I worked with the Tiny Tech steam engine some years ago and developed for Tiny Tech a roughly 30 square meter reflector to generate steam.
Two videos I made about it are here:
The reflector we measured at about 20 KW at high steam pressure, but to run the steam engine at load would haver required 3 such machines.
Instead of building 2 more, 200 000 USD of materials and a whole year was spent building 9 reflectors 3 times bigger, both against my strongest possible advice (smaller modular machines make far more sense to start out commercially and there was no reason to believe a 90m^2 version would be cheaper, due to the exponential cost of building towers and accuracy of reflections decreasing on a cosine curve; I had spent some trouble to work-out that 30m^2 was likely a cost optimum given the fabrication methods available; and refusing my at designing a bigger version so it would at least work).
I have some photos of the 90 m^2, but the reason there's no version online is because it was a disaster and then scrapped. When they couldn't make it work they finally asked me to come to try to fix it. Unfortunately due to the exponential cost of building towers they couldn't make a tower of a proportional size so decided the way to deal with that is make a smaller tower (rendering the machine at best a 50 m^2 machine). The 200 000 USD being wasted the whole development had to be scrapped.
You maybe wondering the reason for this madness, and the answer to that question was that a new machine that I had no part in designing was necessary to cut me out of royalties that I wasn't demanding or asking for. My goal at that time was to transfer the technology to a structure that could continue the development and commercialize it. Had we demonstrated continuous steam power I'm confident enough units would be sold, even just as a development platform.
As for the steam engine itself, it's based on a UK kit that was available at the time, and not even the whole kit, but just the shop drawings to Finnish the machining (the product was unfinished pieces for hobbyists to finish the machining of, or at least one version of the product, but the shop drawings to do the finishing was available for free; dimensions that did not appear in those shop drawings--as not all pieces needed finishing on all the surfaces, or at all--were basically just guessed at).
So definitely could have been made more efficient, but it did at least work and in a configuration where the waste heat is utilized the efficiency does not necessarily matter (they did sell units powered by wood and charcoal, and had a boiler for that).
Generally speaking, if you want to get steam power really going the focus should be on making the cheapest possible solar thermal energy, which then creates the market for steam engine, and for developing markets where cottage industry scale makes commercial sense in a lot of industries but also the labour to build and maintain this kind of technology isn't a problem (automation can then be perfected over time to deploy in higher-labour cost environments).
What you see in the video above is in my opinion the cheapest way to produce solar thermal energy for the scale of a small steam engine.
In publishing the video I thought people would anyways easily reverse engineer and build copies of the machine, but probably seeing Tiny Tech not continuing the development dissuaded copy cats.
However, if there's interest I could find and put together all the designs and, most importantly the original open source software as mentioned in the patent:
Which, again, I wrote to protect and encourage continuing the open source development, but it simply faded off the internet.
My goal with Lytefire was to prove the commercial use of the technology (the main obstacle in my view to inspiring copy cats is proving the commercial use), and even after succeeding in that ... well there are a few copy cats I became informed about but they did not understand what to do, trying to make their own improvements that are counter productive.
Unfortunately Lytefire was infiltrated and then taken over by organized crime as it was key in a fake consortium used to launder hundreds of millions, if not billions, of dollars out of Africa and elsewhere.
So feel free to enquire if you want to see if you can help get all the above knowledge documented and open sourced. The original open source designs are key to avoid patent harassment, and the author of the patent re-publishing everything that was open source at the time (and therefore still open source) would be of immense legal benefit; so yes, do feel welcome to just go ahead and reverse engineer the technology anyways, but even if you're as good as me you're work would have legal risks that are really best to solve by a proper documentation of what was open source at the time (some such material literally pampthlets published for various events, which I'm pretty sure I have the only copy).
What's absolutely critical to really break the cost barrier to solar thermal energy is building things out of wood.