Mounted on its distinctive, wide-base "pyramid" shaped sheet-steel cabinet stand the first 5.75" x 20" Chipmaster, left the production line on July 24th, 1957 - though it had been announced as early as 1956 (Serial numbers are shown at the bottom of this page). The lathe was to occupy an important niche in the Colchester model range (production rates varied between 300 and just over 400 units per year) and remained in the catalogue long after other models from the 1950s had disappeared - the last example being dispatched during 1983. Badged for some export markets as the Harrison 10AA, with its infinitely variable-speed drive and 3000 r.p.m top speed, it was intended as a small but versatile high-speed precision lathe and was to form the basis of the later, cheaper and conventionally-driven Bantam. Being relatively complex, the Chipmaster was always an expensive machine - in some years it cost over 46% more than a similarly-specified but larger "Student" model for example - and in 1959 Colchester announced a short-lived, lower-cost alternative version, the "Six-Speed Model" stripped of the variator and with its spindle driven directly from a 3-speed motor rated at 5 h.p., 2.5 h.p. and 1.5 h.p. to gave spindle speeds of: 75, 150, 300, 750, 1500 and 3000 rpm. The lathe was fitted with a distinctive carriage-handle-shaped control (connected to a Stanton electrical switch) in place of the rotating dial of the variable-speed type. The design, construction and detail finishing on a Chipmaster was of an exemplary standard - as it had to be with a top speed of 3000 rpm - and was heavily built, the approximate weight of a standard machine being some 1204 lbs (545 kg), only very slightly less than a Student. The bed, hardened to order only at first but later fitted as standard, was particularly stiff, with chip clearance through elliptical holes that passed to the rear. On its introduction, the Chipmaster was fitted with a two-speed motor, but this was quickly dropped in favour of a single-speed version - though the speed range was virtually unaltered. At least two generations of this early variable-speed model were made: the first had its 3 h.p. electric motor (though often 1.5 h.p. on those sold to training establishments) and a comparatively small Kopp swash-plate variator (with 3 rather than the later 6-ball bearings) constructed as one integral unit with a tall aluminium casting bolted to its side to support the shaft coming from the speed-change handwheel. From the variator the drive passed, using twin V-belts, to a first design multi-plate clutch (thought to be from Colchester's drawing boards) mounted on the headstock input shaft (immediately behind the main spindle) and from there to the headstock spindle by three V-belts on early machines and a toothed belt on all others - the latter element of design then surviving unchanged for the rest of the lathe's production life. The clutch was not the world's best, and was to be replaced by a much more efficient Matrix unit, the original being all too easily knocked into and out of engagement. However, the novel illuminated ball on the end of the clutch lever was retained - this glowing a pleasing red colour when working. Early versions of the Chipmaster are rare and problems with the drive system, probably concerned with its reliability, must have forced a serious rethink by the designer. On second-generation versions the motor was mounted facing inwards and positioned underneath the plate that held the variator. The latter was now larger, with six ball bearings in the ramps, obviously more durable and driven from the motor by a toothed rather than ordinary V-belt (the change to a proven multi-plate clutch by Matrix also helping). The next stage in the model's evolution was to improve the manner in which variator and motor were mounted: originally, just a simple flat steel plate had been used, supported on three height-adjustable, thick-walled steel tubes; the simplicity and comparative lightness of that design suggests that vibration may have been a problem, for its replacement was a large and very rigid U-shaped cast-iron housing mounted on anti-vibration pads.
It appears that a third generation model was also made - though these are very rare - this having the Kopp variator replaced by a pair of pulleys, one opening while the other closed with the wide drive belt moving up and down their flanks - and so able to vary the drive ratio. If you have one of these machines, the writer would be keen to see some photographs.
Final drive on all models was through the virtually-indestructible Matrix clutch - although on later machines the unit was increased in size - a unit that then allowed the motor and variator to continue running with the spindle stopped. On the first two models the cast-aluminium speed-selector dial had an extra plate, positioned above it, indicating that the readings should be divided by 10 when running in slow speed Electrical control was by a handle protruding from the top of the headstock and, if wired correctly, moving it to the right from the central "off" position caused the spindle to run forwards, moving it left engaged reverse (this was the lever fitted with a transparent plastic ball that glowed red when the machine was running). The Matrix multi-plate clutch was controlled by single lever on the front face of the headstock: moving the lever to the right operated the clutch while pressure to the left engaged a spindle brake. On a safety point, early machines had no safety catch and accidentally catching the lever could engage the drive; later models were fitted with an improved, spring-loaded mechanism that required the operator to first pull the lever outwards before it could be moved sideways.
If the clutch appears not to be working correctly, it might be that a previous owner has misunderstood its construction and adjustment; the answers are in the high-quality and comprehensive data pack we can supply and also outlined below.
On the end of the central, splined shaft is a collar, this being concentric with another larger diameter splined shaft - on which fits the clutch body. On the inside of the collar are three inward-facing lugs, these fitting over three rollers carried on the clutch body. The lugs must engage fully over the rollers, if they don't the clutch will not operate--but the unit will fit together and give the appearance of being assembled correctly. As the clutch lever is operated the rollers are moved inwards, up the ramps, this action engaging the clutch. In order to line things up correctly the inner and outer elements have to be rotated relative to each other, one spline at a time, until the collar will go over the rollers. On the outside of the clutch is a knurled ring that slides on a spline; this is pulled out, rotated one spline to the left or right then released to seat again - and by this means the clutch clearance is altered until the collar goes back over the rollers with a positive action - yet the clutch will not drag or slip. If the collar will not seat over the rollers, back off the knurled ring adjustment until it will. Having used the clutch under power several times to bed things in, it might be necessary to adjust the knurled ring one more time; it should then be set for many hours of hard use. Pictures of this assembly may be here and if not, are reproduced, with thanks, here.
Continued below:

Continued:
The 15/16" (35 mm) bore headstock spindle ran in Gamet "micro-precision" taper roller bearings (made by another company in the Colchester group) with the lower part of the speed range obtained using hardened and ground gears. Because the high-speed range was directly by belt, and the comparatively short and rigid spindle so well supported, the lathe had a well-deserved reputation for being able to produce unusually smooth surface finishes. An American-type D1-3" Camlock nose fitting was used, that allowed the lathe spindle to be safely reversed at high speed, and the spindle was sleeved with a hardened 4 Morse taper short sleeve bored to accept a standard No. 2 Morse taper centre (this fitting is often missing, and expensive to replace - though sometimes cheaper replacements are available). Unfortunately, chucks with an integral D1-3" mounting are pricey to manufacture and it is often cheaper (although with only a little saving) to mount a replacement chuck on a new Camlock backplate. Certain precautions are necessary when mounting new D1 accessories on the spindle nose and it may be necessary, in order to achieve maximum grip, to re-set the Camlock studs within them - it's essential to read the maker's instructions on this point if you are unsure.
All Chipmasters were fitted with a full screwcutting gearbox that offered a greater range of threads than the Bantam. The drive to the gearbox was by changewheels for screwcutting or, for extra-fine feeds, via a V-belt on early lathes and a toothed belt (for a more positive drive) on later models. A lever on front of the headstock - annotated with a belt and gear symbols - selected the appropriate drive. For many years only an Imperial (inch pitches) gearbox was offered, with conversion gears to cut metric threads. However, from the end of 1970 onwards, a version with a dedicated metric gearbox was offered - though at extra cost this box could be provided with inch-pitch transposing gears, the set comprising: 20t, 24t, 28t, 30t, 36t, 42t, 44t, 46t, 52t, 56t, 57t, 60t, 65t, 69t and 70t. The Imperial box can be recognised by its sliding lever control with indent positions (outboard of the box to the right) whilst the metric box was similar to the unit as used on the Bantam with a joy-stick control in addition to levers. At the end of 1970 an "all-metric" version of the Chipmaster became available and carried the identification "Continental"; again, fitted with the appropriate transposing changewheels English to metric and metric to English conversions were possible.
Identical to that used on the Colchester Bantam, the tailstock spindle was equipped with a zeroing micrometer dial and has been found engraved with inches, mm and dual inch/metric markings. Listed as having a No. 3 Morse taper (and all examples seen by the writer have been on the size) some reports indicate that a slightly large size was used, though it's specification remains, for the moment, unknown.
Once in private hands, after hard industrial service, the Chipmaster could suffer problems connected with the drive system; the variator has always been enormously expensive to overhaul and, if it started to produce any untoward noises, would rapidly assume the volume made by a washing machine filled with ball bearings and scrap steel - and if you didn't stop using the lathe immediately, that's more or less what the variator turned itself into. If you are thinking of buying a Chipmaster it's absolutely essential to hear it run - if the seller cannot arrange this or has "disconnected the electrics" assume that the variator is finished and value the machine accordingly - no matter how smart the paint may look. However, ever cloud, etc. for, if the variator appears to be faulty, don't worry, you don't really need it. Dump it into the scrap bin (take it apart first and have a look, just in case) and couple the 3-phase drive motor to a 1-phase to 3-phase inverter. Inverters, unlike simple capacitor-based "converters", provide a variable-output frequency (giving variable-speed drive) and hence obviate the need for any mechanical speed-changing device. They are also inherently reliable - and prices keep falling. The Matrix multi-plate clutch fitted a standard to all Chipmasters not only makes the machine much more pleasant to use but also gives any 1-phase conversion (or electronic control, system) a much easier time.
Some few years ago a friend bought a new, still-crated Chipmaster which had lain forgotten in a store (that's another story). He removed the variator and arranged for the spindle to be driven directly from the electric motor through a 3-phase to 3-phase variable-speed electronic controller. He over-rated it, to give a top speed of over 4000 rpm, and rigged the DC injection braking to make it stop it in a couple of rpm or so (the first time he tried it the motor stopped dead, but the inertia of the rotating parts stripped the drive belt of its teeth.). This machine ran reliably for years, turning out thousands of critically-dimensioned and beautifully-finished aero-space components.
Although it took Colchester a little time to get the drive system right, the rest of the lathe (apart from the drive system) remained essentially unchanged throughout its production life - proof that the original design was well thought out and the concept - a machine that could be marketed as a lathe suitable for toolroom, production and general workshop duties - was absolutely right. Although the main mechanical parts of the lathe stayed the same the variator did not - this going through several undocumented changes with which the handbooks issued did not, sadly keep pace. The main difficulty appears to be getting the oil level correct - and its vital to put in only the amount stated on the body of the unit (if so marked) or 1 pint if the data plate is missing. Completely filling the unit will cause damage, as will the incorrect oil. The numbers assigned to the proper oils have changed over the years; the last-named in the following list being the one to use.
Headstock: Shell Tellus 15 = Morlina 10  = Morlina S2 BL 10
Gearbox: Shell Tellus 33 = Tellus 68 = Tellus S2 M 68
Variator: Shell Vitrea 21 = Vitrea 22 = Process Oil PB 33 = Morlina S2 BL 10
From this it can be seen that headstock and variator oils are now the same - and consistent with recommendations from Allspeeds UK, the manufacturers of the Kopp variator (and suppliers of Kopp variator oil)..
Some Observations:
In addition to those drive systems illustrated and described there was also at least one "odd" or "interim"  model - on this the mounting plate and variator was of the same style as the first models, but with the motor sitting under the variator plate - a second dual-belt arrangement being positioned towards the back of the variator. Arguably, this arrangement was as per the second-generation type, but with the output by dual V-belts - and with no oil filler on the front of the variator. The system was certainly a factory-built type, the mounting plate being of the original design and not having the newer-type support bracket. The earlier doors (i.e. covering the belt/headstock end) are in aluminium but on later models in fibreglass.
Inside the base, additional strengthening cross-sections are present in later models. This may relate to a very recent discovery found in the top of the base section: the later ones have a flat surface, while the earlier have ridges inset from the outside edge to locate on the inside edges of the bed section.
Later aluminium beds, while externally dimensionally identical to that of the earlier, had walls almost twice as think and, at 75 kg, were therefore substantially heavier and - one assumes - stiffer Annoyingly, because of the ridges, the later beds will not fit the early stands (unless you somehow grind away the excess metal...). Late stands, though similar in appearance, were constructed from welded sheet steel.
On the bed section, under the headstock, the later models lack the two large holes found in the earlier - a good thing as the holes caused coolant to draw swarf into the bed section Some Chipmaster beds have been found with triangular holes on both end faces - although that at tailstock end is obscured by the manufacturer's plates Additionally, on these beds are recesses both front and back; their purpose is unknown, but possibly used to clamp the bed down when the ways were finish ground.
Early beds are believed to have had their serial number stamped into the little circles of red-coloured aluminium and not, as on later examples, into the bed itself. It might be that early spindles have a far longer section of thread around the gear together with a toothed locking plate.
It has been noted by some users that removing the headstock spindle and bearings is largely a non-destructive and simple process. However, the oiling system might be a possible cause of premature bearing failure for, located at the top of the housing is a pipe that runs in the same orientation as the spindle. The pipe has its upper half cut off, this catching splashed oil and redistributing it, through small holes, into the bearing space.  However, the recommended hydraulic oil tends to congeal with age and block this hole with one owner reporting that his rear bearing appeared to be receiving too little oil - although it had survived in good condition. Unfortunately, there seems to be no official guidance as to how the spindle bearings should be pre-loaded - and fitting can be difficult. However, heating the headstock casting and cooling the spindle overnight in a domestic freezer set on maximum might just help..

The bed, hardened to order only at first but later fitted as standard, was particularly stiff with chip clearance through elliptical holes that passed to the rear.

Second-generation Chipmaster with motor and variator mounted on the bottom and top faces respectively of a simple 3-point support steel plate and drive to the clutch by a toothed belt.