The Holy Grail of WWII Aviation Engines - The Rolls Royce Crecy

Of all the piston-arrow engines throughout WWII, one engine above all others remains shrouded in mystery. While books and articles surrounding other power units can be hunted down and scavenged, one engine, the holy grail of piston-arrow engines, eludes rediscovery. That engine is the Rolls-Royce Cressy. This experimental two-stroke sleeve-valved V12 monstrosity was so powerful that it would have bridged the technological gap between the piston engine and the jet had there been any delay at all in the jet program. As it turned out, the jet was adopted extremely quickly and thus the Cressy was stillborn. However, regardless of the fact that it never properly flew, it deserves to be remembered as the technological masterpiece it truly was. Employing various exotic engineering solutions which were seldom used before and never combined into an arrow engine, the Cressy was the brainchild of famous engine designer Harry Ricardo, who also brought design elements to the Cressy that he had been using since he was just a school boy. Like an artist with a unique style, the technological quirks Ricardo preferred followed him like a signature, and by taking a step backward, we can see the idea of the Cressy forming throughout Ricardo's career as an engineer. Ricardo's first engine, the Dolphin, was a redesign of a pre-existing engine on which Ricardo implemented the first design feature he would continue to use throughout his designs. At the time, designers were troubleshooting the apparent inefficiency introduced whilst throttling an engine via reducing air and fuel together, a phenomenon that had been traced back to the reduction of air flowing into the cylinders rather than the reduction in fuel. As such, engineers of the day were attempting to find ways to throttle petrol engines through fuel adjustments alone. One way to partially circumvent this issue was the stratified charged method which Ricardo incorporated into his Dolphin design. In this method, by using a bulbous combustion chamber, a rich mixture was preserved near the spark plug even when the mixture was made excessively lean beyond the point a normal chamber would no longer combust. In this way, a petrol engine can be throttled using a wider, more versatile range of fuel-air ratios. Another key design feature Ricardo became known for prior to WWII was his love for sleeve valve designs. After Ricardo finished his education at Cambridge, he went on to work for a civil engineering firm, but he still kept a massive machining garage behind his house so he could continue to work on various engine projects. When WWI broke out, Ricardo found himself designing an airship engine for the Royal Naval Air Service and, additionally, he helped design the 150 horsepower engine used in the British Mark 5 tank. At this time, Ricardo also became one of the first engineers to study the phenomenon of knock and ways to circumvent it. In discovering knock, Ricardo was also one of the first to begin developing methods for measuring knock. Prior to the widespread adoption of the American octane number system, Ricardo and his colleagues measured the likelihood of fuel detonating in terms of toluene number. This knowledge, that of the relationship between fuel quality and the ability to delay the onset of the knock phenomenon, led Ricardo down the path that would eventually lead to his love of sleeve valve designs. In his own words, In those days, we had to deal with very inferior fuels, with an octane number of less than 60, and I was much impressed by the fact that, as compared with our overhead poppet valve engine, we could employ just one hole ratio higher compression on the same fuel and gain a very big advantage. It was for this reason that Ricardo became such a fervent proponent of the sleeve valve designs, and as such, he was responsible for convincing Sir Roy Fedden of the Bristol Airplane Company to build a complete engine. While Fedden at first had his misgivings, he eventually had a change of heart and began investing huge sums of money into developing sleeves able to tolerate the incredible power their later engines would produce. For perspective, Bristol and Vickers tried between 50 and 60 compositions of steel and about 1,100 heat treatments to get a sleeve that was right for production. Cost of this research was about 2 million British pounds, and incidentally, this was about twice what the Ministry of Aircraft Production spent in World War II on Whittle and the power jets in development of the jet engine. Concerning sleeve valve designs, it was perhaps unfortunate that the only firm which was sufficiently interested to take an active part was wedded to air-cooled radial engines. That so many years should have elapsed between the completion of the research and its practical development was bitterly disappointing to me. All the more so, since the advantages of the sleeve valve aero engine were most apparent in the days when we were using relatively low octane fuels. Such a statement sheds light on the previous video we did about the Napier Sabre. It would appear that the sleeve valve's biggest advantage was its ability to use inferior fuel, and when quality fuel production took off, the sleeve valve was already outclassed before the engines that featured this technology entered widespread production. Nonetheless, the choice to use the sleeve valve in the Cressy would have other advantages that we'll touch on in a bit. It was during this point in history that the promises of the diesel engine also became well-known. As diesel engines are virtually immune to knock, it was briefly thought that a compression engine, as they were called, may hold the key to the engine of tomorrow. As petrol engines of the time, again, were limited by the low octane number of their fuel, and aviation diesel had recently begun achieving higher and higher specific power ratings, it made sense at the time to consider that a diesel design may soon overtake a petrol engine in terms of viability. The first time Ricardo incorporated these two ideas, that is, a diesel using sleeve valves, was an experimental engine dubbed the E40. In this early design, a cylinder projection of the cylinder head was fitted into the top end of the sleeve and carried piston rings for sealing. However, in this two-stroke diesel, Ricardo noted, these rings were never really trouble-free, and in spite of their own separate supply of lubricating oil, they were prone to gumming and carbon deposition, which rendered them useless. To remedy this, the E40 was redesigned into the E44, which removed the cylinder head rings altogether and exhausted over the top edge of an open sleeve. In this way, the cylinder head became a lid containing only the combustion chamber, and this idea, that is, exhausting directly over the top of the sleeve, would result in the massive exhaust thrust the Cressy would later enjoy. But before the Cressy's development would properly start, Ricardo needed to shift from diesel design back to petrol engines. That wouldn't occur until war loomed on the horizon, and the method of enemy aircraft detection was revolutionized by the invention of radar technology. By incorporating radar into the tactical picture, fighters in the air war would no longer need to constantly loiter in case enemy aircraft were detected. Instead, they could be scrambled upon enemy detection. However, an engine able to haul a fighter up to the required altitude in time would be necessary. This need gave rise to the proposal of a sprint engine. A proposal was first made by Henry Tizard at the English Subcommittee's December 1935 meeting. In that meeting, Henry Ricardo raised the question of whether a high fuel consumption could in fact be permissible under certain circumstances, and if that was the case, he believed that the two-stroke petrol appeared to be an attractive design. Two months after this meeting, Ricardo officially submitted his report entitled High-Power Two-Cycle Engine for Special-Purpose Machines. In it, it starts, The present note relates to the possibility of developing a special-purpose aero engine of very high performance for short flights only, without the usual regard for fuel or oil consumption, but in which high specific output and small frontal area are taken as being the dominating conditions. Essentially, Ricardo, in this report, planned on redesigning his experimental E-44 design into a spark ignition and running it as a petrol engine. And, as we now know, this design, when scaled up to a full-sized V-12 and manufactured by Rolls-Royce, would eventually be named the Cressy. It's also important to note that while Henry Tizard had suggested to Ricardo that he develop such a sprint engine, he had not only placed a bet on a radar-directed fighter's powerplant with Ricardo, but with Hayne Constant, who would design the first British axial-flow turbojet. After these meetings with Tizard, Constant would later recall, I remember a conversation with Mr. Tizard, the director of the Imperial College, in which he asked me what was the most worthwhile research to be done in the engine field. I told him of my hopes for the turbine and he gave an encouraging reply. There is no doubt in my mind that it was Tizard's imagination and enthusiasm that provided the drive during these critical days. Apparently, Whittle, too, had the same experience, later saying, Our best friend was Henry Tizard. It was Tizard who made the connections that led to the Air Ministry funding of power jets. In the future, I'll probably be covering these early jets, but the reason I bring them up is that it is important to put the Cressy project into context. You see, it wasn't, as some might think, some off-the-wall project made to investigate the feasibility of strange technology. Rather, it was one of several bets made by the British at a time of imminent threat to quickly develop new types of power plants that would be well-suited for what would be a reactive type of defense. So, if the Cressy was supposed to be this radically powerful engine capable of scrambling fighters in short order, how did it work? In rough terms, the Cressy engines were liquid-cooled, upright, 90-degree V-12 two-strokes. Each cylinder had a single sleeve valve and petrol injection directly into the combustion chamber. Again, ignition was spark driven with charge stratification, and the final version of the engine used an exhaust-driven turbine to recover some of the power losses incurred by the single-staged supercharger. However, as we'll see, even after the turbine, there was plenty of exhaust gas to spare. Additionally, the Cressy was directly fuel injected, used oil-cooled pistons, was power controlled via fuel metering, although some throttling was used because the stratified charge wasn't enough for full power band operation, and again, it was supercharged by necessity because two-stroke engines won't operate unless the inlet manifold pressure exceeds the exhaust pressure to drive out the exhaust products and introduce fresh air into the cylinder. As such, the blower was the only means of filling and emptying the cylinders of the Cressy. To give a quick power overview, by the end of the Cressy's development, Henry Ricciardo had his single-cylinder E65 test engine running at a power output equivalent to had it been converted to a full-scale Cressy. Now, that's a lot of power for the time considering the engine only displaced 26 liters. In terms of the engine's particulars, the Cressy featured a 5.1-inch bore, a 6.5-inch stroke, 1,536 cubic inches or 26 liters, a compression ratio of 7 to 1, a 12-inch rotor diameter on the supercharger which was geared at 8.77 and made about 18 pounds of boost using 100 octane fuel, and its dry weight with the turbine and the contra-rotating reduction gears was about 1,900 pounds. So, now that we know the specifics, how did this thing actually function? Well, we're going to cover it, but first, let's do a quick review of some of the differences between a two-stroke and a four-stroke engine. Remember, in the four-cycle engine, the exhaust is removed entirely by the piston then recharged with the requisite mixture of fuel and air through the inlet. In a two-cycle, however, the cylinder cannot be emptied because there aren't enough strokes to go around, so the residual exhaust must be displaced by the incoming air, but if the air supply is throttled, the result is that the less exhaust is displaced and, in the extreme, the cylinder will be just left full of exhaust gas. Because of this, the power of the engine cannot be efficiently controlled by air throttling but only by adjusting the quantity of fuel delivered. Essentially, power control is more similar to a diesel or a jet engine than a typical four-stroke. Also, in a four-cycle engine, air displacement is limited by the volume the piston is capable of displacing, whereas in a two-cycle engine, there's a period in which both inlet and exhaust valves are open at the same time, creating a period in which air can escape. Because of this, at any given boost pressure, the same weight of air will pass into and through the cylinder per unit of time, whether the engine speed is 300 or 3,000 revolutions per minute. In the 300 rpm case, maybe 10 cylinder volumes of air move through the per cycle, whereas in the 3,000 rpm case, perhaps only one. Regardless, it's important to remember that without positive blower pressure, a two-stroke will not scavenge exhaust gases, nor will it create any power. Again, remember that exhaust pressure is one of the biggest advantages of the two-stroke because it has double the number of power strokes as a four-stroke. So, while a Merlin was expected to provide about 150 pounds of exhaust thrust, a Cressy delivered 300. This will be important to remember when we compare the Cressy to the Griffin and the Merlin later on. And while we're on the topic of exhaust, let's talk about the ridiculousness that was the Cressy's exhaust note. You see, because the Cressy exhaust ports opened almost instantly, unlike a poppet valve design, and it opens 91 degrees after the top dead center, when the pressure in the cylinder is still high, you're hearing 36,000 pulses per minute at 3,000 revolutions per minute. That's twice as many as a four-stroke V12. This combination of rapid opening, high pressure exhaust, and the high frequency of the pulses produced an exhaust sound like no other, which incidentally is the most memorable feature of the Cressy. Apparently, when the engine first ran, the air raid sirens went off in Nottingham, 15 miles away. Now, we don't actually have any recordings of the Cressy, unfortunately, and while it wasn't a diesel, the closest engines we have recordings of are high-displacement two-strokes, such as the Detroit diesel. While the Cressy would have no doubt been louder than a Detroit diesel, the exhaust note would have been somewhere in the ballpark. So, without further ado, re-familiarized ourselves with two-stroke operations, and maybe pulled the cotton out our ears, how did the power cycle of the Cressy with its fancy sleeve valve actually work? Well, fortunately, the folks at Rolls-Royce Historical Society have provided us with this excellent graphic that splits the two-stroke power cycle of the Cressy into six steps. In the first image, the piston is at top dead center, while the sleeve has begun a downward stroke 15 degrees ahead of the crank. As combustion occurs, we move on to step two. In the second image, we can see the top of the sleeve is now cutting across the cylinder ports. At the same time, the piston skirts, for cooling, are open to blast air from the intake valve. In phase three, the inlet opens 52 degrees prior to the bottom dead center, as the top of the piston cuts across the sleeve ports. At this time, both the piston and the sleeve are traveling downwards, and scavenging has begun. At this time, the exhaust ports remain three-quarters open, and the sleeve port is still drawing in air, which is cooling the rings and the top of the piston, and moving upwards in a column to clean out the bulb of exhaust residuals. All the while, the exhaust port continues to fully scavenge the remaining exhaust. In the fourth image, the piston is at bottom dead center, and fuel injection begins. At this point, the extreme end of the spray catches air as it enters and eddies around the sleeve port bars, breaking up the fuel particles. As the exhaust closes at 59 degrees after bottom dead center, the injection step ends with an intentional dribble of fuel to ensure the stratified charge of the bulb remains rich. Now, the top end of the ascending sleeve closes the exhaust ports, and the cylinder fills with air until the incoming blower pressure is attained inside. In the final step, before combustion, the top end of the piston cuts across the sleeve port, and the sleeve and the piston are moving upwards. Compression finally commences with the richest charge retained in the bulb. And that's it. While the sleeve makes it seem slightly more complicated, the engine's total cycle is actually more simple than a comparable four-stroke. But that being said, if it was so simple and so powerful and so amazing as we were making it out to be, why didn't it succeed? That's a great question. In the end, the deck was stacked against the Cressy. Due to a lack of priority status, not enough qualified engineering talent, typical teething problems, and the rapid acceptance of the gas turbine, the Cressy program was cancelled in 1946. Interestingly, David Birch, one of the authors of my primary research source, claims that if only Messrs. Whittle and Von Ohain had been born ten years later, what a glorious swan song there might have been for the piston-aero engine. However, while we can't appreciate any actual service from the Cressy, many engineers of the day were keenly interested in performing hypothetical tests and collecting data that I think is extremely interesting. To start, we know that, in fact, the Cressy was actually intended to be mounted originally in the Hawker Henley, and one was actually delivered on March 28, 1943 to be converted to a Cressy power unit. But unfortunately, due to delays in the program, the Henley sat without an engine for the remainder of the war. Interestingly, this was not the first aircraft intended to be mounted with a Cressy, as the Spitfire II was fitted with a Cressy mock-up in the summer of 1941. However, tragically, this exercise proved futile as well, as production of Spitfire III never occurred. That being said, though, the performance of the Spitfire with a Cressy engine was compared with the Merlin 61 in a report prepared by Witold Schellier, who was Hucknall's chief performance engineer. In the report, we can see three estimates of performance, two by the Royal Aircraft Establishment, featuring both single and two-stage supercharged applications, and the Rolls-Royce figures. In this instance, the Spitfire used for comparison, like I said, was using a Merlin 61 operating at 15 pounds of boost. As the engineers discovered, the most notable trade-off of using the Cressy was a slight reduction in climb rate for a huge boost in top speed. With a two-stage supercharger, we can see that the RAE expected a 58 mph increase in top speed over the Merlin. However, in that configuration, we can see that the Spitfire would lose a considerable amount of its climb rate. In this instance, I think a more solid trade-off would be the single-stage Cressy, which offered an impressive top speed of 352 mph across the ground and 431 mph at 19,000 ft, while still making a respectable 3,600 ft per minute in climb. At the end of this report, Schellier states, It will be seen that the estimated performance is very high, both speed and ceiling. The high speed is due to the large ejector thrust power, which is much greater than in a four-stroke engine and amounts to over 25% of the engine power at high speeds and altitudes. So, in this report, we can see this data sort of backing up the later claims that the two-stroke should have been the bridge between the jet and the four-stroke piston-aero engine, as the exhaust jet thrust was a massive advantage. There was also an interesting comparison made between a Griffin-powered Spitfire and the Cressy. However, it's important to note that Farnborough based its estimates on a two-stage supercharged Cressy, which was a hypothetical configuration that was never achieved in reality, as Rolls-Royce opted to use the single-speed supercharger in a power recovery turbine instead. In any case, the resulting comparison is interesting, to say the least. What we see in this chart is that their performance is nearly identical. However, it's critical to remember that the Griffin was 37 liters, while the Cressy only displaced 26. What's also interesting is to see, again, how much exhaust thrust is contributing to top speed. In the Griffin, we're only seeing a boost of about 44 mph due to exhaust thrust, while the Cressy's exhaust is increasing the top speed by 88. Later, as Ricciardo kept making improvements to his experimental E65 engine, upon which the full-sized Cressy was based, Farnborough made an updated estimation of performance. As Ricciardo was getting 270 pounds of brake mean effective pressure while running at 3,500 rpm with a shorter stroke, and, real quick, if you don't know what BMEP is, it's the pressure in pounds per square inch which, if exerted on the piston for the full stroke, would produce the measured by the dyno with zero frictional losses. It doesn't matter if that doesn't make any sense, just know that he was making big power, and he was also forcing enough air into the cylinder to fill it 2.89 times per cycle, so he was using a big blower. In any case, the Farnborough boys extrapolated this new data out to the full 12-cylinder engine, and the results were insane, estimating a maximum output of 3,500 brake horsepower on 26 liters, which would propel a Spitfire to, they estimated, about 535 miles per hour at 32,000 feet. Again, these numbers were thought to be achievable primarily due to the jet thrust of the Cressy. However, it was later determined that the Cressy was probably too powerful for the Spitfire, and interest moved toward mating this beast instead with a Mustang. After the Allison was swapped with Merlin, a study was conducted concerning the performance of a Mustang fitted with a Cressy engine. This study actually planned on mounting the engine mid-fuselage and entertained three possible exhaust configurations. In the first, the exhaust provided ejector thrust in the normal manner, just through exhaust stacks. In the second configuration, an exhaust-driven turbine was fed back into the crankshaft to relieve the energy that drove the propeller and the supercharger in a typical turbo compound configuration. But in the third proposal was when things got interesting. In this configuration, an exhaust drive turbine, called a thrust augmenter, was used along with an outer annulus that was fed with the air from the radiator and oil cooler. The resultant mixture was exhausted into a jet pipe, and a combustion was assisted by the ignition of any unburnt fuel within the air purged through the exhaust ports during the scavenging cycle. Essentially, it was a bypass jet, or the start of the very first motor jet, which is awesome and also a topic for another video. Just know that essentially at this point we're so close to just being a pure jet engine that the reciprocating bit is maybe superfluous, but it is pretty awesome and pretty crazy when we look back at it. So in closing, while it is perhaps a disappointment that the Cressy's amazing performance was never fully realized, it isn't surprising. Like many of the ingenious, wonderful piston engine innovations of World War II, the death knell of the Cressy came alongside the first pure jet engines. However, it should be remembered that, in the words of Henry Tisard himself, the research had revealed that the two-stroke sleeve valve cylinder was capable of a maximum power output of nearly double that of a contemporary four-cycle engine, nearly 200 horsepower per liter. Thusly, it seems probable that, had the turbine not materialized, the two-stroke would have proved the next step in the development of the aero engine. As always guys, thank you so much for tuning into Flight Dojo. If you made it this far, we hope that you enjoy our content enough to hopefully subscribe. Also, you asked and we answered, the Flight Dojo channel now has its own official Patreon. Admittedly, it isn't too fancy at this point, mainly it's just a digital tip jar, but if you enjoy the research going on here at Flight Dojo and want to be involved in deciding which content we cover next, you can hop on over to the Patreon at patreon.com slash flight dojo and send us a few bucks. That being said, thanks again for watching and we will see you guys next time.