Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

SpaceX Internet Satellite Network: Starlink

This site may earn commission on affiliate links.
I'm going with Funny on that one - Funny as in the driver, the person that decided to attach a starlink dish to their car :)
It would make a lot of sense to me for mobile purposes. Such as putting it on top of an RV or private "bus" vehicle especially when getting to a location at night to use it. I've certainly seen bigger things on rec & tour vehicles. But on a car? yup, that would look strange.
 
View attachment 684668

from https://starlink.sx/

gives you an idea how close to complete the first layer of the net is.

Red is wrong altitude, yellow is wrong position compared to the rest of the net.

just a few more launches and it'll fill in nicely.

3 more launches in August and 3 more in September should fill the majority of the gaps.

while the ones launched in September won't be coming on line immediately the ones from June, July, and August will be filling in major gaps right as the ones in September launch.

3 weeks later and almost all the gaps are filled in, today I realized the proper spacing is every 5 degrees and the bottom bar chart should only have 72 bars if sats were all in final positions. As of this morning there were 6 bars out of position mostly with one sat each. The actual gaps in the top portion of the image are a better representation of actual gaps in service.

I didn't mark all of them but notice how the short bar that looks like a huge gap near 260 is really just one sat sitting between two basically full orbits, not an actual huge gap in service like the mostly empty bar implies.

The one big gap is around 141 and a single launch would provide enough sats to fill that and then some.

1628352234151.png
 
Last edited:
"The second generation is elliptical. See E33/34 field: 0.29 m minor and 0.48 m major diameter. That's the dimensions of the antenna radiating elements not its housing."

Interesting. Caveat that I am in no way an antenna person let alone a phased array person, but this, In general:

1. Reduces the number of radiating elements
2. Reduces the number of unique beams that can be formed
3. Reduces the beam forming processing required
4. Increases the size of each beam
5. Changes the shape of each beam from ~conical to ~elliptical
6. Reduces the FOV/half angle is the array (can see less sky)

I would guess that #3 is most important here—the number of elements is likely ~linearly proportional to the processing (and effectively, the number of processors), and the beam forming chips (ASICs, IIRC) are almost certainly the most expensive line item on the BOM. So, fewer chips = lower cost. It’s unlikely this also enables them to drop in an ~OTS fpga, but opening that option up as a result of lower processing demand would manifest as even lower cost.

The downside of fewer and larger beams is likely a minimal issue (that’s rather self-evident since it’s what they’re doing) but theoretically that means each of the UT beams will see more sky and, importantly relative to my speculation here, a higher probability of more satellites in each beam. To be clear I have no idea what their frequency reuse plan or protocol actually is so, again, heavy speculation here, but in general this phenomenon would likely lead to lower user throughput.

Flip side, while the smaller FOV/half angle of the array means it can’t look at low elevation satellites, that’s also the worst case geometry for seeing more sats/beam and the hardest to form useful beams. Low angles are also the most ranging loss (and so, theoretically the hardest to maximize throughput)…so…from an abstracted ‘value to the system’ perspective, those low angle geometries are exactly what you’d eliminate first if you were trying to iterate for total system efficiency. Clever.
 
  • Like
Reactions: scaesare
3 weeks later and almost all the gaps are filled in, today I realized the proper spacing is every 5 degrees and the bottom bar chart should only have 72 bars if sats were all in final positions. As of this morning there were 6 bars out of position mostly with one sat each. The actual gaps in the top portion of the image are a better representation of actual gaps in service.

I didn't mark all of them but notice how the short bar that looks like a huge gap near 260 is really just one sat sitting between two basically full orbits, not an actual huge gap in service like the mostly empty bar implies.

The one big gap is around 141 and a single launch would provide enough sats to fill that and then some.

The reddit hive mind feels my last statement there is misleading as there are already enough sats from the May 2021 launches to fill those gaps.

The number of gaps is less than 60 so if you use one launch as a unit of measurement to mean 60, then my statement stands with a different connotation, they don't have to launch any more sats, but if they did "a single launch would provide enough sats to fill that and then some."

The point wasn't to say they need to launch more sats ASAP, the point was to compare how close to a full first shell they are. Out of thousands of positions filled they have less than 60 to go.
 
ookla_satellite_internet_comparison_us_0821.png


and
ookla_satellite_internet_monthly_comparison_us_0821.png



Starlink was the only satellite internet provider in the United States with fixed-broadband-like latency figures, and median download speeds fast enough to handle most of the needs of modern online life at 97.23 Mbps during Q2 2021 (up from 65.72 Mbps in Q1 2021). HughesNet was a distant second at 19.73 Mbps (15.07 Mbps in Q1 2021) and Viasat third at 18.13 Mbps (17.67 Mbps in Q1 2021). None of these are as fast as the 115.22 Mbps median download speed for all fixed broadband providers in the U.S. during Q2 2021, but it beats digging twenty miles (or more) of trench to hook up to local infrastructure.

from speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/
 
  • Like
Reactions: mongo
Covid hospital cases are preventing SpaceX from launching as many rockets as they would like. SpaceX can't buy enough oxygen because it's being used for covid patients.


Shotwell confirms that SpaceX paused Starlink launches until it can get laser intersatellite links in place on all future Starlink satellites. Next Starlink launch now planned in about three weeks. Shotwell says one supply chain issue for SpaceX is a lack of liquid oxygen because of demands to treat COVID-19 patients. Will impact launch plans, she says.


SpaceX president Gwynne Shotwell says the company has “two big issues” right now from global shortages. 1. Chips & 2. Liquid oxygen (for launches)

Shotwell: “We certainly are going to make sure hospitals have the liquid oxygen they need – but for anybody that has liquid oxygen to spare, would you send me an email?”


Shotwell says one supply chain issue for SpaceX is a lack of liquid oxygen because of demands to treat COVID-19 patients. Will impact launch plans, she says.