Starlink Internet Connection equipment - Home/Office user

[envy887]

Detroit area -

Just received invitation a few minutes ago and signed up immediately ($99 deposit). Received confirmation email - "Starlink will begin offering service in your area beginning mid to late 2021. Orders will be fulfilled on a first-come, first-served basis. You will be notified via email prior to shipment, and you will be charged the remainder of your balance once your kit ships."

I'm happy but very concerned about reports that the dishy requires 100W.  If that's continuous 24/7/365 I'll be signing off quickly.  Does anyone have any deeper info on power consumption?

That 100W sounds not much more than what my Dish/Router/Modem pulls right now for standard satellite, so it's competitive. I also think, but do not know for certain, that Starlink does not pull it's full wattage rating unless it is running its anti-ice heater.

But all in all, we're still talking to things in orbit here, so it's going to use some power.

My Starlink has averaged 98.4 watts over the last 11 days since I put it on the power monitor, with a peak draw of 185 W. It's been very cold, between -10 and 24 F, and snowy so I think the dish heater has been running most of that time.

it will be the same in the summer. (probably even higher consumption "because of physics"). Transmission modules generate plenty of heat. In fact Starlink dish is quite economical and is "top end" in the current design line.

That doesn't make much sense. If the dish already drew enough power during transmission to melt snow, why would they bother with "snow melt mode"? How would they implement an operating mode that melts more snow without drawing more power?

[Nomadd]

That doesn't make much sense. If the dish already drew enough power during transmission to melt snow, why would they bother with "snow melt mode"? How would they implement an operating mode that melts more snow without drawing more power?

It's not transmitting all the time. And not always at full power when it is transmitting.

[envy887]

That doesn't make much sense. If the dish already drew enough power during transmission to melt snow, why would they bother with "snow melt mode"? How would they implement an operating mode that melts more snow without drawing more power?

It's not transmitting all the time. And not always at full power when it is transmitting.

No doubt true, but if snow melt mode involves increasing broadcast power or transmit time, shouldn't that increase power draw?

[Nomadd]

That doesn't make much sense. If the dish already drew enough power during transmission to melt snow, why would they bother with "snow melt mode"? How would they implement an operating mode that melts more snow without drawing more power?

It's not transmitting all the time. And not always at full power when it is transmitting.

No doubt true, but if snow melt mode involves increasing broadcast power or transmit time, shouldn't that increase power draw?

No idea. Heat mode might be uneeded at full Tx power and just something to draw more current if it's available. Maybe even have one of those little solid state heat pumps you find in coolers.

[thirtyone]

Heaters are really cheap to do. Since they're custom ICs, I wouldn't be surprised if there's just a register / command per-chip into on-die joule heating to warm up the antenna. Would be pretty cheap, uniform, and efficient due to selectivity since you can individually control the temperature of each die on the board. Probably adds little to nothing to cost (marginal or NRE given that the dies are already custom) in this arrangement. If you're really fancy you could inefficiently bias the RF power transistor on board with no carrier signal to reuse the RF transistor for heat. Might save a few fractions of mm^2 on die. Can be difficult to characterize across bias conditions though. Again best part is no part, which is partly why I don't expect seeing a separate component on the board to perform this function. This is especially true on a very high count board like the Starlink antenna, where each extra component adds up for assembly cost.

Still doubt there are or will be any solid state heat pumps in the antenna. That's from the perspective of what I believe is Elon/SpaceX's engineering philosophy. Solid state heat pumps are awfully inefficient, not incredibly robust (the can in fact crack after extended use), and you still have to throw that waste cold somewhere (basically, maybe you melt the front of the antenna but you risk freezing the rear unless there's a good heatsink). Best part is no part.

[DistantTemple]

Heaters are really cheap to do. Since they're custom ICs, I wouldn't be surprised if there's just a register / command per-chip into on-die joule heating to warm up the antenna. Would be pretty cheap, uniform, and efficient due to selectivity since you can individually control the temperature of each die on the board. Probably adds little to nothing to cost (marginal or NRE given that the dies are already custom) in this arrangement. If you're really fancy you could inefficiently bias the RF power transistor on board with no carrier signal to reuse the RF transistor for heat. Might save a few fractions of mm^2 on die. Can be difficult to characterize across bias conditions though. Again best part is no part, which is partly why I don't expect seeing a separate component on the board to perform this function. This is especially true on a very high count board like the Starlink antenna, where each extra component adds up for assembly cost.

Still doubt there are or will be any solid state heat pumps in the antenna. That's from the perspective of what I believe is Elon/SpaceX's engineering philosophy. Solid state heat pumps are awfully inefficient, not incredibly robust (the can in fact crack after extended use), and you still have to throw that waste cold somewhere (basically, maybe you melt the front of the antenna but you risk freezing the rear unless there's a good heatsink). Best part is no part.

Erm? Its a while since I read on NSF about the chips.... are there any that are "processors", where just running some intensive code would heat them up! And if so a software update, or maybe a firmware update could do it! .

[daavery]

the problem is the 100W POE power budget - the dish already uses all of it

[Redclaws]

the problem is the 100W POE power budget - the dish already uses all of it

Given that they’re making the equipment, they could likely pretty easily do a small redesign to exceed that spec.  Not sure what the limit would be, but they don’t have to be interoperable with other POE stuff.  They might not be able to raise it much without signalling issues, though, and it could drive up costs.

Just basically indicating that the POE spec is not necessarily a constraint, since they’re not interoperating using it.  It’s within their own system.

[TheRadicalModerate]

the problem is the 100W POE power budget - the dish already uses all of it

Given that they’re making the equipment, they could likely pretty easily do a small redesign to exceed that spec.  Not sure what the limit would be, but they don’t have to be interoperable with other POE stuff.  They might not be able to raise it much without signalling issues, though, and it could drive up costs.

Just basically indicating that the POE spec is not necessarily a constraint, since they’re not interoperating using it.  It’s within their own system.

I don't think Starlink wants to be in the business of designing custom PoE chipsets.  I don't think they want to have to supply thicker than 23-AWG wires in their cables.  I don't think they want to fight with UL and the CE marking people in the EU.

This isn't spacecraft design; it's consumer electronics.  You can't go wild and crazy.

[Slarty1080]

the problem is the 100W POE power budget - the dish already uses all of it

Given that they’re making the equipment, they could likely pretty easily do a small redesign to exceed that spec.  Not sure what the limit would be, but they don’t have to be interoperable with other POE stuff.  They might not be able to raise it much without signalling issues, though, and it could drive up costs.

Just basically indicating that the POE spec is not necessarily a constraint, since they’re not interoperating using it.  It’s within their own system.

I don't think Starlink wants to be in the business of designing custom PoE chipsets.  I don't think they want to have to supply thicker than 23-AWG wires in their cables.  I don't think they want to fight with UL and the CE marking people in the EU.

This isn't spacecraft design; it's consumer electronics.  You can't go wild and crazy.

It will depend on how much a bespoke chip gives them and how much money it would save. If both answers are "not a lot" then its not going to happen. But if it gives them a lot and saves them a lot then they absolutely would do it.

[Faerwald]

Peak power draw may be the same but time of peak power draw may not be. Could just be as simple as making the integrated processors do some wasted calculations.

[OTV Booster]

Peak power draw may be the same but time of peak power draw may not be. Could just be as simple as making the integrated processors do some wasted calculations.

Lol. Hook em up to do SETI work with waste cycles.

[dondar]

Detroit area -

Just received invitation a few minutes ago and signed up immediately ($99 deposit). Received confirmation email - "Starlink will begin offering service in your area beginning mid to late 2021. Orders will be fulfilled on a first-come, first-served basis. You will be notified via email prior to shipment, and you will be charged the remainder of your balance once your kit ships."

I'm happy but very concerned about reports that the dishy requires 100W.  If that's continuous 24/7/365 I'll be signing off quickly.  Does anyone have any deeper info on power consumption?

That 100W sounds not much more than what my Dish/Router/Modem pulls right now for standard satellite, so it's competitive. I also think, but do not know for certain, that Starlink does not pull it's full wattage rating unless it is running its anti-ice heater.

But all in all, we're still talking to things in orbit here, so it's going to use some power.

My Starlink has averaged 98.4 watts over the last 11 days since I put it on the power monitor, with a peak draw of 185 W. It's been very cold, between -10 and 24 F, and snowy so I think the dish heater has been running most of that time.

it will be the same in the summer. (probably even higher consumption "because of physics"). Transmission modules generate plenty of heat. In fact Starlink dish is quite economical and is "top end" in the current design line.

That doesn't make much sense. If the dish already drew enough power during transmission to melt snow, why would they bother with "snow melt mode"? How would they implement an operating mode that melts more snow without drawing more power?

I don't work in SpaceX so I don't know the  details of the implementation but the preheating (full load with closed antenna gate) is rather standard feature on synthetic radars. I expect that they do continuous "static" load which is the best way to rise ambient temperature generally. And they have to do it regularly because the snow/ice don't wait until you connect to Internet.

[dondar]

the problem is the 100W POE power budget - the dish already uses all of it

Given that they’re making the equipment, they could likely pretty easily do a small redesign to exceed that spec.  Not sure what the limit would be, but they don’t have to be interoperable with other POE stuff.  They might not be able to raise it much without signalling issues, though, and it could drive up costs.

Just basically indicating that the POE spec is not necessarily a constraint, since they’re not interoperating using it.  It’s within their own system.

I don't think Starlink wants to be in the business of designing custom PoE chipsets.  I don't think they want to have to supply thicker than 23-AWG wires in their cables.  I don't think they want to fight with UL and the CE marking people in the EU.

This isn't spacecraft design; it's consumer electronics.  You can't go wild and crazy.

You are talking about Musk here. If he (in his "basement") will calculate that it is cheaper for SpaceX do it in-house they will design, find and build respective factories to make "it" in-house. Consumer electronics or not.

[Keldor]

Well that solar has to power two fridges, a freezer, a convection oven, induction hob, lights, instruments and nav gear, autopilot, pumps, water maker, toilets, washing machine, hot water maker, inverters, radios and on an on.

100W would make it the second biggest power draw so anything they can do to reduce it would be great for anyone living "off the grid". Same applies to RVs and cabins in the mountains.

I hope they do because this is where Starlink with struggle on boats. Bigger than mine (50foot) and you're probably ok as power generation is easier but for small boats 100W is a big draw.

I have 1400W sunpower solar, 9KW genset and pushing into 9.5 KWh Lithium batteries and feeding out 12/24/220 volt. My Iridium go draws 10W max (I know you can't compare the two) so Starlink might need to be relegated to off and on when needed.

I am surprised that with 1400W solar power you are concerned about 100W for Starlink. I was thinking of putting a 220W solar panel up and that would be enough to power the Starlink tranciever and my laptop. Am I being unreasonable? I don't have space for more solar power on 32ft sailboat.

Keep in mind, 100W draw is almost certainly NOT continuous.  It's peak draw, when the device is actively transferring data back and forth at maximum data rate.  Idle wattage is the interesting number, since unless you're streaming 4K video continuously (and even then, it's going to be operating in bursts, so the average is lower than peak), the device will be mostly in an idle state, occasionally powering up to transfer a few packets of data.  Keep in mind that as far as computers are concerned, "occasionally" means "no more than a few thousands of times per second".

[kevinof]

Well that solar has to power two fridges, a freezer, a convection oven, induction hob, lights, instruments and nav gear, autopilot, pumps, water maker, toilets, washing machine, hot water maker, inverters, radios and on an on.

100W would make it the second biggest power draw so anything they can do to reduce it would be great for anyone living "off the grid". Same applies to RVs and cabins in the mountains.

I hope they do because this is where Starlink with struggle on boats. Bigger than mine (50foot) and you're probably ok as power generation is easier but for small boats 100W is a big draw.

I have 1400W sunpower solar, 9KW genset and pushing into 9.5 KWh Lithium batteries and feeding out 12/24/220 volt. My Iridium go draws 10W max (I know you can't compare the two) so Starlink might need to be relegated to off and on when needed.

I am surprised that with 1400W solar power you are concerned about 100W for Starlink. I was thinking of putting a 220W solar panel up and that would be enough to power the Starlink tranciever and my laptop. Am I being unreasonable? I don't have space for more solar power on 32ft sailboat.

Keep in mind, 100W draw is almost certainly NOT continuous.  It's peak draw, when the device is actively transferring data back and forth at maximum data rate.  Idle wattage is the interesting number, since unless you're streaming 4K video continuously (and even then, it's going to be operating in bursts, so the average is lower than peak), the device will be mostly in an idle state, occasionally powering up to transfer a few packets of data.  Keep in mind that as far as computers are concerned, "occasionally" means "no more than a few thousands of times per second".

I hope its not continuous but looking at some of the feedback  from beta testers it doesn't appear to change much during the day. Min I've seen is 90 Watts.

We will have to wait and see.

Sent from my SM-G975F using Tapatalk

[envy887]

Well that solar has to power two fridges, a freezer, a convection oven, induction hob, lights, instruments and nav gear, autopilot, pumps, water maker, toilets, washing machine, hot water maker, inverters, radios and on an on.

100W would make it the second biggest power draw so anything they can do to reduce it would be great for anyone living "off the grid". Same applies to RVs and cabins in the mountains.

I hope they do because this is where Starlink with struggle on boats. Bigger than mine (50foot) and you're probably ok as power generation is easier but for small boats 100W is a big draw.

I have 1400W sunpower solar, 9KW genset and pushing into 9.5 KWh Lithium batteries and feeding out 12/24/220 volt. My Iridium go draws 10W max (I know you can't compare the two) so Starlink might need to be relegated to off and on when needed.

I am surprised that with 1400W solar power you are concerned about 100W for Starlink. I was thinking of putting a 220W solar panel up and that would be enough to power the Starlink tranciever and my laptop. Am I being unreasonable? I don't have space for more solar power on 32ft sailboat.

Keep in mind, 100W draw is almost certainly NOT continuous.  It's peak draw, when the device is actively transferring data back and forth at maximum data rate.  Idle wattage is the interesting number, since unless you're streaming 4K video continuously (and even then, it's going to be operating in bursts, so the average is lower than peak), the device will be mostly in an idle state, occasionally powering up to transfer a few packets of data.  Keep in mind that as far as computers are concerned, "occasionally" means "no more than a few thousands of times per second".

I've seen peaks of 165 W on my system. That's AC in, so that includes the router, PoE injector, and dish. Average is around 100 W and doesn't seem to vary too much with use.

[Nomadd]

I've seen peaks of 165 W on my system. That's AC in, so that includes the router, PoE injector, and dish. Average is around 100 W and doesn't seem to vary too much with use.

How are you measuring peaks? Some gear will register the 100ms or so it surges on power up or hardware connection to charge filter caps.

[envy887]

I've seen peaks of 165 W on my system. That's AC in, so that includes the router, PoE injector, and dish. Average is around 100 W and doesn't seem to vary too much with use.

How are you measuring peaks? Some gear will register the 100ms or so it surges on power up or hardware connection to charge filter caps.

A cheap AC monitor, so I don't know the time scale of the peak. It might be a surge, but the highest I saw was a couple minutes after powering up.

[niwax]

Well that solar has to power two fridges, a freezer, a convection oven, induction hob, lights, instruments and nav gear, autopilot, pumps, water maker, toilets, washing machine, hot water maker, inverters, radios and on an on.

100W would make it the second biggest power draw so anything they can do to reduce it would be great for anyone living "off the grid". Same applies to RVs and cabins in the mountains.

I hope they do because this is where Starlink with struggle on boats. Bigger than mine (50foot) and you're probably ok as power generation is easier but for small boats 100W is a big draw.

I have 1400W sunpower solar, 9KW genset and pushing into 9.5 KWh Lithium batteries and feeding out 12/24/220 volt. My Iridium go draws 10W max (I know you can't compare the two) so Starlink might need to be relegated to off and on when needed.

I am surprised that with 1400W solar power you are concerned about 100W for Starlink. I was thinking of putting a 220W solar panel up and that would be enough to power the Starlink tranciever and my laptop. Am I being unreasonable? I don't have space for more solar power on 32ft sailboat.

Keep in mind, 100W draw is almost certainly NOT continuous.  It's peak draw, when the device is actively transferring data back and forth at maximum data rate.  Idle wattage is the interesting number, since unless you're streaming 4K video continuously (and even then, it's going to be operating in bursts, so the average is lower than peak), the device will be mostly in an idle state, occasionally powering up to transfer a few packets of data.  Keep in mind that as far as computers are concerned, "occasionally" means "no more than a few thousands of times per second".

I hope its not continuous but looking at some of the feedback  from beta testers it doesn't appear to change much during the day. Min I've seen is 90 Watts.

We will have to wait and see.

Sent from my SM-G975F using Tapatalk

Once the dish is set up, reconnecting to the network takes about a minute. So it's perfectly feasible to turn it off when you don't need it.

If they improve the connection speed with more satellites overhead, I expect them to add an option to do this automatically.