-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
SpaceX is continuously upgrading their satellites, so I don't understand this distinct phase nomenclature. Once the laser sats are validated, it's not like they are planning to still send up nonlaser ones to fulfill some 'Phase 1' goal. The plan was always to have lasers, development just lagged launches.
Yeah--I think it was really just a statement of the obvious that the majority of the existing sats will not be compatible with the ISL network.
More interestingly, it will be interesting to see how SpaceX fills out the constellation around the non-ISL sats and how they advertise latency in the interim. Presumably the ISL sats will fill in the planes between the presumably evenly spaced non-ISL planes; The obvious issue is that until full ISL coverage (and likely redundant ISL coverage) is available at a latitude, ISL-level latency won't be available for all users all the time, since their link may have to go through a non-ISL satellite. Given the relatively limited coverage provided by the 6 or 700 or whatever on-station Starlinks I'd hand wave that 1000-1500 ISL sats will be required to cover high latitudes (more or less the beta latitudes) and 2500-3000 ISL sats or more will be required to cover down to the equator.
Non-sequitor, given the to-date rate of ~1000 sats per year (+ acceleration, and accounting for orbit raising) that's probably beginning of '22 for high latitude ISL coverage and end of '22 for full ISL coverage.
Anyway, the potentially more complicated issue is that with operational non-ISL sats AND some guaranteed ISL-level latency for the user, SpaceX will need to know the latency requirement of a user's data package (eg, Netflix = slow, Zoom = fast) in order to route the data through the proper satellite. If that's an issue--I'm not a network engineer but it seems like it could be a privacy problem--the bookend is that ISL level latency can't be guaranteed for users until the non-ISL sats are deprecated.
Buckminster
Well-Known Member
Always happy to see someone passionate about a topic produce and publish their thoughts, but honestly, not a lot of insight here. Mostly just a compilation of existing information (tweets, articles, etc.) that's pretty well disseminated here in this forum. Disappointingly, the competition slides boiled down to "OneWeb, but I don't think their credible. (emphasis on period)" and "Elon thinks Kuiper is a copy cat".
Looks like Jupiter 3 is getting delayed. (Again) As it turns out designing and building ridiculously complex mega GEOs is hard and unpredictable even without a pandemic. Among other things, this highlights the massive (if not self evident) upside to Starlink's approach of building a bunch of relatively simple satellites. J3 was ordered in 2017; looks like its going to be the better part of four years before its in service. That's a LONG time for business models to shift (Starlink was pretty fledgling in 2017) and, even though GEO technology is super lagging, that's still a long time for technology to advance.
Flip side, 500gbps is no slouch (to be fair, the three Viasat-3's under construction at Boeing are 1tbps each), and this single $400-500M J3 sat will probably provide CONUS bandwidth equivalent to ~1/3 of Starlink's initial ~1400-1500 satellite constellation.
In another article on the same topic Echostar is predictably leaning on their initial and recurring customer costs as an advantage over Starlink.
Somewhat related I recently read an analyst report that concluded essentially the same--LEO internet success is going to be all about bringing down the cost of user equipment. The good news for Starlink and others is that there's probably a good bit of opportunity in the technology/manufacturing of phased arrays.
In unrelated not-really-Starlink-competition-yet news, folks in the UK are taking a page out of Loon and are trying to moonshot airborne connectivity.
Flip side, 500gbps is no slouch (to be fair, the three Viasat-3's under construction at Boeing are 1tbps each), and this single $400-500M J3 sat will probably provide CONUS bandwidth equivalent to ~1/3 of Starlink's initial ~1400-1500 satellite constellation.
In another article on the same topic Echostar is predictably leaning on their initial and recurring customer costs as an advantage over Starlink.
Somewhat related I recently read an analyst report that concluded essentially the same--LEO internet success is going to be all about bringing down the cost of user equipment. The good news for Starlink and others is that there's probably a good bit of opportunity in the technology/manufacturing of phased arrays.
In unrelated not-really-Starlink-competition-yet news, folks in the UK are taking a page out of Loon and are trying to moonshot airborne connectivity.
Last edited:
Starlink approved to operate in Canada
Canada Approves SpaceX Starlink Internet to Operate - TeslaNorth.com
Canada Approves SpaceX Starlink Internet to Operate - TeslaNorth.com
dhanson865
Well-Known Member
I've posted in multiple threads how I did not expect starlink to work on a moving vehicle and got told by various posters that the dish can adjust thousands of times a second and motion of a vehicle wasn't an issue.
How about video proof instead of my word vs yours.
This is on an island near the US/Canada border so starlink sat coverage is already solid there. To be clear it's not me, someone else posted this video but it shows that the dish doesn't move quickly and the antenna array can't compensate for random movement.
How about video proof instead of my word vs yours.
This is on an island near the US/Canada border so starlink sat coverage is already solid there. To be clear it's not me, someone else posted this video but it shows that the dish doesn't move quickly and the antenna array can't compensate for random movement.
Last edited:
dhanson865
Well-Known Member
fwiw I've used the app and found that my back porch rail isn't an ideal mount location, obstruction partially from roof to the west and trees to the north. Walking around the yard I didn't have any unobstructed views. I'll need to either ridge mount it at the south end of the house (weighted mount to avoid penetration), J mount it on the south end of the house (penetrate wood that is part of the overhang), or just not do starlink at my house (I do have cable internet that is fast enough and reasonably reliable).
It's really nice to have the AR feature in the app to figure out where to mount.
The down side is that J mounting or ridge mounting is an overhang that's 2.5 stories above asphalt so it'd probably be a one way trip if I fell. I'd probably pay a professional to do that one for me if I do starlink. Also because of that inaccessibility (the risk of falling off the roof) I can't get my cell phone in that position to see if the view is fully unobstructed or partially obstructed ahead of time. I could get up on the roof and slowly back towards the drop off holding a phone up to the sky but that just isn't my idea of how I want to leave this earth.
I suppose if I rented a ladder truck from the local fire dept I could easily climb a ladder and hold the phone up.
Oh, yeah I suppose I could pole mount it on the porch to get it above the roof edge but the pole/mount might be more expensive than the j mount and paying an installer to just do it the obvious way on the end of the house. And again I can't get my phone into position to easily check for obstructions from whatever height the pole would add.
Or I can just dual WAN it and have the starlink on the back porch rail with minor obstructions if I'm avoiding the proper install.
It's really nice to have the AR feature in the app to figure out where to mount.
The down side is that J mounting or ridge mounting is an overhang that's 2.5 stories above asphalt so it'd probably be a one way trip if I fell. I'd probably pay a professional to do that one for me if I do starlink. Also because of that inaccessibility (the risk of falling off the roof) I can't get my cell phone in that position to see if the view is fully unobstructed or partially obstructed ahead of time. I could get up on the roof and slowly back towards the drop off holding a phone up to the sky but that just isn't my idea of how I want to leave this earth.
I suppose if I rented a ladder truck from the local fire dept I could easily climb a ladder and hold the phone up.
Oh, yeah I suppose I could pole mount it on the porch to get it above the roof edge but the pole/mount might be more expensive than the j mount and paying an installer to just do it the obvious way on the end of the house. And again I can't get my phone into position to easily check for obstructions from whatever height the pole would add.
Or I can just dual WAN it and have the starlink on the back porch rail with minor obstructions if I'm avoiding the proper install.
Last edited:
That test indicates nothing other than a system meant for fixed mounting doesn't handle non fixed mounting well.
Satellite tracking is primarily via the phased array beam steering, only gross pointing is done mechanically. Mechanical steering would drop the connection every time a sat goes out of field and it needed to reaquire.
An antenna system meant for static terrestrial use would not have the correct control loops nor IMU bandwidth (if it even has an accelerometer or gryo) for a dynamic situation.
Roughly, a 90 minute orbital period means 4 degrees per minute angular change for a satellite being tracked (really, it's faster, but I'm lazy). Rocking the vehicle is much more than that. (See also GPS receivers with static, vehicle, and plane modes). The base station beamwidth is only 1.5 degrees. https://licensing.fcc.gov/myibfs/download.do?attachment_key=1190019
Note: file requires appending .pdf
During the drive, it appears the vehicle ended up pointing a different direction (shadow of head cast on passenger door). The software would not normally be set up to handle this situation, especially if it is using a magnetometer for azimuth and an accelerometer for elevation.
Starlink (properly configured) works with planes.
SpaceX's Starlink satellite internet was tested by the US Air Force and the results are in
A simple scan technique would allow reduced data rate functionality in dynamic situations with existing hardware, but that software would be a liability since the system should not be moving. In terms of emissions and safety, an induced movement should be flagged as a sensor error and result in termination of transmission until position and lock can be reestablished (or for x time, or until reset).
Yeesh. Nonetheless, RIGHT NOW, Starlink does not work on a moving vehicle. And as you correctly point out, it seems the current hardware AND software cannot work on a moving vehicle. As far as the Air Force test goes, do you know the parameters of the test? Was it a simple straight line flight test? How many turns did it do?
Now, starlink may very well get hardware and software upgrades in the future to allow service on moving vehicles, but if so, passenger vehicles would probably be the last platform to be supported, if ever. I can see service for ships (once they get ISL links working) and airplanes, eventually, but I doubt passenger cars will ever happen.
dhanson865
Well-Known Member
Wow that's a lot of hand waving you are doing there. The dish in the video didn't move mechanically during the portions it lost signal and the electronic steering didn't keep up.
Then you say they could design a different setup to address that need. Sure, but the average person on these threads isn't looking for what might be, they are looking for what is. And what is doesn't handle motion at all right now.
And then you bring up planes, you know why it works on planes? Because planes fly pretty close to straight and level. If they turn they turn in gentle arcs. This plane isn't a fighter jet, it's not a biplane for an airshow. It's not going to be doing loops, rolls, spins, and hammerheads. It's just going to climb to the desired testing altitude at a comfortable rate and fly mostly level for a while during the test.
dhanson865
Well-Known Member
back on the subject of the dish the surface that faces the sky is practically flat but thermal imaging gives a peek at whats under that outer layer and it seems to do nicely for melting snow off the face.
the person that took the thermal image said Ambient Air Temperature: 20° F Terminal Front Temperature: 32° to 40° F Terminal Back Temperature: 5° F
someone else did a rough count and said it is 44 elements across and about 1500 total.
the person that took the thermal image said Ambient Air Temperature: 20° F Terminal Front Temperature: 32° to 40° F Terminal Back Temperature: 5° F
someone else did a rough count and said it is 44 elements across and about 1500 total.
Last edited:
As an ex off-shore worker, starlink just sounds amazing on paper.
I'm just imagining being in the middle of the ocean, with an internet connection you can actually use for downloading, streaming maybe even gaming to pass the time.
Hope they succeed, as this could be literally revolutionary...
I'm just imagining being in the middle of the ocean, with an internet connection you can actually use for downloading, streaming maybe even gaming to pass the time.
Hope they succeed, as this could be literally revolutionary...
Yeah, thst's what I wrote: the fixed terrestrial system does not work on a moving platform. Nor was I making a case for widespread Tesla support of moving vehicles.
Rather, I was addressing this statement:
The system would not respond to that rocking via mechanical movement. The beam steering can happen on a pulse by pulse basis, that means instead of a 1.5 degree beam width, one could scan a 4.5 degree section at 1/10 the rate and recenter based on max strength (akin to MPPT on solar). That is a simple (but wasteful) method to handle platform movement with zero extra hardware.
Wow that's a lot of hand waving you are doing there. The dish in the video didn't move mechanically during the portions it lost signal and the electronic steering didn't keep up.
Then you say they could design a different setup to address that need. Sure, but the average person on these threads isn't looking for what might be, they are looking for what is. And what is doesn't handle motion at all right now.
And then you bring up planes, you know why it works on planes? Because planes fly pretty close to straight and level. If they turn they turn in gentle arcs. This plane isn't a fighter jet, it's not a biplane for an airshow. It's not going to be doing loops, rolls, spins, and hammerheads. It's just going to climb to the desired testing altitude at a comfortable rate and fly mostly level for a while during the test.
It's not hand waving, it's how phased array systems work. (Sude bar, here's one that tracks space debris Eglin AFB Site C-6 - Wikipedia)
If your argument is that the current setup meant for fixed ground mounting does not support a moving platform (other that a linear rail), I agree that is likely the case. The steering algorithm would not need nor have a motion compensation component.
If you are saying that lack of support indicates the same basic system is incapable of supporting a moving vehicle, I disagree. Even with no additional sensors (they would be <$10, if not already included), the system could handle motion with only a software change.
The plane reference was only to show that such system do work. You may call a turn a gentle arc, but the reality is, a standard bank turn involves a 30 degree roll along with 3 degrees per second of yaw/ heading change (which also changes the effective elevation angle relative to the reference frame/ aircraft). Sure, we don't know how the test was run which is why it was not a primary source.
The wording of beamed to the cockpit makes me think they actually just put the antenna in the cockpit to avoid any FAA modification issues. The standard bank angle is more than the normal beam steering which would limit coverage for a single antenna system during manuvers. A straight and level flight is practically a non-test since there is virtually no change in orientation due to velocity.
Edit: in the case of a low density constellation which requires a specific azimuth/non 90 elevation, the current antenna would need to physically steer to remain grossly pointed at the correct section of sky during turns. That would need a more robust mechanical system (and slip rings) than the current point once systems likely have. The maneuvers in the video did not appear to require that level of correction.
Last edited:
Ouch. Melting snow. So Starlink would not work well in an off-grid location with limited energy available then? No hope of the user terminal using 10-20W or so like a 4G modem or router?
No way there would be enough gain to close a useable link.
I don’t know if anyone posted any nameplate ratings for the UT (let alone current draw) but I’d guess nominal power is somewhere in the low hundreds of watts with the ability to ramp up close to a kw maybe (?) in bad weather or with otherwise unfavorable links.
The video a few posts up showed it at around 112 Watts.
Are you including heating or high transmitter inefficiency in your kw high end range?
