Crankshaft power stroke

[EngineerMiltonKeynes]

I've just been reading the posts on Epitrochodial crankshafts and the Atkinson cycle. Interesting that people have had tries at correcting what seems to be one of the major disadvantages of the crankshaft in an engine. Cranks seem to do a great job on the compression stroke as the effective lever length decreases as the cylinder pressure increases towards TDC. A few degrees later, though, just after TDC on the firing stroke, we're back to effectively a short lever just as the cylinder combustion pressure is at maximum. By the time the crank is at 90 degrees ATDC, the effective lever is at it's longest, but the cylinder pressure has dropped something like 75%, if I remember it right.
Has anyone ever come up with a system which converts the thrust on the piston to rotary output right from firing stroke TDC to BDC? Looking at typical graphs of cylinder pressure after TDC, it seems that there is a lot more power to be had from that explosion that we take so much trouble to create and control. I've had a go at some of the maths (which I'm not anything like expert enough at) and I can't work out exactly what the power increase would be. Are there any engine designers out there who can help me out on this?

[Greg Locock]

The power increase would be slight. One way of thinking about it is that the fuel burn creates a compressed spring. The exact details of how the spring is allowed to expand don't really affect how much energy can be extracted, it'll still be 1/2*k*x^2.

There are second order effects such as friction and the thermodynamics of the combustion that will have some effect, but they are small potatoes.

[Tommy Cookers]

@EMK
there's millions of engines (Prius, Fiat etc) that use more-complete-expansion (than compression) via a conventional crank
and Honda has its own patented crank approach (not allowed in F1 of course)

the conventional crank has as you say a short effective length eg around tdc
this is useful giving enough time for the combustion heat release to occur before expansion
short 'rod ratio' is chosen eg in high rpm engines to contribute to this
shorter 'rod ratios' mean less time around tdc but more time around the 60-70-80 deg atdc, so matching the combustion rate
(the 1.5 litre F1 BRM was stretched to 2.1 litres using a lower rr, accidently giving a better match to combustion rate and good power)

conventional cranks and rods seem less impressive at blowdown to ambient pressure
the EV is necessarily opened apparently early eg at 7 or 8 bar (compromised for mep/throttle variation ?)
maybe VVT helps with this

EDIT drens post later on this page mentions Hondas offset design, (this was a big feature in early engines aka desaxe)
this presumably has motion effects additional to those headline mechanical factors he mentions
there will be (handed) offsets in narrow V engines eg the VR5 etc and old Lancias, some must be adverse ?


sometimes it's not clear (to me) what the designer of an improved crank design had in mind
there's chassis and aero expertise on this site, but IMO we are rather thin on the engine side

[EngineerMiltonKeynes]

Many thanks for your replies. This all came about from me looking at graphs of crankshaft torque and cylinder pressure. I noticed that the crankshaft torque is a at max at 90 degrees ATDC (as you'd expect withe the longest effective lever) in a sort of inverted bell curve, but the graph of cylinder pressure , if overlaid, would give a totally different output curve, certainly higher up to 50 degrees ATDC, or so. It was the area up to 50 degrees ATDC that I was looking at. There are two fairly easy ways to build an engine which takes the torque output from the piston in a linear fashion (there are other big advantages which I won't go into here). Maybe I'll have to take a couple of little same model single cylinder engines and modify one so I can compare outputs. That's easy to do if you drive a 240v alternator from them and apply load and compare output curves.

[hollus]

The power increase would be slight. One way of thinking about it is that the fuel burn creates a compressed spring. The exact details of how the spring is allowed to expand don't really affect how much energy can be extracted, it'll still be 1/2*k*x^2.

There are second order effects such as friction and the thermodynamics of the combustion that will have some effect, but they are small potatoes.

I don't think that is a valid analogy. That particular "spring" will follow more or less the same expansion trajectory whether it has a controlled explosion in it or not. In the second case there is no work to be extracted at all except what you put in in the compression stroke. So when, what and how you expand, has to matter.

[Greg Locock]

I don't think that is a valid analogy. That particular "spring" will follow more or less the same expansion trajectory whether it has a controlled explosion in it or not. In the second case there is no work to be extracted at all except what you put in in the compression stroke. So when, what and how you expand, has to matter.

I don't think you understood what I meant. The spring is the lump of heated compressed gas, after combustion has taken place. If you start with a lump of gas at P1 V1 and T1, and end with it at P2 V2 and T2, then the work that can be extracted from the gas is very strictly defined, no matter what path is taken between the two end points. Check out Carnot on wikipedia, and heat engine theory. Or more simplistically, given that people have been trying to improve the efficiency of internal combustion engines for 110 years, why have they stuck with crank and slider for their most efficient engines?

[strad]

You adjust the rod length to stroke ratio to achieve close to 1.5:1 . Pretty sure thats the ratio,,,been a long time Then the crank pin is at about 90º when ignition takes place and less power is lost.

[Tommy Cookers]

...... The spring is the lump of heated compressed gas, after combustion has taken place. If you start with a lump of gas at P1 V1 and T1, and end with it at P2 V2 and T2, then the work that can be extracted from the gas is very strictly defined, no matter what path is taken between the two end points.

in reality the combustion heat is not added at constant volume, or constant anything else
the compression, combustion and expansion phases always overlap to a significant extent, this varies with rpm
that's why even with a conventional crank, in a given situation the rod ratio can have effects even on efficiency
because it affects the time-variation of combustion volume around tdc, to an extent it affects the thermodynamics
the historical trend to lower RRs helped making engines compact, but is important thermodynamically as rpm has risen

I imagine that car manufaturers always see 'optimisation' of behaviours around tdc as accessible via existing methods
improved DI seems able to in effect modify the combustion rate relative to the time-variation of combustion volume
in this way (as well as by injection later relative to compression) to raise effective CR
raising the CR is also to some extent equivalent to reducing the imperfection of the conventional crankshaft nearer to bdc
improved DI also can expand the envelope of near-zero NOx operation, allowing more lean-mixture running and less throttling
(where NOx is not near-zero we must run the catalyst in 3 way mode ie use a richer mixture and so more throttling)

all this without replacing the conventional crankshaft
....... but Honda does have available something new (for increased expansion ratio) ?

[dren]

Honda uses offset cranks (crank center to piston center) in their L series engines, and I think in their R series. This is so at TDC there is leverage.

Also, the RR is often optimized for lower piston to cylinder forces at high RPMs.

[EngineerMiltonKeynes]

There's plenty of good feedback being written about this and it's good to hear all the opinions.
This all started from me looking at the crankshaft torque graphs of a 4 stroke engine. Torque is directly proportional to HP at any given rpm, so I was looking at ways to increase the torque during the firing stroke.
If you imagine the crankshaft being at 90 degrees during the whole power stroke, then what the output torque would be is what I'm trying to find out about. I know it's kind of a mind bending concept, but that's what a proper linear output would do. Certainly, from TDC down to where the piston has travelled halfway down the cylinder, there must be a lot more torque (force x distance) than compared to a crank where the distance starts at pretty much zero. I would guess there would be more torque from there down to BDC, but not so pronounced as the cylinder pressure is pretty much gone.

[olefud]

There's plenty of good feedback being written about this and it's good to hear all the opinions.
This all started from me looking at the crankshaft torque graphs of a 4 stroke engine. Torque is directly proportional to HP at any given rpm, so I was looking at ways to increase the torque during the firing stroke.
If you imagine the crankshaft being at 90 degrees during the whole power stroke, then what the output torque would be is what I'm trying to find out about. I know it's kind of a mind bending concept, but that's what a proper linear output would do. Certainly, from TDC down to where the piston has travelled halfway down the cylinder, there must be a lot more torque (force x distance) than compared to a crank where the distance starts at pretty much zero. I would guess there would be more torque from there down to BDC, but not so pronounced as the cylinder pressure is pretty much gone.

There’s something to be gained from a non or modified sinusoidal piston motion. However, the gains are primarily during the intake stroke by better matching port/valve events and flow to piston speed. And, as mentioned, if piston dwell could be increased at BDC during blow down, the power stroke could be enhanced by a later blow down. But, as also mentioned, varying the piston speed and associated leverage on the crank won’t materially change the power stroke work done, i.e. force X distance. It’s a matter of gaining at one stage and losing at another.

Increasing the CR, however, causes the power stroke to start with a smaller combustion gas volume and higher pressure that remains relatively higher, but diminishingly so, during the power stroke.

http://www.yellowbullet.com/forum/showt ... p?t=464436

[aussiegman]

Many thanks for your replies. This all came about from me looking at graphs of crankshaft torque and cylinder pressure. I noticed that the crankshaft torque is a at max at 90 degrees ATDC (as you'd expect withe the longest effective lever) in a sort of inverted bell curve, but the graph of cylinder pressure , if overlaid, would give a totally different output curve, certainly higher up to 50 degrees ATDC, or so. It was the area up to 50 degrees ATDC that I was looking at. There are two fairly easy ways to build an engine which takes the torque output from the piston in a linear fashion (there are other big advantages which I won't go into here). Maybe I'll have to take a couple of little same model single cylinder engines and modify one so I can compare outputs. That's easy to do if you drive a 240v alternator from them and apply load and compare output curves.

Maybe you've considered this maybe not, however maybe you should think that piston, rod & crankshaft motion is not symmetrical throughout its movement in the 180° from TDC to BDC and then the 180° back from BDC to TDC.

For a 2.4Lt V8 F1 engine, if you assume max stroke allowable under the regs of 39.8mm (given max bore allowable is 98mm), a rod length of 110mm (for a rod/stroke ratio of 2.76:1), then the movement from TDC to 90° after TDC results in movement of the piston of 21.72mm or 54.57% of the stroke as a result of the sideways movement of the rod on the crank pin. So for the remaining 90° to BDC the piston moves only 45.43% or 18.08mm. For the piston to move 50% of the stroke or 19.8mm, the crank only rotates 84.53° and so is 5.47° short of the half stroke distance.

The proportional cosine effect of the crank angle effectively shortens the length of the rod which moves the piston the extra 4.57% over the 50% rotation angle. My engine builder calls that “dynamic shortening of the rod”. As the crank rotates the 90° back to BDC, the crankpin moves back to the centerline of the cylinder and completes the complete length of the rod. He calls that "dynamic lengthening" of the rod.

So as the piston moves away from or toward TDC or BDC, the apparent length of the rod is lengthened or shortened with regard to the motion of the piston/rod assembly by calculating the center-to-center rod length x cosine of XX° between the crankpin and the bore centerline. As such, piston movement is asymmetrical and not a simple sinusoidal motion.

This asymmetrical motion affects many engine variables.

You adjust the rod length to stroke ratio to achieve close to 1.5:1 . Pretty sure thats the ratio,,,been a long time Then the crank pin is at about 90º when ignition takes place and less power is lost.

Rod/stroke ratio's can have a large impact on the profile of the piston/rod velocity curve as well as where max velocity occurs during stroke event. A few thoughts and general observations on rod/stroke ratio's are below..

A longer rod engine with a larger rod:stroke ratio can change an engines profile markedly. With a higher rod ratio, the piston will have a longer dwell time around TDC due to a slower acceleration rate towards and away from TDC. Max piston velocity generally occurs closer to the middle of the stroke event. Longer rods also reduce the side loading on the piston and piston skirt which can reduce friction and wear. Longer rod ratio engines with a longer dwell time @ TDC can make better use of the combustion pressures converting this into torque on the crank pin. Lower piston acceleration aids VE at higher RPM's however can negatively affect VE at lower RPM's due to a reduced intake pulse from slower piston acceleration speeds away from TDC.

The reverse is true for engines with shorter rods (smaller rod:stroke ratios). Shorter rod ratio engines see the maximum piston velocity moving closer to TDC (both before and after TDC) resulting in a reduction in dwell time. A shorter ratio will create a stronger intake pulse as the piston accelerates away from TDC faster which can effect intake valve opening durations.

Rod;stroke ratios also affect vibration and balance characteristics of engines due to the alterations of the piston accelerations, speed and crankpin angles.

[olefud]

Many thanks for your replies. This all came about from me looking at graphs of crankshaft torque and cylinder pressure. I noticed that the crankshaft torque is a at max at 90 degrees ATDC (as you'd expect withe the longest effective lever) in a sort of inverted bell curve, but the graph of cylinder pressure , if overlaid, would give a totally different output curve, certainly higher up to 50 degrees ATDC, or so. It was the area up to 50 degrees ATDC that I was looking at. There are two fairly easy ways to build an engine which takes the torque output from the piston in a linear fashion (there are other big advantages which I won't go into here). Maybe I'll have to take a couple of little same model single cylinder engines and modify one so I can compare outputs. That's easy to do if you drive a 240v alternator from them and apply load and compare output curves.

Maybe you've considered this maybe not, however maybe you should think that piston, rod & crankshaft motion is not symmetrical throughout its movement in the 180° from TDC to BDC and then the 180° back from BDC to TDC.

For a 2.4Lt V8 F1 engine, if you assume max stroke allowable under the regs of 39.8mm (given max bore allowable is 98mm), a rod length of 110mm (for a rod/stroke ratio of 2.76:1), then the movement from TDC to 90° after TDC results in movement of the piston of 21.72mm or 54.57% of the stroke as a result of the sideways movement of the rod on the crank pin. So for the remaining 90° to BDC the piston moves only 45.43% or 18.08mm. For the piston to move 50% of the stroke or 19.8mm, the crank only rotates 84.53° and so is 5.47° short of the half stroke distance.

The proportional cosine effect of the crank angle effectively shortens the length of the rod which moves the piston the extra 4.57% over the 50% rotation angle. My engine builder calls that “dynamic shortening of the rod”. As the crank rotates the 90° back to BDC, the crankpin moves back to the centerline of the cylinder and completes the complete length of the rod. He calls that "dynamic lengthening" of the rod.

So as the piston moves away from or toward TDC or BDC, the apparent length of the rod is lengthened or shortened with regard to the motion of the piston/rod assembly by calculating the center-to-center rod length x cosine of XX° between the crankpin and the bore centerline. As such, piston movement is asymmetrical and not a simple sinusoidal motion.

This asymmetrical motion affects many engine variables.

You adjust the rod length to stroke ratio to achieve close to 1.5:1 . Pretty sure thats the ratio,,,been a long time Then the crank pin is at about 90º when ignition takes place and less power is lost.

Rod/stroke ratio's can have a large impact on the profile of the piston/rod velocity curve as well as where max velocity occurs during stroke event. A few thoughts and general observations on rod/stroke ratio's are below..

A longer rod engine with a larger rod:stroke ratio can change an engines profile markedly. With a higher rod ratio, the piston will have a longer dwell time around TDC due to a slower acceleration rate towards and away from TDC. Max piston velocity generally occurs closer to the middle of the stroke event. Longer rods also reduce the side loading on the piston and piston skirt which can reduce friction and wear. Longer rod ratio engines with a longer dwell time @ TDC can make better use of the combustion pressures converting this into torque on the crank pin. Lower piston acceleration aids VE at higher RPM's however can negatively affect VE at lower RPM's due to a reduced intake pulse from slower piston acceleration speeds away from TDC.

The reverse is true for engines with shorter rods (smaller rod:stroke ratios). Shorter rod ratio engines see the maximum piston velocity moving closer to TDC (both before and after TDC) resulting in a reduction in dwell time. A shorter ratio will create a stronger intake pulse as the piston accelerates away from TDC faster which can effect intake valve opening durations.

Rod;stroke ratios also affect vibration and balance characteristics of engines due to the alterations of the piston accelerations, speed and crankpin angles.

Very well put. The consensus seems to be that performance can be shifted primarily in tilting the maximum performance towards broader torque or maximum power somewhat, but efficiency isn’t affected.

[riff_raff]

The real goal is to extract as much net work from the heat energy produced by combustion of the fuel as possible. Designing the engine to allow a greater expansion ratio of the working gas within the cylinder is one possible approach, but it can be difficult to achieve in practice. Another more practical approach would be to use an exhaust turbine to extract additional work from the combustion gas outside of the cylinder. Something like a turbocharger.

[olefud]

The real goal is to extract as much net work from the heat energy produced by combustion of the fuel as possible. Designing the engine to allow a greater expansion ratio of the working gas within the cylinder is one possible approach, but it can be difficult to achieve in practice. Another more practical approach would be to use an exhaust turbine to extract additional work from the combustion gas outside of the cylinder. Something like a turbocharger.

There are (at least) two ways to optimize work from the power stroke. Low RPM allows for fairly complete fuel charge combustion to take place near TDC thus avoiding the increase in pressure during the compression stroke from early ignition timing and loss of pressure during the early power stroke due to combustion kinetics. Low RPM allows both late ignition timing and near complete combustion at TDC by slowing the compression/expansion events. Lower RPM also allows for EVO later in the power stroke thus avoiding blow down losses. Again this is due to the longer time period per unit piston travel afforded for the event at low RPM.

Also, as previously mentioned, a higher compression ratio confines the combustion products to a smaller volume thus providing higher pressures on the piston throughout the power stroke.