Starlink has bufferbloat. Bad.
 help / color / mirror / Atom feed
From: Nathan Owens <nathan@nathan.io>
To: Ulrich Speidel <u.speidel@auckland.ac.nz>
Cc: "starlink@lists.bufferbloat.net" <starlink@lists.bufferbloat.net>
Subject: Re: [Starlink] Starlink power use & satellite tracking
Date: Thu, 16 Feb 2023 15:14:07 -0800	[thread overview]
Message-ID: <CALjsLJvzVWAW8CmUJe45ExWnFBL+bx9VfCDOYRmAbLoX8uX80A@mail.gmail.com> (raw)
In-Reply-To: <CALjsLJvy66hU6wBaegQ2q7Q05s0VjpEhj6yQAv76x4Z8rdK0sA@mail.gmail.com>


[-- Attachment #1.1: Type: text/plain, Size: 5192 bytes --]

I should add, I can actually run the HP dish w/o the router using the same
setup, but when it's obstructed it spikes to ~190W AC, which if it lasts
more than 60s does trip my car E-Fuse. I'll be adding a small battery as a
buffer, which should be able to pull 150W DC from my car, and handle the HP
dish continuous draw of 65-95W.

On Thu, Feb 16, 2023 at 3:12 PM Nathan Owens <nathan@nathan.io> wrote:

> I ran a Round Dishy in my car for a long time, along with the router,
> using a 300W inverter. My car is rated for 12A continuous, 16A peak. Per my
> Kill-A-Watt, it drew on the order of 30-50W, sometimes spiking to 80-90W
> when obstructed/booting. My inverter is 90% efficient, and never tripped my
> car 12v E-fuse. The V3 dish is more efficient, and uses less power.
>
> On Thu, Feb 16, 2023 at 3:08 PM Ulrich Speidel via Starlink <
> starlink@lists.bufferbloat.net> wrote:
>
>> In the aftermath of our cyclone here, I got dragged out for a bit of
>> media comment and, downstream, a few questions came up on dishy power use.
>> Here's what I know and can glean - comments welcome:
>>
>>    - Starlink's own specs say 50-75 W "Average Power Usage":
>>    https://www.starlink.com/specifications. But that's average, not
>>    peak, and peak is what matters when people start recommending that Starlink
>>    could be run out of a small inverter and a car battery in a disaster.
>>    - Small inverters usually come with cigarette lighter cables, and
>>    cigarette lighter sockets are typically fused with 8 or 10 A fuses. That
>>    puts maximum safe power outputs in the 96W to 130-something W range
>>    depending on battery voltage.
>>    - Our lab's "RV" subscription rectangular dishy & router regularly
>>    clocks in at around 80-100 W, and I've seen it go as high as 108 W on one
>>    occasion. I've also seen it go as low as 30 W for the first time last night.
>>    - I have a user report from an older round dishy owner having seen up
>>    to 200 W on occasion.
>>    - Assuming conservatively 90% inverter efficiency, that could mean up
>>    to ~120W and maybe more for the rectangular version and over 220 W for the
>>    circular one.
>>    - If dishy goes over cigarette lighter fuse capacity, people may lose
>>    their ability to charge phones from their car - also a critical capability
>>    in a disaster.
>>       - Not everyone takes kindly to the suggestion that advising
>>       inverter + car battery use could potentially be counterproductive.
>>       - So, what's the peak power use you have seen on your version of
>>    dishy?
>>    - It appears that the current mode of operation here is that dishy
>>    uses several satellites in parallel if these can all see a gateway and have
>>    capacity to carry traffic.
>>    - So for us in urban Auckland with few Starlink users in the cell and
>>       three gateways in the vicinity, our dishy is spoiled for choice and usually
>>       gets to use maybe three or more satellites at once. That takes a
>>       corresponding amount of power but also means great data rates a lot of the
>>       time.
>>       - For a rural user with more Starlink users in the cell and
>>       further away from gateways, the satellites that the cell can see and that
>>       can also see a gateway may be fewer in number. This means dishy only gets
>>       to talk to maybe one or two birds at a time and so uses a lot less power,
>>       and you get more average data rates there.
>>    - If this is so, then it begs a question:
>>       - If Starlink could cap the number of satellites dishy can use in
>>       an emergency area, they would be able to keep your fuse intact. Should they
>>       aim for that, even if it means that you might see lower data rates in a
>>       situation when many people depend on one unit?
>>
>> Starlink is currently being touted as THE comms solution for emergencies
>> as large swathes of NZ's northern and eastern North Island remain without
>> terrestrial or mobile Internet coverage after cyclone Gabrielle. This is
>> the outage map of one of the larger mobile phone providers just for these
>> areas:
>>
>> Most of these are due to power outages to sites, but there are quite a
>> few backhaul cable issues as well.
>>
>> One of the biggest problems is that electronic payment systems don't work
>> without Internet. In our largely cashless society, this is leading to
>> situations where emergency services can't refuel their vehicles because
>> their fuel cards won't work, supermarkets and other stores can't sell
>> anything because customers have no means of paying, and the air force is
>> flying in hard cash in order to help the locals buy food.
>>
>> --
>> ****************************************************************
>> Dr. Ulrich Speidel
>>
>> School of Computer Science
>>
>> Room 303S.594 (City Campus)
>>
>> The University of Aucklandu.speidel@auckland.ac.nz http://www.cs.auckland.ac.nz/~ulrich/
>> ****************************************************************
>>
>>
>>
>>
>> _______________________________________________
>> Starlink mailing list
>> Starlink@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/starlink
>>
>

[-- Attachment #1.2: Type: text/html, Size: 6883 bytes --]

[-- Attachment #2: T7iUkprxXR60409x.png --]
[-- Type: image/png, Size: 172632 bytes --]

[-- Attachment #3: F2CEH9TnkDh1Wrc0.png --]
[-- Type: image/png, Size: 366212 bytes --]

  reply	other threads:[~2023-02-16 23:14 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-02-16 23:08 Ulrich Speidel
2023-02-16 23:12 ` Nathan Owens
2023-02-16 23:14   ` Nathan Owens [this message]
2023-02-16 23:25     ` Jonathan Bennett
2023-02-16 23:23 ` David Lang
2023-02-16 23:36   ` David Lang
2023-02-17  1:24 ` Bruce Perens
2023-02-17  5:27   ` Ulrich Speidel
2023-02-17  5:31     ` Nathan Owens
2023-02-17 15:43     ` Michael Richardson
2023-02-17 19:13       ` Bruce Perens
2023-02-18 10:25       ` Ulrich Speidel
2023-02-18 10:52         ` David Lang
2023-02-18 12:36           ` Ulrich Speidel
2023-02-18 20:13             ` Bruce Perens

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.bufferbloat.net/postorius/lists/starlink.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CALjsLJvzVWAW8CmUJe45ExWnFBL+bx9VfCDOYRmAbLoX8uX80A@mail.gmail.com \
    --to=nathan@nathan.io \
    --cc=starlink@lists.bufferbloat.net \
    --cc=u.speidel@auckland.ac.nz \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox