The rod journal on a crankshaft goes around in a circle. This seems like a simple fact, and it is. This means that a reciprocating piston that rides on a crankshaft hangs essentially motionless near top dead center and near bottom dead center for a rather long period of time. This has many ramifications for the functional operation of reciprocating piston internal combustion engines. One often overlooked consequence has to do with what intake valve closing times work best. The simple explanation is that it works very well for the intake valve to stay open somewhat after bottom dead center. Even on very slow turning engines hardly any cylinder filling is lost so long as the intake valve gets fully closed before the piston substantially begins to move upward. This also is fairly simple and easy to understand: The intake valve can stay open a bit past bottom dead center for better flow up to elevated engine speeds without compromising cylinder filling at lower engine speeds. The conundrum comes from the geometry of a circle and how it relates to stroke length in gasoline engines.
The Ideal Intake Valve Closing Time
Stroke Length in Gasoline Engines
The Non-Functional Compromise
High Speed Diesel Engines
At 15 degrees ABDC the piston has moved up only a very small amount. Really an insignificantly small amount, the piston has hardly started to move at 15 degrees ABDC. At 30 degrees ABDC the piston has moved up much more substantially, almost five times as far as at 15 degrees ABDC. Five times as much piston movement for only twice the length of time. That is a pretty dramatic difference.
Just looking at the geometry of a circle it is easy to see that an intake valve closing time of around 15 or 20 degrees ABDC works very well over a rather wide range of engine speeds. The extra intake valve opening duration past bottom dead center allows the intake charge to continue flowing in at elevated engine speeds.
At 45 degrees ABDC the piston has moved up about two and a half times as much as at 30 degrees ABDC. That's two and a half times as much piston movement for only another 15 degrees of crankshaft rotation. This is not quite as dramatic a difference as from 15 degrees ABDC out to 30 degrees ABDC, but it is still substantial.
The reason that extra intake duration is important at elevated engine speeds is that it takes a while for the intake charge to begin to flow. As the piston begins to move down there is a delay as the intake charge accelerates. For a period of time the actual flow into the cylinder remains rather low, then as the intake charge accelerates the rate of flow increases. At elevated engine speeds this delay means that by the time the intake charge has reached it's maximum rate of flow the piston is already slowing towards bottom dead center. The extra intake duration past bottom dead center allows last part of the accelerated intake charge to actually get into the cylinder at elevated engine speeds.
An intake valve closing time of around 15 or 20 degrees ABDC works well at very low engine speeds where cylinder filling is determined only by the intake valve closing time, and the 15 or 20 degree ABDC intake valve closing time also provides a substantial amount of additional duration for the intake charge to continue flowing in past bottom dead center at elevated engine speeds. Going up to a 30 degree ABDC intake valve closing time provides only an additional 10 degrees of intake duration, but much more substantially reduces peak cylinder filling at low engine speeds.
An intake valve closing time of 15 or 20 degrees ABDC would be a camshaft with 180 to 190 degrees of duration at 0.05" valve lift and a 106 degree lobe separation. Where the conundrum comes from is that this is an excessively small camshaft for a gasoline engine. Gasoline engines have to spin up to more than about 3,000 or 3,500RPM to be able to run in late compression ignition mode. For a reasonably wide range of good running engine speeds in late compression ignition mode the minimum size camshaft that works well for a gasoline engine is about 210 or 230 degrees at 0.05" valve lift with a 106 degree lobe separation which is an intake valve closing time of 30 or 40 degrees ABDC. The ideal camshaft for a reciprocating piston internal combustion engine is smaller than the minimum size camshaft for a good running gasoline engine. That is the conundrum.
Reciprocating piston engines with fixed valve timing work best over the range of engine speeds that can be provided for with intake valve closing times around 15 to 20 degrees ABDC. That is about 1,000RPM up to about 4,000RPM. A substantially wide range of engine speeds. Unfortunately this range of engine speeds is far too low for a gasoline engine running in late compression ignition mode. The upper end of this range of engine speeds from 3,000 to 4,000RPM can work reasonably well for late compression ignition mode, but then most of the range of engine speeds where the camshaft is working well from 1,000 to 3,000RPM is totally wasted. And that is just the latest possible time of late compression ignition at about 15 or 20 degrees ATDC that can work reasonably well down to 3,500RPM. To make matters worse gasoline engines actually work extremely well at the earlier and easier to hit 5 degree ATDC time of late compression ignition up at even higher engine speeds. It is usually considered that this earlier 5 degree ATDC time of late compression ignition works best up above 6,000RPM, mostly because the time of late compression ignition also so easily falls over to 5 degrees BTDC which really needs to be up well above 6,000RPM to work at all well. The earlier and easier to hit 5 degree ATDC time of late compression ignition can sort of work down to 5,500 or even 4,500RPM as long as it is not falling over to a top dead center or earlier time of late compression ignition.
Where the stroke length comes in is that longer stroke gasoline engines can be made to work down to slightly lower engine speeds. A long four inch stroke gasoline engine seems to run fairly well at 3,000RPM and can sort of work down to 2,700RPM. An even longer five inch stroke gasoline engine can sort of work down to even slightly lower engine speeds around 2,500RPM. The much higher mean piston speeds allow late compression ignition to work down to slightly lower engine speeds, but that is only part of what is going on. Longer stroke length engines appear to be able to sort of work down to lower engine speeds in late compression ignition mode simply because the mean piston speed gets so high that efficiency drops off dramatically at higher engine speeds. A five inch stroke length gasoline engine appears to work fairly well down at 2,500 to 3,000RPM in late compression ignition mode simply because it can't attain good efficiency at higher engine speeds. The earlier and easier to hit 5 degree ATDC time of late compression ignition just does not work down to those low of engine speeds and the mean piston speed on a five inch stroke engine is quite a bit too high for the latest possible time of late compression ignition at 3,500 to 4,500RPM. With hot burning gasoline a five inch stroke gasoline engine can make some substantial power up at 5,000 and 5,500RPM, but efficiency remains very poor. Especially reduced load efficiency is just abysmal on a five inch stroke gasoline engine. Compared to the horrendous medium load efficiency up at higher engine speeds a five inch stroke gasoline engine appears to work fairly well down at 2,500 to 3,000RPM. The longer stroke length not only allows slightly lower engine speeds to work better, but even more significantly the longer stroke length simply hides the fact that late compression ignition does not work well down at those very low engine speeds bellow about 3,500RPM.
There are of course also other factors that have resulted in automotive gasoline engines having excessively long stroke length, but the fixed valve timing is quite significant. Longer stroke length engines have also been popular simply because higher engine speeds make so much screaming noise. Lower engine speeds are easier to muffle, where screaming 6,000 and 8,000RPM gasoline engines tend to be more difficult to muffle down to acceptable levels of noise. Then there is also the simple fact that there is less time for the intake charge to flow in and for the exhaust to flow out at higher engine speeds. Higher speed engines are inherently more difficult to get to flow well. Lower speed engines also require less reduction to drive land vehicles, making transmissions simpler and somewhat easier to build. All of these things and others have added up to automotive engines using first extremely excessive four to five inch stroke lengths and later moderately excessive three and a half to four inch stroke lengths.
The three and a half inch stroke length automotive engines are a compromise on stroke length, but it turns out to be a rather non-functional compromise. Reciprocating piston internal combustion engines with fixed valve timing would best run down at 1,000 to 4,000RPM, and that is far too low of an engine speed for any gasoline engine. The four and a half and five inch stroke length automotive and light truck gasoline engines of the 1920's and 1930's mostly just masked the real problems by working so poorly in late compression ignition mode that they were forced to run almost entirely in full flame front travel mode.
The shorter three and a half inch stroke length automotive engines of the 1950's onwards were a dramatic improvement in that much more power could be produced in late compression ignition mode over a range of higher engine speeds. The reason that the three and a half inch stroke length automotive engine is a failed compromise though is that the stroke length is really still way too long for late compression ignition and the stroke length then is also too short for even full flame front travel mode operation down at 1,000 to 2,000RPM. The three and a half inch stroke length automotive engine can run in full flame front travel mode at 1,500 to 2,000RPM, but the mean piston speed is so low that efficiency is compromised under anything other than a really very light load. Light loads and low mean piston speeds go along well together, but efficiency is pretty poor. The three and a half inch stroke length automotive engine gives up the lower 1,000 to 2,000RPM portion of the range of engine speeds that would tend to work best with fixed valve timing because the mean piston speed is too low, and the three and a half inch stroke length automotive engine still won't run efficiently over a wide range of engine speeds and engine loads in late compression ignition mode because the mean piston speeds get very high. A three and a half inch stroke length engine running on hot burning gasoline can make some substantial power up to about 8,000RPM but the mean piston speeds are so high that efficiency even under a full load is rather poor at all engine speeds above about 4,000RPM. This leaves just a very narrow range of engine speeds from about 3,200 to 4,000RPM where the three and a half inch stroke automotive engine can run fairly efficiently under a full load in late compression ignition mode. Again part of the problem is that the mean piston speed gets too high before a high enough engine speed for the earlier and easier to hit 5 degree ATDC time of late compression ignition can be obtained. It is the same problem as with the five inch stroke length gasoline engines, just not to nearly as large an extent. Under somewhat reduced loads the problem is even worse, and three and a half inch stroke length automotive engines remain abysmally inefficient under most operating conditions.
The obvious solution is an even shorter stroke length to deliver a reasonably wide range of engine speeds and engine loads in late compression ignition mode. Going down to a two inch stroke length yields a much broader 3,500 to 6,000RPM range of engine speeds where the latest possible time of late compression ignition will work well. The two inch stroke length is also short enough that the engine can easily spin up to a range of engine speeds where the earlier and easier to hit 5 degree ATDC time of late compression ignition works well. At least under a full load the two inch stroke length engine can run fairly efficiently up to around 8,000RPM. What is really most significant is that the two inch stroke length allows for a range of reduced loads in late compression ignition mode at moderate mean piston speeds around 3,500 to 5,000RPM.
It is the higher engine speeds that allow late compression ignition to work well. With the engine always operating up at higher engine speeds later intake valve closing times are required. This is where some difficulty arises for an engine with fixed valve timing. A camshaft that delivers later intake valve closing times for higher engine speeds always tends to narrow the range of engine speeds where high cylinder filling can be obtained when compared to the 15 or 20 degree ABDC intake valve closing time that works well from 1,000 to 4,000RPM. What it comes down to is that gasoline engines just don't work as well with fixed valve timing as slower turning diesel engines.
The simplest partial solution is a competent valve train that can flow as well as possible over a wide range of elevated engine speeds with a somewhat later intake valve closing time. Canted valve two valve per cylinder engines work much better as gasoline engines than parallel valve two valve per cylinder engines. Four valve per cylinder engines work much better as gasoline engines than two valve per cylinder engines. Canted valves is a step in the right direction, and four valves per cylinder makes a substantial difference in getting a wide range of elevated engine speeds for a gasoline engine. How those valves are opened also makes a big difference.
The other big thing that can be done to get fixed valve timing to deliver high cylinder filling over a wider range of elevated engine speeds is to open and close the valves more rapidly. A heavy pushrod valve train tends to result in a camshaft with mild slow opening and closing ramps being used. A lighter valve train works much better with a more aggressive camshaft profile that opens and closes the valves more rapidly. An overhead camshaft eliminates not only the push rods, but also either the lifters or the rocker arms. The difference in valve train weight is quite dramatic between an overhead camshaft engine and a pushrod engine.
Once a light enough valve train can be provided for then an aggressive camshaft profile can be used to rapidly open and close the valves. How fast the valves can open and close depends on the shape of whatever rides on the camshaft. A flat lifter limits the maximum rate of valve opening and closing. Larger diameter lifters on a larger diameter camshaft can somewhat increase the maximum rate of valve opening and closing, but there is a severe limit to how big of a lifter will fit on a DOHC four valve per cylinder engine. Roller lifters not only reduce friction for more efficient operation of a heavily loaded camshaft, roller lifters also allow faster maximum rates of valve opening and closing simply because of the round shape compared to flat lifters. An overhead roller camshaft can rather easily allow for both a light weight valve train and very fast rates of opening and closing of the valves.
There are some compromises with faster opening and closing valves at high engine speeds. Opening and closing the valves more rapidly does increase the reciprocating losses of the engine slightly, but this isn't any kind of real problem. The valves typically open and close less than one fifth of the stroke length of the engine and the valve train also typically weighs considerably less than the pistons and rods. One half the weight and one fifth the travel at half speed means one twentieth the reciprocating losses of the pistons and rods. Even if this 1/20th reciprocating losses value is substantially increased with much faster opening and closing valves it still remains very low as a proportion of the total reciprocating losses of the engine.
The most extreme case for high reciprocating losses in the valve train would be a radically over square two valve per cylinder canted valve engine. With a huge 1.5:1 bore to stroke ratio and giant valves canted away from each other at an 80 degree included angle the valves get pretty heavy and they also lift quite a bit compared to the stroke length of the engine. As an example 2.2 inch diameter intake valves for a big block Chevy used on a 250F dirt bike with a 270cc big bore kit gives an idea of what this would look like. That would be a 1.55:1 bore to stroke ratio, and the valves might lift as much as 1/3 of the stroke length. Those two giant valves might weigh as much as 370g together. This is obviously rather high compared to the approximately 425g weight of the piston, rings, wrist pin, connecting rod and bearing for a big bore 250F. The total valve train weight with springs, retainers and the effective weight of the rocker arms could easily come up to slightly more than the weight of the piston and rod assembly. With the valves lifting 1/3 of the stroke length and the valve train weight way up at 150% the piston and rod weight the reciprocating losses of the valve train would however still tend to be not much more than 1/4 of the reciprocating losses of the piston and rod. Even with a very aggressive roller overhead camshaft this extreme case radically over square canted valve two valve per cylinder engine still ends up with essentially insignificantly low reciprocating losses from the valve train. On this extreme example it can be seen that a four valve per cylinder engine can certainly have some small efficiency advantage over a two valve per cylinder engine just on the reciprocating losses of the valve train, but it takes a very extreme example to get to where the reciprocating losses of the valve train are worth considering at all.
A more relevant concern with rapidly opening and closing valves is valve train wear. Just the right gradual ramp at the very bottom of the valve lift can very easily keep the wear of the valve faces and valve seats down to a reasonable level as long as the valves don't float and crash closed. Faster opening and closing valves does however put larger loads on the valve train, especially when stiffer valve springs are required to prevent valve float on an aggressive camshaft. Stiffer springs and a more aggressive lobe profile will more heavily load the valve train, and this can result in elevated wear or even failure on a poorly designed valve train. Roller followers and roller camshaft bearings go a long way to reducing rates of wear and increasing reliability and longevity. When ratio rockers are used there is still the problem of the rocker end and valve tip wearing, and more aggressive camshaft profiles certainly do increase this rate of wear. Roller tip rockers are one solution, but this then introduces difficulty in providing a means of adjusting valve lash. A fully roller valve train that does not wear rapidly can however go for a long time between valve lash adjustments. Grinding down valve stems or changing out different diameter rollers may sound like a difficult way to do a valve lash adjustment, but if it only needs to be done very infrequently then it would be a realistic service procedure.
Perhaps the ultimate solution is variable valve timing. With a variable valve timing system reasonably high cylinder filling can be obtained over a range of elevated engine speeds without the need for extremely fast opening and closing valves. Obviously a large number of automotive engineers have had this very idea what with the enormous prevalence of variable valve timing systems on automotive engines of the past ten years. The disadvantage of variable valve timing systems is simply the complexity. Even the most basic electro-hydraulic camshaft phasor still requires a control system of some kind. This has typically been provided by an electronics or computerized engine management system. A pair of centrifical switches could also be used to control an electric solenoid valve on a hydraulic camshaft timing mechanism. There are a large number of possibilities, but they all add substantially to the complexity of a basic gasoline engine. Most mechanics would probably vote for a fully rollerized single overhead camshaft over electro-hydraulic camshaft phasors for ease of service and reliability.
Forced induction is also a way that flow limitations can be overcome at elevated engine speeds. A mechanically controlled turbocharger tends to not work well on a gasoline engine because the boost curve is too steep. A variable vane geometry compressor can help a lot with leveling out the boost curve, but turbochargers still tend to have a steep boost curve at the lowest engine speed where boost is provided. Gasoline engines also generally don't work well with large amounts of boost pressure. Small amounts of boost can be beneficial for assuring good flow up to high engine speeds, but large amounts of boost require dramatically reduced compression ratios that severely reduce efficiency. A computer controlled variable vane geometry turbocharger can provide whatever smooth boost curve is required, but again there is the complexity issue. The variable vane geometry turbocharger itself is a substantially complex sub-system and the computerized engine management system required to operate it is a whole higher level of extreme complexity.
A roots blower can also do a good job of providing small amounts of boost with a nice smooth mostly linear boost curve. Really a roots blower is a much better fit for a gasoline engine than a turbocharger. Not only is a roots blower much simpler than a turbocharger, but the roots blower also inherently operates fairly well over a wide range of engine speeds. Roots blowers might not be quite as efficient at providing boost as a turbocharger, but when only small amounts of boost are required the drive power for the blower remains low.
Whatever solutions are used to improve performance two things remain universally true. Gasoline engines need rather short stroke lengths to operate up at a range of elevated engine speeds where late compression ignition can be used, and those elevated engine speeds tend to require some additional valve train features to get a wide range of engine speeds where good cylinder filling can be obtained.
The simplest gasoline engine for delivering big power over a wide range of engine speeds seems to have a stroke length of around two and a half to three inches. If efficiency is important then the stroke length needs to be even shorter. Particularly for good reduced load efficiency the stroke length of a gasoline engine needs to be down bellow two inches. A three inch stroke length racing engine might operate up to just as high of engine speeds as a two inch stroke engine for practical purposes, but the shorter two inch stroke length will deliver both somewhat higher peak efficiency and spectacularly higher reduced load efficiency.
Also for diesel engines the ideal intake valve closing time tends to be around 15 or 20 degrees ABDC. Larger slower turning diesel engines certainly can work rather well also, but there are some reasons why a 15 or 20 degree ABDC intake valve closing time is somewhat better for diesel engines. The combustion event itself favors engine speeds in the 750 to 2,000RPM range of engine speeds. Longer stroke length slower turning diesel engines have been used, but peak efficiency appears to be easiest to attain up at somewhat higher 1,100 to 2,000RPM engine speeds.
The main reason that a late 15 or 20 degree ABDC intake valve closing time works a bit better on diesel engines is simply that increasing cylinder filling up to the higher engine speeds tends to deliver the highest overall operating efficiency. Slightly less air in the cylinder for light load operation at low engine speeds reduces pumping losses and also tends to increase combustion efficiency. Up at maximum engine speed efficiency tends to be highest when output is high, more air allows more fuel to be burned efficiently.
For a diesel engine operating over any range of engine speeds it would be true that a slight increase in cylinder filling up to maximum engine speed would yield the highest overall operating efficiency. The range of engine speeds where this gradually increasing cylinder filling can best be provided by fixed valve timing is in the 1,000 to 4,000RPM range with about a 15 or 20 degree ABDC intake valve closing time.
Increasing cylinder filling as engine speed is increased can also be provided by later intake valve closing times up to higher engine speeds, but the widest range of gradually slightly increasing cylinder filling is obtained with about 15 or 20 degree ABDC intake valve closing times.
For a diesel engine to run well up to 4,000RPM the stroke length needs to be less than about four inches. High output turbo charged diesel engines can run up to somewhat higher mean piston speeds, and normally aspirated diesel engines tend to need to stay at somewhat lower mean piston speeds to attain high efficiency and run clean. Depending on the application a diesel engine that runs up to 4,000RPM might have a stroke length anywhere from two to four inches. Since 4,000RPM on a diesel engine is best reserved for intermittent high output there is little reason to use stroke lengths less than about three inches for most normal diesel engines. Really the only reason for a diesel engine to use a stroke length less than about three or three and a half inches is when highest possible output is desired from a small fast spinning engine. Because diesel engines tend to run dirtier with higher unburned hydrocarbon emissions at engine speeds above about 2,500RPM there really is hardly any reason to use stroke lengths shorter than about three or three and a half inches on diesel engines. The only application for a shorter stroke length diesel engine would be where high power output up to high engine speeds is required from diesel oil running in a simple diesel engine. For most purposes high power output to high engine speeds is better provided by a gasoline engine, and this can even be done on diesel oil with a dual fuel system and good engine controls.
Late compression ignition in a gasoline engine is far cleaner than a diesel engine, but it is somewhat difficult to run diesel oil in a gasoline engine. Diesel oil will run in late compression ignition mode in a gasoline engine, but when the engine falls off of late compression ignition and runs in full flame front travel mode the spark plug is likely to foul badly and cause the engine to stall in short order. Running diesel oil in a gasoline engine requires a dual fuel system so that the engine can start on gasoline and then switch to diesel oil for high power output operation. A gasoline engine running on diesel oil can often fall into full flame front travel mode for a short period of time, but the spark plugs do begin to gunk up essentially immediately. It almost goes without saying that a gasoline engine is extremely unlikely to start from cold on diesel oil, so again a dual fuel system is the only way to run diesel oil in a gasoline engine.
A carbureted gasoline engine with a dual fuel system is of little use because the response time is so slow. The engine has to be switched back to gasoline before power is reduced so that the oil in the carburetor bowel is used up and replaced by gasoline before the engine is brought down to low idle. For a well managed marine, generator or industrial engine a dual fuel system on a carbureted gasoline engine certainly could be used. Also for sustained highway cruising a dual fuel system might be used, but an unexpected stop could require spark plug cleaning and carburetor bowel draining before restarting the engine. A power bowel flush valve could be used to speed fuel transition on a carbureted dual fuel engine, but this would result in some mixing of gasoline with the diesel oil tank each time an unexpected stop was made. Dual fuel carbureted engines were apparently used in small numbers around the middle of the 20th century, but obviously they did not catch on.
Electronic fuel injection systems could do much better on a dual fuel system. The simplest system would still use a manual switch to select the fuel being used, but the response time could be made instantaneous with no mixing of the two fuels. It would just be two sets of injectors and two parallel EFI systems. A dedicated set of injectors would also have the advantage of being able to provide better atomization of the fuel oil. A fully automated system would also be possible, but this would require a rather robust computer system to keep track of engine operating parameters for a determination to be made about when to switch back to gasoline. Even though an automated dual fuel system could be made to work rather well it would tend to stay stuck on gasoline in variable conditions where lots of light load operation in full flame front travel mode was used.
Dual fuel gasoline engines are a promising technology that could be very good for overall efficiency and cleanliness of land vehicles. In the absence of dual fuel gasoline engines diesel oil is burned only in diesel engines, and for the most part there is no reason for diesel engines to have stroke lengths less than about three and a half inches.
What really stands out here is that the three and a half inch stroke length automotive engines have many design features that look like they are supposed to be on diesel engines. The three and a half to four inch stroke length looks a lot more like a diesel engine than a gasoline engine. The 190 degree at 0.05" valve lift traditional automotive camshafts also seem more like something that would be used in a high speed diesel engine that operates up to about 4,000RPM. It is however only the stroke length and camshaft duration at 0.05" valve lift that point to a diesel engine. The automotive camshafts themselves typically have very long total durations that would not be well suited for use in a diesel engine. Opening and closing the valves very slowly down at the last 0.05" of valve lift allows the excessively small 190 degree at 1mm valve lift camshafts to work somewhat better in a gasoline engine as cylinder filling drops off more dramatically bellow about 2,5000RPM. Essentially the 190 degree at 0.05" valve lift traditional automotive camshafts had a lobe profile designed to help prevent late compression ignition bellow about 2,500RPM while maintaining a short 190 degree at 0.05" valve lift specification that would seem to be compatible with the three and a half and four inch stroke lengths. In a diesel engine a three and a half inch stroke length works rather well all the way down to about 1,500RPM. A three and a half inch stroke length diesel engine can attain considerably higher peak efficiency up at around 2,000 to 2,500RPM, but high torque generation is still attainable down to 1,500 and even 1,000RPM if the injection flow rate can be reduced for those lower engine speeds. That is much different than a three and a half inch stroke length gasoline engine that looses torque dramatically as engine speeds are reduced bellow 3,000RPM regardless of what camshaft it has in it.