While we prepare our next batch of tera® Quad-Cam V12 engines, I thought you may be interested in a little background to the project ...
Before the V12, Jaguar’s racing and practically all road cars were powered by the powerful and renowned XK straight-six double overhead-cam unit. This engine had its origins in pencilled sketches drawn during the London blitz by Sir William Lyons and his engine designers; William Heynes (Chief Engineer), and Claude Baily. These sketches and original designs were followed by working prototypes as early as 1943. The first 3,442cc production unit saw the light of day in the beautiful XK120 of 1947. The same basic engine continued production into the 1990s.
As can be seen from the original drawing from my own archive reproduced below, Jaguar ended up with a design where both inlet and exhaust valves were inclined towards the centre-line of the hemispherical combustion chamber at 35°. This was changed to 30° inlet and 45° exhaust for the ultimate "wide angle" head used in racing engines. The valve angle was modified simply to allow the use of larger inlet valves.
Original Drawing of the very first 6-cylinder "XK" 3½-litre 35°/35°
© Neville Swales
In the 1950s/60s this hemispherical type of combustion chamber was considered ideal for high-performance engines because of reduced valve "shrouding" compared to a "flat-head" design and a low surface-area to volume ratio. As can be seen from the following photo taken of a head I sectioned, the ports and valves were arranged more or less in-line across the engine. However, Weslake worked closely with Jaguar when the engine was being designed and he introduced a curvature to the inlet port in an attempt to allow charge movement inside the cylinder ("swirl"). This was done to aid combustion efficiency and is evident in the following photo.
Sectioned Jaguar XK Cylinder Head - "Curved" inlet port on right
© Neville Swales
Weslake's modification, whilst introducing swirl, was compromised by the need to place the spark-plug off to the side so as not to interfere with the valves. A central spark-plug would have been ideal in this situation. Many designers of similar engines tried to improve the situation by introducing a second spark-plug on the other side of the chamber but this was never really successful.
As owners of XK-engined cars will testify, these engines seem to prefer richer mixtures and rather a lot of ignition advance (10° and more). This generally indicates combustion is not as good as it could be. In the end, Jaguar's "wide-angle" racing head probably reached the end of its potential because it could breathe better than it could burn.
Food for thought ...
A bit more food for thought
See the very limited water passages in the above photo as well as the large amounts of metal in the casting? Square exhaust ports? Nowadays, and especially with the advent of 3D visualisation using tools such as CAD, it is possible to design optimal ports and heads with far greater and more efficient cooling surfaces - as well as optimal air flow characteristics. Wheras the thinnest port walls in the XK head are more than 10mm thick, today's cylinder heads tend to be closer to 4.5/5.0mm with considerably increased cooling surfaces. Whilst I don't pretend to be any sort of expert in this field, it seemed to me that the port shapes, by today's standards, could also be improved.
Baily's prototype quad-cam engine
showing down-draft arrangement
© Neville Swales
One thing which did work in the 6-cylinder engine's favour may have been the side-entry and curvature of the inlet port which assisted combustion. In common with other engine designers of the period, Jaguar's Claude Baily anticipated that further improvements could be made to this basic design by making use of down-draft porting. Baily adopted this when he designed his successor to the XK engine - the quad-cam V12. In theory, there just had to be a benefit of down-draft porting but Baily (and other designers) found these benefits weren't achieved in practice. Flow may have been excellent but this arrangement simply didn't allow useful swirl/charge-movement within the cylinder and combustion suffered as a result. Others who wandered down this cul-de-sac included Ferrari, BRM, Matra and Ford - Jaguar wasn't alone in this.
Heron Head
A breakthrough came from work carried out by GM's Sam Heron in the 1950s (Heron became famous for his work with aircraft piston engines and the sodium-cooled exhaust valve). Rover were probably the first to adopt his scheme which consisted of a flat cylinder head with the combustion chamber in the piston crown - a feature later adopted by Jaguar in their first SOHC V12 engines. These heads became known as "Heron Heads". In the mid-1960s Ford (of England) adopted the Heron layout for their entire range of engines. A close relative of the Heron layout was the very successful Repco V8 engine that powered Jack Brabham's team to two F1 World Championships in 1966/67.
Jaguar's "Fireball" HE Combustion Chamber
Why was this basic layout found to be better? The increased combustion efficiency as a result of increased swirl and charge-movement may have pointed the way. It is all very well having superb flow, but this is to no avail unless the charge can be adequately and rapidly burnt.
Jaguar later improved the efficiency of their V12 further by adopting Michael May's "Fireball" combustion chamber. In this design, the exhaust valve is deeply recessed into the head, forming a compact oval chamber with the spark plug at one end. As the piston advances up the bore it forces some of the charge into a shallow channel around the flush inlet valve from where it is squeezed tangentially. This creates a high-speed vortex in the combustion chamber - LOTS of movement and "squish" here ...
Could it be possible to design a down-draft, hemispherical head with the necessary degree of charge-movement to allow combustion to match this design's superior flow characteristics?
More food for thought ...
The Project
So where was all this leading to? Can't you guess?
Having installed my unique quad-cam prototype V12 in my re-creation I thought it would be rude not to attempt to produce at least one engine of my own and thought it might be interesting to go the whole hog and perhaps include a quad-cam V12 for someone wanting the ultimate XJ13 replica.
Ground Rules
The plan was to end up with a quad-cam V12 which bears a close resemblance to Jaguar's prototype "XJ13" V12. We weren't trying to re-invent any wheels here or produce anything approaching "state of the art" but, instead, a reliable fast-road/race engine which shares the same basic architecture of Baily's prototype quad-cam and Jaguar's legendary XK 6-cylinder engine.
To this end, we are setting ourselves some basic ground rules:
- We are producing cylinder heads only. These heads will bolt directly onto Jaguar's later SOHC V12 block. This means the heads may have applications in cars other than the XJ13. - Quad-Cam Series 3 E-Type, XJ12 or XK120 anyone?
- Whilst remaining true to the basic architecture of Baily's quad-cam, opportunities to improve gas flow, combustion and overall efficiency in the light of current knowledge were taken. Whilst cosmetically similar, these are not exact copies of Jaguar's quad-cam prototype engine.
- As was the case with the original XJ13 engine, cam drive is via duplex chain.
- Two-valve, hemispherical head design.
- Fully programmable fuel injection & ignition (prototype quad-cam uses a pair of 6-cyl distributors and Lucas mechanical fuel metering unit). Alternatively, downdraught carbs can be considered.
- Normally aspirated (although we are currently looking at a blown dry-sump version - watch this space ...)
First Steps
Because we started with an almost clean sheet of paper, we had the opportunity to go back to first principles and consider things such as optimal valve sizes, port configuration, charge movement and spark-plug positioning etc.
The first step was to draw up a pair of heads combining a SOHC V12 mounting face with the basic 6-cyl DOHC design just to see if everything could be made to fit. After all, we didn't want to end up with head studs coinciding with inlet/exhaust ports! Also, we needed to make sure it was practicable and possible to mate up with existing SOHC V12 oil and water passageways. There were also practical considerations to consider such as being able to access head nuts - bearing in mind each SOHC V12 head is fastened down by four rows of head studs but only two in the prototype quad-cam and XK 6-cyl.
This is what we started from:
SOHC V12 Head and 6-cyl XK engine accurately captured in CAD
© Neville Swales
The V12 head is not only longer than the 6-cyl head, but the bore positions are different. (Incidentally, the SOHC V12 shares the same bore spacing as the prototype quad-cam - a SOHC crankshaft fits). Positions of water and oil passages are very different between the XK 6-cyl and SOHC V12. The biggest challenge was combining the two heads so that the V12 stud pattern was maintained. It became evident very early on that the new quad-cam engine will have unique cam covers as well as custom cams. Fortunately, items such as XK cam buckets & valve guides etc can be sourced "off the shelf".
The following pictures of the prototype quad-cam V12 show what we hoped to end up with - or, at least, something close:
Prototype Quad-Cam - RH Head
© Neville Swales
Prototype Quad-Cam - LH Head
Note offset inlet ports and recessed access to spark-plugs
(a pig to get at when fully assembled!)
© Neville Swales
Prototype Quad-Cam - LH Head Detail.
© Neville Swales
In contrast to the 6-cyl XK head and the SOHC V12, oil is fed to the quad-cam heads via a drilling passing from the gallery to each head. The later SOHC V12 scheme is adopted for the new engine. The next batch of tera engines will use separate cam bearings as with the 6-cyl XK head (no cam bearings in the SOHC V12).
The following pictures show the general layout of the initial designs. They were just preliminary designs with no attempt to optimise things like port configuration, spark-plug location etc. They showed it would be possible to design our own heads which would bolt straight on to the SOHC block. Discussions with a local foundry highlighted considerations we need to build into the design. Further discussions with pattern-makers confirmed it would be possible to produce the necessary patterns etc.
Initial design - no attempt at this stage to optimise port location/configuration
© Neville Swales
Experimenting with spark-plug location - improved access as well as better positioning in combustion chamber.
© Neville Swales
Note positions of outer studs for SOHC block on exhaust face. These were later opened up to allow more "swing" on a spanner.
Round exhaust ports as opposed to rectangular ones on the XK 6-cyl head.
© Neville Swales
Experimenting with design for maximum coolant flow and reducing overall weight. 4.5mm wall thicknesses.
© Neville Swales
Further investigation of inlet port location and optimal valve angles to give maximum valve size. The final layout, by coincidence, comes very close to Jaguar's "wide-angle" layout.
© Neville Swales