Ford built most of the R-2800 in house. However that does not mean that basic castings/forgings were not subcontracted.
Ford sent some engineers to P & W Hartford to look at the factory layout and they basically duplicated it. So I assume, whatever P & W was doing as far as subcontracting went Ford was planning on a similar operation (trying to add an aluminium foundry
inside an existing factory layout doesn't seem like it would work well).
To hit that 2400 engine per month mark (and they only did it once) required expanding the original Ford factory (purpose built in 1940-41) to just about 3 time is original floor space. The Original contract for the factory was for 800 engines per month.
As mentioned by
X
XBe02Drvr
, one of the key points to increasing the power of aircooled engines was the great increase in the area of the cooling fins on each cylinder and for most of the 30s and 40s it was a case of
HOW do they increase the number/size of the fins, NOT
if they should increase them. It required new casting and forging techniques in addition to new machining techniques. You can hand them a set of drawings for a cylinder with massive finning but if they don't know how to make it in quantity it doesn't do them much good, they
already know they need more fin area than what they are making.
A great resource for this discussion is :
https://www.enginehistory.org/References/WWIIEngProduction.pdf
I would note that P & W was rather free with granting licences, but apparently was
not disappointed by licensee quality. P & W at one point was charging a licence fee of $1.00 per engine and at least point in the war waived even that. Wright on the other hand was much more reluctant to grant licences and relied more heavily on subcontractors with final assembly by Wright.