Frame & Engine numbers
Peter Cornelius has been doing research on the engine and frame number system and the assembly date/month system for many years. Here are some of his observations:
"Engines, frames, gearboxes and other parts were made in different parts of the factory - known as 'Shops', which was short for 'workshop'. So there was an Engine Shop, Frame Shop, Machining Shop, Wheel Shop - and an Assembly Shop.
A central storage for these various parts was The Stores.
FROM MY RESEARCHES - As parts were completed in the various stores an apprentice lad (probably a 14 year old fresh from school) would transfer those parts to stores with a barrow/trolley. The storeman would stack those parts on shelves and probably hang the frames up. The frames and gearboxes were serial number stamped when they were made but the storeman would not move items along the shelves and racks to keep them in number sequence.
Engines MAY have been number stamped, but I think not as these numbers certainly went into frames in a sequence. Also into stores would have gone other parts made elsewhere, such as magnetos, control levers and fuel tank caps.
Another lad (probably would have been an older and more experienced apprentice who knew which items would be needed) would transfer parts from the stores to the Assembly Shop. Here it didn't matter about serial numbers, as long as the right frames and gearboxes went together for the various models, and there would have been separate 'lines' for different models. Either engines were used in their numerical sequence, or more likely they had not been numbered, for on assembly the date or date code was stamped on the engine and MOST LIKELY the number was also applied to the engine at this time. The line foreman would keep records and he would tell the men which number and code was to be stamped on each machine.
It's difficult to be accurate with engine/frame relationships. I can often say that a frame was from a certain period, but occasionally there would have been early numbered frames hanging at the far end of the stores rack which wouldn't have been used until near the end of a production run. (The storeman wasn't going to walk the length of the stores to select a frame when one which came in 'yesterday' was much closer.) Engine codes also occasionally 'throw up a wobbly'. Ws and Ms can be stamped upside down, and sometimes a tired workman selected the punch next to the one his hand was aiming for!
Occasionally I find an engine number totally out of sequence with the date code - and the date code is later than in 'should' be. In the event of a replacement engine or crankcases being required it would have been normal to stamp the same engine number as that which was being replaced. This way the documentation (log book) did not need to be changed. However, the factory or the Agency doing the work would have been told what code to stamp, and the new code would indicate the true date of that replacement.
(Or they may have been stamped prior to despatch from the Works.) As log books didn't record the letters these were not affected.
the dating code
Although the machine assembly dating code is well circulated in books and
internet Web pages as TRIUMPH CODEX, where T=1, R = 2, etc. through to 11 and 12,
for indicating the month of assembly at the Triumph Works this is not correct
and has caused much confusion amongst owners of veteran Triumphs who will never
have seen an X on their bikes, but may have seen another letter not included
within the twelve above. The code was actually TRIUMPH CODEY, and was first used
in September 1910 as a two letter code where that month was coded as OD
(9,10). These letters were stamped further along the same crankcase surface as
the engine serial number. Prior to September of that year the actual day of
assembly was stamped, i.e. 13-8-10. The 1910 code proceeded through to December,
1910 as YD (12,10).
The code was fine as two letters until 1913 was encountered. The code then
became three letters and January 1913 became TTI (1.13) and YTI (12,13) was
(It was not until well past the veteran era, in 1920 that the ‘X’ was introduced
as a ‘zero’, and January 1920 was indicated by TRX (1,20). Y and X appeared
together in December 1920 as YRX (12,20)."
Peter also remarks that the widely published list of frame and engine numbers (see thumbnail) is not an official works document. In various years bicycle frame numbers are included in the runs. In the thumbnail table I have put question marks for the years where no numbers seem to be available. On a number of pre-September 1910 survivors the day-month year code is not complete; note also that cylinder base was stamped with engine number. Not all my recorded survivors have this. See thumbnails.