The Velocette Parts Numbers - the system is explained by 1999/2000 Club PRO, Colin Goodwyn

Article first published in Fishtail 299

Has it ever occurred to you why, when you ordered a part from Veloce Ltd., they asked you to state, besides the part number, the part description, and the engine and frame numbers of your machine? Surely the part number itself was sufficient; the inclusion of all the extra details is a bit like writing out a full postal address and then adding the post code - why should both be needed? Well, there's no explanation for the Post Office's peculiarities but as far as Velocette parts go the answer is that the number often isn't enough. It should be, but it isn't. You pedantic types might have thought that if, for example, you ordered an A28/7 you would get an MAF rear chain of 108 links but, if you had wanted it for a MAC, that would be a link too short. Which part did you really want? The demand for full details stems from the facts that Veloce themselves sometimes used the same number for different parts dependent upon which model it was for (e.g. "K1/2 - cylinder head for Model K or with polished ports for KSS") and that they published spare parts lists which were not always, or even often, correct.

Sometimes our hard-pressed volunteer spares staff (of whom more are always wanted and warmly welcomed) receive an order which comprises only part numbers and, sometimes, these numbers, which the customer has taken from his genuine Veloce Ltd. parts list, don't make sense. When this happens it is necessary to turn to the only place where parts are to be found in numerical order which is Veloce's old Price Lists. Those can prove that such-and-such a number did once exist but don't, unfortunately, describe the part so the next step is to search through old Illustrated Parts Lists - the date of the price list having provided a clue as to where to look. All this is no fun at all and it is because there was no means of simply looking up a Velocette part number that I have recently produced a combined parts list compiled from all the published spares lists that I could find. Velocette themselves must have had a similar document or perhaps, because some of their numbering is a bit odd, it is probable that they had a card index which occasionally fell on the floor.

The early parts lists didn't give numbers for all replacement components after all, if you wanted a frame for your 1913 Model DI you just asked Veloce for one; well, of course, it wasn't you who actually did that, but your dealer who had probably supplied your machine and maintained it for you. At that period it's unlikely that the general public would have ever seen spares lists or even knew they existed. In a manufacturer's youth, its small scale and uncomplicated nature meant that illustrated lists and even numbers weren't really required because the big bits of a machine could be easily identified both by the customer and the maker or his storeman but the smaller parts presented obvious difficulties. Hence, from about 1918 parts numbers came to be introduced and before 1925 all Velocette parts, big and small, had them.

The original plan is not exactly clear. There must be some early lists which we don't have in the Library and which I have not seen elsewhere but, from the beginning, it seems that parts were categorised by letter prefixes. Thus F was for Frame, W for Wheel, FB for Front Brake, FS - Front Stand, C for Clutch, P - oil Pump, S - rear brake (Stopper?), A for frame Accessories (cycle parts, cables, pipes, tools etc.), T was for engine parts (Two-stroke?), G for all Gearbox components - but which was soon supplemented by GC to cope with all the Gear Change parts - and, by 1922, G (which had principally covered the D2 and D3 models) had been replaced by B (for the gearbox of the Model G two-stroke).

Apart from the prefixes, the early numbers are interesting in that the T and G series reached five figures - no less! - but they are not so straightforward as they might appear. For example, T1521 was not the one thousand, five hundred and twenty-first part in the T series, but referred to a new variation of T15 (a gudgeon pin) which was introduced in 1921 and it was thus spoken of as a "tee fifteen, twenty-one". By 1922 the stroke (/) had been introduced to emphasise the pronunciation; hence T15/21, and a five figure number such as T11625 (left hand crankshaft pin) became T116/25. After 1923 strokes became commonplace.

By 1925 it had occurred to someone that the method of giving a modified part a year suffix (as in T33/23) wouldn't work if there was more than one variation of a component in a year, the only alternative being to issue a completely new number which would spoil the numbering sequences. For example, if a new silencer was introduced in 1924 and became T76/24 but, before 1925, another variety was needed, how could it be numbered? The answer was to abandon these year numbers and use a variation suffix number. Therefore T76 would be followed by T76/1, T76/2 etc. though, curiously, it was only in the SL series (and for pistons) that the /1 was used. From a variation suffix, therefore, it should not be implied that T71/19 is the nineteenth variety of chaincase inspection cap - it was simply introduced in 1919 (or, rather, for a 1919 model - probably from the autumn of 1918) but KA18/27 really is the twenty-seventh type of spoke!

When the overhead cam models arrived in late 1925 their parts were prefixed by K (for Kam!) and the related frame, gearbox, rear brake and cycle parts were prefixed FK, BK, KS and KA but why were the brake and cycle parts not SK and AK? Its a very good question and made more pointed by the fact that the clutch parts were prefixed KC.

That is, most of them were, but there are also Model K clutch operating parts prefixed by CK. There aren't many of those, however, and 1 suspect that either someone simply forgot or, in the case of CK, that it was the result of a misprint which had to be maintained because of the importance of the parts and their continued use. The misprint theory is supported by the fact that, apart from CK21, all the other CK numbers, of which there are about ten, are missing from the KC series. It was probably in an attempt to keep these new K-part prefixes simple that there were no KFS or KGC classes, the OHC models' front stands and gearchange parts being given a K suffix, e.g. FS 1/2K. The use of a suffix to denote the model the part was for had been employed in the early '20s; for example an A16A was an A16 front mudguard modified for the Model A and an A28AC was an A28 rear chain for the Model AC. This should not be confused with the practise of labelling the two parts of a rear chainguard A and B like the F60/7A and F60/7B for the MOV/MAC. After 1934 the M suffix was occasionally used to identify, for instance, the speedometer KA268/AS/M for an MSS compared with KA268/AS/K for a KSS/KTS but most new or altered parts for the MOV, MAC, and MSS went into their own series. In most spheres, the letter X is usually reserved for desperate matters and, happily, was rarely used by Veloce, only occurring as a suffix to the Model K's oil hardened gears (which must NOT be run with the case hardened variety) and as a prefix, on its own, as X8 which was a 12:1 KTT piston. What, I wonder, were XI to X7?

Around the time that the K series appeared so did E (for Electrics) and, with the arrival of the KTP came IG, for IGnition parts. The old D, E, B, G, H, A etc. two-strokes were superseded in 1930 by the GTP engine whose numbers were prefixed by RT (for R --- Two-stroke? Answers on a postcard ... ) and also into the lists came the SL series, which starts with piston rings. (SL for SLug, perhaps, as in "What's black and 'orrible and slides up and down?"). In fact, the majority of the SL numbers refer to bolts, washers, nuts, studs and threaded rods, most such fasteners having been simply ignored in pre-1927 lists. That the SL3 series should comprise piston rings and SL91 to 95 series be sprockets seems to be something of an anomaly. GTP rings should logically have been under T or RT and why they're not is, like most of the SL series, a mystery.

The introduction of the MOV in 1933 created the M prefix series which grew as the MOV, MAC and MSS engines were developed. Simple and straightforward, like the engines, these part numbers were not supplemented by subsidiary prefixes such as MB, MF, ME or whatever for the associated gearbox, frame, electrics etc. though there were eventually eighteen types of cylinder barrel and, coincidentally, eighteen types of valve.

But it was the M numbers which later inspired a whole new approach. Velocette had long known that their customers would order, for example, a piston complete with rings, gudgeon pin and circlips or, perhaps, a sleeve gear complete with bush, so instead of asking for all the different parts by number the request would be, effectively, for an assembly as in the speedo example already quoted which included the cable, bracket, drive gear and gearbox. In the case of the sleeve gear, a BK8/2 and a BK7 became a BK8/2AS. Similarly, an MOV con rod and bush (M28 and M29) became an M28AS and this numbering developed from the mid-1930's.

It is obvious that during the last war someone had time on his hands to make a reappraisal of the Spares Lists and part numbers. The numbers by the illustrations in the post-war lists were replaced by code letters (and, from the introduction of the 1950 front fork, by plain numbers) which meant that the specification of a handlebar, perhaps- could be changed, and the new part and its number listed, without having to amend the printed illustration. This sort of simplification (for Veloce) probably resulted from the firm's experience in dealing with military requirements. When the MAF replaced the W.D. MAC the factory had not only to issue a new spares list but had to redraw the illustrations getting them more in proportion and group them in a logical manner and also had to list assemblies likely to be ordered by the customer (e.g. M4AS was an MAF engine complete, B42/2AS was a gearbox complete, A11/9 was a full set of forks). These sorts of items had not previously been listed as such. It had been possible to order, say, all the parts of a wheel but there was no published number for a complete one. Henceforth, by ordering an assembly by its number, the "misleading word, complete" (as Veloce described it) could be abandoned.

During 1950 and ’51 the M --- AS numbers suddenly blossomed into the MAS series, and a perfectly logical run it was too but, before long, the brains behind the master plan seem to have got carried away. A timing gear with its bush could be called an assembly as could a complete oil pump or the oil tank check valve (MAS 14) or a set of Dowty forks (MAS 10) - but could a frame? Well, yes, a frame is assembled from various bits but you can't buy those bits (the frame tubes and lugs) separately and the same applies to an oil tank whose pressings are welded together. A toolbox (with knob) is hardly an assembly (you couldn't buy the lid alone), nor is an exhaust pipe even with its bracket welded on. A pushrod, I suppose, has been assembled (though I never have managed to get the ends out) as has a dual seat but a "dynamo strap assembly" pushes credulity near the limit as does MAS159. MAS159 is a wingnut. Not just any wingnut, however, for it fits on the Thruxton models which I think provides a clue as to this peculiar "Assembly" philosophy. After its initial logical beginnings it looks like psychology took over for, by the early 'fifties most dealers' stocks of spares were being bought and fitted by increasingly knowledgeable and skilled customers (many of whom had lately had excellent MT training), and the value (and hence the cost) of the parts could be better implied by describing relatively simple items as "Assemblies". From 1950 and throughout the next two decades when the only single cylinder machines in production were M models, more and more new parts fell into the MAS group and the use of the correct prefixes was apparently given up although some Thruxton and Clubman parts entered the hallowed realms of the FA series which had hitherto been almost exclusively the domain of the Mk.VIII.

Following the introduction of the LE, a few of their parts found their way into the single cylinder world, the SL57 series of spring washers becoming LE366 to 369 keeping close company with a few felt washers, bushes and number plate parts, the last being, of course, like the Thruxton battery box, assemblies, and therefore prefixed LAS (LE-ASsembly).

Although the foregoing outlines the development and meaning, of Velocette part numbers there are several unsolved mysteries and innumerable gaps in the sequences. The latter can, in part, be explained by our shortage of parts lists from the early days (if, indeed, any were published) and also by numbers being allocated to parts which were in use for so short a time that they never got into a printed list. Another reason is that, because their business had grown in size and complexity (and very rapidly in 1928 when production doubled) it was necessary for them to number everything that they had in their components stores prior to fabrication which included every type of frame tube, lug, pin, the two separate bits of tube which they welded together to make a gear pedal; everything. Obviously these numbers would never make it into the illustrated spares lists and, except in the case of the MK.VIII, they are lost to us. Also amongst the little curiosities is the Andre steering- damper which, with all its components, was prefixed SD (Steering Damper, I bet!) but for the MK.VIII it was instead labelled KAI IO and, amongst the peculiarities of the MAF, its piston rings were prefixed NIV - all suggestions gratefully received - were they old New Imp. stock?

One major problem with Velocette part numbers is that misprinted lists are very common. I doubt that any of their spares lists was printed without at least one error and usually there are several. The MAF list is a star. The demands of the government resulted in a much better laid-out list but the apparent haste with which it was produced resulted in many mistakes. It is unfortunate that the Introduction specially claimed a degree of simplification intended to assist storekeepers who, having been reminded to check both the printed lists and the illustrations before placing an order, must have been left wondering if it was going to be worth the resulting trouble. The abandonment of certain odd sizes of bolt which were not strictly necessary couldn't have been much compensation and, if all Allied manufacturers had adopted Velocette's inaccuracy, there is little doubt that the war would have been considerably shortened though the result would not have been in our favour.

Apart from illustrating something of the manufacturing Organisation and thinking at Veloce, there are several other interesting lessons to be learned from their Parts Lists. One is that so many parts continued in use, unchanged, for such a long time. That the clutch thrust bearing, and its thrust cup C29/26, date from 1926 is well known; the chaincase inspection cap T71/19K is a 1925 modification of a 1919 part and its spring clip and pillar also date from eighty years ago. It's incredible that any part, let alone an important component such as the clutch operating mechanism, should not only still be in manufacture after so long but still be standard on the last single to roll out of the factory in 1971. What other makers could claim interchangeability over forty-five years? These are extreme examples but interchangeability amongst later parts was of a high degree (until the later 1950s anyway). So many KTT designs, and actual parts, were incorporated in production machines that a claim that racing improved their products simply cannot be denied. Both engine and gearbox components and also cycle parts were common to several models over the years resulting in some strange survivors such as that odd nut (CK20/2) which, in 1953, still held a rigid MAC oil tank to its battery platform but had performed a similar service on the KTP twenty-four years before and had also been used as a locknut on the old adjustable clutch cable stop. The pre-1940 MOV/MAC rear brake plate anchor bolt S36/23AC was a 1923 item modified in 1925 for the Model AC. Another item which apparently could travel through time was the brake pedal F39/4 which graced the Clubman models as it had once done for GTPs and MACs in a previous existence.

Both restorers (and autojumblers) benefit from this kind of thing and my conclusion is that if I wanted to build up a cheap but fast and desirable model it would probably be easier to go for a replica KTT rather than try to Thruxtonise an MSS. It is amongst these later rip-snorters that interchangeability breaks down. Their construction demanded so many different odd bits, especially of tinware and brackets, brakes, bars, seats, stays, footrests, etc. that they must have been a manufacturing nightmare. Was the production of one of these ever actually costed, I wonder? Perhaps Dr. Kelly could tell us the answer to that one, but then, we probably know it already.

Another result of this exercise has been a growing reverence for whoever it is in the VOC's Spares Co. who determines what parts should be remanufactured when he has to choose from 21 types of toolbox, 20 types of number plate and at least 14 types of handlebar (besides the Amal Tidy Handlebar fitted in 1932 and early 1933 only). And as for control cables, bearing in mind that many machines have had different bars fitted and that their owners have changed the levers over the years - they must be an absolute minefield.

It is rare that one can survey a situation without concluding that "it could be worse" and, on the whole - and taking into account the added complexity of printing errors - the Veloce parts numbering system wasn't too bad, and - apart from the development of the MAS group - it never suffered a reorganisation. What sort of twisted mind, I wonder, did it take to come up with Triumph's word-code system?