"A lot of people will sell you cubesat subassemblies. One thing that the
more enthusiastic cubesat promoters seldom mention is that a lot of
cubesats are dead on arrival or quite short-lived; component quality is
undoubtedly part of that, although plugging things together is not
always
simple either. :-)"
This is very true! Just because they sell cubesat stuff does not mean
you will get what is advertised!
Comms performance is very low thus far.
Pointing of cubsats is very difficult. (attitude control)
It is advancing but it is not what the funny papers will tell you.
There is a need for a network on the ground
Check out SatNOGS
-------- Original Message --------
Subject: [AR] Re: Pilot, ALASA, and cubesat launch (was Re: some
interesting developments)
From: Henry Spencer <hspencer@xxxxxxxxxxxxx>
Date: Wed, January 06, 2016 5:30 pm
To: Arocket List <arocket@xxxxxxxxxxxxx>
On Wed, 6 Jan 2016, Valtteri Maja wrote:
In short, I think a realistic commercial "cubesat" launcher *should* beI've been toying with this idea for quite a while, but the last stage
aiming at something close to ALASA's payload. I don't think you're going
to make it pay at 3kg.
and the payload have a lot of potential for shared hardware (and
software too). In larger rockets, it's not worth the trouble...
The original Proton fourth stage, Blok-D, had no brains of its own --
after third-stage separation, the payload had to operate the rocket! The
later Blok-DM accepted the mass penalty of providing its own GNC, mostly
for launching Western satellites that wanted to be pure passengers.
On a somewhat smaller scale, Pegasus had an option (PegaStar?) to let you
build your payload around the third-stage avionics package, and that was
used at least once.
... This is of course if we assume the upper stage is not a
dumb-as-possible spinning solid with an egg timer release.
A reasonably safe assumption. The problem with a spinning-solid final
stage is that orbit insertion can be pretty imprecise. And since one big
reason why a nanosat owner might want a dedicated launch, rather than
going up as a secondary payload, is to pick his own orbit, I think you
want at least the option of precision insertion.
For example the upper stage has telemetry (radios, antennas), attitude
and position sensing (gyros, accelerometers, star cameras, GPS), and it
also has to have some kind of batteries, and of course a computer.
Typically not a star camera -- gyros are normally good enough for launch,
and there are concerns about the rocket exhaust plume (which at high
altitude can envelop the entire vehicle) crudding up optical sensors.
And the first choice for batteries would most likely be primary lithium
batteries (non-rechargeable), for minimum mass. But yes, shouldn't be
that big a deal to revise the design a little to permit other uses.
The main things you'd need to add to produce a satellite would be solar
arrays and long-term thermal control (for a 15-minute ascent you can
pretty much just rely on a bit of insulation plus thermal inertia, but to
spend many days in orbit, rather more attention to thermal is needed),
plus perhaps attitude actuators like reaction wheels and magnetorquers.
Might also need a radio uplink -- telemetry only needs a downlink, and
aside from range safety (often constrained to be separate, and often
located on the next-to-last stage rather than the last one) there's
typically no perceived need to uplink commands to a launcher in flight.
Of course then you can say that integration is really hard...
There *is* an awkward problem of who's in charge of the design, especially
if a screwup by the customer can compromise control of the rocket (which
is an issue for your reputation and probably for launch licensing too).
Might be better to provide the whole satellite bus, including solar arrays
etc., and the customer just provides the satellite payload. Of course,
this means you're doing much more work and undoubtedly having to charge
rather more too. And it's drifting away from the idea of being able to
launch satellites designed to a launcher-independent spec; some customers
won't want to tie themselves to a single launch provider (especially when
it's a new startup).
For example Clyde Space already provides cubesat platforms and components,
A lot of people will sell you cubesat subassemblies. One thing that the
more enthusiastic cubesat promoters seldom mention is that a lot of
cubesats are dead on arrival or quite short-lived; component quality is
undoubtedly part of that, although plugging things together is not always
simple either. :-)
Henry
