h a l f b a k e r y
Veni, vedi, fish velocipede
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
Solar aircraft with photo-voltaic cells are baked. Recent
increases in solar cell efficiency, battery & motor
technology mean that they're moving from "only just
possible" toward "could do something useful, maybe".
One thing against the photo-voltaic route is the focus on
certain types of
efficiency. W/m2 or W/$ are the goals,
weight isn't a consideration for the most part, and if they
could make them out of sand, they probably would.
Thermoelectric generators are fussy and horrendously
inefficient... sort of. In terms of heat in, electricity out,
10%, maybe, dreadful. But, in terms of Watts out/kg,
they're really quite good. I just weighed a 250W* unit at
something grams, and it's clearly not optimized for low
weight, at least 10 fold better than PV panels. How does
this concern aircraft? Well, they have a couple of
resources available, intense, cloud-free sun and cold air.
So let's design our aircraft.
First, lots of low powered aircraft look like gliders,
sensible as this is efficient. Gliders have a lot of wing
and often a bit of dihedral... in some solar craft, the
is flexible and bends into a convenient parabola <link>.
So, let's have a great big, mirrored parabolic wing
on a point somewhere above the central fuselage. At
point, the hot sides of two thermoelectric generators
each wing. The reflected sunlight, all of it, supplies
200kW of total energy, assuming a 90% efficiency mirror
and 200m2 wing. The two cold sides facing each other
cooled by flowing -40C air over a shrouded heatsink**
It's possible that even with the weight saving, the
thermoelectric generator will prove insufficient, which
10% conversion is only 20kW peak, less than half that of
Solar Impulse's 45kW photovoltaic setup. In that case, we
move to 30% efficiency with a Stirling engine at the focal
point. There should be enough power left over to make
tea, which is important.
**with good design, a Meredith-type thrust/drag
should be possible.
[bs0u0155, Aug 27 2019]
Body heat generator, uW/cm2
[bs0u0155, Aug 28 2019]
||//I just weighed a 250W* unit at 60 something grams// Is
that a Peltier? And is 250W the maximum power output (and,
if so, with what temperature differential) or the power
consumption for heat-pumping?
||Peltier, TEG, Seebeck device etc. 250W comes at a delta of
410C, which isn't trivial. I've no idea how it performs in
pumping mode, presumeabley with dreadful efficiency. It
was hanging around in a recently closed lab and I can make
a reasonable case for it being useful in a temperature
||Wow. 250W from 60 grams is HUGE! I mean, yes, I guess
there's not much to a Peltier, but even so that's impressive.
||Out of curiosity, do you know what the difference is between
Peltiers intended for TEG and Peltiers intended for heat-
||Yes, the manufacturer's part numbers are different.
||Thank you, [8th], that was very insightful.
||We wouldn't like to be accused of being unhelpful.
||Try telling that to the Intercalary. He says the ocelots are
still off their food.
||//difference is between Peltiers intended for TEG and
Peltiers intended for heat- pumping?//
||There's a bunch of chemistries, essentially optimized for
the temperature range. Garden variety Peltier coolers are
Bismuth Telluride I think, which is great at low temps.
Generators, like Lead Telluride and whatever TAGS is, go
up to the 400-800C range, which is obviously a richer
source of energy.
||//250W from 60 grams is HUGE!// There's more problems
when you dig deeper, it turns out thermal cycling
semiconductors through 500C routinely isn't good for
reliability. I think there is potential for a huge jump if
some new chemistry is found that can interconvert
temperature and electricity at anything like the levels we
can do for say, light. Thermal logic gates are possible,
huge jumps in machine efficiency, slightly thinner fridges,
||//thermal cycling semiconductors through 500C routinely
isn't good for reliability// Yes, I can imagine. The Peltiers
in PCR machines die young because of multiple temperature
cycles of only 100°C or so.
||But for your thermosolar application, that needn't be an
issue since it needn't temperature-cycle quickly (or often).
||As an aside, my skin temperature is currently close to 500°C
thanks to DNP; I am wondering if I could be running my
laptop off of me.
||// Try telling that to the Intercalary //
||We don't deny that we are actually, by and large, unhelpful. It's just that we resent people pointing it out.
||That would probably take a full body coating. But a
watch would be trivial, and the kind of gimmick that
would go over well with watch fans.
||Presumably the watch fan is needed to keep the cold side of
the watch cool.