If you are not aware of this website I strongly suggest it.
New Memories of Old Machines
I'm aware, thank you. And yes, it's very good!
One of the many articles that pertains to this discussion.
In 1943, Lycoming began work on a powerful and efficient engine intended to power the next generation of very large aircraft. The 5,000 hp XR-7755 ran in 1946, but such a large engine was no longer…
Yes, that project was unsuccessful. And so were things like RR Vulture and Jumo 222, which to an extent can be seen as stepping stones on the path to liquid cooled radials, if not already fullblown such. But I think the takeaway from these failure isn't that the fundamental concept of a liquid cooled radial is unsound, but that there was never the time to develop this branch of engine design to maturity before turbines made high power piston aero engines obsolete. For comparison, how many V-12 aero engines did RR make before the Merlin? And even so, it took a long time before they worked out the various issues in the design and it became the famous war winning engine. Making complex things work just takes time and a lot of elbow grease, no reason to expect that the liquid cooled radial would have been able to somehow take a shortcut on this path.
A major problem is that while you could make engines bigger, you could no longer change the power to weight ratio very much. That needed new materials.
If you had a 2000hp engine that weighed 2000lbs there were only certain applications that called for a 6000hp engine of 6000lbs. There is an economy of scale to be sure but for such things as fighters (and not bombers calling themselves fighters) a 6000lb engine needs 3 times the wing area, and 3 times the fuel (a bit less when cruising) which means 3 times the wing area for the fuel, and 3 times the landing gear weight, etc, etc, etc, The weight savings of using 2 6000hp engines vs 4 3000hp engines is there but it is not great.
Jets won because a jet offered much more power for the same weight.
They were also running into problems with the efficiency of propellers the upper 400mph and low 500mph speed range.
They could not turn the extra power into higher speed or at least effectively.
I fully agree there would have been challenges. And yes, the lower power/weight achievable with piston engines compared to turbines would have limited what was achievable.
However, I'm quite sure that without jets we'd have seen faster prop fighters, as the advantages of higher speed were just too large to be ignored. The historical 'superprops' immediately after WWII wouldn't have been the end of the line if jets wouldn't have been ready. Say something like a single engine fighter using the same tricks the Tu-95 does with swept wings, area ruled, counter-rotating props, and other tricks in the book, should be able to make 900+ km/h like the Tu-95.
I agree with the point that there's an upper limit to how big engines you can find a market for. I guess one reason for the success of engines like R-1820, R-1830, and R-2800 was that they were widely used in everything from single-engine fighters to bombers and civilian aircraft. Still, I think there would be a market for substantially more power than the historical late-war aero piston engines delivered. How much power would the above swept wing single engine fighter capable of 900 km/h need? That might be a suitable size, and that size might be very large indeed. Bigger aircraft could then slap however many of those engines on the wings as they needed.
But perhaps the Tu-95 has help from the Undead Hand of Stalin pushing it along at that ludicrous speed for a propeller aircraft.
Speed record planes do not make money
Military planes seldom do.