Engine throttle/resistance question 12

[Yster]

I have been given the job of creating a dynamic car model in software at the company I work for. Being a computer engineer, it has been quite an effort for me to get to know cars and the dynamics thereof. As guidelines, I have the following three books to help me:
Fundamentals of Vehicle Dynamics - Gillespie
Motor Vehicle Dynamics - Genta
Race Car Vehicle Dynamics - Milliken & Milliken

These books have proven to be invaluable, but there are still some questions that aren't answered which I just can't figure out for myself.

First question: how does the throttle affect the torque delivered by the engine at a specific engine speed? I assume the torque table gives the max torque at a certain engine speed, thus if the throttle is at maximum, that is the torque delivered. I further assume the torque is scaled linearly with the throttle, thus if the pedal is pushed only half way, you only get 50% torque at the specified rpm value, etc. If this is incorrect, please inform me.

My second question is this: if the car is idling, the throttle is probably at some minimum level, just enough to keep the engine running. I assume this level probably can't be computed mathematically, and each car manufacturer just approximates this value by trial and error. Also, in the above mentioned books several equations are given to determine the engine torque, the torque at the clutch, the torque to the drive shaft, the torque to the axles, and the torque to the wheels, which are all pretty straight forward. The only problem I have is to get my engine to idle correctly. I know that is my car is in neutral and the engine is idling, that there is still some resultant torque delivered from the engine, but why isn't the engine speed increasing all the time? According to the equations I have, this is what should happen. So I think one of the following is possible:
1. I shouldn't scale torque linearly with throttle
2. There is some engine resistance factor that is not shown in the equations, that works against the torque delivered by the engine to keep the engine rotating at a constant speed.

It seems as if all this should be pretty straight forward, since the books don't really bother all that much with the engine and drive train, but instead focuses on the tyres and suspension dynamics.

A pointer in any direction at all would be greatly appreciated!

 

[patprimmer]

First question.

The change in torque is not linear with throttle position, and in fact is far from it. It will be somewhat closer to linear if you compare manifold pressure to torque, but this is still not linear. The proportional difference between throttle position and torque will be dependant on quite a few variables, such as rpm, cam timing, ignition timing, fuel properties, throttle plate size, plenum volume and shape, air quality, inlet manifold runner length and cross sectional area to name just a few.

For a particular engine, it is probably possible to get repeatable data for engine output vs manifold vacuum if the fuel and engine management system are controlled, and corrections are made for air quality.

The idle will be set by 1 of 2 methods.

The Engine Management System will control the throttle plate or a throttle bypass air bleed to control idle speed with it's computer.

The throttle plate or idle bypass will be adjusted until the idle speed is acceptable. The engine will then run at a speed where resistance and torque power are in equilibrium. If the supply of air or the resistance change, the idle speed will change to maintain equilibrium.

Regards
pat pprimmer@acay.com.au
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[GregLocock]

To amplify Pat's answer, on the idle question.

Typically an engine runs a MAP of about -70 kPA at idle, or, if you want to think of it that way, 1/3 of the WOT indicated power (or torque) at that speed.

However, the throttle butterfly will be barely cracked at that condition, to such an extent that a bypass is usually used instead.

From this 1/3 figure you can work out the frictional losses in the engine, which is why it does not continually speed up. The friction power required will probably rise proportional to engine speed^(just over 1)





Cheers

Greg Locock

 

[franzh]

My two cents worth:
Lets not forget the increased motoring power loss of the engine at closed throttle. This can be a significant amount of load, easily exceeding the driveline and radiant loss at low speeds. Visualize pulling six 75mm suction cups 90mm, five times per second.
Franz

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[GregLocock]

Good catch. That's roughly um, 70000*.002/2*900/60 for a 2 litre, at 900 rpm, or 1 kW

Wow








Cheers

Greg Locock

 

[Yster]

Ok, thanks for all your replies. It has helped quite a bit. Just to make certain that I understand exactly what you are trying to say:
First of all, there are too many factors to take into consideration than to just model an engine of a particular car from some information on a website giving details such as max power and torque at certain rpm values, number of cylinders, cylinder capacities, etc.
Second of all, the resultant or net torque at the output of the engine is zero, since the engine management system ensures that the engine doesn't stall. If there were a resultant torque, the engine would keep on accelerating, since Torque=Inertia*Alpha ==>> Alpha=Torque/Inertia.
Thirdly, if the car is in neutral and I step on the throttle partway, the rpm value eventually stabilises on some arbitrary value and doesn't keep on accelerating to the redline due to the friction power rising to rpm^(>1). (If you can maybe supply an equation I would appreciate it.)
Fourthly, since I don't have all the detailed information concerning the engine, would it be, from a simulation point of view, a bad idea to assume the throttle to be linear? How badly would that affect my simulation?

That is quite a lot of information I need just to simulate a car in software. Can I just create the following test scenario:
Let’s say my car (engine) is idling at about 800 RPM, with a max torque of 160 N.m at that engine speed. I know I read somewhere during idling the typical engine produces about 50 N.m of torque (what a really bad assumption, I know, but I don’t have anything else to use), so the resistive forces of the engine must be generating -50 N.m of torque, or equal to the idling engine torque to bring the system in equilibrium. Suddenly I push the throttle to the max, thus the engine now delivers 160-50=110 N.m resultant torque to accelerate the car. This torque goes to the clutch->transmission->driveshaft->axle(s)->wheels, which eventually accelerates the car. Is this more or less accurate? Also, if I rev the engine up to some higher rpm value, does the engine management system see that the engine isn’t going to stall at that rpm value, and consequently doesn’t add any input to the engine? The engine then decelerates due to the internal engine friction, until the rpm value is low enough that the engine management system kicks in again and supplies input to the engine?

If you can maybe recommend a good book or informative websites that give more detailed information about engines (from an engineering perspective), I would greatly appreciate it! I hate having to read how some people who created car simulators fudged this and that just to get it to work more or less.

To Greg:
What are all the parameters you used in that equation? I gather that the 900/60 gives you the revolutions per second.

Thanks again for your help!

 

[GregLocock]

That's power =(change in pressure)*(volumetric flow rate) in the appropriate units, and is an approximation not strictly valid for large relative changes in pressure.

By all means assume a linear relationship for throttle vs torque, for a racing game it should make very little difference. The characteristic is heavily non linear in real cars anyway.

Your model of the system is slightly over complex, you are treating it as a governed system such as a diesel. An SI engine runs open loop, so far as demand and output speed/torque goes.



Cheers

Greg Locock

 

[AboveRedline]

Yster,

It would help us if we knew what you expected to learn from the software (or your customers) once its in use.

There is torque available from an engine at idle, just put it in gear and let out your clutch with no throttle (even a non-computer controlled engine), the car will move, although engine speed will drop.

 

[Yster]

Redline,

I would like to have several C++ classes that, in the end, describe a car in terms of engine, suspension, wheels/tyres, and steering. Obviously it’s going to take a fairly large amount of work just to get all the information about a specific type of car read in. Eventually, you're supposed to be able to drive the car around that you just created by means of a large amount of parameters. I don't what to unnecessarily fudge certain aspects of the car (as is done in almost all driving games), since we create accurate simulators where I work, not racing games or similar. I'm just having a hard time finding the required information in the text books we have in our library, and the Internet isn't all that useful either in terms of accurate car models. The only software that have accurate models of engines/cars are commercial products.

I don’t think that there is torque available at the engine at idle. I’m pretty sure of this, because I emailed the author of Motor Vehicle Dynamics, and he said that at idle the engine gives exactly enough power to overcome internal losses in the engine. If not, the engine would just keep on accelerating (Torque=Inertia*Alpha)..

I’ve got my virtual engine more or less up and running now; it’s idling correctly, drops back to the correct idling frequency if it was revved a bit, and delivers the correct output torque at the axles after multiplying with gear and final drive ratios as well. So far so good. I’ve also got the engine placed in a preliminary software car model/class. Most of the required parameters are hard coded for testing purposes at the moment, but the engine is causing the car to accelerate forward, and I can get the correct engine speed from the wheel rotational velocity. My next objective is to properly implement the tyres, suspension, and steering. So I’m busy reading up on that. It seems that most the books go into very great detail about tyres, and just ignore the details of the engine. If you can maybe recommend a good book on the dynamics of car engines, it might help me a lot.

 

[ivymike]

I don’t think that there is torque available at the engine at idle.

Define "torque available at the engine at idle" more clearly for a more accurate answer. At idle, the engine is still driving accessories, pumps, at least part of the transmission, etc. This means that there is necessarily some torque output beyond what is required to keep the engine alone spinning, otherwise the equation you mentioned above will tell you that the engine decelerates and stops.

 

[GregLocock]

There is torque available. As I said, the throttle is almost closed at idle, and the MAP indicates that roughly 3 times the torque is available, than is used to overcome friction and drive the accessory load at idle (which is quite substantial by the way), and overcome the pumping losses in the butterfly - which, incidentally, will probably vary proportional to n^3 by the way.

Ugly long sentence, reparse it at your leisure!

Cheers

Greg Locock

 

[Yster]

Define "torque available at the engine at idle": The torque finally delivered to the clutch, after all the internal losses have been accounted for. The final output delivered to the transmission. The car is in neutral (hence idling), thus the transmission isn't even connected to the engine. The entire engine system is in equilibrium, and no components are accelerating or decelerating. Does that make sense or am I still talking nonsense?

You have to excuse the poor way at which I describe what I mean. I'm a computer engineer who doesn’t know all that much about cars and now suddenly I have to know a lot. Any of you guys know anything about a good textbook giving more detailed information about car engines and drive trains? Come on all you mechanical and automotive engineers! You had to learn all this somewhere?

Thanks for the replies! I greatly appreciate this!

 

[TTmotors]

Very simply all throttle assemblies are not treated equal. Depending on the attention to drivability by the manufacturer the throttle body will cause an inverted logrithmic effect. As the throttle increases numerically in degrees the effect on engine power will go down in perportion. at 50 percent of the 90 degree throttle angle 75 percent of engien power mayby available. At 75 percent of maximium throttle angle 90 percent mayby available. At 99% of throttle angle 100 percent of engine power mayby available. However each potentiometer works very differently. As the starting angle and bore are changed the responsiveness of the throttle to the engine also changes.

For instance at a high starting angle (a plate set at 30 degrees less then 90) the throttle response will be very quick. Because of its large permiter very little movement is needed to unviel a large flow area. So 10 percent of throttle may give you 40% of the engines maximium power.

Larger bores increase throttle response because they flow more at a lower throttle angle like the higher angle plates. However they do not make complete use of the entire throttle body. Once the engine has achieved an air velocity low enough not to cause too many parasitic losses over the throttle plate there is very little to be gained. In some cases power is even lost at WOT over 98 or 99 percent of maximium throttle angle.

It wouldnt be that hard to design a simulation that is static. Some simple dyno runs on your own car might allude to some realistic and generally acceptable potentiometer lines. Map sensors and 02 sensors as well as the torque sensing equipment would give you accurate values in MPG and part throttle performance and acceleration. Some simple timed tests would allude to its aerdynamics and mechanical losses. GTEC units mounted in the front and rear will give good lateral grip lines coupled with a regimen of traction testing involving differing kinds of corners and road cambers.

But building a simulation that is dynamic and responds to individual changes to a vehicle would nearly be impossible. The shear number of inputs required to make a vehicle behave realistically would overwelm even the largest computers.

-Travis-

 

[MaxRaceSoftware]

>>>I further assume the torque is scaled linearly with the throttle, thus if the pedal is pushed only half way, you only get 50% torque at the specified rpm value, etc. If this is incorrect, please inform me.<<<

all the Source Code i've seen or come across at various sites , just use exactly like what you 1st were going to do

50 % Throttle = 50 % Torque

its sure the easiest method to use, i doubt if anyone would notice that your Simulation wasn't exactly mimicking
real-world results ???



Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[MaxRaceSoftware]

here's a Website that might be of Help to you ?

http://www.gamedev.net/reference/list.asp?categoryid=28

Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[Yster]

Just one more question: If you guys would recommend a good book on internal combustion engine dynamics, what would it be? The books I have don't really go into engines all that much, and mainly discusses tyres, suspensions, and aerodynamics.

Thanx for all your help, you guys have been really helpful!

 

[TTmotors]

4 Stroke Performance Tuning by grham a bell
(general n/a tuning Carb and EFI)

Corky bell`s maximium boost
(Turbo tuning)

Corky bell`s superchargers
(Supercharger tuning)

Those are all good basic titles that cover all the spectrum of performance engines.

 

[Tmoose]

When the old Volvo is revving under 3000 rpm the last half of the throttle motion does not seem to increase the acceleration appreciably. I think that means whatever angle the throttle plate is open lets >>enough<< air seep in. I kind of think the manifold vacuum might be close to zero at that point also.

 

[GregLocock]

Yes TMoose, that's how it works. The pressure drop across the butterfly is given by Bernouilli (1/2 rho v^2), v is volumetric flow rate divided by area, and at 3000 rpm the volumetric flow rate is about half what it'll be at 6000 rpm. So for a given pressure drop your open area in the throttle should be roughly proportional to engine speed.

Fundamentals of engine design are covered by two excellent books, Charles Lafayette Taylor and Heywood. They are both called something obvious, the internal combustion engine, I think.


Cheers

Greg Locock

 

[SBBlue]

Just some thoughts to muddy up the waters. . . . .

One way to look at this, assuming stoichometric combustion (which isn't always the case) is that the power output of the engine will be proportional to the volume of air flow through the engine. The more the throttle is open and the higher the pressure in the intake manifold, the higher the engine power. The higher the engine RPM (within some limits), the more air will be pumped and the higher the power output of the engine.

Consider when the intake manifold pressure is one-half atmosphere and the engine speed is at 3000 rpm. The same amount of air (again, I'm ignoring all that stuff about valve timings, volumetric efficiency, etc just to make some rough calculations) will flow as if the engine rpm is 1500 rpm and the throttle is wide open (intake manifold pressure is one atmosphere).

Since Torque is equal to (engine power)/rpm, if we increase the engine rpm while keeping engine power constant, we will decrease the torque. In our example, the engine will have one-half the torque at 3000 rpm that it does at 1500 rpm. If we had a perfect gear set, we could reduce the speed of the shaft from 3000 rpm to 1500 rpm and. . .surprise!! The torque is now twice as much.

But. . . . .things aren't quite that simple in River City. Consider the amount of work the engine does to pump air through the engine. If the intake manifold pressure is constant, the amount of work to create the vacuum will be proportional to the mass of air pumped times the intake manifold pressure. In our example, the amount of air pumped is the same at both 1500 and 3000 rpm. However, the vacuum that the engine works against at 1500 rpm is one atmosphere; at 3000 rpm, it's only 0.5 atms. The pressure difference for the piston for the first case will be 0 atms; for the second case, 0.5 atms. It should be intuitively obvious to the most casual observer that the amount of vacuum work will be greater for the engine going at 3000 rpm than it will be for the engine going at 1500 rpm.

It is reasonably easy to show that amount of vacuum work for an engine with an intake manifold pressure of 0.25 atms is about 10% of a gas engine's power output. As engine speed increases (if the power is held constant) the amount of vacuum work will increase -- similar to the manner in which the amount of friction work will increase.

So you need to also consider the effect of intake manifold pressure on the vacuum work.

Hope this has raised the level of confusion!!!