To understand Bruce's figures it is important to note that the keyway at the 6 o'clock position in the above diagram is aligned with the point of the "bottom" tooth of the sprocket. If somebody machined fourteen keyways aligned with the points of each tooth of the sprocket, they could slip any one of them over the crankshaft key and the valve timing would not change at all. To get some slight valve timing adjustment a keyway has to be machined such that there is a difference in angle between a line going from the centre of the sprocket through the keyway and a line going from the centre of the sprocket through the point of the adjacent tooth. This keyway could be machined adjacent to any of the other teeth (besides the bottom tooth in the diagram). I guess to make it roughly symmetrical, Ariel chose to put their keyways about 120° apart. So, again referring to the above diagram, the point of the fifth tooth up to the right from the bottom is at 128.55°(5 x 360°/14 teeth) (or 5 x 25.71°= 128.55°,roughly 2 o'clock). To create the offset needed for timing adjustment, they machined a keyway at 123° or 5.5°offset from the point of the tooth (128.55° - 123° = 5.5°). When the sprocket is placed on the crankshaft using this 123° keyway, ALL the teeth are offset 5.5° vis-à-vis the original (or bottom or 0°) keyway. To get the tooth adjacent to the 123° keyway into the same link of the timing chain as the 0° one was in, one would have to turn the crankshaft 5.5° in its normal clockwise direction therefore putting it "ahead" of the camshaft or, in other words, retarding the valve timing 5.5°. Whew!
1946 4G, 1950 NG and 1951 VH