[Starlink] lumen orbit

[Starlink] lumen orbit

[Dave Taht]

[-- Attachment #1: Type: text/plain, Size: 303 bytes --]

They plan to build a 5GW 4k solar square for data centers in orbit,
launched on starship:

https://www.lumenorbit.com/

How are they going to cool this thing?

-- 
Artists/Musician Campout Aug 9-11
https://www.eventbrite.com/e/healing-arts-event-tickets-928910826287
Dave Täht CSO, LibreQos

[-- Attachment #2: Type: text/html, Size: 717 bytes --]

Re: [Starlink] lumen orbit

[tom]

[-- Attachment #1: Type: text/plain, Size: 738 bytes --]

It may be possible to create lower resistance circuits in space to reduce the heat that needs to be dissipated. The backside of panels could be radiative surfaces.

 

From: Starlink <starlink-bounces@lists.bufferbloat.net> On Behalf Of Dave Taht via Starlink
Sent: Wednesday, September 4, 2024 10:49 AM
To: Dave Taht via Starlink <starlink@lists.bufferbloat.net>
Subject: [Starlink] lumen orbit

 

They plan to build a 5GW 4k solar square for data centers in orbit, launched on starship:

 

https://www.lumenorbit.com/

 

How are they going to cool this thing?


 

-- 

Artists/Musician Campout Aug 9-11

https://www.eventbrite.com/e/healing-arts-event-tickets-928910826287

Dave Täht CSO, LibreQos


[-- Attachment #2: Type: text/html, Size: 3402 bytes --]

Re: [Starlink] lumen orbit

[Sebastian Moeller]

Mmmh, the white paper states only this sketch:


Orbital data centers can leverage lower cooling costs using passive radiative cooling in space to directly achieve low coolant temperatures. Perhaps most importantly, they can be scaled almost indefinitely without the physical or permitting constraints faced on Earth, using modularity to deploy them rapidly. All of this will have a net benefit on the environment - a recent study by the European Commission concluded that orbital data centers will significantly reduce greenhouse gas emissions from grid electricity and eliminate fresh water usage for cooling.3


No shit, Sherlock for the last sentence... I wonder what kind of stawman design they envision for fresh water cooling in space ;) 




> On 4. Sep 2024, at 16:48, Dave Taht via Starlink <starlink@lists.bufferbloat.net> wrote:
> 
> They plan to build a 5GW 4k solar square for data centers in orbit, launched on starship:
> 
> https://www.lumenorbit.com/
> 
> How are they going to cool this thing?
> 
> -- 
> Artists/Musician Campout Aug 9-11
> https://www.eventbrite.com/e/healing-arts-event-tickets-928910826287
> Dave Täht CSO, LibreQos
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink

Re: [Starlink] lumen orbit

[Dave Cohen]

[-- Attachment #1: Type: text/plain, Size: 1811 bytes --]

She's going to need a whole lot of glycol

On Wed, Sep 4, 2024 at 11:01 AM Sebastian Moeller via Starlink <
starlink@lists.bufferbloat.net> wrote:

> Mmmh, the white paper states only this sketch:
>
>
> Orbital data centers can leverage lower cooling costs using passive
> radiative cooling in space to directly achieve low coolant temperatures.
> Perhaps most importantly, they can be scaled almost indefinitely without
> the physical or permitting constraints faced on Earth, using modularity to
> deploy them rapidly. All of this will have a net benefit on the environment
> - a recent study by the European Commission concluded that orbital data
> centers will significantly reduce greenhouse gas emissions from grid
> electricity and eliminate fresh water usage for cooling.3
>
>
> No shit, Sherlock for the last sentence... I wonder what kind of stawman
> design they envision for fresh water cooling in space ;)
>
>
>
>
> > On 4. Sep 2024, at 16:48, Dave Taht via Starlink <
> starlink@lists.bufferbloat.net> wrote:
> >
> > They plan to build a 5GW 4k solar square for data centers in orbit,
> launched on starship:
> >
> > https://www.lumenorbit.com/
> >
> > How are they going to cool this thing?
> >
> > --
> > Artists/Musician Campout Aug 9-11
> > https://www.eventbrite.com/e/healing-arts-event-tickets-928910826287
> > Dave Täht CSO, LibreQos
> > _______________________________________________
> > Starlink mailing list
> > Starlink@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/starlink
>
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink
>


-- 
- Dave Cohen
craetdave@gmail.com
@dCoSays
www.venicesunlight.com

[-- Attachment #2: Type: text/html, Size: 3077 bytes --]

Re: [Starlink] lumen orbit

[Brandon Butterworth]

On 04/09/2024 15:48:51, "Dave Taht via Starlink" 
<starlink@lists.bufferbloat.net> wrote:
>They plan to build a 5GW 4k solar square for data centers in orbit,
>launched on starship:
>
>https://www.lumenorbit.com/

Maintenance call outs are going to be interesting

>How are they going to cool this thing?

With some big radiators. From
https://www.nasa.gov/wp-content/uploads/2021/02/473486main_iss_atcs_overview.pdf

it seems that the ISS units take about 23 * 3.5m area for 35kW so
about 440 W/m^2, if they are about the same then 5GW will need
around 11 million sq metres.

That seems a totally unfeasible number so either I messed up the
calculation, or they have way more efficient radiators than NASA,
or this is wishful thinking/investor grift.

Compare it to the solar panels needed to collect that much power,
current ISS panels are 190W/m^2 @ 14% efficiency, assuming they can
get to earth efficiency of around 20% they might get 270W/m^2 so
they are going to need more space. At lease one could be on the
reverse of the other.

Anyone got better numbers?

brandon

Re: [Starlink] lumen orbit

[David Lang]

On Wed, 4 Sep 2024, Brandon Butterworth via Starlink wrote:

> On 04/09/2024 15:48:51, "Dave Taht via Starlink" 
> <starlink@lists.bufferbloat.net> wrote:
>> They plan to build a 5GW 4k solar square for data centers in orbit,
>> launched on starship:
>>
>> https://www.lumenorbit.com/
>
> Maintenance call outs are going to be interesting
>
>> How are they going to cool this thing?
>
> With some big radiators. From
> https://www.nasa.gov/wp-content/uploads/2021/02/473486main_iss_atcs_overview.pdf
>
> it seems that the ISS units take about 23 * 3.5m area for 35kW so
> about 440 W/m^2, if they are about the same then 5GW will need
> around 11 million sq metres.
>
> That seems a totally unfeasible number so either I messed up the
> calculation, or they have way more efficient radiators than NASA,
> or this is wishful thinking/investor grift.

the efficiency of radiators goes up drastically as the temperature of the 
radiator goes up (there is a T^4 factor in the equasion, T in kelvin) so if they 
can get the coolant hot enough, they can shrink the radiator area by increasing 
the temp.

but since 0c = ~273K, running the temp up to just over the melting 
point of aluminum would let you get to about 1000x efficiency for the radiators 
(@1255c) but would still need a massive radiator. 11k sq meters is still a lot. 
the ISS radiators are a little of 6Kg/sq meter, so you are up to around 70T 
worth of radiators (possible with a starship, but still massive)

designing a heat pump that can pump heat from chips into molten aluminum is left 
as an excercise for the reader :-)

David Lang

Re: [Starlink] lumen orbit

[David Lang]

David Lang via Starlink wrote:

> On Wed, 4 Sep 2024, Brandon Butterworth via Starlink wrote:
>
>> On 04/09/2024 15:48:51, "Dave Taht via Starlink" 
>> <starlink@lists.bufferbloat.net> wrote:
>>> They plan to build a 5GW 4k solar square for data centers in orbit,
>>> launched on starship:
>>>
>>> https://www.lumenorbit.com/
>>
>> Maintenance call outs are going to be interesting
>>
>>> How are they going to cool this thing?
>>
>> With some big radiators. From
>> 
> https://www.nasa.gov/wp-content/uploads/2021/02/473486main_iss_atcs_overview.pdf
>>
>> it seems that the ISS units take about 23 * 3.5m area for 35kW so
>> about 440 W/m^2, if they are about the same then 5GW will need
>> around 11 million sq metres.
>>
>> That seems a totally unfeasible number so either I messed up the
>> calculation, or they have way more efficient radiators than NASA,
>> or this is wishful thinking/investor grift.
>
> the efficiency of radiators goes up drastically as the temperature of the 
> radiator goes up (there is a T^4 factor in the equasion, T in kelvin) so if 
> they 
> can get the coolant hot enough, they can shrink the radiator area by 
> increasing 
> the temp.
>
> but since 0c = ~273K, running the temp up to just over the melting 
> point of aluminum would let you get to about 1000x efficiency for the 
> radiators 
> (@1255c) but would still need a massive radiator. 11k sq meters is still a 
> lot. 
> the ISS radiators are a little of 6Kg/sq meter, so you are up to around 70T 
> worth of radiators (possible with a starship, but still massive)
>
> designing a heat pump that can pump heat from chips into molten aluminum is 
> left 
> as an excercise for the reader :-)

correction, sloppy reading (F vs C), molten aluminum is only 660c, to get the 
1000x efficiency, you would be up to the temp of molten iron.

just don't let if freeze in the pipes :-)

David Lang

Re: [Starlink] lumen orbit

[Michael Richardson]

[-- Attachment #1: Type: text/plain, Size: 318 bytes --]


David Lang via Starlink <starlink@lists.bufferbloat.net> wrote:
    > designing a heat pump that can pump heat from chips into molten aluminum is
    > left as an excercise for the reader :-)

No chips.
The entire structure will be "smart aluminium" for computation, as featured
in the Terminator 2 documentary :-)



[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 511 bytes --]

Re: [Starlink] lumen orbit

[Sebastian Moeller]

If we only had a snappy name for the network that links such LEO data centers in the sky...

Maybe "skynet" would be a decent proposal?



> On 5. Sep 2024, at 19:58, Michael Richardson via Starlink <starlink@lists.bufferbloat.net> wrote:
> 
> 
> David Lang via Starlink <starlink@lists.bufferbloat.net> wrote:
>> designing a heat pump that can pump heat from chips into molten aluminum is
>> left as an excercise for the reader :-)
> 
> No chips.
> The entire structure will be "smart aluminium" for computation, as featured
> in the Terminator 2 documentary :-)
> 
> 
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink

Re: [Starlink] lumen orbit

[Brandon Butterworth]

On 05/09/2024 20:37:42, "Sebastian Moeller via Starlink" 
<starlink@lists.bufferbloat.net> wrote
>If we only had a snappy name for the network that links such LEO data centers in the sky...
>
>Maybe "skynet" would be a decent proposal?

A fan in the uk military already grabbed it

https://en.wikipedia.org/wiki/Skynet_(satellite)

given the current AI and drone trajectory it seems appropriate
they have it.

brandon

Re: [Starlink] lumen orbit

[Dave Taht]

[-- Attachment #1: Type: text/plain, Size: 181 bytes --]

I was thinking that at least some of the research into molten salt reactors
would apply.

https://en.wikipedia.org/wiki/Molten-salt_reactor

Without the radioactives. Good to 700c.

[-- Attachment #2: Type: text/html, Size: 330 bytes --]

Re: [Starlink] lumen orbit

[Dave Taht]

[-- Attachment #1: Type: text/plain, Size: 610 bytes --]

This is the business case they are making for putting DCs in orbit.

https://lumenorbit.github.io/wp.pdf

It is hard for me to cope with the prospect of pouring so much energy into
AI.


On Thu, Sep 5, 2024 at 9:31 PM Dave Taht <dave.taht@gmail.com> wrote:

> I was thinking that at least some of the research into molten salt
> reactors would apply.
>
> https://en.wikipedia.org/wiki/Molten-salt_reactor
>
> Without the radioactives. Good to 700c.
>


-- 
Artists/Musician Campout Aug 9-11
https://www.eventbrite.com/e/healing-arts-event-tickets-928910826287
Dave Täht CSO, LibreQos

[-- Attachment #2: Type: text/html, Size: 1414 bytes --]

Re: [Starlink] lumen orbit

[Michael Richardson]

Dave Taht <dave.taht@gmail.com> wrote:
    > This is the business case they are making for putting DCs in orbit.

    > https://lumenorbit.github.io/wp.pdf

    > It is hard for me to cope with the prospect of pouring so much energy into
    > AI.

"Can you go through all the old pitch decks and replace the word 'crypto'
with 'A.I.'?"