Fundamental explanation for different torque curve SI-CI
Fundamental explanation for different torque curve SI-CI
(OP)
Diesel engines reach their maximum torque at lower engine speeds than SI engines.
What is the fundamental explanation for this difference?
Is it just because a diesel is built heavier than a SI engine? Or is it caused by the different combustion?
What happened to the "SI vs CI" thread that is mentioned in other questions?
What is the fundamental explanation for this difference?
Is it just because a diesel is built heavier than a SI engine? Or is it caused by the different combustion?
What happened to the "SI vs CI" thread that is mentioned in other questions?





RE: Fundamental explanation for different torque curve SI-CI
The engine operates unthrottled, no suction throttling losses;
Diesel fuel has a much higher btu rating than gasoline (roughly 145,000 versus 120,000 per gallon);
Higher compression ratio (ranging from about 16:1 to near 24:1);
The pressure curve of diesel has a longer duration than on an SI engine.
The weight has nothing to do with power, but to make the engine durable against the CI pressures.
Franz
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of Eng-Tips Fora.
RE: Fundamental explanation for different torque curve SI-CI
I would turn the question around to answer it. Why are SI engines able to maintain torque at higher speeds?
RE: Fundamental explanation for different torque curve SI-CI
RE: Fundamental explanation for different torque curve SI-CI
With a spark ignition engine, you have a ready-to-fire mixture of air and fuel during the compression stroke. The mixing is (substantially) already done. When ignition occurs, the only thing limiting combustion is the flame speed.
With a diesel engine, since it is compressing air only, fuel injection and ignition and combustion all have to happen in a few degrees before and after top-dead-centre. After the fuel injection starts, there is a delay before ignition occurs. The delay is in the millisecond range, but it is nevertheless a delay, and a millisecond is 24 crank degrees at 4000 rpm. Then there is the time for adequate air and fuel mixing to occur, while combustion is taking place.
What's found with high speed (relatively) automotive diesels is that even though the engine may be mechanically capable of going 6000 rpm, any attempt to use such high revs results in combustion taking too long and not going to completion, so the result is too much smoke and high exhaust temperature but not much power.
There is an upper boundary to the amount of fuel that can be injected before the exhaust opacity ("blackness") goes beyond an acceptable limit, and that amount of fuel tends to drop off at higher and higher revs even if the boost pressure is kept the same.
RE: Fundamental explanation for different torque curve SI-CI
I understand that the ignition delay of CI engines limits the amount of fuel that can be injected for a fully completed combustion at high engine speeds. This causes the torque curve of CI engine to drop faster than that of a SI engine at high engine speeds.
But what happens at low engine speeds? I thought that CI engines could produce a higher torque than SI engines at low speeds, in other words, that the torque curve of SI engines dropped faster there.
Is it the lower flame propagation speed that reduces the produced torque? From the explanation above, it seems that a CI engine doesn't have any trouble with the lower speed as there is more time available for a fully completed combustion and more fuel can be injected, causing a higher torque. Is this correct?
RE: Fundamental explanation for different torque curve SI-CI
The SI engine has to advance the timing to complete the burn. Some times 40* or more and that is what steals the torque as the piston has to overcome the approaching flame front. I have had 2 strokes in bad tune actually start and run backward because of the approaching pressures of advanced ignition timing.
The other point is the burn is more complete in DI especially at low RPM's and AFAIK franzh is correct about BTU values.
I don't know anything but the people that do.
RE: Fundamental explanation for different torque curve SI-CI
RE: Fundamental explanation for different torque curve SI-CI
The BMEP is a function of compression ratio. Diesels have a higher compression ratio.
RE: Fundamental explanation for different torque curve SI-CI
That'a an oversimplification. The main indicator of BMEP is energy release per unit volume; a secondary indicator is thermal efficiency, which is of cource influenced by compression ratio. Due to frictional and thermal losses during the compression stroke, there is little net efficiency gain above 13-14 CR, IIRC. Diesels typically run ratios of 16 or higher in order to support compression ignition under adverse conditions (namely cold starting & warmup).
RE: Fundamental explanation for different torque curve SI-CI
If we had identical SI and CI engines (bore, stroke, Vs) running at the same CR - say 14:1, would the diesel still produce more torque at WOT at low rpm? Yes.
Diesel engines are built to take advantage of their slower running speeds and shorter rpm ranges. They have longer strokes and ports that promote high turbulance and swirl. These things promote low down torque production. There are components from the slighlty higher CR and also the energy in the fuel helping to produce better torque, but also the fact that the fuel can be kept burning for longer in the low rpm range - past the point that the piston is exerting the most force on the crankshaft. Like FranzH said, the pressure curve is of a longer duration. This actually helps at low rpm to produce torque.
Remember the Therm Eff for the diesel cycle includes beta- the cutoff ratio for injection, not just CR.
RE: Fundamental explanation for different torque curve SI-CI
al1
RE: Fundamental explanation for different torque curve SI-CI
As the piston descends in the combustion stroke, the angle of the connecting rod and big end relative to the crankshaft increases. Resolving forces shows that the push on the piston will generate more torque up to the point when the piston is half way down the bore and the big end is at 90 degrees crankshaft rotation. After that the 'leverage' is decreasing again. If the pressure in the cylinder can be kept as high as possible, later into the power stroke, this extra 'leverage' generates an increase in output torque. Diesels have a longer burn duration, allowing the pressure to be kept higher in the cylinder during combustion for longer, therby utilising more of this 'leverage' effect. The same longer burn duration is what also hampers the diesel engine in the higher rpm ranges that a conventional SI engine reaches.
There are of course other components that give diesels the low down torque, something that GDI engines have yet to match completely.
RE: Fundamental explanation for different torque curve SI-CI
- Steve
RE: Fundamental explanation for different torque curve SI-CI
I agree, laughter it makes.
Next they'll be telling us the engine goes round at constant speed.
Cheers
Greg Locock
Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of Eng-Tips.
RE: Fundamental explanation for different torque curve SI-CI
What is wrong with using the term 'leverage' to illustrate the condition at any one time window pray tell? Is the answer to the question fundamentally flawed?
And of course the dyno will cycle average the torque, you expect that I though that it gave a series of discrete measurements for one cycle?
And also, from the above posts WHY would you think that the extension of this would be that I thought the engine speed was constant through the cycle?
Intelligent input guys.
RE: Fundamental explanation for different torque curve SI-CI
Because the magnitude of instantaneous torque peaks is largely irrelevant (unless you are concerned with stress calcs). It's the total PdV integral that's important.
- Steve
RE: Fundamental explanation for different torque curve SI-CI
RE: Fundamental explanation for different torque curve SI-CI
RE: Fundamental explanation for different torque curve SI-CI
For the record, I disagree with the statement about the optimal compression ratio for a diesel. There is efficiency to be had at higher compression ratios, if you can handle higher cylinder pressure and still meet emissions. Nearly all of the 2007 on-hwy diesels seem to have CRs in the 16-18 range:
http://www
I don't think it was cold start that put them there; most of those are available with a preheater.
RE: Fundamental explanation for different torque curve SI-CI
I do not have cam timing specifications in front of me so please correct if I'm wrong here.
RE: Fundamental explanation for different torque curve SI-CI
SI - low engine speed throttle is typically closed so very low air flow, and further the low engine speed reduces air draw by the engine. High engine speed the throttle is typically open and the engine has a fast stream of air coming in (ram air as well if in a moving vehicle).
CI - torque is whatever you want as long as you have air. At high engine speeds, you are limited by the rate of combustion due to the slower flame mechanism as mentioned previously. However, the real torque peak is typically programmed a little lower than it could be to encourage drivers to drive at particular speeds.
RE: Fundamental explanation for different torque curve SI-CI
This too is an oversimplification. Other posts above have alluded to this already.
At the very least the fuel rate also needs to be constrained for this comparison. If you do that (i.e. make the chemical LHV delivery rate the same for both engines, then I tend to agree. But it's hard to reconcile that constraint with WOT operation, since if we're talking about a normally aspirated engine, the spark ignition engine may well have the advantage (assuming this identical engine design is a compromise between SI and CI optimum design), since it will be running at or near stoich, which is difficult to do with a diesel as we all know.
I also think the 14:1 CR proposed for this comparison is unfair to the SI engine, assuming pump fuel. Perhaps a better way to pose the comparison would be to allow each engine to optimize CR (as I assume is the case for ignition/injection timing) while not materially affecting swirl/turbulence etc.
RE: Fundamental explanation for different torque curve SI-CI
The integral is just the area under the curve as already mentioned a few times in the statement that the pressure is held higher for longer over the whole cycle than in an SI engine... Nothing new there! If you are considering the PV diagram, how do you assume you translate that pressure into a torque? Through geometry of the engine! Would you care to give YOUR different explanation to the OP?
RE: Fundamental explanation for different torque curve SI-CI
- Steve
RE: Fundamental explanation for different torque curve SI-CI
- Steve
RE: Fundamental explanation for different torque curve SI-CI
RE: Fundamental explanation for different torque curve SI-CI
Apparently there is some kind of semantics going on here regarding the word leverage. Whether you like it or not, the crankshaft is in fact a lever arm --- but you can pick the word.
Regarding rod/crank kinematics & BMEP being one size fits all, why do you think that instant torque is typically around 42 degrees ATDC, even though the typical best crank angle is around 72 degrees (18 degrees of angle for the rod makes 90). And why do you think that no matter how high the pressure is no torque is produced at TDC? And yes I know that this all averages out through out the power cycle.
But to best understand why some engine combinations are better than others, I believe you have look at the details on say a 1 degree basis. The peak torque at the 42 degree area is because it’s the best combination of available pressure at the most advantageous (like that word better) crank position at this point in the stroke. Obviously at 72 degrees the crank is in the best position, but the pressure is much lower at this point. And if the highest peak pressure takes place at exactly TDC there is no leverage (there is that word again) and no work is produce as only rotational inertia makes it all happen. As a matter of fact, if the peak pressure takes place to much before TDC the engine will turn backwards. As a result, most engines are calibrated to have the peak pressure around 15 degrees ATDC.
Yes I understand the laws of energy conservation, but you can change the power profile characteristics of an engine by changing the mechanical configurations. And when you skew the pressure later in the power cycle you get a combination of higher pressure with a more advantageous crank angle and thus more torque.
The best example I can think of at this time is how do you think a supper charged engine can double the power over a normally aspirated engine without doubling the peak pressure? Yes the peak pressure is higher, but not twice as high. It’s the charge density being higher in a boosted engine and therefore taking longer to complete the combustion process is one of the real reasons for the additional power. This increases combustion time can permit twice the pressure at 72 degree crank angle over a normally aspirated engine, but still have lower pressure at this point then the peak pressure of a normally aspirated engine. The results are twice the power potential without twice the structural mass and/or integrity of the engine all because of the combination of best combustion pressure with the most advantageous crank angle.
al1
RE: Fundamental explanation for different torque curve SI-CI
RE: Fundamental explanation for different torque curve SI-CI
RE: Fundamental explanation for different torque curve SI-CI
- Steve
RE: Fundamental explanation for different torque curve SI-CI
You are laboring under a common mis-conception about engine torque.
Engine power is the amount of work performed by the engine in a given time. Engine torque is the amount of work performed by the engine in a given crank angle (typically, the average work performed over one radian).
Therefore, the "leverage" and actual TORQUE exhibited by the engine on the crankshaft is irrelevant for calculating torque of the engine as that term is used in industry. The supercharged engine has double the area under the curve and/or a higher engine speed to produce double the power.
Area under the P-V curve, minus efficiency losses, is proportional to the engine "torque". That's it.
RE: Fundamental explanation for different torque curve SI-CI
- Steve
RE: Fundamental explanation for different torque curve SI-CI
Absolutely.
RE: Fundamental explanation for different torque curve SI-CI
Speaking of efficiency, not only does the real world diesels have more torque at low RPM, they also have more efficiency as well. A spark engine can be calibrated to have the same peak pressure later in the cycle, but the efficiency becomes even poorer all things being equal.
al1
RE: Fundamental explanation for different torque curve SI-CI
Could you explain your definition of Torque and Power in greater detail please?
I thought Torque was a force multiplied by a distance hence its units being 'Nm', therefore implying that for a given force an increase in the length away from centre of rotation will increase torque.In this case the force is used to turn a shaft. As the engine is of the reciprocating type, the length away from centre of rotation changes every degree.
Thus the torque changes every degree(regardless of cylinder pressure). The mean of the calculated torque for the cycle will be the same as the measured torque.
Is this wrong??
RE: Fundamental explanation for different torque curve SI-CI
If you were a physicist, you would be right on the money. However, as torque has been used in the engine business (I'll call that e-torque), it's a measure of how hard the engine tugs on the transmission. It happens to have units of force*distance, which is either torque or work.
If you could actually calculate the instantaneous net torque (let's call that p-torque for physics-torque) on the crankshaft, you could possibly determine the "mean of the calculated torque for the cycle" as you suggest. However, you have to remember that there are typically multiple cylinders acting on the crankshaft at any moment, some putting work onto the crankshaft, and some taking work off of the crankshaft. Therefore as a practical matter e-torque cannot be determined by your method - at least not without exotic modeling that no one would bother to perform. Realize that "peak torque" is a steady-state concept referring to the net action of all cylinders acting on the crankshaft at the maximum fueling per stroke, and not the maximum p-torque exerted by any cylinder on the crankshaft.
But back to the original point - e-torque is defined as the amount of work the engine puts onto the crankshaft over a given rotational angle of the crankshaft, and not the physics based concept of rotational force exerted on the crankshaft at any particular moment.
In my opinion, the term torque for engines as it is used is a bit misleading. However, it is what it is, and it's been that way for so long I doubt anyone is going to change it. It is useful to think of it as p-torque when designing transmissions, because the transmission sees the entire net effect of all cylinders acting - which is probably why we are where we are. However, it's not helpful to imagine the p-torque provided by a particular cylinder at a particular moment in time, because that's just not the same concept.
RE: Fundamental explanation for different torque curve SI-CI
Just because the automotive world confuses the engine torque with the phisical torque exerted on the shaft does not give an answer to the OP in my opinion.
Imagine the case of a one cylinder diesel engine, how would you explain the higher torque there compared to SI?
If we had a one cylider engine connected to a brake with a line plotter showing torque throughout the cycle at a low rpm, the CI engine would hold a higher value for longer than the SI engine on the plotted line through the combustion stroke, this is agreed.
If you then averaged the torque value for the whole trace, the diesel would still come out on top. Your concepts of e-torque and p-torque are not mutually exclusive.
Although the averaged value (e-torque) includes the time frame for the average, the axis units of the dyno plot will cancel this out, so effectively, at a certain rpm the theory states that you could stop the engine dead with a bar and a weight long enough to match the torque being produced.
But on this one cylinder engine, reading the averaged torque value how would you know when the engine was producing exactly this value? It may be more or alot less at any one time.
The torque is always changing due to the geometry of the engine.
Diesels benefit from this by being able to hold the pressure for longer.
The overall area under the curve (minus pumping loop and friction) on a diesel will be higher, yes. But if you could find a diesel engine and a gas engine that had the same overall areas, the diesel WOULD STILL PRODUCE MORE LOW RPM TORQUE!
Its not the magnitude of the force on its own, but rather the effectiveness of the timing of the push on the piston to generate maximum torque at the crankshaft, there are two separate components that combine to give THE SAME TORQUE READING. Torque is torque, no matter the application.
RE: Fundamental explanation for different torque curve SI-CI
If you build them to have the same area under the curve (and all that other stuff you said), they will in fact produce the same torque (e-torque).
I think what you're getting at is that the gasoline engine would have more crank angle degrees where it didn't "push", which is true. That does mean the gasoline engine would be more prone to stall, but not because it was producing less torque. That's why they have flywheels.
I agree that e-torque and p-torque are not exclusive. It's just that getting e-torque from p-torque is difficult, because you have to sum the p-torque for all cylinders through an entire engine rotation (probably two to get a combustion event on each cylinder since the cylinders won't be identical), then divide the total by 2-pi to get your e-torque. That's not exactly what people are thinking of when they start talking p-torque.
RE: Fundamental explanation for different torque curve SI-CI
The gas engine will still use the same work on the piston earlier on to produce a lower value of torque, but can obviously maintain this at higher rpm. How COULD it have the same value of torque if the force produced is the same yet it has less distance?
If you say that this is a function of time and that the force is applied over a shorter period of time though at a smaller angle which gives an average torque for the whole combustion period similar to the diesel (which is what I assume you are getting at) then a simple integration of instantaneous torque by time within the combustion window will show that is not true.
I imgaine it like spinning a bike wheel; spinning by hand either the tyre or the hub of the spokes. You can obviously start it spinning easier at the tyre, but can only maintain acceleration to a point, then you are better off spinning at the inside with the same work input you can acheive a much higher speed. The work you are doing to spin the wheel is exactly the same. Diesel at the tyre, gas at the hub.
Greg
RE: Fundamental explanation for different torque curve SI-CI
- Steve
RE: Fundamental explanation for different torque curve SI-CI
What, so cylinder volume varies independently of crankshaft position? Bull-oney. Assuming we all believe conservation of energy applies to internal combustion engines (speak up if you disagree), the the area under the P-V curve HAS to add up to the amount of work out of the crank over a cycle. The amount of work out of the crank over a cycle is torque integrated over a cycle. As stated above, If you build them to have the same area under the curve, they will in fact produce the same torque.
RE: Fundamental explanation for different torque curve SI-CI
I think I've identified the source of your confusion.
This statement: "The work done on the piston by the gas is converted into the plotted torque line by multiplication of the distance of the acting force to the crank centreline"
illustrates that you are talking about P-torque still rather than e-torque. That is not how the plotted torque line is created. The plotted torque line is measured by a dyno, or you could conceptually imagine measuring the acceleration of a flywheel. The E-torque is measuring the sum of all cylinder effects dampened by the flexible crankshaft, and does not see individual combustion events occuring against the crankshaft.
E-torque is defined as the work produced by a given turning angle of the crankshaft. This is required to be proportional to the work performed by a combustion event because the same number of combustion events occur in a single turn of the crankshaft for gasoline and diesel (assuming 4-cycle, which I think we're not disagreeing about). Therefore, if you build me a gasoline engine with the same useful combustion work per event, same number of combustion events per turning angle, same torque.
The way in which the work gets to the crankshaft - i.e. in high p-torque bursts or low p-torque longer events - is irrelevant from the perspective of e-torque.
RE: Fundamental explanation for different torque curve SI-CI
“What, so cylinder volume varies independently of crankshaft position? Bull-oney.”
It’s not the cylinder volume per se, it’s the pressure from the combustion process that can be different i.e. CI verses Si. The same principal applies to why diesel have more low speed torque by having more of the pressure curve at a more favorable crank position. This is not challenging the laws of energy conservation as typically only 30 to 40% of the heat energy is only being used to turn the crank any way.
al1
RE: Fundamental explanation for different torque curve SI-CI
RE: Fundamental explanation for different torque curve SI-CI
- Steve
RE: Fundamental explanation for different torque curve SI-CI
I beg your pardon - am not confused! Please re-read my post that you replied to for the answer to your last post. I can clearly see where you guys have hit a brick wall in terms of relating the now used terms of p and e-torque. I understand you wish to stick to your guns in terms of conventions and current use of the term torque (or e-torque) when defining the characteristics of an engine.
The fact is that the 'dyno plot' is an average of the 'plotted line' for one cycle. Please forget about multiple cylinders and flywheels at the moment, that just clouds the water.
The plotted line (torque output) changes as a function of the angle of rod to crank and pressure. The 'dyno' graph takes a moving average of the torque over a cycle, i.e. the value displayed on a typical engine dyno will have the average torque over that sapce of time. Hence why I said it cancels out!
I am not going to explain it all again, but it is frustrating that several people here seem to reach the point where they understand this principle and then fall back on the 'thats not torque' answer. IT IS.
One way to end this is to do a simulation on say GTPower for a one cylinder diesel and petrol and set the parameters so that you have the same BTU's of petrol and diesel injected at the same low rpm and have complete combustion at stoich. This should fix the area under the curve in terms of pressure. Then run the simulation and watch the diesel produce more torque. The program will calculate torque by multiplying through by the engine geometry.
I am certainly not going to to this simulation as I have been down this road before, and I happily convinced that the later pressure availiable in a diesel engine cycle provides a greater torque at low rpm compared to a gas engine due (in part) to the geometry of rod and crank.
That was the last post from me in this topic.
RE: Fundamental explanation for different torque curve SI-CI
Ivymike, you are just plain not reading the words on the screen.
V2 is the BDC piston position. If the pressure is higher toward that end of the diagram, then the combustion event has kept the pressure higher, later in the stroke. But thanks for totally colluding what I had said.
RE: Fundamental explanation for different torque curve SI-CI
In a SI engine you have ignition and burn before TDC you will not only loose that part of the energy but you will have to overcome it to get past TDC. Just calculate the force on the piston at 15* ATDC and now apply that same force at 20* ATDC.
Disregarding the BTU delta IMHO is the torque difference
Cheers
I don't know anything but the people that do.
RE: Fundamental explanation for different torque curve SI-CI
I apologize for misinterpreting the question of why CI has more torque then SI. Of course if you could have the same pressure under the curve and all other things are equal, they should be the same --- but in the real world they are not and that is why diesels have more torque. I also apologize for assuming gross heat energy verses net cylinder pressure on the debate of energy conservation.
This is one of my favorite subjects as I’m working on a true constant volume during combustion engine that has a different mechanical kinematics then a classic engine. There are major improvements to be had when you get away from what I call the 1.57 classic engine mentality. All engine are 1.57 no matter what size, meaning that the crankshaft travels 1.57 further than the piston does. Because of the 1.57, it also permits the PV charts to apply to all engines, however when you change that, it no longer applies.
al1
RE: Fundamental explanation for different torque curve SI-CI
RE: Fundamental explanation for different torque curve SI-CI
Yes, but... in a well tuned engine this energy is not very significant.
The faster the burn rate (the more net energy release right after TDC, versus later in the cycle), the greater the net area under the PV curve. This is easily seen on a PV diagram. Think about effective expansion ratio (EER).
So maximizing torque at any engine speed is all about getting the fastest possible burn rate right after TDC (within the physical, material & NVH constraints). To accomplish this, diesel as well as SI engines do well to trade-off the small energy loss from initiating combustion slightly before TDC.
RE: Fundamental explanation for different torque curve SI-CI
I assumed based on the OP that we were talking about the torque curve on a speed-torque plot. I agree with your comments in the context of a crank angle-torque plot.
RE: Fundamental explanation for different torque curve SI-CI
We ARE talking about a speed vs torque plot.
If you agree with the fundamental reasoning behind the fact that the diesel will make more instantaneous torque at lower rpm, and that you agree with the fact that diesels will make more 'e-torque' at lower revs as an SI engine then you agree with the whole thing.
I have become totally disillusioned by this whole thread.
RE: Fundamental explanation for different torque curve SI-CI
- Steve
RE: Fundamental explanation for different torque curve SI-CI
Regards, Ian
RE: Fundamental explanation for different torque curve SI-CI
RE: Fundamental explanation for different torque curve SI-CI
Maybe enough is enough.
Regards
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of Eng-Tips Fora.