AJ6 Engineering specialize in EFI (fuel injection) equipped Jaguar engines. Use the direct link above to visit their website and view some of the superb EFI related performance products they have available.

The article is reproduced here by kind permission of Roger Bywater himself.
 

If it starts out good, how do you make it better?

The definitive XK has to be the original 3.4 litre of 83 mm bore and 106 mm stroke (fig 1). This is the size it was intended to be and all other variants were compromised in some way from this, the original design. (Actually, to be pedantic, at a late stage to gain some extra torque the stroke was lengthened from the proposed figure of 98 mm, which would have given 3.2 litres. 40 years or so later a Jaguar 3.2 litre did appear - itself an offspring of the XKs successor, the AJ6.) A four cylinder variant also reached a late stage of development but, in a manner repeated later by the eight cylinder version of the V12, was rejected for lack of refinement and never entered production.
 A feature of any really good product is the difficulty of making it better. In the case of the XK this certainly applies and of the changes it underwent during its production life relatively few stand out as being worthwhile improvements on the original. As an indicator of how rare this situation is, consider another legendary engine - the Rolls Royce Merlin aero engine - which made the Spitfire, Hurricane, Mustang, Mosquito and Lancaster so outstanding during the war years in which the XK was conceived. In its original form the Merlin was plagued with problems and needed an enormous amount of development to become a world beater. The XK was pretty good right from the start.

The most innovative feature of the XK for a volume produced engine was its twin overhead camshaft layout at a time when side valve engines were still commonplace (fig 2). Overhead camshaft drives had a reputation for being noisy but careful development ensured that the XK was quite acceptable in this respect and the twin chain drive lasted with only minor, though numerous, alterations right through to the end. Close attention had been given to proper lubrication of the camshafts and chains and this contributed to the lack of problems in this respect all through the XKs life. Some have questioned the manual adjustment of the top chain but the worst that can truly be said of it is that it was a minor inconvenience. Quite early on the original spring blade tensioner for the bottom timing chain (fig 3) was replaced by a Reynold hydraulic tensioner which itself changed in detail over the years as did the chain guides. 5/16" lift cams were initially used because of a fear that clumsy mechanics used to simpler engines would bend valves during repairs. This caution proved to be unnecessary and the intended 3/8" lift cams were introduced, first as an upgrade option then as standard. Another early change was from a gear oil pump to the Hobourn Eaton rotor type.

Some other changes were clearly improvements - for instance the much revised long stud block gave a more direct load path between head and main bearings, provided more secure clamping pressure on the head gasket and eliminated the bore distortion which had become apparent from stretching the block to 4.2 litres. (This lesson was not forgotten when the AJ6 block was designed - the head clamping stresses being taken down via heavy ribbing in the outer wall of the structure). Mechanical noise was reduced by the introduction in 1969 of modern parabolic cam profiles applying less severe accelerations to the valve gear than the simple three arc cams used previously.

I hesitate to mention the dreadful sludge traps in the XK crankshaft - probably its worst feature. Any minute debris which could get that far into the engine could not possibly be harmful yet by collecting it together a potential problem was created for the later life of the engine.

Cylinder Head Evolution

Debatable are the merits of the so-called "straight port head" with which the XK ended its days. In the original cylinder head design the inlet ports had quite noticeable curvature to impart swirl to the income charge so that it would rotate around the cylinder axis (fig 4). Such swirl is considered useful as an aid to efficient combustion and was very much the right thing to do in the days before the greater benefits of horizontal swirl, a feature of modern short-stroke 4 valve engines, was recognised - this taking more of the mixture through the vicinity of the spark plug to achieve better flame propagation. There was of course an uprated version of the curved port head derived from the C Type racer but confusingly known as the B type head. Unconcerned as it must be with part-load operation a racing engine requires maximum flow, this in itself providing ample charge turbulence (arguably more effective than swirl) under full throttle conditions, and so straight port heads were soon introduced for racing and then subsequently appeared on the higher performance road cars. Gas velocities are higher with larger engine sizes so as the XK was stretched to 3.8 and finally 4.2 litres so the advantage swung more in favour of the straight port head and it made some sense to rationalise on the one type. The XK cylinder head, in its various forms, always possessed above average gas flow properties but the valve included angle was much wider than would now be thought desirable and the resultant deeply hemispherical combustion chamber was, with hindsight, less than ideal for good combustion. Although no comparison was ever made it is quite probable that the original curved port head would have provided better mid-range torque, better part-throttle economy, and lower exhaust emissions. The larger bore of the 3.8 and 4.2 engines created a peripheral squish band to push the outer layers of mixture back into the combustion chamber but this was not of much consequence. With regard to the deficiencies of part throttle combustion it is interesting to consider how far back the catalyst was placed (almost behind the transmission) in the carburetter emission cars, all of which had straight port heads. This indicates that a fair amount of burning took place downstream following air injection into the exhaust ports and the catalyst was moved back to keep its temperature within bounds.

Of course, the ultimate racing head was the so called wide-angle 35/40 version (fig 5) used on the later D Types and the light weight E Types with various sizes of valves and in single and (rare) twin plug versions, the latter showing no measurable benefit. The numbers relate to the valve angles away from vertical and in this case the exhaust valves were shifted outwards by 5 degrees to permit fitment of larger (2" then 2 3/32") inlet valves. At the same time the tappets were increased in diameter to allow higher lift cams (7/16") to be used. In the continuing pursuit of power the 3 twin choke Weber carburetters of the early D types gave way to Lucas fuel injection with carefully developed ram induction tracts and slide throttles (fig 6). Technically this was a great leap forward so much so that the same basic injection system survived in Formula 1 racing into the 1980s before it was finally ousted by electronic engine management.

Aluminium Cylinder Blocks

The later Ds and the "light weights" also had aluminium cylinder blocks so were a very substantial departure from the production engine. There were also some aluminium block race engines built to the 3 litre class rules with 85 x 88 mm bore and stroke rather than 83 x 91 (or 92) mm used with equivalent cast iron blocks. Initially head gasket failure was a problem, caused by the cylinder liners sinking and relaxing the gasket "nip", but this was cured by adding flanges to secure the liners at the top. The original aluminium main bearing caps proved to be inadequate and were replaced by steel items but at the time an aluminium cylinder block was quite innovative so some difficulties could be expected.
By the standards of the time these were pretty powerful engines though the project lacked the commitment of earlier days, and they never really came to terms with Ferraris all conquering V12s.

Once the aluminium block had been made reliable there were plans to develop a production version and one or two 4.2 engines were built. They proved to be unacceptably noisy and were abandoned after a short time.

Fuel and Sparks

A number of carburetter arrangements were used over the years, the twin SU set up being the most common although the carbs themselves varied in type (H, HD, HS, HIF) and size over the years from 1.75 inch on the original XK120 and most early saloon engines to 2 inch in later years as most enthusiasts will know. 24 mm Solex down-drafts were used on most 2.4s although 1.75 inch SUs were an option which became standard on the later 240s and of course the MK10 and E Type had the triple 2 inch SU setup. From the late 1960s emissions legislation in the USA meant that SUs in any configuration had to be abandoned for that market, to be replaced by a pair of CD Strombergs with their greater accuracy and more sophisticated control of warm up fueling. Accurate warm up fueling was never a strong-point for the SU company - the electric choke devices used with their early carbs were barely satisfactory whilst the later AED was really pretty hopeless.
Finally even the Strombergs were superseded by Bosch/Lucas L Jetronic Fuel Injection which made it possible for the XK to meet ever tighter emission regulations in the late 1970s using Lambda feedback of exhaust oxygen content and a three way catalyst. This EFI system relied on air mass flow measurement rather than manifold pressure/speed as used with the V12 and there was much debate about how best to apply the calibration measurements from the test bed. It was not that there were errors in the work, just that it could be interpreted in such a way as to introduce errors so that the overall fueling could be about 4-5% richer than it should have been. All this was resolved however long before reaching production.

A characteristic possibly unique to injected engines using the twin tank system of Jaguar series 2 and 3 saloons was that of weak back-fires indicating that a tank had run dry. On early development cars this could blow the inlet elbow off the throttle thereby immobilising the car. When this was made more secure these back-fires found the next weakness, slamming the airflow meter flap shut with sufficient violence to damage the pivot bearings. A spring relief valve mounted in the flap proved ineffective and the cure was a rubber buffer for the flap to close against.

 Incidentally, the 4.2 EFI engine, aided by the largest of all production inlet valves at 1.875", and rated at just 200 b.h.p. DIN (that means certified) was almost certainly the most powerful production XK ever. One might say that there was a degree of optimistic exaggeration about the 265 b.h.p. claimed for the earlier triple SU engine which was never verified under similar conditions.

Less well known are those carburetter arrangements which never made it beyond the experimental department. Various down-draft and side-draft configurations with as many as four carbs were tried but I understand that the most effective by far was a sort of reversed triple SU setup with long ram-stacks extending out to three small plenum chambers from which the carbs pointed inwards between the pairs of stacks. Apparently this even had advantages over the triple Weber setup used for racing but it lacked the tremendous visual appeal of the conventional triple SUs and was abandoned largely for that reason. Another promising long ram system that did not see the light of day was developed with the aim of giving the 2.4 a much needed torque boost.

Ram length has always been crucial to getting the best out of any Jaguar engine but sadly the production versions always seem to be compromised by the available space. This was never the case with the racing XKs, other than very early ones with twin SUs, as would be obvious from a glance at the long trumpets of the triple Weber and Lucas Injection (fig 6) systems mentioned earlier.

For many years ignition was by a conventional contact breaker but in 1978 US emission engines began to use a version of Lucas OPUS, successful enough in top level racing and a little troublesome on the Jaguar V12, yet not far short of disastrous on the poor old XK. The act of cramming everything inside the distributor body resulted in an unexpected sensitivity to heat having the effect, if not of failing permanently, of periodically shutting the engine down for about 20 minutes until the critical components cooled down again. OPUS was hurriedly replaced by the new Lucas Constant Energy system featuring automatic control of coil current to give consistent spark energy over a wide range of engine speeds. This lasted until the end with centrifugal and vacuum advance mechanisms so we never had the chance to see how much the XK might have been improved by full electronic engine management.

The XK Joins the Army

Anyone wandering around certain areas of Jaguar in the 1970s or 80s would from time to time see a rather odd looking XK painted all over in a bland pale green, with a huge black carburetter on top of the inlet manifold. This would be the military version used to power the Scorpion light tank and its derivatives. Aside from the enormous down-draft Solex carburetter, which would have looked more at home on a wartime aero engine, another external difference would have been the massive distributor, completely screened and water tight, with heavy screwed-in braided leads trailing away and screwing onto the bodies of unusual looking spark plugs. Understandably, military folk dont like ignition systems which cause interference or cant stand a bit (perhaps a lot) of damp.

A revised breather system with flame traps was another demand of the military mind which not surprisingly has a serious aversion to engines fires.

Internally these engines had low compression (7.75:1) pistons so were not dependent on having a supply of high octane fuel. Also the tappets were larger than those used in normal production engines so that they could accommodate a collapsing spring and washer mechanism to ensure positive rotation of the valves, with the aim of making valve seat wear less likely in arduous conditions using doubtful fuel.

More Mechanical Changes

Coming back to earth with production changes, one which served no purpose at all was that from two to four bolt fixing for attachment of the vernier sprockets to the camshaft flanges. Two bolts had given no trouble whatsoever so why the change? Believe it or not, it was the result of a cost cutting exercise which went wrong. The serrated vernier plate is a cumbersome thing to produce and the plan was to replace it with a simple pressing rather like a pastry cutter. This would not work with two bolts so four were tried before the futility of the scheme became obvious, but the four bolt fixing remained from that day on, thought by some for no better reason than to protect the reputation of the perpetrator.
Actually there is something of an anomaly regarding the vernier sprockets. They permitted very precise adjustment of cam timing, yet the accuracy of phase angles between individual cams along the shafts was not always as good as it should have been and as far as the 5/16" lift cams were concerned might best be described as "vague". The 3/8" lift cams were much better in this respect.

The 87 mm bore 3.8 litre was the maximum practical stretch within the original cylinder block and was a good example of a speculative modification proven under racing pressure feeding back into production. The 2.4 (83 x 76.5mm) was really rather a disappointment in terms of specific performance because the ports and valves were just too big for that size of engine. In any event the twin cam / hemispherical combustion chamber layout never really worked so well with short stroke engines partly because of the piston shape needed to achieve a reasonable compression ratio. The later 240 engine was encumbered with the straight port head and only retention of the mild 5/16" lift cam enabled it to produce acceptable mid-range torque. At least the weight of these small capacity engines, including the later 2.8 litre, was reduced by use of a shortened cylinder block accompanied by shorter conrods.

Prior to the 4.2 all XK cylinder blocks were cast with space for coolant to circulate all the way around each cylinder (fig 1). The 92.07 mm bore of the 4.2 left little room for such a luxury and also required the bore centres to be moved apart so the inner and outer pairs became offset from the combustion chambers in the cylinder head which remained unchanged. Within the casting the cylinders were now joined together in two groups of three, a so-called siamesed block. Coolant flow between cylinders was still thought desirable and was achieved by the rather cumbersome means of machining a hole through from one bore into the next just below the top deck. Of course this meant that another change had to be introduced - linered bores to seal off what now became narrow coolant passages. It was to be many years before the slotted block was introduced to permit coolant flow between bores without the need for liners. In this respect the 4.2 was a bit of a bodge driven by the need for extra capacity.

When the 3.4 engine size was resurrected in the mid-seventies it was really just a small bore version of the 4.2 with siamesed bores, liners and straight port head and was some way removed from its illustrious ancestor. This 3.4 always had S.U. carbs like its namesake but was cursed with the appalling AED which the 4.2 was fortunate to escape from for its final years. Nothing highlights the superficial nature of some aspects of Jaguars much hyped "Quality Drive" of the early 1980s more than the fact that the AED was never junked from the 3.4.

The first obvious signs that the XK was becoming fragile must have been when 2.8 XJ6s began burning pistons. Strangely, a hard driven 2.8 would run for ever, and the Jaguar test fleet in those days would have been hard driven in the belief that it was the sure way to find a weakness, yet customers found that after a few weeks of gentle motoring in town, a full throttle burst to get up to motorway speed could "smoke" a piston in very short order. The reasons for it are somewhat obscure and there are varying opinions on the matter but my understanding is that extra piston dwell around TDC on account of the shorter stroke might have played a part allied to a humped piston crown that came into close proximity with the hot exhaust valve. In essence, if the geometry is such that the piston is unduly slow in accelerating away from TDC then pre-ignition can be more easily promoted by soft deposits laid down in the combustion chamber during light load operation. The official name is "Deposit Induced Pre-ignition" and there can be a narrow margin between survival and failure for the brief time it takes for the deposits to burn off. Unfortunately, in the case of the 2.8 the condition could, on occasions, last long enough to melt a piston.

The Manufacturing Facilities

Perhaps this is time to look at how the XK was made. It is fairly common knowledge that (Sir) William Lyons bought the machinery on which the XK engine was to be produced from Sir John Black of Standard Cars in circumstances in which it was thought prudent to whizz the payment cheque round before Black could change his mind.
 So, an engine of advanced design was to be made on second hand machinery. An astute person can profit from such a situation if the machinery is good enough, but problems arise if the machinery wears out and can no longer maintain accuracy. I doubt if it was even considered that much of this same machinery would still be in use 30 years and more later although of course some of it had to be renewed.

In fact the XK engine was never a true mass produced engine - it was selectively assembled with various grades for cylinder bores, pistons, and even gudgeon pins. The cost of skilled labour increased with time, whilst the ability of the machinery to produce accurate components declined as it wore out, so by the 1970s the poor old XK really had its back to the wall. Who would want to invest in new machinery to keep a near 30 year old engine going? - certainly not British Leyland, who then had control - and would a wise man advise otherwise? Of course, BL was never awash with money and Jaguar was probably regarded as having a need for substantial investment which they were not in a position to provide. In more recent times I am sure Ford were unpleasantly surprised to find how much it was going to cost to get Jaguar up to date. Perhaps it was easier for BL to just let it sink or swim and apply pressure to keep production up at all costs in the interests of desperately needed short term cash flow, regardless of any effect on future sales. It is easy to blame BL for Jaguars ills, but in fairness BL did not have a total monopoly on bad decision making. If you doubt that just consider the bland and unimaginative body colours offered by Jaguar in the 1970s.

What comes next?

Of course, an XK replacement should have been initiated at a much earlier date, but the intended V8 derivative of the V12 was just not good enough, neither were the 2 and 4 valve slant six prototypes based on one cylinder bank. Had this been known at the time it is quite possible the V12 would never have made it to production. Actually one could easily look at the now defunct 2.9 variant of the AJ6 engine and imagine that it descended from the V12 and Slant Six, but in fact the AJ6 evolved from a late 1970s four valve engine based on the XK block (fig 7). The existing head stud pattern necessitated the curious feature of having the head bolts passing through the camshaft caps which carried over onto the AJ6 and later AJ16 engines. As the project progressed the cylinder bore spacing was increased to be the same as the V12 so that the May head could be more easily adapted for the 2.9.
70s Decline.

As the 70s progressed so the shortcomings of the poor XK became more exposed and it began to wilt. Bore grades were reduced in number, as were piston grades, so piston/cylinder fits considered too sloppy to be acceptable in the 1950s became OK for the 1970s. Piston slap from cold became noticeable, as did "little end knock" which, strangely, could be best heard by standing a few yards in front of the car. Most engines were not too bad but the worst were getting to be a bit clanky for a "Quality Car".
Around this time the sump was altered and this introduced further problems. To meet the ever tightening exhaust emission regulations in the U.S.A. the XK was equipped, as we saw earlier, with L Jetronic fuel injection and a three-way catalytic converter, which needed to be placed nearer to the engine than had been the case with the earlier oxidising catalyst. The only way the catalyst could be accomodated was by cutting off the "ear" of the sump which was in the way. Somebody said "why not cut both ears off the sump, simplify the casting and just raise the oil level" - so that was done. Sometime after that it was noticed that if a car was left idling on a hill then one of the crank throws would hit the oil and make a noise easily confused with "big end" knock. Just to compound the problem an epidemic of real big end failures started around the same time. Spates of big end troubles were not unknown on the XK but the outbreak in 1978 was the worst by far.

A situation now existed in which production engines could display several faults:- piston slap, little end knock, oil slap, and big end knock. The latter was further complicated by variability in the surface finish from the crank grinding process and because some engines were not noisy but could suffer sudden bearing failure, whilst others knocked quite audibly but did not fail. Around the same time it had been decided to relax the interference fit of the tappet guides, which seemed all right for a while until they started working loose in service. (This problem gave graphic proof of the quality of Jaguars chill cast camshafts which could happily bash a loose tappet and guide to smithereens without suffering in any way.) Even the well proven skew gear drive to the distributor began to fail following a minor modification. Another change was that the block settling time was cut from, I think, 6 months to 3, referring to the practice of leaving castings outside to stabilise and for stresses to be relieved naturally before machining. There is no such thing as a totally rigid engine and it was known that deflections of as much as 0.010" could take place around the centre bearing structure of the XK block under high speed loads, which gives some idea of the stresses generated even in normal use. The reduction of settling time for the block may well have contributed to the formation of cracks between bores which now became common with the slightest overheating provocation. The so-called slotted block mentioned earlier made its appearance around this time to simplify production. A shaft carrying a gang of circular cutters machined transverse slots between each pair of head studs to create coolant passages between bores without need for liners. For some reason a few of the early slotted block engines burnt oil heavily whilst others were fine - by XK standards - although no XK ever set records in this respect.

Really, apart from the 2.8 situation, most of the problems were largely the consequence of the XKs age and the state of the tooling with which it was made - not to mention a dispirited workforce aggravated by myopic management from on high. The exception was the bearing trouble - and to find the origin of that we need to go back in time - to THE Mk X SALOON.

The Bearing Saga

Now it turned out that the Mk X bodyshell could resonate in sympathy with, and amplify, normal main bearing noise from the crankshaft, so rather than redesign the bodyshell it was easier to alter the bearings to detune the crank vibrations to a different frequency. Actually a very effective way of eliminating main bearing noise is to bend the crank slightly (or even deliberately machine it out of true as some manufacturers have been known to do) but this was thought impractical as a production process. The method chosen was to switch from a full oil groove main bearing to a half groove arrangement with the lower shell being plain over most of its surface. The XK always had cross drilled main bearings and therefore would still have oil feeding into one end or the other so an adequate supply to the crank pins should still be assured, one might think The trouble is that a whirling crankshaft generates centrifugal forces in the oil within its passages, and the oil has to overcome this force to enter through the main bearings before being flung out to the crankpin, any entrained air tending to remain at the centre of the crank. The centrifugally generated pressure at the crank pin rises as the square of the speed of rotation, and at high speeds will be considerable, so any excessive clearance in the bearing will allow too much leakage. In fact some early development was aimed at curbing centrifugal pressures and resulted in the right angle drillings at the crank pin but the situation had now changed for the worse.

The oil entering the crank drillings from the main bearing grooves is also subject to centrifugal force which opposes this flow and a certain critical speed will eventually be reached, hopefully well above normal operating speed, when the quantity of oil flowing into the crank will not make up for the leakage at the crankpin, with catastrophic results. Very high revving race engines use high oil pressure and have the drillings only just below the surface of the bearing journal, in any case of much smaller diameter, to alleviate this problem. Cosworth coined the term "low pressure crankshaft" when they applied the principle to the legendary DFV, but as far as I am aware, the idea originated in the BRM V8 of the early 1960s (fig 8).

At low speeds the situation is quite different because then there is negligible centrifugal pressure so gallery pressure alone feeds the oil into the crankpin bearing. The success of the entire bearing system relies on a delicate balance between gallery pressure, centrifugal pressure, oil passageways and bearing clearances and whilst most engines have a generous margin of safety, in the case of the XK changing to a half groove main bearing considerably reduced it. It is perhaps now easier to see how the oil flow into the cross drilled main bearing, far from being steady, actually must undergo constant reversals as the feed changes from end to end thereby reducing the average flow. To make sure that oil would still be able to form an adequate lubricative film on the crankpin Jaguar did something rather unusual - THEY MADE THE BIG-END BEARINGS OVAL.

The idea of this was that by providing extra clearance in the lighter loaded areas around the sides less oil pressure would be needed to fill the bearing yet the average clearance would still be close enough to prevent too much leakage. Of course we are not talking about out-of-round crank pins, the bearing shells were formed to give the required clearance (fig 9). The idea worked - most of the time - but now and again tolerance conflicts would stack up, possibly aggravated by lapses of crank grinding quality, to cause a spell of bearing trouble. Such events were often signified by a change of the drilling arrangement of the crankpins. Some were cross drilled, some had a single drilling at 90 degrees, others had a single inclined drilling, and some had a slight relief across the drilling. Each change was sufficient to shift the delicate balance of the bearings back to a safe condition to suit the circumstances. The primitive manufacturing arrangements made such changes easy. XK crankshafts were drilled on a sequence of ancient radial arm drilling machines (really a jobbing type of machine) locked in position over a jig in which the crankshaft was placed.

The bearing ovality was not great - about 0.002-3"- and continued for many years until around 1977-8 when it was decided that the bearing shells could be made more easily (cheaply?) if they were circular and the necessary side clearance achieved by machining chamfered reliefs at the mating faces (fig 10). Clearly this change upset the delicate balance on which the bearing depended for survival and small variations of clearance in the critical relieved area, in combination with normal bearing tolerances, could make the difference between insufficient oil flow into the bearing and too much leakage through it. In fact the change had gone ahead under pressure without being thoroughly tested and the end result was that for a short period almost every single engine was a potential failure and some could not even make it through the factory gate. In retrospect one has to wonder why, after the Mk X and 420G were discontinued, nobody thought about returning to a conventional bearing system. Really the problem could have been brought under control by tufftriding the crankshaft, as on the V12, to give it a hard wear resistant surface better able to tolerate marginal lubrication conditions, but it was 1983 before this happened. The delay was because of fears that the treatment would add to the problems by causing distortion of the crankshaft.

In a small way I was personally involved in these bearing problems of the XK engine during 1978. I knew then that it was a fragile engine but it was only after I left Jaguar and encountered cars not long out of warranty with serious engine problems in other ways as well that I realised how bad things really were. Most engines would survive the warranty period but after that they could be really bad news, yet no knowledge of this had seemed to penetrate the factory environment. It wasnt so much that the engine would expire without reason, it had just become very unforgiving of the slightest hint of neglect. An all too common example was that if the wrong antifreeze were to be used the radiator would lose efficiency due to deposition inside the tubes, causing the engine to overheat. Most other engines would be reasonably tolerant of such abuse for a short time but an overheating XK from around 1980 could be a total wreck in minutes and a simple hose failure could be catastrophic. Strangely, whatever damage occurred, be it loose tappet guides, gasket failure, bearing failure, block or head cracking, (often all of these together!) always seemed to start with the next to back cylinder. This would indicate that this cylinder was more thermally loaded than the others, probably because of local stagnation of coolant flow in extreme conditions, but such things happen to some extent in most engine designs.

Did nobody question the high number of exchange engines that must have been sold during those times? - or did they just rub their hands and take the money?

Ones judgment is surely coloured by circumstances and if I had been party to the XK successes of the 1950s then I am sure my attitude would be different. As it is, I experienced the XK at its worst so it is not within me to regard it with much affection. It is too much like the proverbial curates egg for my taste.

Roger Bywater.
AJ6 Engineering