The reason I asked Marty about his Engine number pad is because the casting date is June 3. This means they were 1. still using the same sand box mould blanks in June to cast 664, 2. the codes are not dates at all (unlikely) or 3. the casting date quoted in the VIN registry is incorrect. If the latter 2 are not appropriate then we have a real dilemma.
OK now I am sounding like a stuck record, maybe the engines were not stamped until they were tested and close to the frame assembly. This theory answers some questions on how the E numbers don't always match-up closely to the frame numbering in several instances. This is all speculation as we don't know:
1) When the engines tested fine there was a close frame/engine pairing, but in the few instances when problems occurred the engines were repaired or scrapped.
2) My guess is frames were almost always good, but engines had some problems.
3) Honda would have needed to make every engine they assemble a good engines if they stamped them right after the casting.
4) As noted, 7414 Frames had Sandcast engines (plus the know French Sandcasts).
5) I would think some issues occurred once engines were assembled, so are there big gaps in the engine numbers? Are engine numbers not sequential? I would think not, but maybe.
In most production assembly operation, the serial numbers are not placed on the final product until all components are proven to operate and meet quality standards. The Japanese had some of the highest, if not the highest quality standards in the world at this time, thanks to Deming.
We will more than likely never really know what happened on Honda's assembly line, but I am sure as more factory frame/engine pairing are known, we will gain additional insights.
DW