This is what they actually do:

> Parker Solar Probe is powered by two solar arrays, totaling just under 17 square feet (1.55 square meters) in area. They are mounted to motorized arms that will retract almost all of their surface behind the Thermal Protection System – the heat shield – when the spacecraft is close to the Sun.

https://blogs.nasa.gov/parkersolarprobe/2018/05/07/solar-pow...

I guess that's exploiting the temperature differential for power systems but it's not exploiting the temperature differential for power generation.

Solar panels are just more reliable than any heat engine and actually using the temperature differential in a heat engine (stirling or whatever) would decrease the heat shields ability to reject heat. And it'd add mass in the form of radiators with space already at a premium.

---

But more in general, I'm super excited about Solar Probe+ (Parker Probe) as well as the completion of the Advanced Technology Solar Telescope (Daniel K. Inouye Solar Telescope (DKIST)). They're both going to be exposing never before sampled, very exciting, territory.

The Solar Probe with actually in-situ measurements of the magnetic field of the corona (as opposed to inferences based on optical Hanle scattering or radio emission). It is going to put a lot of constraints on previously weakly constrained flare and CME models. Maybe in the 9 radius passes it might even detect (or more likely not!) the stupifyingly unphysical entire electron contents of the corona dumping down during flares most models call for.

And the ATST/DKIST with it's diffraction limited imagining and fast optical spectroscopy of the sun at 100s of km scales projected on the photosphere. There's so much going on in the photosphere and chromosphere that right now is just blurry blobs. When resolved I think there's going to be a lot to see.

Another way to think of it is that they already have too many solar panels because the probe has to power itself from Earth orbit to its final orbit near the Sun. So, anything that uses what you've already got (solar panels) with minimum additional mass and risk (the little motors to move the panels mostly behind the heat shield) is a winner.

In Watts' Echopraxia the powersat Icarus does that, exploiting the temperature differential between the hot and dark sides.

There are ways, piezoelectric devices can generate power off temperature differential. And mechanically you could put a stirling engine on although the torque and rotational force likely would not be ideal.

The Sodium Heat Engine is an ingenious device: https://en.wikipedia.org/wiki/Alkali-metal_thermal_to_electr...

Best thing: no moving parts required!

Thermoelectric, not piezoelectric ^_^

WRT Stirling generators, there is already work ongoing for arrangement ssuitable to space applications, through new research on nuclear power sources (stirling radioisotope generators, and the Kilopower nuclear reactor). Generally the solution is to mount several Stirling generators such that their torques cancel each other out. Search for "managing the stroke of the pistons" at https://beyondnerva.wordpress.com/2017/11/19/krusty-first-of...

That seems really complicated for something that the solar arrays do with no moving parts.

Note the quote from a while ago:

> They are mounted to motorized arms that will retract almost all of their surface behind the Thermal Protection System – the heat shield – when the spacecraft is close to the Sun.

Yes, there's a moving part. It only has to move twice (deploy, mostly retract) but it's still a moving part.

A motor that predictably moves an object from one point to another can be much more reliable than an array of heat-driven reciprocating engines that have to be in constant motion and rely on conducting heat toward the internals of the spacecraft in order to function.