Thank You for using CarTest 2000. I are sure that you will find this program both useful and instructive. CarTest is an automobile straight-line, maximum acceleration computer simulation. The user enters the specifications for an automobile and CarTest will mathematically accelerate the car from a standing or rolling start until maximum speed is reached. The simulation is time-based and event-driven. That is, the acceleration times are not explicitly calculated but, rather, time progresses at small increments and the usual events that take place during acceleration: clutch dump, wheelspin, shifting, etc. happen in proper course. The acceleration times are arbitrary milestones that are reached during the car's response to these events.

To understand what CarTest can do it may be interesting to know why the program was written in the first place. Some time ago, when researching a new car purchase, I was comparing specifications and performance figures from various car magazines, I discovered that performance (acceleration) results for the same model often varied substantially from one publication to another, sometimes as much as 20% in 0-60 acceleration times. (Insignificant differences are considered to be about 5% in 0-60 times.)

Some explanations for this could be that there are actual differences in the manufacturing of each car, variations in test conditions (air temperature, humidity, track surface smoothness and temperature, altitude, etc.), or the driving technique of the road tester (using aggressive launch techniques that may be hard on the test car's mechanicals). Whatever the reasons are, it becomes difficult to definitively determine an automobile's performance. I designed CarTest to use a strictly mathematical model to evaluate a car's acceleration performance. The advantage of such a tool is that it is strictly objective and consistent as performance results are based strictly on specifications. This approach filters out differences due to manufacturing variances, test conditions, testing techniques, and driver skill. It also eliminates the lurking suspicion that a specially "souped-up" car was used for testing and that the car you might buy off the showroom floor will not perform nearly as well.

CarTest, therefore, is designed to augment, not replace, published automobile actual road test results. Whatever variations exist in road test results, they accurately reflect how that car performed on that day, in that state of tune, at that test site, with that driver.

Although CarTest does have its limitations (described in a separate section) the accuracy of the model is very good and can be used with confidence in assessing the absolute performance of an automobile as well as its performance relative to its competitors and other cars on the road. You may judge CarTest yourself by comparing its results with performance results available in the road test pages of the various car magazines. For the best comparisons, select cars tested by more than one publication.

It was decided that for CarTest to be useful, it must be based solely on commonly published specifications for a car such as those found in the road test specifications pages of car magazines. All other parameters upon which the model is based are stored separately and, although are made available to the CarTest user, need not normally be modified.

CarTest is supplied with data for thousands of foreign and domestic cars, but more cars can be added to the data base. CarTest users will want to evaluate cars being considered for purchase.

In addition, CarTest can be used as an educational tool to understand how performance varies with changes in various specifications. For example, CarTest allows you to:

The accuracy of an automobile performance model is primarily limited by the data that can be readily obtained about a particular automobile. In particular, the shape of the engine power curve is limited by the knowledge of only the peak horsepower and peak torque points. An idealized 'shape' is derived solely from the engine type (turbocharged, supercharged, or normally aspirated) and this data only, but in reality, power curves do not follow an ideal average shape (if you do know the power curve shape exactly, you can replace the idealized shape with the known curve using 'car-specific' model parameters). Further, power losses due to engine auxiliaries, transmission, drive shafts, and differentials are also generalized by single percentages and are really much more complex than can be modeled using only road test specifications data. Also, some tires are 'stickier' than others and the coefficients of friction specified will not be representative of the performance of all tires on all road surfaces on all road conditions. Hence, standing start acceleration tests are subject to this limitation, rolling start tests, however, are not.

Wheelspin starts do not model any left/right side imbalances which are caused by uneven left/right side weight distributions, uneven road surface characteristics or uneven tire characteristics. Hence, high rpm wheelspin starts which would, in actuality, cause a car to swerve and skid off a straight line are not modeled. Also, any suspension related acceleration effects such as wheel hop or axle tramp off the line are also not modeled.

Top speed estimates in CarTest assume an even and well-defined environment. In actuality, the top speed of a car can vary greatly due to a number of factors: local wind and weather conditions, road surface variations, engine cooling system performance, wheel alignment, state of engine tune, turbocharger intercooler temperature buildup, tire condition and temperature, etc. None of these complex real world factors are modeled in CarTest.

Skidpad and slalom tests are based mostly on tire specifications and do not take into account suspension design. Therefore, these numbers should be used only as rough estimates of handling performance.

The world's race tracks are defined approximately without variations in elevation and with flat bank angles. The track surface is also assumed to behave as would a typical road surface. Race track surfaces (not city street tracks) are designed to enhance friction between the surface and the tires without consideration for wear of either the track or the tire. The car's braking potential is assumed to be optimal and does not account for variations in efficiency in braking systems. Also, no attempt has been made to optimize the driving technique for a particular car around a particular track.

CarTest is supplied with a preloaded data file containing pertinent data on thousands of domestic and foreign automobiles. Every effort has been made to ensure accurcy, but it is not guaranteed. Further, data for test weight, ground clearance, and coefficient of drag is not always available and was estimated for some cars.

CarTest 2000 differs substantially from the DOS version in its user interface capabilities. You should have no trouble using the buttons and menu bars to navigate through the various functions. CarTest 2000 is written as a Java Application. It should run equally well on any computer that has a compatible Java Virtual Machine, an interpretor that executes the Java bytecodes for the computer and operating system you are using. That includes Windows 95/98/NT/2000/XP/Vista, Mac OS Linux, Solaris, etc.

Since Java is a multithreaded language, you can have things happening simultaneously all over the screen, limited only by available memory and processing power. Because CarTest displays lots of information on the screen at the same time, it was designed to require a minimum screen resolution of 1024x768 pixels. You can also choose to leave some windows up on the screen after computations are completed and perform other tasks to perform comparisons.

For example, CarTest 2000 has no limit on the number of cars that compete in a drag race. All selected cars will appear in a scrollable window. If you select too many cars, however, performance will drastically suffer. This holds for all comparison tests.

The CarTest 2000 database uses curb weight for a car rather than the test weight as in the DOS version. The model parameters contain additional fields for driver weight and on-board fuel weight that you can modify separately.

You can now specify a greater range of values for a custom horsepower curve, up to 20,000 RPM. You can now also specify that curve as being at the driving wheels as well as at the flywheel. You can also specify measured maximum driving wheels power and torque without the entire curve if that is all the data you have. Custom power curve use between specified values can be evaluated using a cubic spline or linear interpolation.

The Parameter Sensitivity function optionally re-optimizes the launch method and engine RPM for each data point in the analysis. The option to use the same starting method as for the baseline data is the default.

The Parameter Sensitivity feature has a Horsepower Finder function. The user specifies the performance trial such as the 1/4 mile, enters the observed trial time and CarTest 2000 will compuite the required horsepower to produce that result.

The Parameter Sensitivity feature has an Optimum Gear Ratio Finder function. CarTest 2000 will determine a set of optimum gear ratios that produce the minimum time the specified acceleration trial.

The time-to-speed and time-to-distance milestones are not fixed but, rather, are definable by the user. This is true for both the individual car acceleration as well as for the Performance Comparison Analysis.

CarTest 2000 now reports 1/4 mile speed as the average speed over the 66 foot interval preceding the finish line. This is to be more in conformance with drag strip common practice.

Data displayed graphically should be more useful due to the inclusion of a number of additional features related to these plots. First, passing the mouse arrow over a plot acts as a digitizer showing the position of the mouse in data units. You can place the mouse over a position on the curve and instantly read the (x,y) coordinate values displayed in the frame title bar. This function is currently functional only when using the Metal Look & Feel. Second, you can draw a box on the plot by dragging the mouse. You can then redraw the plot using the bounds of that box as the new axes. You can zoom in as many times as you like. The original plot can be retrieved at any time. Third, you can toggle the display of the actual data points used to construct the plot.

CarTest 2000 now includes a capability to model Continuously Variable Transmissions (CVT). The program will optimize the ratio profile for best acceleration. An extra field has been added to enter data for the new 7 and 8-speed transmissions.

Combined Hybrid, Parallel Hybrid, and Electric cars can be specified and analyzed in CarTest 2000

To select a car from the list, simply place the mouse arrow over the car name and click the (normally left) mouse button. To select an additional car, place the mouse arrow over that car name, hold down the Ctrl key and click the mouse button. Two cars are now selected. You may select as many cars as you wish in this manner. To select a range of cars, select one car then place the mouse arrow over a second car name, hold down the Shift key and click the mouse button. The entire range of cars between the two will be selected. You may select various cars from the database using combinations of these techniques.

You may select a single car and immediately bring up the data window for that car by double clicking on a car name.

After you have 'selected' the car list by clicking at least once on the list area you can shortcut immediately to cars beginning with a letter by pressing a letter key. If you press the 'N' key, for example, cars beginning with 'N' will appear at the top of the list.

You may clear all selections by clicking on the 'Clear All Selections' Button below the list of cars.

You have a choice of screen "look and feel" selections. There is the "metal" look and feel chosen as the preferred alternative for Java programs. You can also return to the familiar "Windows" look and feel as well as a "Motif" selection. Mac users can also choose a familiar look. Not all selections may be available on all platforms.

You can also choose to enable or disable random background images on the main screen. These photos were taken at various Concours car shows. These images are interesting but consume memory. If you find you are experiencing odd performance, disable background images to free up some RAM.

The data file maintenance functions are accessed from the File menu on the main task bar.

ONLINE DATABASE UPDATE:

This function enables the user to keep up with the latest version of the distributed database. Press the "Display Additions Only" button to preview a list of cars that can be added to your database from the master database updated on the CarTest Software home web page. On the window that appears, press the "Update with File Additions" button to view this list and to perform the actual database updates on your data file. Note that no entries will be deleted from your database and no existing entries will be modified. An active internet connection is required.

REFORM DATA FILE:

This function allows you to 'fix' any structural errors in the data file that may occur as a result of power interruptions during critical data update procedures or due to disk media problems. The window shows the current capacity of the data file and permits you to overwrite this field with a number of your choice.

The current capacity of the file is the number of data records that can be contained before the automatic self-expansion procedure occurs when the file becomes full. If you wish to expand or contract the capacity of the file, enter a new number. Click on 'OK'to proceed with the reform process or 'Cancel' to exit. If you enter a different value for the file capacity, that number will be adjusted by CarTest to verify that there is enough space to hold the current number of entries and will be increased to the closest 'prime number' greater than or equal to the value you entered.

MERGE DATA FILES:

The merge data files function allows you to consolidate two CarTest data files into a single master file. This is useful if you have received a new version of CarTest with an updated data file and you wish to preserve any customizations or additions you have made to the data file of the previous version. This function will take the cars in the 'other' data file and add them to the main data file unless they already exist in the primary file. The 'other' file will be unchanged.

Be careful with this function and note the following:

You may chose any valid file name for the 'other' CarTest data base and the merge process will occur. CarTest always uses the file 'CarTest.Data' for its analyses. Therefore, you will probably want your customized 'CarTest.Data' to be the main data file and the new version of 'CarTest.Data' supplied with the CarTest upgrade to be the 'other' data file. Since the two names are the same, you will have to place the secondary 'CarTest.Data' on a different drive or subdirectory or rename the file as you copy it to the same subdirectory containing your customized, primary 'CarTest.Data'. After the merge process, the primary 'CarTest.Data' will have no existing data changed but will now contain any additional cars and data that were provided in the new (other) 'CarTest.Data'.

MERGE DOS DATA FILE:

This function worls identically to the 'Merge Data File' function but the data file to be merged into the main data file, 'CarTest.Data', must be a data file from the DOS version of CarTest (named ct.dat).

CarTest 2000 uses RAM memory as it is needed up to that amount available on your computer up to 256 MB. If you have more memory than that you can alter the startup command in the shortcut used to start the program (right click on the CarTest 2000 startup icon with the little arrow, not the .ico file). The -Xmx256m parameter defines that maximum. Change the 256 to whatever value is appropriate. If you experience sluggish performance or nonfunctionality of some buttons, it is possible that you have consumed all available memory. CarTest will begin using virtual memory (hard disk) after the maximum has been reached. Performance at that point will be drastically reduced.

You can check this by selecting Help - About from the main menu bar. A system information window will appear showing, among other things, memory utilization information. You can click the 'Garbage Collection' button to try to free up some memory although this will normally be done automatically by the program. In any event, you can see if you have a memory consumption problem. Closing open windows will release memory. Screen graphics use alot of memory. This includes the background images on the main screen which you can disable/reenable at any time.

Screen resolution plays a large part in memory consumption. At 1024x768 a full-screen plot uses nearly 3.2MB assuming 32-bit true color. At 1600x1200 that same image consumes about 7.7MB. With the buffering techniques used to generate the plots, these values actually go quite a bit higher.

You may work in either English or Metric units. The units mode is selected from the main menu bar. Whatever mode is selected at the time cars are selected and individual or comparison tests are started will be the operating mode for those analyses windows. Changing the mode from English to Metric while windows are on-screen will not change the units on the fly. What you can do is start an analysis in one units mode, change the mode selection, then start a simultaneous analysis for the same car(s) using the alternate units.

You can experiment with this technique when adding a new car to the database where your data source is a mixture of units. Simply enter some units in one window, others in the other. Be careful when saving to the file however that you don't overwrite good data with zero's.

There is also a handy Units Conversion calculator available from the Help menu. A number of common units can be converted into others in several categories, power, torque, velocity, weight, volume, area, and length. Enter the value on the left, use the boxes to select the 'from' and 'to' units, then press the convert button. All the values in the 'from' fields in each category are converted each time you press the button.

The calculator is available from the Help menu item in the main menu bar. To use the calculator, enter a mathematical expression on the working line then click 'Calc'. The numerical answer will be placed in the R register. To perform another calculation, enter another expression, click on 'calc', and that answer will be placed in the R register while the previous contents of the R register will be moved to the X register. Any contents of the X register will move to the Y register, the Y will move to the Z register, the Z will move to the T, and the previous contents of the T register will be lost.

The following symbols are valid for use on the "Expression" line:

All exponentiations will be done first, followed by multiplication and division, then addition and subtraction. This order of precedence may be altered by using parentheses to specify higher priority operations. Further, the contents of the r,x,y,z, and t registers may be used in the expression. The following are examples of valid expressions:

The results will be maintained to approximately 16 digits of precision. Division by zero is not permitted.

You can use either the buttons on the keypad to enter information or use the keyboard directly. The 'Clear' key erases pending expressions on the working line.

You may also use the menu bar Copy, Cut, and Paste clipboard operations to transfer results to the car data entry screens. Do this in the same way as any other program.

There are two ways to add a car to the data file.

The 'Add New Car' function brings up the quick data entry window. In this window you enter the car name and data for each of the 33 data fields. Here you have the flexibility of selecting various units for several of the data fields. Also, here the cursor will jump from field to field as you press the Enter key or use the up and down arrow keys. Use the menu items to store the data in the file, compute the optimum 0-60 mph launch parameters, and compute the optional tire circumference parameter after you have entered the remainder of the data. The optimum launch parameters computation can take some time, so be patient here.

To compute the optional tire circumference data field, you can use the data fields in the bottom panel. Enter both the Engine rpm and the car's speed at that rpm in top gear. You can then use the menu item at the top to perform the computation. If you don't know both of these two values, you can specify the car's theoretical speed at 1000 rpm in top gear. This last value is sufficient to perform the computation.

CarTest will remember your selection of data entry units so you don't have to reset them each time. Lastly, you may have more than one 'Add New Car' window on the screen at a time. If you want to add car-specific parameters, you must go to the analysis window.

The second way to add a car to the database is directly in the analysis window. Click on the 'Add New Car in Analysis Window' button to invoke this function. You will see the data entry window. Enter a name for the new car in the car name field. At this point you can bring down the File menu and select 'Save As New Car Name' to store the new car in the data file. The data for the car can be added by selecting each data field and entering the numbers and making the correct combo box selections. Be sure to bring down the File menu again and select 'Save Changed Data' to store the data in the file.

You can create car-specific parameters from the file menu and modify the contents as you wish. Be sure to bring down the File menu while these parameters are displayed to save them to the file. The delete option in the File menu applies only to the car-specific parameters if they are displayed. Click on the tabs to switch back to displaying the car data.

To copy a car, select a car to copy from the main list, click the 'Individual Analysis' button to bring up the car data and analysis window, enter a new name for the car, bring down the File menu, then click on 'Copy To New Car name'.

There are two ways to delete a car. You can select the car from the main displayed list, then click on the 'Delete All Selected Cars' button at the bottom of the list. You can also delete a car which is being displayed in its individual analysis window by bringing down the File menu and selecting 'Delete'.

To rename a car, select a car to rename from the main list, click the 'Individual Analysis' button to bring up the car data and analysis window, enter a new name for the car, bring down the File menu, then click on 'Save As New Car Name'.

The data that defines a car is separated into three sections: engine, drivetrain, and chassis and body. This data is normally available from the specifications page or the text of a published road test and are described in detail in separate topics.

ENGINE DISPLACEMENT: Enter the engine size in cubic centimeters.

ENGINE LOCATION: Use the Combo Box to select Front, Mid, or Rear engined cars.

ENGINE TYPE: Check the Combo Box item for normally aspirated, turbocharged, supercharged, sequentially (twin) turbocharged, combined hybrid, parallel bybrid, or electric car.

A combined hybrid is a car that can operate either with electric power, gas engine power or both like the Toyota Prius. The electric motor is connected directly to the wheels with the gasoline engine contributing to powering the car through planetary gears. See more under TRANSMISSION. A parallel hybrid typically runs always with the gasoline engine driving the wheels with the electric motor used to supplement power, An example of a parallel hybrid is the Honda Insight. A car that uses a gasoline engine to strictly run a generator that powers an electric motor that powers the car is a series hybrid. For purposes here a series hybrid is an electric car since the gasoline engine does not contribute directly to powering the wheels.

HORSEPOWER: Enter the peak SAE standard J1349 measured net horsepower and the engine rpm at which it is developed. Horsepower measured in DIN are equivalent (close enough) to S.A.E. net. Horsepower measured in ps (metric horsepower) should be multiplied by .968363636 to obtain SAE net numbers. You can add horsepower directly in ps by using the quick 'Add New Car' function.

TORQUE: Enter the peak engine torque (in lb-ft or N-m) and the engine rpm at which it is developed.

COMPRESSION RATIO: Enter the engine compression ratio (e.g. 10.2:1).

REDLINE: Enter the maximum engine speed in rpm.

STARTING ENGINE SPEED: Enter the engine speed at start of acceleration in rpm.

WITH CLUTCH: This describes the method used to launch the car. Select clutch dump, or clutch slip for manual transmission cars. Automatic transmission cars will use Brake Torque Launches.

An exception to this is that Combined Hybrids will use a Brake Torque launch even though its transmission type should be defined as a Manual.

Note: The above two parameters can be optimally determined by CarTest. See the help information for OPTIMIZE STARTING METHOD.

For hybrids and electric cars additional data fields must be completed as follows:

MOTOR HORSEPOWER: Enter the horsepower and the electric motor rpm at which it is developed.

MOTOR TORQUE: Enter the peak torque (in lb-ft) the electric motor produces. Electric motors always produce maximum torque at zero rpm but that torque is relatively constant for a range of rpm starting at zero. Here you should enter the upper rpm value of that flat torque region.

MOTOR REDLINE: For combined hybrids and electric cars, enter the maximum speed of the electric motor. For parallel hybrids this field is not used as the motor is assumed to turn at the same speed as the gasoline engine.

TRANSMISSION: Select the number of gears and specify if the car has a manual, automatic, or continuously variable transmission (CVT).

Combined hybrids are typically touted as having a CVT transmission. While this may be strictly correct as the planetary gears provide an infinite range of variable speed (but no torque multiplication) between the gasoline engine and the electric motor (which is connected to the driving wheels), CarTest 2000 allows for this separately and a transmission may be defined between the electric motor and the driving wheels. For example, the Toyota Prius is defined as having a manual transmission with one gear of ratio 1.0. The Prius is, in effect, always in top gear. Since the electric motor drives the wheels directly and starts out a zero rpm (matching the stationary wheels) there is no clutching at the start rpm like a conventional manual transmission, but always a direct connection. Other hybrids may have a conventional or real CVT transmission to the driving wheels in addition to the planetary gear set and this may be defined here.

Note that Combined Hybrids are assigned a launch clutch mode start method of Brake Torquing even though the transmission should be defined as a Manual type.

GEAR RATIOS: Enter the gear ratios for each of the transmission gears. Leave any unused gears blank or zero. Note that these are the individual gear ratios NOT the overall ratios that include the final drive ratio. These are expressed as nn:1 (e.g. 3.45:1, etc.). CarTest will verify that the car will not stall in 1st gear at idle speed and will compute the power curve accordingly (non-custom power curves). This means that if you numerically lower the 1st gear ratio, it may affect the shape of the power curve and, hence, the performance of the car in other gears. With regards to CVT gear ratios, you must enter the range of gear ratios as gears 1 (low gear, high ratio number) and 2 (high gear, low ratio number). CarTest will use 1st gear as the starting ratio but will use the 2nd gear ratio only as a limit. Most CVT's range from about 2.4 to 0.45.

FINAL DRIVE RATIO: Enter the driving axle gear ratio as nn:1. Include any transfer gear ratios in this overall number if you like or enter that separately in the car-specific parameters you can define.

DRIVING WHEELS: Select Rear wheel drive, Front wheel drive, or All wheel drive.

CAR CURB WEIGHT: Enter the curb weight of the car excluding the driver, passengers, or fuel.

% WEIGHT ON FRONT WHEELS: Enter the vehicle weight distribution as the percent of the car's weight over the front wheels to the nearest whole number percent.

WHEELBASE: Enter the wheelbase in the appropriate units.

TIRE SECTION WIDTH: Enter the tire section width in millimeters. Tire specifications are normally given in this example format: P205/50VR-15. The 205 is the tire section width in millimeters. The 50 is the tire profile percent and the 15 is the wheel diameter in inches (see below). Use the tires for the driving wheels if the sizes are different from front to rear.

TIRE CIRCUMFERENCE (opt.): Enter the rolling tire circumference in the appropriate units. This parameter is optional and will be estimated from the tire/wheel parameters if it is omitted. However, a more precise value can be calculated by you from published road test data as follows: Use the 'Engine Revs at 60 mph in top gear' figure in the expression:

Tire Circumference = (5280)(axle ratio)(top gear ratio) / (Engine Revs, 60 mph, top gear)

Alternatively, if the 'mph per 1000 rpm' figure is given, use that number in top gear as follows to obtain the engine revs at 60 mph and then plug that into the above equation.

Engine Revs at 60 mph in top gear = (60000) / (mph per 1000 rpm in top gear)

The  rolling  tire  circumference  may  differ  from  the   measured stationary circumference in that it includes the effect of hot  tire pressures, the flat area at the bottom that forms the contact patch, tire slippage, other deformation effects, etc.

WHEEL DIAMETER: Enter the diameter of the driving wheels (not including the tires.)

TIRE PROFILE: Enter the tire profile as a percent (see example above under Tire Section Width). The tire profile is the ratio of the sidewall (or section) height to the section width (not the tread width).

COEFFICIENT OF DRAG: Enter the coefficient of drag (e.g. 0.32) Although this parameter is not always published with each road test for a car, it may be estimated with a reasonable degree of precision. The easiest way to estimate this value is to use drag coefficients for similarly styled cars. Some important styling features to look for include hidden windshield wipers, front air dams, aerodynamic outside mirrors, general rounded shape, windshield rake angle, etc.

FRONTAL AREA (opt.): Enter the frontal area of the car in the correct units. This is optional and CarTest will estimate it if it is not known more precisely.

OVERALL HEIGHT: Enter the overall height of the car in the appropriate units.

OVERALL WIDTH: Enter the width of the car.

GROUND CLEARANCE: Enter the ground clearance of the car.

Model parameters control the car acceleration simulation and are not necessarily specific to any one car but apply to all cars. Default values are supplied. You may elect to specify and customize a set of model parameters for any specific car you wish.

Default values will be shown in black, custom values in red.

General parameters or the car specific parameters are presented in a tabbed pane behind the car data when requested.

No changes to the data will be permanently saved until you choose to save the data using the menu bar items in the car window.

You can replace all the current values with the supplied Original Parameters by requesting this function from the menu bar in the car window. The Model Parameters are described below.

WEIGHT OF DRIVER: Enter the weight of the driver. This value (and the weight of fuel below) is added to the curb weight to obtain the total vehicle weight. You may also use this field to add any additional arbitrary weight to the vehicle.

WEIGHT OF FUEL: Enter the weight of additional fuel on board. This value (and the weight of the driver above) is added to the curb weight to obtain the total vehicle weight. Gasoline weighs 6 lb/gallon. You may also use this field to add any additional arbitrary weight to the vehicle.

AIR TEMPERATURE: Enter the ambient air temperature for the test conditions.

BAROMETRIC PRESSURE: Enter the barometric pressure for the test conditions. This should be the pressure as reported by the local weather station uncorrected for altitude. At higher altitudes, the atmospheric pressure will be generally reduced.

RELATIVE HUMIDITY: Enter the relative humidity for the test conditions.

ELEVATION: Enter the elevation of the test location.

HEADWIND/TAILWIND: Enter a value for a headwind (positive number) into which the car is driving, or a tailwind (negative number) helping push the car.

ROAD GRADE: Enter the road grade slope in percent. The value is the amount of vertical rise per unit length of road. For example: a 10% grade is a road that rises 10 ft. for every 100 ft. of road length (this is equivalent to an approximate 6 degree slope angle; a 50% grade is a road that rises 50 ft. per 100 ft. of road (about 27 degrees); and a 100% grade rises 100 ft. for every 100 ft. of road (45 degrees). Use negative values to define downhill roads; -10% for a road that drops 10 ft. per 100 ft. of road length, etc.

SHIFT TIME, MANUAL (sec.): Enter the time required for a manual transmission shift not including clutch engagement time after the shift.

SHIFT TIME, AUTO (sec.): Enter the time required for an automatic transmission shift.

ENGAGE TIME, MANUAL (sec.): Enter the time required for clutch dump at start and clutch engagement after a manual transmission shift.

ENGAGE TIME, AUTO (sec.): Enter the time required for an automatic transmission to engage after a gear shift.

CLUTCH SLIP MAX. TIME, MANUAL (sec.): Enter the maximum time allowed for clutch slip starts with a manual transmission. CarTest will determine the optimum clutch slip time with a maximum value bounded by this parameter.

BRAKE RELEASE TIME AT START (sec.): Enter the time required for brake release with an automatic transmission using the brake torque technique at the start.

TRANSMISSION ENGAGE METHOD: Specify the technique used to engage the transmission for both manual and automatic transmissions: a linear engagement over the engagement time; a progressive engagement, a small amount of engagement at first, increasing progressively over the engagement period; a regressive engagement, a large amount of engagement at first, slowly completing the engagement over the engagement period.

SHIFT POINTS: CarTest calculates the Optimum shift points for overall maximum acceleration. To cause CarTest to shift at redline rather than the optimum shift points, specify Redline shift points. Note: An acceleration time for 0-60 mph may be faster when shifting at redline (which may occur just after 60 mph) instead of shifting at a lower, but optimum overall, rpm (which may occur just below 60 mph). To shift yourself, specify Driver shift points. This will allow you to control when gears are shifted during acceleration using a button. To use user-specified shift points, specify Programmable shift points and enter the values in the area after the custom horsepower curve. ("Drag Racing" will use Optimum shift points and "Optimize Launch RPM" will use Redline shift points if Driver Sift points are specified.)

ENGINE FREE DECEL RATE: This specifies how fast the engine decelerates upon lifting the accelerator and pushing in the clutch or during an automatic transmission shift. The number is rpm per second.

ENGINE BOG DOWN DECEL RATE (MANUAL TRANSMISSION): This specifies how fast an engine will decelerate at launch when the engine is turning more rapidly than allowed by the driving wheels and the clutch is fully engaged and may or may not be slipping. The result is an engine that is braked and decelerating. This parameter determines the rate of that deceleration.

ENGINE WHEELSPIN DECEL RATE (MANUAL TRANSMISSION): This specifies how fast an engine will decelerate at launch when the engine and driving wheels are turning more rapidly than is allowed by the available grip between the tires and the road surface with the clutch fully engaged with no clutch slip. The result is an engine that is braked and decelerating. This parameter determines the rate of that deceleration. Note: A negative number may be entered to simulate engine rpm increase during wheelspin launches.

ENGINE CLUTCH SLIP DECEL RATE (MANUAL TRANSMISSION): This specifies how fast an engine will decelerate when the engine is turning more rapidly than allowed by the driving wheels and the clutch is partially engaged during a controlled slip start launch. The result is an engine that is braked and decelerating. This parameter determines the rate of that deceleration. Note: A negative number may be entered to simulate engine rpm increase during controlled clutch slip launches.

CVT/HYBRID ENGINE ACCELERATION RATE: CVT Transmissions and combined hybrid gasoline engines have an infinitely variable gear ratio to the driving wheels and hence may increase in rpm during acceleration uncoupled to the rear wheels. Enter an rpm rate value here at which the engine accelerates under maximum vehicle acceleration.

ENGINE TRANS. SLIP ACCEL RATE (AUTOMATIC TRANSMISSION): This specifies how fast an engine will accelerate when the engine is turning more rapidly than allowed by the driving wheels and the automatic transmission or hydraulic torque converter is slipping during launch. The result is an engine that is being released and accelerating. This parameter determines the rate of that acceleration until the car catches up. Note: A negative number may be entered to simulate engine rpm decrease during brake torque launches.

ENGINE WHEELSPIN ACCEL RATE (AUTOMATIC TRANSMISSION): This specifies how fast an engine will accelerate at launch when the engine and driving wheels are turning more rapidly than is allowed by the available grip between the tires and the road surface with an automatic transmission car launched with the brake torque method. The result is an engine that is released and accelerating. This parameter determines the rate of that acceleration. Note: A negative number may be entered to simulate engine rpm decrease during automatic transmission wheelspin launches.

MAXIMUM COEFFICIENT OF STATIC FRICTION: Enter the maximum coefficient of static friction between the tires and the road surface. The maximum value occurs at zero tire loading. CarTest will compute the working value to account for actual tire loading. This parameter specifies the adhesion of the tire to the road at launch and controls maximum possible acceleration. A value of 1.0 is about the maximum for passenger car tires on ordinary dry roads; the value drops to approximately 0.5 on wet roads. The value for tires on ice can be as low as 0.1.

MAXIMUM COEFFICIENT OF KINETIC FRICTION: Enter the maximum coefficient of kinetic friction between the tires and the road surface. The maximum value occurs at zero tire loading. CarTest will compute the working value to account for actual tire loading. This parameter specifies the power transfer capability of a spinning tire during wheelspin starts. A rule of thumb is that this value will be approximately 75-85% of the coefficient of static friction.

COEFFICIENT OF ROLLING RESISTANCE: Enter a value for the coefficient of rolling resistance assuming rubber tires on dry pavement at low speed.

TIRE EXPANSION FACTOR: Enter the tire expansion factor as a percent circumference growth per unit of speed increase.

HOT TIRE PRESSURE: Enter the hot tire inflation pressure.

TIRE TREAD TO SECTION WIDTH RATIO: Enter the percent relationship between tire section width and actual tread width. The ratio specified is the percent of the tire tread width to the larger tire section width.

WHEEL AND TIRE WEIGHT: Enter the weight of a single wheel/tire assembly. The wheel is assumed to be 15" in diameter with a tire of 200mm section width and a profile of 60% (P200/60VR-15). Adjustments will be made for wheels and tires of different sizes.

MINIMUM ENGINE SPEED: Enter the engine idle speed in rpm. There is an assumed stall if the engine is forced to slow down to this speed.

ZERO H.P. ENGINE SPEED: Enter the engine speed in rpm at which zero horsepower is developed. This will be below idle speed. This parameter affects the low speed power an engine produces, a lower rpm value increases low speed power.

TURBO START SPEED: This is the engine rpm at which the turbocharger first provides significant boost. The engine power curve used in CarTest reflects the low speed characteristics of turbocharged engines. This parameter controls the point at which the turbocharger provides increase in engine power.

MAX. TORQUE CONVERTER SLACK UNTIL: Enter the engine speed at which an automatic transmission torque converter no longer exhibits slack during launch. Engine rpm will climb rapidly, slowly decreasing until this value is reached waiting for the car to catch up to match engine speed and road speed.

LOW MAX. BRAKE TORQUE SPEED: CarTest estimates the maximum engine speed that can be reached just prior to brake torque launches with automatic transmission cars. This parameter places a minimum value on this estimate.

TIME INCREMENT: Enter the time increment to be used in the simulation in seconds. Smaller time increments will increase the simulation accuracy but will run slower, and larger increments will reduce the accuracy but will run faster. A fixed 0.01 second time increment is always used during the critical initial launch period.

START TIME AFTER CAR MOVES: A common road test or drag race procedure is to start recording the acceleration times after the car has 'rolled out' or moved a small distance, say, 1 foot. Enter this value here in feet. Note that this will cause the listed acceleration times to differ slightly from the acceleration curves plotted. This can have a profound effect on short distance acceleration times if deep staging is used by the vehicle enabling a good running start at the launch trial.

ELECTRIC MOTOR DISPLACEMENT: Enter the capacity for the rotational mass of the electric motor for purposes of accounting for the rotational inertia.

EFFECTIVE ELECTRIC MOTOR RADIUS: Enter a value that specifies the radius of the electric motor modeled as a spinning cylinder. This parameter is used to calculate the rotational inertia of the electric motor. A larger value results in increased power required to accelerate (increase the rpm) the motor itself.

EFFECTIVE ENGINE RADIUS: Enter a value that specifies the radius of an engine modeled as a spinning cylinder. This parameter is used to calculate the rotational inertia of an engine. A larger value results in increased power required to accelerate (increase the rpm) the engine itself.

DRIVE SHAFT RADIUS: Enter a value that specifies the radius of a transmission drive shaft. This parameter is used to calculate the drive shaft rotational inertia.

CENTER OF GRAVITY FROM BOTTOM: Enter the location of the center of gravity of the car as a percent distance from the bottom of the car. If the center of gravity lies one fourth of the distance from the bottom of the car to the top of the car, then 25% is the value.

MECHANICAL LOSSES-AUXILIARIES: Enter the engine horsepower losses in percent due to engine auxiliaries that are a part of a assembled automobile but are not included in the horsepower dynamometer tests upon which the published power ratings are based. These can be power steering, air conditioning, additional emission control equipment and restrictive exhaust systems, etc.

MECHANICAL LOSSES - MANUAL TRANSMISSION: Enter the engine horsepower percent loss due to the mechanical efficiency of a manual transmission.

MECHANICAL LOSSES - AUTOMATIC TRANSMISSION: Enter the engine horsepower percent loss due to the mechanical efficiency of an automatic transmission.

MECHANICAL LOSSES, CVT: Enter the engine horsepower percent loss due to the mechanical efficiency of a CVT transmission.

MECHANICAL LOSSES, COMBINED HYBRID DRIVE: Enter the engine horsepower percent loss due to the mechanical efficiency of the combined hybrid drive gearset.

MECHANICAL LOSSES-DIFFERENTIAL: Enter the engine horsepower percent loss due to the mechanical efficiency of a differential. Note: CarTest knows that a four-wheel drive car may have more than one differential.

MECHANICAL LOSSES-AXLES & SHAFTS: Enter the engine horsepower percent loss due to the mechanical efficiency of the various drive axles and shafts. Enter a value applying to a single drive linkage for a front-engine, rear-wheel drive car. CarTest will make its own allowances for cars with other configurations. No allowances are made, however, for transverse mounted engines. Cars with engines mounted transversely can tend to have more efficient drive efficiencies, perhaps as much as 3%. These adjustments must be made by entering a lower loss value here for this parameter.

MECHANICAL LOSSES-TORQUE CONVERTER: Enter the engine percent horsepower loss due to the mechanical efficiency of a torque converter. CarTest assumes a lockup type torque converter that engages in top gear.

MAX. CLUTCH SLIP LOSSES (MANUAL): Enter the maximum power loss due to a fully engaged but slipping clutch during launch for a manual transmission car. CarTest computes a power loss with an upper bound set by this value.

MAX. TRANS SLIP LOSSES (AUTO): Enter the maximum power loss due to a fully engaged but slipping automatic transmission and torque converter during launch. CarTest computes a power loss with an upper bound set by this value.

MAX. TORQUE CONVERTER TORQUE MULTIPLIER: Enter a value for maximum torque multiplication from a torque converter. Maximum torque multiplication typically occurs at launch when the speed difference between the engine and the drive shaft is a maximum.

HYBRID DRIVE TORQUE SPLIT TO DRIVE GEAR: Enter the percent of the gasoline engine torque used to drive the wheels in combined hybrid cars. The remainder is used to drive the generator. The percent is determined from the design of the planetary gear set.

In addition to the general model parameters, car-specific model parameters include additional data. In addition to a set of suggested parameters for automobiles, a set of suggested values is also available to model a motorcycle. Make sure the 'Parameter Use' menu item radio button is selected to use car-specific parameters. If not, the general model parameters will be used and the following will be ignored:

CUSTOM HORSEPOWER CURVE USE: Specify 'None' if not using a custom horsepower curve, 'Engine Power' if entering horsepower data at the engine flywheel. The maximum horsepower and torque from the main car data will also be a part of this curve. Specify 'Driving Wheels' if the power curve is from data measured at the driving wheels. In this case, also enter data for 'maximum driving wheels horsepower' and 'maximum driving wheels torque' and the engine speeds at which they occur as described below. Specify "Max Values Only' if you have only the measured driving wheels maximum horsepower and torque. In that case you must, of course, enter those values and any entered custom power curve data will be ignored. You should still enter the maximum engine power and torque on the car data screen as well. The rpm values entered there will be used for the maximum driving wheels power and torque rpm.

CUSTOM POWER CURVE INTERPOLATION: If you have specified a custom horsepower curve you can select how CarTest will interpolate between the specified data points. You may choose from a Cubic Spline or a Linear Interpolation. The Cubic Spline maximizes the smoothness of the curve but may create artificial peaks and valleys between data points to achieve this if your data contains sudden sharp increases or drops.

MAX DRIVING WHEELS HORSEPOWER: Enter the horsepower value that is the maximum reached at the driving wheels. Enter the rpm at which this occurs in the main car data window in the field for maximum power rpm. The value there for maximum horsepower will not be used but the rpm value will.

MAX DRIVING WHEELS TORQUE: Enter the torque value that is the maximum reached at the driving wheels. Enter the rpm at which this occurs in the main car data window in the field for maximum torque rpm. The value there for maximum torque will not be used but the rpm value will.

HORSEPOWER CURVE VALUES: If you are using car-specific parameters, you can override the idealized engine power curve shape that CarTest prepares with the exact engine power curve if it is known. You may enter a horsepower value for the engine or drive wheels power at 500 rpm intervals beginning at 500 rpm and ending at 20000 rpm. Enter only those values which are specifically known. Leave any unknown values blank or zero. In addition to the above parameters, the engine power curve shape will use the ZERO HORSEPOWER ENGINE SPEED parameter, and the maximum HORSEPOWER and TORQUE values and the RPM at which they occur from the Engine Data describing the car (unless you are using this curve as driving wheels data). These latter values take precedence over any values specified in the 500 rpm interval list occuring at the same rpm. Engine power at engine rpm occuring between the data points are interpolated using a mathematical algorithm known as a cubic spline or a linear interpolation depending on your selection. A spline minimizes the curvature of the smooth line needed to fit the data points on the engine power curve. Because of the characteristics of the spline, you MUST specify a power curve value at engine redline (if it falls on an even 500 rpm multiple) or estimate a value at the next greater 500 rpm multiple. For example, if redline is at 6500 rpm, you need a power value at 6500 rpm, but if redline is at 6700 rpm, you need to estimate a value at 7000 rpm. All data must be SAE corrected.

PROGRAMMABLE SHIFT POINTS: Specify the RPM at which the transmission shifts for the 1-2, 2-3, 3-4, 4-5, 5-6, and 6-7 gear change. Use these values when you specified Programmable shift points. These values are only for car-specific parameters.

PRIMARY RED RATIO: Enter any transmission primary reduction ratio here if you didn't include it as part of the final drive ratio. A 0. means to ignore the field. A 1. is a direct ratio.

DEFINED LATERAL ACCELERATION: Enter a lateral acceleration for use in the track testing function if you feel that the computed value does not accurately reflect the performance of the car.

Individual analysis options include:

In the title bar of each plot following the title will be displayed the x and y coordinates of the location of the mouse as it is placed over the plot. The coordinates are in the units of the data displayed. In this way, the mouse is acting as a digitizer on the plot. If there is no room on the title bar to see all the numbers, you should use the 'resize' menu bar function to enlarge the plot. The real time display of the coordinates is currently function only when using the Metal Look & Feel.

You can select an area of a plot to zoom in on. Use the mouse to place the cursor over one corner of a rectangle of the area of interest. While holding down the left mouse button, drag the mouse to the opposite corner and release the left button. You can restart the rectangle at any time and you can erase a displayed rectangle by clicking the right mouse button. The enclosing rectangle can be viewed as an entire plot by using the 'Crop to Box' menu item under the 'Replot' menu. A new plot will be displayed using the approximate bounds of the rectangle as the region of the plot. Note that a plot is a series of connected data points. The new plot will contain only those data points falling within the bounded rectangle and, hence, the new plot curves will not be flush with the borders of the new plot. You can return to the initial plot by selecting the 'Restore Full Plot' menu item under 'Replot'.

You can also use the 'Points' menu to toggle the display of the actual data points used to construct the line on the plot. If you see a blank plot, chances are the plot consists of a single data point. Select 'Display Data Points' to view this point. You can erase the data point display by selecting 'Hide Data Points'.

On plots that have many data curves, the curve labels may be obscured. You can use the 'Labels' menu item to bring up a separate window that lists the curve labels for the plot. Options are to show the window and then to hide the window after it has been displayed.

You may notice a slight difference between listed numeric acceleration times and plotted values. The difference is because of the 'Start Time After Car Moves' model parameter. Listed numeric values take this parameter into account while the plots do not. This will result in listed numeric values to indicate slightly faster acceleration times than are shown on the plots.

Select Standing Start Acceleration to monitor the progress of a standing start acceleration simulation. Individual windows depict a simulated analog tachometer (with the initial launch rpm indicated by the initial needle position) and speedometer. The orange arc on the tachometer is the maximum torque peak and the green area is the horsepower peak. These windows can be closed at any time to increase the computational speed at which the simulation occurs.

The launch rpm can be increased or decreased by successively pressing the appropriate button. For automatic transmission cars you cannot specify a starting rpm greater than the estimated maximum rpm reachable using the brake torque launch method. For manual transmission cars the launch mode, clutch Dump or Slip can be selected by pressing the appropriate button. The initial vehicle status will show the currently selected mode. Automatic transmission cars use a transmission engagement algorithm simulating brake torque only.

Other parts of the screen will show the acceleration and distance times at user-defineable milestones, the acceleration curves (which will automatically rescale as needed), current time, speed, and distance traveled, estimated top speed, shift points, and the vehicle status. The 1/4 mile milestone speed will be the average speed over a 66 foot interval preceding the 1/4 mile mark. This is to be in conformance with common drag strip practice. The 1 km terminal speed is the instantaneous speed at the crossing of that marker.

The time to speed curve will show the speed of the car versus time and the time to acceleration curve will show the acceleration of the car in G's (one G is the acceleration due to Gravity of a free-falling object.) Both the plot windows can be closed at any time to speed up the simulation. Pressing the launch button will start the simulation. After the simulation has begun, pressing the pause button will halt the simulation; at this point you may press: the resume button, the reset button, and the Exit button. The simulation will halt by itself when vehicle acceleration has stopped and the actual top speed has been reached.

If you specified a driver shift option in the model parameters, you will be shifting gears yourself by pressing the shift button when you desire the simulation to shift gears. The section of the window for shift points will be filled in with the engine rpm and vehicle speed at which you chose to shift gears.

You can choose menu items to cause the simulation to run at no faster than real time and to print the results.

Use the 'Goals' menu item to access the window displaying the tables of 0-Speed and 0-Distance milestones. Use the text fields to specify which milestones you wish analyzed, then use the 'Update' menu item to display the new acceleration milestones in the analysis window. Remember to use the 'File' menu to 'Save' the results if you wish the altered values stored for future use. Speed values may be three digits long, distance values may be four digits in length. You cannot enter digits that cause the displayed lengths to be longer. You must delete a digit first. The last English units distance value is fixed at 1320 feet (1/4 mile) and cannot be changed. That is because of the special manner in which the terminal speed is computed for 1/4 mile runs. The speed and distance values need not be in numerical order. These user-defined goals are the same goals that are used in the Performance Comparison Analysis. So, if you change them here, you also change them for the comparison test runs.

Rolling start acceleration mode is similar to standing start acceleration simulation except that the car is assumed to be traveling initially at 5 mph in first gear prior to maximum acceleration. The initial speed can be increased in 5 mph increments up to 90 mph by the use of the appropriate button. For higher speeds, the initial gear will change automatically for best overall acceleration. The speed can be decremented in 5 mph increments back down to 5 mph by the use of the appropriate button. After selecting the initial speed, you may select the initial gear by pressing ther increase or decrease gear button. You may not select a gear that would cause the engine rpm to fall below idle speed or faster than the engine redline. If you then vary the initial speed after selecting a gear, you may have to go back and reselect the gear you want since the gear may be changed by the initial speed selector.

Both the 1/4 mile and 1 km terminal speeds are the instantaneous speeds at the crossings of those markers.

A rolling start simulation can test the low-speed power of the car and can be more representative of performance during actual driving conditions encountered in everyday traffic.

Use the 'Goals' menu item to access the window displaying the tables of 0-Speed and 0-Distance milestones. Use the text fields to specify which milestones you wish analyzed, then use the 'Update' menu item to display the new acceleration milestones in the analysis window. Remember to use the 'File' menu to 'Save' the results if you wish the altered values stored for future use. Speed values may be three digits long, distance values may be four digits in length. You cannot enter digits that cause the displayed lengths to be longer. You must delete a digit first. The speed and distance values need not be in numerical order. These user-defined goals are the same goals that are used in the Performance Comparison Analysis. So, if you change them here, you also change them for the comparison test runs.

This function determines the combination of launch engine rpm and transmission launch method (clutch dump or slip) that produces the fastest acceleration times (automatic transmission cars use a transmission engagement algorithm simulating brake torque). All combinations of launch techniques are attempted to determine that method which results in the fastest acceleration time. Ranging from 1000 rpm through engine redline, the car is launched using both clutch dump and clutch slip (manual transmission only) launch methods. High rpm launches with automatic transmission cars simulate brake torque starts and this function examines starts over the range of that algorithm only. Brake torque is holding the car with the brake with the transmission selected while increasing the engine rpm, then quickly releasing the brake when the engine reaches the desired rpm. This method takes up the slack in the torque converter and reduces launch delay. CarTest estimates the maximum initial engine rpm that can be reached using this launch method. The launch rpm and transmission engagement method which results in the fastest acceleration time is then passed back to the main analysis screen. The launch rpm and transmission engagement method which results in the fastest acceleration time is then passed back to the car data window.

You may also produce plots of the best launch RPM versus various acceleration criteria, either zero to a velocity or zero to a distance.

When you enter this function, CarTest will optimize for a speed of 60 miles per hour. You may change this criteria by using the buttons.

Since this process can take quite a bit of computer time on slower machines, you may check the curves occasionally to check the progress of the computations. If the curves show that the optimum has already been reached and the trend of the curves as launch engine rpm is increased is toward longer acceleration times, you may elect to terminate the process and send the results back to the main data screen before the search procedure terminates by itself.

Alternately, you may press produce a plot of the best launch RPM versus terminal velocity in time-to-velocity acceleration trials. You may also produce a plot of the best launch RPM versus terminal distance in time-to-distance acceleration trials. For manual transmission cars, two lines will be shown on the plot, one using clutch dump starts and one using clutch slip starts. For automatic transmission cars, one unlabled plot will be shown, using brake torque starts. These plots are useful in that they show the sensitivity (or lack of it) to the selected acceleration criteria in determining the optimum starting RPM. When looking at the resulting plot, watch for horizontal lines indicating no sensitivity to acceleration criteria in determining the optimum starting RPM. This means that it does not matter what criteria you select, the optimum RPM will be the same for all criteria. Also, watch for the horizontal lines along the bottom or top axes of the plot, they may be hard to see. If the plot is not a horizontal line, then the optimum starting RPM will depend on the desired acceleration criteria. Short distances or low terminal velocities may tend to have different RPM optimum starts than long distances or high terminal velocity accelerations.

The data file supplied with CarTest contains the optimum 0-60 mph launch method for each car using either the default or 'car- specific' Model Parameters.

The horsepower curve is a plot of engine horsepower versus engine rpm. The curve shown is either the modeled idealized curve or the spline-fitted or linearly interpolated exact power curve using the car-specific parameters you created. Briefly, power is the rate at which the engine is doing work. Work is a measure of the force produced by the engine acting over a circular distance. One horsepower is the rate of work done when 550 pounds of force is exerted against an object which is moved a distance of one foot in one second.

The torque curve is a plot of engine torque versus engine rpm. The torque curve is derived from the engine horsepower curve above. Briefly, torque is a measure of twisting force applied on the end of a lever arm intended to cause rotation. A force of one pound applied at the end of a lever arm one foot long produces a torque of one lb-ft.

This function simulates a top speed run as typically done on the Bonneville Salt Flats. The course is straight and five miles in length. Cars may be pushed by another vehicle to start them and then they are released. You may input the distance the vehicle is pushed and the velocity at time of release (a minimum of 5 mph is assumed). Also specify the gear selected. Adjustments will be made on initial gear selection to ensure engine operating limits.

The results are shown on a time slip which shows the following:

- Average speed over the distance from mile 2 to mile 2 1/4 - Average speed over the distance from mile 2 to mile 3 - Average speed over the distance from mile 3 to mile 4 - Average speed over the distance from mile 4 to mile 5 - Terminal speed at the end of mile 5

The timing slip also shows the current weather parameters for that run: wind speed, temperature, humidity, and barometric pressure. These values are taken from the general or car-specific model parameters depending on your selection. The resultant density altitude is then computed and displayed in the timing slip window. Engine power and atmosphere corrections are made per the SAE J1349 (2004) standard.

At very high speeds a vehicle may experience unwanted front end lift that needs to be corrected through additional aerodynamic front end downforce. This function can assess the penalties in terms of speed that would result from the addition of a front end inverted wing.

The Bonneville Salt Flats provide low coefficients of friction between the surface and the tires at Bonneville. Values for the Coefficient of Static Friction on the order of 0.6 are typical. The value for the Coefficient of Dynamic Friction can be 0.5. Because of this traction can be a problem and racers look to increase downforce through the use of inverted aerodynamic wings.

This function accepts configuration input data for both rear and front inverted wings. A wing's effectiveness is defined by its Coefficient of Lift and plan-form (projection on a flat surface like its ground shadow at noon) area. As a wing produces downforce it also produces drag so the Coefficient of Drag must be supplied. Rear mounted wings located aft and above the rear axle produce a moment about the rear axle as the vehicle accelerates under power. For a coasting vehicle the torque will be about the center of mass. This moment causes a weight transfer from front to rear and can cause a front end lift instability. To cure this a front mounted wing can compensate. Enter the data for a optional front mounted wing. The location is the aerodynamic center of pressure. If a front end instability is detected during the run an information message will be displayed aborting the run.

The Coefficients of Lift and Drag for a wing are normally experimentally determined in a wind tunnel and the values vary by angle of attack. Further, the values can be for a wing of infinite width. For wings with end caps the Coefficient of Drag must be adjusted according to a formula dependent on wing chord length, aspect ratio, etc. to account for this. Coefficents of Lift for wings are typically in the 0.2 to 2.0 range with a record value of 4.0 being reported. Coefficients of Drag for wings are typically in the .006 to .020 range. There is of course a whole science to this.

Not considered are low speed wing effects, pitching moments or the effects of side wind gusts.

Clicking the Wing Effects Plot button displays the time slip as well as a plot of the:

1. Rear Wing Drag 2. Front Wing Drag 3. Rear Wing Downforce 4. Front Wing Downforce 5. Weight Transfer from Front To Rear due to moments about the axles

The drag or weight transfer, in lbs, is plotted against vehicle road or ground speed. Tiny gaps in the plot occur when the vehicle shifts gears and the speed decreases in that time interval.

This is a plot the curves of modeled horsepower loss versus road speed due to each of the following effects: transmission (accelerating car showing transmission shift effects), tires, aerodynamics, and the total of these three.

This is a plot the curves of drive wheel force versus road speed in each transmission gear. These curves are used to determine optimum acceleration shift points. Total power losses versus road speed is also plotted. The intersection of this loss curve with the force curves gives the vehicles predicted top speed in each gear. No intersection means the vehicle speed is limited by the maximum engine speed. This plot is for constant road speed. Hence, the effect of torque multiplication of automatic transmissions in 1st gear where input shaft speed does not match output shaft speed is not reflected in the curve.

This is a plot a family of curves for theoretical car speed versus engine rpm in each gear. These curves ignore any losses due to friction or aerodynamics.

This analysis estimates the fuel economy of the car during various driving conditions including the EPA City and Highway driving schedules. Curves of estimated fuel economy in miles per gallon versus constant vehicle speed can be plotted. The fuel mileage is shown as a separate curve for each transmission gear.

The fuel economy function estimates what can be expected from the car under various conditions. After you select this function, windows appear showing the computed EPA cycle estimates. CarTest mathematically drives the car along the EPA City and Highway driving cycles and computes the amount of gasoline that will be consumed during the various acceleration, constant speed, deceleration, and idling portions of each cycle. The City and Highway miles per gallon estimates are shown separately along with the Combined value which is the harmonic average of the City and Highway numbers.

Also shown in a window are curves of estimated fuel economy versus constant vehicle speed in each gear. These curves are interesting and illustrate some of the complexities of fuel economy calculations. Fuel economy is dependent on a number of factors, including engine size, engine power, compression ratio, transmission gearing, vehicle weight (during acceleration), power transmission efficiency, drag losses, required throttle opening, engine thermal efficiency, etc. Note that for the lower gears, the full range of the curve represents only a part of the throttle opening range. Redline in 1st gear, for example, can usually be reached without a full throttle opening. Generally, a 3/4 throttle opening produces the greatest fuel efficiency.

Shown in separate windows are the EPA City and Highway driving schedules. These schedules show the vehicle speed versus time for each standard cycle. The City driving schedule is approximately 7.45 miles long and the Highway test is about 10.25 miles long. Note that the EPA does not actually drive the cars along such a route and measure the amount of fuel used. Rather, they simulate the cycles by placing the car on a dynamometer which produces driving loads for the specific vehicle being tested. The fuel mileage is then computed by measuring the chemical composition of the engine combustion products at the tailpipe.

This function estimates the maximum lateral acceleration (in G's) around a circular skidpad and the maximum speed through a slalom course with pylons spaced at 100 foot intervals. It should be emphasized that these numbers are estimates only and are based on expected results for an average car with factors to account for specific tire specifications, weight distribution (horizontally and vertically), and vehicle width. Suspension characteristics are not modeled and a vehicle that has a suspension design oriented toward fine handling will do better than the average. Vehicles with suspensions designed more towards ride comfort will do worse than the average.

In modeling the lateral acceleration of a car traveling around in a circle, the physics reveal that the actual radius of the skidpad does not matter (ignoring steering effects for a very small circle and aerodynamic effects for a very large circle). The only important quantity is the lateral coefficient of friction between the tires and the test surface. In fact, if you ignore the variation of the coefficient due to such factors as uneven loading and weight shifting among the four tires during the skidpad run, the lateral acceleration, A (in G's) is mathematically equal to the coefficient of friction, u, (A=u).

CarTest models changes in the coefficient of friction due to weight shifting from the inside of the circle to the outside (gains in the inside tires do not fully compensate for losses in the outside tires) taking into account the static front/rear weight distribution also. Additional general suspension effects for an average car are also modeled. What is not modeled are variations in the coefficient of friction due to different brands of tires (tread design, softness or hardness of rubber compounds, etc.) or the condition of the tires (shaved tread tires do much better than new tires). Further, finely tuned suspensions designed for handling that are harsher but keep more of the tires level on the road will do better than the average suspension assumed by CarTest. Suspension designs that slightly steer the rear tires or outright four-wheel steer cars will also do better than average. Cars with suspension designs that result in excess wheel camber or non-independent rear suspensions may do worse than average.

Slalom tests are modeled similarly to skidpad trials but under different conditions. Cars that are narrow in width will tend to do better due to the slightly reduced amplitude of the swerving maneuvers required to negotiate the slalom cones.

The accuracy of the CarTest skidpad and slalom estimates should be viewed in statistical terms. You can generally expect CarTest to deliver results in lateral acceleration that fall within the range of +/- .03 G's from published measured results about 68% of the time (one standard deviation). Similarly, slalom speeds will generally be within +/- 1.5 mph of published measured data also about 68% of the time. These results are comparable to the variation of test results from one automotive publication to another (which also vary in test surface and procedure).

Finally, the slalom and skidpad are assumed to always be on level ground, therefore, no grade effects (if specified) are considered.

This function estimates the maximum lateral acceleration (in G's) around a circular skidpad and the maximum speed through a slalom course with pylons spaced at 100 foot intervals. It should be emphasized that these numbers are estimates only and are based on expected results for an average car with factors to account for specific tire specifications, weight distribution (horizontally and vertically), and vehicle width. Suspension characteristics are not modeled and a vehicle that has a suspension design oriented toward fine handling will do better than the average. Vehicles with suspensions designed more towards ride comfort will do worse than the average.

In modeling the lateral acceleration of a car traveling around in a circle, the physics reveal that the actual radius of the skidpad does not matter (ignoring steering effects for a very small circle and aerodynamic effects for a very large circle). The only important quantity is the lateral coefficient of friction between the tires and the test surface. In fact, if you ignore the variation of the coefficient due to such factors as uneven loading and weight shifting among the four tires during the skidpad run, the lateral acceleration, A (in G's) is mathematically equal to the coefficient of friction, u, (A=u).

CarTest models changes in the coefficient of friction due to weight shifting from the inside of the circle to the outside (gains in the inside tires do not fully compensate for losses in the outside tires) taking into account the static front/rear weight distribution also. Additional general suspension effects for an average car are also modeled. What is not modeled are variations in the coefficient of friction due to different brands of tires (tread design, softness or hardness of rubber compounds, etc.) or the condition of the tires (shaved tread tires do much better than new tires). Further, finely tuned suspensions designed for handling that are harsher but keep more of the tires level on the road will do better than the average suspension assumed by CarTest. Suspension designs that slightly steer the rear tires or outright four-wheel steer cars will also do better than average. Cars with suspension designs that result in excess wheel camber or non-independent rear suspensions may do worse than average.

Slalom tests are modeled similarly to skidpad trials but under different conditions. Cars that are narrow in width will tend to do better due to the slightly reduced amplitude of the swerving maneuvers required to negotiate the slalom cones.

The accuracy of the CarTest skidpad and slalom estimates should be viewed in statistical terms. You can generally expect CarTest to deliver results in lateral acceleration that fall within the range of +/- .03 G's from published measured results about 68% of the time (one standard deviation). Similarly, slalom speeds will generally be within +/- 1.5 mph of published measured data also about 68% of the time. These results are comparable to the variation of test results from one automotive publication to another (which also vary in test surface and procedure).

Finally, the slalom and skidpad are assumed to always be on level ground, therefore, no grade effects (if specified) are considered.

This function allows you to vary the value of one car or model parameter and produce a plot of vehicle performance versus the value of that parameter. You must have accessed either the general or car-specific parameters for this car to have them appear in the selection list.

When you enter this function you are shown a window listing the parameters that define the car. If you have accessed either the car- specific or general model parameters they will be added to the list of parameters that you can vary. Use the scrolling list to select a parameter then click on that parameter to highlight it.

You can access other control parameters for this analysis by clicking on the tabbed pane titles to access other panels in this window. The percentage variation panel allows you to select how much to vary the selected parameter by percentage. A starting or suggested value of +/- 20% is first displayed. The minimum lower limit is 10% and the maximum upper limit is 1000%. Note that this is a plus and minus percent variation. If you select +/- 50% of a parameter with a value of 80, the range of analysis will be 40 to 120. If you select +/- 200% of a parameter with a value of 75, the range of analysis would normally be -75 to +225. But if -75 (or anything less than or equal to zero) is a disallowed value for that parameter, the analysis will exclude values of that parameter in the disallowed range.

You may select the acceleration criteria for the analysis. You may press the buttons to increase or decrease this target velocity. If you attempt to select a time-to-speed greater than the maximum value, 'Top Speed' will be displayed as the criteria. This means that values of the selected car or model parameter will be plotted against the resultant vehicle top speed (not an acceleration criteria) at each parameter value. If you desire a rolling start time-to-speed acceleration criteria instead of standing start, press that button. If you prefer a time-to-distance criteria such as 660' or a 1/4 mile time, press the appropriate buttons to define that as the acceleration criteria.

You may specify the number of data analyses. This is the number of data points on the plot that will span the percent variation of the parameter value. Fewer data points will take less time to compute but more data points will provide greater resolution on the graph. For example, if 10 data analyses are selected for a parameter of value 90 that is to be varied +/- 40%, then data points will be computed for parameter values of 54, 62, 70, 78, 86, 94, 102, 110, 118, and 126. If any of these data points were at disallowed values, then no data would be produced for them. For example, if 8 data analyses are selected for a parameter of value 12 that is to be varied +/- 200%, but any values of the parameter less than zero are disallowed, then the data points at -12, -6, and 0 will be skipped but data points at 6, 12, 18, and 24 will be computed. Use the buttons to make the selection in increments of 1. The lower and upper limits are 2 and 250, respectively.

You must return to the 'Select Parameter' panel to access the 'Start' button to commence the analysis.

The menu item for Launch RPM allows you to choose whether you wish to re-optimize the launch method (starting clutch mode and launch rpm) at every step in the analysis. This will slow down the analysis considerably but will produce more meaningful results. The default is not to re-optimize.

The 'Horsepower Finder' function allows you to enter an elapsed time in the text field shown in that tabbed pane. That elapsed time should be for the acceleration criteria you have defined in that tabbed pane. CarTest will vary the engine and/or driving wheel maximum horsepower plus and minus the percentage defined and in increments determined by the number of analyses specified in that pane. Driving wheel power will be used if you have defined car-specific parameters for the car and entered driving wheel power data. Maximum torque will be varied in the same proportion as power. If the percentage variation is insufficient to determine the power and torque required to produce the entered elapsed time you will receive a message to that effect indicating also whether the result lies in the increased or decreased required power direction. Increasing the number of analyses may result in a finer computation of the required power. As with the primary Parameter Sensitivity function, re-optimizing the starting rpm for each data point greatly slows down the speed of the computations.

The Gear Ratio Finder is one of the most interesting features of the Parameter Sensitivity Analysis. CarTest 2000 will determine a set of optimum gear ratios such that the minimum time will be required to conduct the selected acceleration trial. The solution to this problem requires traversing the multi-dimensional hypersurface defined by each gear ratio as an axis, and the acceleration time as the function value. For a car with a six-speed transmission plus a final drive, the surface has seven dimensions. To completely analyze this surface would require billions of years of computer time. The algorithms used here require a starting point and then traverse the surface following downward paths until a deep well is encountered trapping the search. This deep well may, or may not, be the absolute minimum time. Included refinements to the algorithms attempt to crawl out of these wells in search of deeper wells. This is done by re-starting the searches at various points in hopes that the best descent from a different place will produce a better result. This continues until no better result can be achieved. Hence, the running time of this function is quite long. Your computer speed as well as the difficulty of the problem posed will determine this. If you have selected the mnu item to re-optimize the starting rpm at each data point, the required running time will be that much longer (perhaps by a factor of 10).

The Gear Ratio Finder screen has a number of options accessed by the use of the radio buttons. The top line lets you select the initial starting location for the search. The first is the existing gear ratios for the car. Next is a set of ratios estimated by CarTest using a set of synthesized drive power curves. The CarTest estimate produces an optimum ratio for top speed, determining that top gear ratio that produces the top speed of the car at that engine rpm corresponding to maximum horsepower. The other gear ratios are determined backwards from there attempting to determine the best drive power curve crossover points for each of the gears with consideration for various other criteria as well. The third option is to use the gear ratios listed in the car data fields. These are initially the existing gear ratios for the car but you can use these fields for manual data entry or intermediate storage (see below). These will not be permanently saved unless you deliberately do so using the 'File' functions on the car data menu bar.

The second line on the Gear Ratio Finder screen lets you select which algorithm is used. The first is a simple 'steepest descent' algorithm, not fast or elegant but quite robust and finds very good answers. The second is a 'conjugate gradient' algorithm which is a bit faster and more elegant, sometimes overlooks better downwards paths, but sometimes finds paths the steepest descent method bypasses. The third algorithm is a 'variable metric' technique which employs different criteria for path selection. Three algorithms are made available because sometimes one will find an answer the others won't. If you have a lot of time available or if the answer is critical to you, you should try all three. Further you should use all three with different initial starting points. You can also use the result of one search as the starting point for another search with a different algorithm. Use the 'Send Results' button activated when the search is paused or ends normally to send the current 'optimum' results to the data fields in the car data screen. From there you can use them as the starting point for another search by selecting the third top-line radio button.

There are also radio buttons for fixing the value of the ratio for each gear at the starting value. This may be useful, for example, if you want to keep a certain final drive ratio but select new transmission gears. You may also be able to click these radio buttons as the analysis is running to fix them at their current 'best' value.

You should note that the 'optimized' gear ratios are optimized for the particular acceleration trial you have selected. For example, if you optimize for the 1/4 mile, they will produce the best 1/4 mile acceleration time but nothing beyond that. Cruising speed rpm, top speed, economy, relaxed highway driving, etc. are not considerations at all. There is much more that could be said about this function but that would fill a book. The thing to do is to play around with the function to get a feel for what it does and does not do and proceed from there. You can see what the ratios do by sending them to the data fields and then running an acceleration.

Interpreting the results of these analyses can be difficult. When only one parameter is varied at a time, many unexpected results can occur. For example, as engine power is strictly increased you would expect acceleration times to strictly decrease. This, however, does not always occur. If you are using 0-60 mph as the acceleration criteria, increasing the engine power can result it unnecessary additional wheelspin off the line, which increases acceleration times. You can mitigate this effect by verifying that the menu item under the 'Launch RPM' menu entitled 'Recompute Optimum Launch RPM Each Data Point' is selected. This will cause the optimum launch rpm to be recomputed for each data point. The analysis will run more slowly if this is selected. The alternate menu item entitled 'Use Fixed Launch RPM' will cause the same launch rpm listed with the vehicle data to be used for each data point. Neither of these menu items is applicable if you are specifying a rolling start as the acceleration criteria.

Further, as power is increased, the shape of the power curve may change, resulting in an optimum gear shift to occur at say, 59 mph, where at lower power the optimum shift may have occurred at 61 mph. The change in shift points will show as a sudden rise in the curve at the power level that caused the change. If you then repeat the analysis using 5-50 mph as the criteria, the results may look quite different since you have removed the anomalous effects of launch method and gear shifting. Remember that the optimized launch method (starting clutch method and launch rpm) does not hold for every data point on the curve, only at that data point representing the original car data values specified. This is true unless you selected the menu item for re-optimizing that launch method at each step in the analysis. That said, you may then prefer to perform your analyses using the rolling start option rather than standing start. The shift point variation problem will still occur, however, unless you choose the ending speed of the analysis carefully.

About engine power variation, this function is designed so that if you choose horsepower as the varied parameter, engine maximum torque will be varied proportionally as well. If you have a custom power curve, each of the values will also be varied accordingly. However, if you choose engine maximum torque as the varied parameter, the horsepower will remain fixed (as well as any custom power curve values) and only the engine maximum torque will be varied. This may result in torque values that result in impossible power curves. These values will be identified and discarded in the analysis and not plotted. If you have a custom power curve specified, varying the engine maximum torque may not produce meaningful results. Also, if the engine is turbocharged, engine power variations will affect the low engine speed shape of the power curve thus injecting another element of complexity in understanding the results.

This brings up another concept worth noting. This parameter variation function allows you to specify very wide ranges for parameter variation. At the extremes, and when only one parameter is varied at a time, the computed results may stretch the bounds of validity of the computer models used to simulate vehicular behavior. Although parameter bounds checking is performed to eliminate the obvious violations, many parameters are allowed to vary beyond practical engineering boundaries. For example, wheel diameter may be increased or decreased way beyond where any vehicle would reasonably be built, and while general trends may be seen, acceleration results for very small or very large wheels may not be meaningful. In general, the results for small variations around the specified design parameters will tend to be more significant than those farther out.

As another example of the complexity in interpreting results, when vehicle test weight is varied and top speed is the analysis criteria, those physicists among you would expect to see a horizontal line on the plot. While vehicle weight will determine how long it takes to reach top speed, it should have nothing to do with what that top speed actually is. However, the plot will show a slight decrease in top speed with increasing weight. As it turns out, increased vehicle weight with no corresponding increase in tire pressure to support that weight results in greater tire deformation thus increasing the rolling resistance which results in decreased top speed.

It is interesting to vary the coefficient of drag to see how top speed is affected. The resulting plot shows a horizontal line indicating the region of low drag where top speed is engine rpm limited and drag has no effect, then drops down as drag increases indicating drag limited top speed.

Varying the final drive or top gear ratio to investigate the effect on top speed may reveal a bumpy plot illustrating the result that top speed may occur in top gear at some ratios and in the next-to- top gear at other ratios. On some cars, this plot will reveal the optimum top gear ratio for top speed.

Acceleration times tend to increase (worsen) as the vehicle wheelbase is increased on rear-wheel drive cars. A longer wheelbase reduces the amount of weight transfer to the rear drive wheels. This weight transfer is, in general, desirable since it produces greater traction at the drive wheels. Top fuel dragsters have long wheelbases to prevent front-wheel lift at the small expense of already abundant rear-wheel traction.

Watch out for the effects of the two optional car data parameters, tire circumference and frontal area. If tire circumference is known and specified then it will not be recalculated when the wheel diameter, tire width, or tire profile are varied. If you wish to vary any of these three parameters, you should first delete out the tire circumference and allow CarTest to compute it from these three parameters to get a better picture of the effects of varying that parameter. The same goes for the frontal area which, if not known and specified, is estimated from the vehicle height, width, ground clearance, and tire width. Again, if any of these four parameters are to be varied, you should first delete out the frontal area and allow CarTest to compute it as the parameters are varied. Of course, if the tire circumference itself is known and specified, that parameter may be varied directly. The same is true for the frontal area also.

Car parameters that specify the wheels and tires will tend to result in a slower accelerating vehicle whenever the wheel and tire radius is increased. This includes the tire circumference and, if that is not specified, the wheel diameter, tire profile and section width, and the weight distribution. Model parameters that affect wheel and tire radius include the tire expansion factor, hot tire pressure, and tread to section width ratio. Increased wheel and tire radius reduces the net pounds of force that accelerate the car at the contact area between the tires and the road.

Note that if a parameter has an initial value of zero, as elevation or road grade may, then the percent variation specification will apply to the maximum range the variable is allowed to span. About road grade variations, remember that as the grade gets too steep, the optimum standing start launch parameters for level ground may cause the car to stall and no data points will appear for those grades. A rolling start may provide a better view of the effects of grades.

One more thing to watch out for are the values along the vertical (y-axis) of the plot. If they appear to all read the same value then the actual result is a horizontal line at the y-value displayed (no change in performance with variation of the chosen parameter.) Numbers all appearing to have the same value actually differ in the third or greater decimal point which is not shown. Any deviations on such a plot are very small and may, in fact, be meaningful but sometimes can also be attributed to numerical approximation and computational precision variations.

This menu contains functions to conduct various performance comparison tests on as many cars as you selected. You may run on on- screen drag race, compare standing start acceleration curves, view the engine flexibilty of various cars as measured by rolling start acceleration in selected gears and at selected initial speeds, display and/or print a comparison table of acceleration results for the selected cars, or run the cars around one of the world's race tracks to compare lap times. You may also rank all of the cars in the database according to various acceleration performance criteria, top speed, fuel mileage, skidpad acceleration, slalom speed, or race track lap times.

CarTest allows you to select as many cars as desired for a standing start drag race. This feature allows you to compare the performance of the selected cars by viewing them as they race side-by-side. You can see which cars are faster, and by how much, in a single trial.

Press the 'Launch' button on the comparison test window to proceed with the drag race. Drag Race always uses the car-specific parameters for each car if they exist. However, to conduct a fair race, rollout distance and environmental parameters from the general model parameters (elevation, temperature, pressure, humidity, road grade, and headwind) will override.

Prior to pressing the 'Launch' button, you may adjust the time increment of analysis by use of the tine increment buttons. Note that a fixed time increment of .01 seconds is used until all cars have completed the computationally sensitive launch phase. A shorter analysis increment will increase the accuracy of the simulation but at the expense of taking a longer time to run.

For the 1/4 mile race, the terminal speed displayed will be the average speed over a 66 foot distance preceding the line. This is to be in conformance with common drag racing practice. For the 1 km race, the terminal speed is the instantaneous speed at the end of the run.

This analysis shows you plots of vehicle speed, distance, and acceleration (in G's) versus time, and speed versus distance, as it accelerates from a standing start. As many curves are shown on the graph as cars selected. A steeper time-to-speed curve or higher G's means a faster accelerating car. The interruptions in the curves, shown as nearly horizontal lines are the vehicle gear shifts during which no acceleration occurs (actually slight deceleration). The acceleration is conducted until top speed is reached. If you wish a shorter plot you may Pause the run at any time. Note that the beginning of the time-to-speed curves does not occur at zero speed. This is due to the use of a rollout distance which is nominally one foot (see Model Parameters). The time clock does not start until the car has 'rolled out' this distance. The initial speed at zero time is the vehicle speed after it has travelled this distance.

Engine flexibility curves illustrate the ability of the car to accelerate from a selected speed in a selected gear. The acceleration proceeds from the speed selected all the way to that speed at which the engine is at redline or for an elapsed time of 60 seconds, whichever occurs first. This analysis shows you the ability of the car to respond in everyday driving conditions. The ability to accelerate from a low speed in a middle gear may be more important that a flat out acceleration run during which the engine speed stays at high rpm for the entire run. These curves can also show the passing ability of the car during highway driving conditions. Steeper time-to speed curves or higher G's indicate better accelerating cars and longer curves indicate greater range in that gear before shifting is required.

SELECT GEAR:

Select the transmission gear for the Engine Flexibility Analysis using the appropriate button. If you select Top Gear, the highest gear will be selected for each of the cars regardless of the number of gears in each car's transmission. For example, you may then be comparing 5th gear acceleration for one car with 4th or 6th gear for another. Further, if you select 5th gear for analysis and one of the cars has a four-speed transmission, the highest gear will be chosen for that car, namely 4th, even though the plot will be entitled 5th Gear Engine Flexibility.

Press the 'Acceleration Plot' button to produce the graphs. These plots do not use the initial speed information specified in the other window panel.

Press the 'Rpm/Speed Plot' button to produce a plot of Engine RPM vs. Road Speed comparing each selected car in the selected gear. The initial speed information specified in the other window panel is used here.

SELECT INITIAL SPEED:

Select the starting rolling speed for the Engine Flexibility Analysis using the buttons. A suggested speed is chosen initially depending upon the selected transmission gear. Note that if you select a speed which will cause the engine to stall or is beyond the top speed of the vehicle, you will not see any data on the resulting plot.

Press the 'Time Plots' button to produce the time-based graphs.

SELECT SPEED RANGE:

Select a start speed and an end speed using the buttons. This analysis will produce plots of rolling start acceleration from the start speed to the end speed beginning in the selected gear. Unlike the other analyses here the car will shift through the gears as required to reach the end speed. Shifting will occur at the rpm method defined for each car.

Press the 'Time Plots with Shifting' button to produce the three time-based graphs.

This analysis produces a comparison table of results for user-defined standing start and rolling start acceleration trials for the cars selected. Engine flexibility rolling start trials are run as well, starting in 1st gear with shifting, and in 2nd through 7th gears with either shifting or no shifting as determined by a user-defined option in the Goals window.

Use the 'Goals' menu item to access the window displaying the tables of 0-Speed, 0-Distance, and Gear Flexibility milestones. Use the text fields to specify which milestones you wish analyzed, then use the 'Recompute' menu item to display the new acceleration comparison values. Remember to use the 'File' menu to 'Save' the results if you wish the altered values stored for future use. Speed values may be three digits long, distance values may be four digits in length. You cannot enter digits that cause the displayed lengths to be longer. You must delete a digit first. The last English units distance value is fixed at 1320 feet (1/4 mile) and cannot be changed. That is because of the special manner in which the terminal speed is computed for 1/4 mile runs. The speed and distance values need not be in numerical order. These user-defined goals are the same goals that are used in the individual car analysis. So, if you change them here, you also change them for the single car runs.

Use the radio button in the gear flexibility pane to set the option for whether or not shifting will be allowed in the three gear flexibility trials. With shifting off, the speed trial will not be possible in all gears with redline rpm being the factor that determines the limit. With shifting allowed, the labeled gear is the starting gear. Shifting is done at the rpm method defined for each car.

The second tabbed pane in this frame displays a comparison table of the specifications for each of the cars selected.

CarTest allows you to select cars for a race around a race track of your selection. This feature allows you to compare the performance of the selected cars considering not only their acceleration potential but also considering the lateral acceleration around curves as well. You can see which cars are faster around the track, and by how much, in a single trial.

The results window shows you the results of running the cars around the selected track. Note that these lap times are for the second lap around the track when the car passes the starting line at full speed. For the first lap, the cars are starting out stationary. The second lap, and all subsequent laps, produce more representative race condition results.

This function computes the performance of all the selected cars and then produces a sorted list of the results. Cars may be listed in alphabetical order or ranked by time-to-distance, time-to-speed, top speed, EPA combined fuel mileage, EPA urban cycle mileage, EPA highway cycle mileage, skidpad lateral acceleration, slalom speed, and race track lap times. Use the buttons to determine the criteria for ranking.

The 'Track Test' menu shows the list of 50 of the world's most famous racetracks. Select the track you wish to run a selected car around. You can do this by selecting a track with one click then pressing the 'Track' button or you can double click on the track name.

CarTest takes the first two cars you have selected in alphabetical order for this analysis. The first car will be the test car and second car will be the pace car. You may then swap the pace car for the test car using the menu bar options provided on the track window.

This track window shows the shape of the selected track along with its name and length. The tachometer, speedometer, and gear indicator for the selected car will be shown and engine rpm, vehicle velocity, and the selected gear will be updated as the car is driven by CarTest around the track in the direction the track is normally used. A small image of a race car on the track will update the car's position. Note that this car as well as the track width shown is not to scale. The acceleration 'friction circle' shows the current and historical forces on the car, due either to forward acceleration, braking and downshifting, and left and right cornering forces. The circle size is a radius of 1-G (1-G is normal acceleration due to gravity).

Lap times are shown below the track. Note that the first lap will be for the car starting at zero speed. Subsequent laps will be timed with the car starting at full speed.

Note that CarTest is not trying to drive each car optimally for each track. The shape of the track is being followed. No attempt is being made to find the best driving line or the best braking points around the track. It is believed that CarTest presents a reasonably good simulation of what each car can achieve on each track, not the best possible times. In fact, the tracks are defined approximately with no information known about climbing or descending track segments or even bank angles on turns. Nevertheless, this simulation presents a good evaluation of how one car might fare against another on that race track. Also, race tracks are constantly being modified to keep up with race car technology, safety regulations, and to make them more interesting and challenging. The track shapes included here may not reflect the latest modifications made to each track.

This function determines, for all the selected cars, the combination of launch engine rpm and transmission launch method (clutch dump or slip) that produces the fastest acceleration times (automatic transmission cars use a special transmission engagement algorithm simulating brake torque).

All combinations of launch techniques are attempted to determine that method which results in the fastest acceleration time. Ranging from idle rpm for automatics and 1000 rpm for manuals through engine redline or maximum rpm for brake torque, the car is launched using both clutch dump and clutch slip (manual transmission only) launch methods. The launch rpm and transmission engagement method which results in the fastest acceleration time is then stored in the data file. Use the buttons to determine the optimization criteria. The results are stored in the database for each car.