable as they de facto actually are.
[1] https://www.nasa.gov/pdf/499224main_JWST-ICRP_Report-FINAL.p...
The red arrows indicate the phased-array antennas.
Apparently they're going to attempt to go fish these ones out, too, since the boats didn't head out today because of rough seas.
Starship is the real reusability project; Falcon is now just holding on to revenue while they get Starship ready to take over the entire launch market (including their own).
The fairing is passive, I really wonder if it would have been more cost effective to try and reduce the fairing cost to a (few) hundred thousand bucks or so instead of going to all the trouble of catching it.
Although maybe this is best seen as R&D cost for the eventual Starship architecture?
I can't imagine positive ROI requiring more than 4 or 5 recoveries. Personally, I'm kind of disappointed they didn't go with a skyhook concept, but getting that system to work with something the size and drag of a half-fairing would probably be a nightmare.
https://en.wikipedia.org/wiki/List_of_Falcon_9_and_Falcon_He...
This reads that he was unaware that they were reusing at all.
It would make much more sense for them as a business to try to provide censored service to China so that they could openly sell to the Chinese. Elon has business ties in China now with Tesla and I don't see him trying to be a rebel with SpaceX now. He needs that Chinese revenue to fund his Mars mission.
That is incorrect. SpaceX doesn't need to know where a terminal is for it to work. The Terminal itself though will know where it is, as it needs that information to calculate the pointing angles of the phased array to the satellite orbits.
You need a Starlink terminal in order to communicate with the satellite. China will most likely just block the sales of Starlink terminals in the country if they want to enforce a ban.
Shotwell said the company’s board of directors in 2012 realized the profit margins from the commercial satellites it was launching for customers were “much higher” than SpaceX’s launch business. Musk estimates Starlink could generate more than $30 billion per year – at least 10 times what SpaceX could bring in at best from its launch business.
What other customers businesses will they seek to cannibalize next?
Having said that, my understanding is that BO has nothing even planned for full re-use. They are going for 1st stage re-use only, similar to what Falcon 9 is doing right now.
SpaceX Starship, which is well under development, and has had two very short test hops, is fully re-usable. Musk said that each Starship flight to orbit and back costs a TOTAL of $2m, including fuel and operational overhead. The currently expendable 2nd stage on Falcon 9 costs several times more than that, and Blue Origin's 2nd stage should be in the same cost category.
If you had dozens to hundreds up there, you could probably do some useful interferometry even with small telescopes.
Also, this being a composite image, I'm assuming it is from multiple exposures. Can the software that merges the images together also filter out the satellites? And how is it that just 60 satellites do this, but the 2000 or so currently in orbit don't exhibit this behavior?
Additionally, SpaceX promised to do their best to lower the albedo of this second set of sats, which, based on side-by-side pictures of the two stacks of sats, they seem to have done.
In my opinion, there's definitely reason to be afraid of the effects of Starlink on astronomy, but SpaceX also seems reasonably receptive to making changes in order to lessen the impact.
Communications development for Mars before even having a launch vehicle for Mars payload delivery is putting the cart before the horse. You don't need a global network of satellites to communicate with Mars.