Hacker News new | past | comments | ask | show | jobs | submit login

They used to launch 60 satellites per launch, now they launch 21. I'm guessing this hardware is a big part of the reason why. It needs a folding antenna much larger than the ones used for the broadband service.



This phone texting support hardware is tiny compared to the size and weight increase the v2 sats took on for their primary mission, broadband capacity, which went from from about 20 Gbps per bird in the v1 config to as much as (estimated) 80-100 Gbps per sat in this v2(mini) config. It was this increase that made the sats that much heavier and cut the total number per launch about in half.


The hardware for this, is as far as we know, not on any of the currently flying satellites. They were explicitly downsized to fit on Falcon 9 and thusly named "v2 mini". Most likely the major thing stripped off was the large folding antenna that would be required for cell phone signal.

The move from 60 to 22 (or 21) satellites was to maximize the normal Starlink service on the satellites. (21 are on board when launched from the west coast and 22 are on board when launched from the east coast.)


There was an article recently saying that they'd be testing the service in December [1]. I would've taken that to mean that the hardware is up there and just undergoing preparation, but this is SpaceX, they launch like every 4 days, I guess they very well can have not launched the ones with this functionality yet.

[1] Can't find the one I had read, but this one published just a few minutes ago also mentions this https://cordcuttersnews.com/spacexs-starlink-is-looking-for-...


I’m not seeing your claim in that linked article unless I missed it. It talks about v2 satellites, which have yet to launch.


Search for "December" in the text. Unless you're referring to the bit about if the satellites are already up there, which is just my own thoughts rather than a statement of fact.


Okay I see that now, but I think that's referring back to earlier in the article where it says:

> The companies had targeted beta trials at the end of this year following the launch of Starlink V2 satellites, according to CNET.

Which also contains a link to an article from 2022 that was published during the original announcement of the T-Mobile partnership. However those satellites didn't get launched because of the delays of Starship.


LEO-to-mobile use cases don't usually need a reflector dish: Iridium and Globalstar don't have one either. That's more of a feature of GEO-to-mobile (e.g. Inmarsat, Thuraya, Skyterra etc.)


Iridium and Globalstar don't communicate with unmodified regular cell phones.


They do, these days: The iPhone 14 and above can use Globalstar for bidirectional text messaging (for emergencies only for now), and Qualcomm has announced something similar for Iridium.

Other devices they do communicate with (for text messaging) are also cell phone sized or smaller and don't have external antennas.


As you know, the recent iPhones have special hardware modifications to enable that communication on different frequency bands, with significantly higher transmit power and lower data rates than regular unmodified cell phones, which is what determine the necessary antenna size.


> As you know, the recent iPhones have special hardware modifications [...] with significantly higher transmit power [...]

I don't know that – do you, and if so, from where?

But what I do know is that it's possible for very small devices to directly talk to LEO and even geostationary satellites with transmit powers comparable to terrestrial applications.


https://fcc.report/FCC-ID/BCG-E8141A/6095651.pdf

L-Band is used exclusively for the satellite uplink. The power is significantly more than the majority of LTE/5G bands, but I see it's not the highest, actually that crown goes to GSM. Interesting. Seems like the much bigger reason that they don't need a large antenna is the other factor I mentioned, data rate, which is extremely low, while Starlink is targeting 2-4 Mbps. Plus there's the requirement to point the phone at the sky, while Starlink has stated that their service won't require that and should work from a pocket or inside a car. SpaceX has also specifically described the required satellite antennas as "really quite big".

Also, I'm not suggesting that the antenna is a "reflector dish" as you specified. It's a phased array, just much larger than the ones on previous Starlink satellites.


Oh, thank you for that link! I’ve been struggling to find exactly that FCC filing :)

I’m not surprised that GSM uses higher transmit power than Iridium/Globalstar actually – GSM phones were basically portable microwave ovens compared with modern mobile radio technologies.

But if you look at the table closely, the 5G C-band frequencies have even higher power outputs than Globalstar/L-band too! That's presumably because they are so wide, though (channels in band n77/n78 can be up to 100 MHz wide; compared to > 2W for 0.2 MHz wide channels for GSM, that's nothing).

Also, for bidirectional communications path losses are usually symmetric, and since satellites are usually severely power-limited compared to terrestrial base stations, blasting dozens of watts of transmit power from the mobile device wouldn't help with receiving the satellite's response.

> Also, I'm not suggesting that the antenna is a "reflector dish" as you specified.

Ah, I misunderstood then; I thought that's what you meant by "foldable antenna". I thought phased arrays are mostly solid-state (at least they are on comparable last-gen LEO satellites). But it makes sense for Starlink to do that; every bit of antenna aperture presumably helps when trying to talk to unmodified 4G mobile devices from space.


The iPhone 14 and above are not "unmodified". They have new specific hardware on them, otherwise Globalstar would work with the iPhone 13 and all Android phones.


V2 sats are a lot bigger




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: