Starlink Internet Connection equipment - Home/Office user

[thirtyone]

Okay, so here's the deal.
1) Not sure how many ICs per element. I'm sure SpaceX has some tricks up their sleeve (was this in the patent?)
2) If it's silicon and your design is good (high yield, low test cost), it's going to be about 10-20 cents per square mm at very high volume with a moderately mature process (probably 65nm these days). The difficult thing to pay off is usually the mask set, which is a one time cost but can be in the millions.
3) Any existing IC that costs in the $10s of dollars either
a) is making an absurd profit margin per die and / or paying absurd licensing fees
b) has a huge die (the largest ones are going to be CPUs - i9-7900x is ~300mm^2)
c) has awful yields because it's pushing the process too hard and/or is very expensive to test (test machine time can quickly exceed die production cost)
d) is so low volume that they're still busy paying NREs (e.g. if it's a newer process, that $1million mask will still cost $20 per chip to pay off at 50k parts, a substantial volume for many aerospace parts)
4) If they're using III-Vs (GaN, etc.) it's going to be a hell of a lot more expensive per mm^2 but I don't have high volume numbers. To be honest most people doing high volume products probably avoid these in the first place. If SpaceX is using III-Vs my guess is their first order of business is to figure out how to get those parts usable on silicon.

---
Questions for anyone who has one and wants to tear it down:
- How many ICs are mounted to each element?
- What's the die size?
- Take some die images

Those'll pretty much point you to the actual production cost of this thing

[Zed_Noir]

<snip>
Questions for anyone who has one and wants to tear it down:
- How many ICs are mounted to each element?
- What's the die size?
- Take some die images
<snip>

Pay someone 600 bucks to installed a terminal at Sandy Munro's office. 

[vsatman]

They'd have to be idiots to subsidize to that extent,

StarLink was conceived in 2015, and then much was unclear, in the choice of the orbit altitude, SpaceX reduced it 2 (two) times!!!! (in 2016 they believed that 1100 km was correct), the architecture of the constellation  changed 4 times!
This is the typical path for Elon Musk by "make - test - change"
.
And now, no matter how much the terminal's cost (most important - it works!!!), they need to produce and sell them at the "right market price" at which they can sell 200,000 per year and 1 million devices in max 3 year in order to receive money from the service, and hope that they can reduce the cost of the terminal, enter other markets , RDOF, orders from the government etc.
 Any other option is worse ..

P.S. «satellite communications it is not as simple as rocket science!   

[Robotbeat]

I'm fairly certain SpaceX *has* indeed reduced the costs of these phased arrays dramatically from the industry standard that you'd expect.

5G uses similar frequencies.

SpaceX could potentially get into the 5G base station business.

[su27k]

News Analysis | SpaceX has a lot riding on Starlink’s $499 “UFO on a stick”

Earlier this month, Hughes Network Systems President Pradman Kaul suggested during an EchoStar investor’s call that Starlink’s antenna probably costs SpaceX between $1,000 and $1,500 per unit. Analyst Tim Farrar of TMF Associates, told Business Insider he believes the $499 terminal is actually costing SpaceX closer to $1,500 to $2,000 to produce.

[woods170]

News Analysis | SpaceX has a lot riding on Starlink’s $499 “UFO on a stick”

Earlier this month, Hughes Network Systems President Pradman Kaul suggested during an EchoStar investor’s call that Starlink’s antenna probably costs SpaceX between $1,000 and $1,500 per unit. Analyst Tim Farrar of TMF Associates, told Business Insider he believes the $499 terminal is actually costing SpaceX closer to $1,500 to $2,000 to produce.

Well that article was a let down, certainly not up to usual SN standards. It quotes Tim Farrar, THE most UNreliable source with regards to external "insight" into Starlink.

Tim is well known for being highly critical of Starlink. Just check his Twitter account. He is also not without bias, often basing his (inaccurate) conclusions on the knowledge he retained from his time at Teledesic. But in doing so he regularly fails to take into account that 20 years of progress have been made since then.

[RotoSequence]

News Analysis | SpaceX has a lot riding on Starlink’s $499 “UFO on a stick”

Earlier this month, Hughes Network Systems President Pradman Kaul suggested during an EchoStar investor’s call that Starlink’s antenna probably costs SpaceX between $1,000 and $1,500 per unit. Analyst Tim Farrar of TMF Associates, told Business Insider he believes the $499 terminal is actually costing SpaceX closer to $1,500 to $2,000 to produce.

Well that article was a let down, certainly not up to usual SN standards. It quotes Tim Farrar, THE most UNreliable source with regards to external "insight" into Starlink.

Tim is well known for being highly critical of Starlink. Just check his Twitter account. He is also not without bias, often basing his (inaccurate) conclusions on the knowledge he retained from his time at Teledesic. But in doing so he regularly fails to take into account that 20 years of progress have been made since then.

Biased he may be, but he's got a point; a directional antenna element has to be a very, very low cost device to hit even $1000 per unit with every other part of the dish factored in. I figured $500 was an effective user lock-in price (via sunk costs), with the service cost being used to amortize the remaining dish expense.

[DigitalMan]

News Analysis | SpaceX has a lot riding on Starlink’s $499 “UFO on a stick”

Earlier this month, Hughes Network Systems President Pradman Kaul suggested during an EchoStar investor’s call that Starlink’s antenna probably costs SpaceX between $1,000 and $1,500 per unit. Analyst Tim Farrar of TMF Associates, told Business Insider he believes the $499 terminal is actually costing SpaceX closer to $1,500 to $2,000 to produce.

Well that article was a let down, certainly not up to usual SN standards. It quotes Tim Farrar, THE most UNreliable source with regards to external "insight" into Starlink.

Tim is well known for being highly critical of Starlink. Just check his Twitter account. He is also not without bias, often basing his (inaccurate) conclusions on the knowledge he retained from his time at Teledesic. But in doing so he regularly fails to take into account that 20 years of progress have been made since then.

Biased he may be, but he's got a point; a directional antenna element has to be a very, very low cost device to hit even $1000 per unit with every other part of the dish factored in. I figured $500 was an effective user lock-in price (via sunk costs), with the service cost being used to amortize the remaining dish expense.

Do you suppose Greg Wyler is the only one able to solve this problem?

https://twitter.com/greg_wyler/status/1081989389453066240

[RotoSequence]

Do you suppose Greg Wyler is the only one able to solve this problem?

I'd naively figured it was still unsolved. Guess I was wrong about that.

[woods170]

Do you suppose Greg Wyler is the only one able to solve this problem?

I'd naively figured it was still unsolved. Guess I was wrong about that.

It is one of the reasons why I pointed out that Tim Farrar is a completely unreliable source with regards to insight into Starlink. He estimates a single Starlink terminal to cost between $1,500 and $2,000. That estimate is not supported by any facts.

[gongora]

Do you suppose Greg Wyler is the only one able to solve this problem?

https://twitter.com/greg_wyler/status/1081989389453066240

Did you ever hear more about that antenna after the single tweet?  Did it even work?

[woods170]

Do you suppose Greg Wyler is the only one able to solve this problem?

https://twitter.com/greg_wyler/status/1081989389453066240

Did you ever hear more about that antenna after the single tweet?  Did it even work?

Update by Wafer CEO 3 months after Greg's tweet: https://www.bcsatellite.net/blog/wafer-thin/

[DreamyPickle]

Found a teardown:


Part 1 does not open the actual phased array antenna.

[RotoSequence]

It is one of the reasons why I pointed out that Tim Farrar is a completely unreliable source with regards to insight into Starlink. He estimates a single Starlink terminal to cost between $1,500 and $2,000. That estimate is not supported by any facts.

Not to dispute his wrongness but for the sake of understanding and estimating costs, if one assumes equivalent component costs and linear scaling by element count, the BOM for the Starlink antenna array should be around $100 for that part/sub-assembly.

[su27k]

Some sleuthing about router serial number on reddit, sounds like the serial number format is YYWW + a 6 digit sequential number, with the highest sequential number reported at 37xx.

[vsatman]

Do you suppose Greg Wyler is the only one able to solve this problem?

I'd naively figured it was still unsolved. Guess I was wrong about that.

It is one of the reasons why I pointed out that Tim Farrar is a completely unreliable source with regards to insight into Starlink. He estimates a single Starlink terminal to cost between $1,500 and $2,000. That estimate is not supported by any facts.

there is no point in discussing  with Tim as he is just repeating what other profis in the industry have said. The terminal has 1600 chips. There is information that this is As-Ga MMIC (Monolithic microwave integrated circuit)
(https://www.digitimes.com/news/a20200820PD207.html), WIN Semiconductors (https://www.winfoundry.com/en-US) Taiwan.
And the fact that after the words of Greg Wyler in 2019 (who personally invested in this company and possibly wanted to attract other investors) nothing happened about "the $ 15 miracle in 2020" , tells me that $ 2,000 is possibly a bottom estimate ..
I have been hearing from new companies in their first year for 5 years now about the future breakthrough and prices of 1000 USD per ESA FPA terminal. Only after 2 years the authors of these presentations cannot be found anywhere ..

And my congratulations to Space X for a great job - the terminal is really great!!!!!

[RotoSequence]

Do you suppose Greg Wyler is the only one able to solve this problem?

I'd naively figured it was still unsolved. Guess I was wrong about that.

It is one of the reasons why I pointed out that Tim Farrar is a completely unreliable source with regards to insight into Starlink. He estimates a single Starlink terminal to cost between $1,500 and $2,000. That estimate is not supported by any facts.

there is no point in discussing  with Tim as he is just repeating what other profis in the industry have said. The terminal has 1600 chips. There is information that this is As-Ga MMIC (Monolithic microwave integrated circuit)
(https://www.digitimes.com/news/a20200820PD207.html), WIN Semiconductors (https://www.winfoundry.com/en-US) Taiwan.
And the fact that after the words of Greg Wyler in 2019 (who personally invested in this company and possibly wanted to attract other investors) nothing happened about "the $ 15 miracle in 2020" , tells me that $ 2,000 is possibly a bottom estimate ..
I have been hearing from new companies in their first year for 5 years now about the future breakthrough and prices of 1000 USD per ESA FPA terminal. Only after 2 years the authors of these presentations cannot be found anywhere ..

And my congratulations to Space X for a great job - the terminal is really great!!!!!

If they're running Gallium Arsenide ICs on those things, they're going to be bound by the fabrication cost of the wafer and the size of the IC. A fully developed GaAs wafer, I believe, runs around $1000, but I don't really know for sure; that might be too low since silicon CMOS on moderate processes run around that price. We need IC sizes and wafers to know the real cost of the antenna bits.

Cursory research on the size of antenna elements suggest that a 6" wafer should yield something around 8000 devices, for a guesstimated cost to SpaceX of something between 10 and 20 cents per IC.

[copper8]

Everything gets cheaper with volume.

[RotoSequence]

Everything gets cheaper with volume.

Integrated circuits are bound pretty tightly by any given fabrication line's throughput limits. Volume can reduce costs there, but it'll do more to amortize the cost of photomasks and setup costs than most other things involved with the job.

[LouScheffer]

There is information that this is As-Ga MMIC (Monolithic microwave integrated circuit)
(https://www.digitimes.com/news/a20200820PD207.html), WIN Semiconductors (https://www.winfoundry.com/en-US) Taiwan.

I'd say there is almost surely no GaAs in the user terminal.   Even a cheap GaAs chip is expensive by the standards needed, because GaAs processing is expensive, foundries can't use big wafers, and yield is low.  It's hard to imagine a GaAs chip for less than a dollar.  The cellphone people ditched GaAs the first chance they got, and they only need 1 chip of it.  The key to making the user terminal work will be to make the needed technology in a stock CMOS process.

On the other hand, GaAs in the satellite would be expected (just as satellites use GaAs solar cells).  They make more efficient RF power amplifiers than CMOS, which is critical in a satellite since lower efficiency increases the needed solar arrays, the size of the batteries, and the rejection of dissipated power.  These are less important on the ground.  Plus, of course, cost is less of a concern for a satellite, since there are many fewer of the them.

Overall I'd be extremely surprised to see GaAs in the user terminal, and equally surprised if it was not used in the satellites.