Why didn't Allison work harder on developing it? Did it have some serious flaw? (Hard to believe because the V-1710 seemed to be more reliable and easier to build than the Merlin. I'm beginning to think that the Allison got a really bad rap.) Like the P-38, a 1942 XB-42 COULD have gotten one (or two) good turbochargers... Of course the XB-42 might have been even better if it was a tractor ala the R2Y than a pusher...
There are a lot of erroneous posting in this thread.
First, the US Air Corps wanted a double V-1710 to power the XB-15 back in 1934. Concurrent development of the X-3420 and V-1710 meant slow progress on both. When Bob Hazen took over as Chief Engineer in 1936 he got the Air Corps to back off of the 3420 project while he got the V-1710 through Type Testing by promising them a better V-3420 based upon his improved V-1710E/F models. This was done and the V-3420-1(A1) first ran in April 1938, although the Air Corps continued to support its development, they did not have a project needing the engine at the time. In September 1940, during the Battle of Britain, the Air Corps suspended all V-3420 work to focus on desperately needed fighter engines.
When the B-29 project got into trouble because of problems with the R-3350 the Air Corps restarted the V-3420 project in February 1941. Focus was on engines for the XB-19A and XB-39, both powered by turbosupercharged V-3420A engines. Allison was awarded a contract for 500 engines, enough to support 100 B-39s. When the R-3350 issues were mostly resolved, and the continuing need for V-1710s, the B-39 project was terminated. For a more comprehensive look at the V-3420 see
Vee's For Victory! by Dan Whitney.
The V-3420B engines had contra-rotating crankshafts, ideal for powering similar turning propellers. This engine was adopted for the Fisher P-75 and the questioner is right, it could have been utilized in the later XB-42, which had two V-1710s, each driving parallel drive shafts to the aft mounted coaxial propeller gearbox. The V-1710s had the advantage that one engine could be shutdown and the other operated at full power, a measure of redundancy. This was not possible with the V-3420B.
The single-stage Merlins did have a slightly more efficient supercharger, which also could operate at either of two speeds. This improved its service ceiling.
Not everyone understands that Allison and Rolls-Royce were both working to develop two-stage engine driven superchargers at the same time, and both entered service in 1943.
Allison had wanted to introduce a two-stage supercharged V-1710 in the North American X-73, the Mustang prototype in 1941 (See
V-1710 and V-3420 Designs and Concepts by Dan Whitney and John Leonard). The Air Corps did not want Allison to divert engineering and manufacturing effort into this project, so it was delayed. Their argument was that they had sponsored the development of the successful GE Turbosupercharger since WWI, the V-1710 was designed to work with a turbo, thereby giving it "two-stage" capability, e.g., X/YP-37, P-38, XP-39. Prior to the war they saw no need for mechanical two-stage superchargers.
The V-1710 was a very reliable engine. End of War TBO was 500 hours, while the Merlin never exceeded 400 hours, even on P-51s in the 1950s. Examples of V-1710s achieving over 1,000 hours on the wings of P-38s without removal are available.