One of the keys to any engine development program, from Winston Cup to the Saturday- night racer, is the ability to effectively evaluate changes to the engine. The goal of any engine development program is an improvement in on-track performance, but track testing engine changes can be difficult. The dynamometer is a much more effective way of evaluating engine changes, since non-engine variables can be eliminated from this type test.
The dynamometer allows instrumentation to be used that would not be practical to include in a vehicle test. The ability to measure key performance characteristics such as fuel flow, airflow, and blow-by, is key to increasing understanding of the engine combination. The dynamometer also allows you to analyze the shape of the torque and power curves, something a stopwatch and the racetrack cannot do.
In a Winston Cup engine shop, the dynamometer is not only a development tool but also the final quality control check before the engine is shipped. Every engine that leaves a Winston Cup engine shop is run on a dyno prior to shipment. Therefore, many shops also run the vehicle on a chassis dynamometer to evaluate the complete powertrain and the installation of the engine in the vehicle. Dynamometer testing becomes one of the cornerstones of a Cup engine program.
Types of Engine Dynamometers There are many different types of engine dynamometers, but the two most common types are hydraulic and electric.
Hydraulic Dynamometers--The hydraulic dyno or "water brake" converts energy from the crankshaft into heat. A device similar to a torque converter (fluid coupling) is attached to the crankshaft. Water is fed into the fluid coupling and the shearing of the water between the rotor and stator provides a braking effect to the crankshaft while heating the water. The braking effect of the engine is controlled by the water level in the fluid coupling. A hydraulic dynamometer can only absorb power. It cannot power the engine.
The water brake is the most cost-effective and common engine dyno used in race engine shops today. The advantages of the water brake include simplicity, cost effectiveness, ease of control, and low inertia. Electric Dynamometers--There are three types of electric engine dynamometers commonly used: DC (direct current), eddy current, and AC (alternating current).
The DC dynamometer is simply a big DC electric motor or generator. The energy from the crankshaft is used to generate electricity that must be dissipated by a load bank or converted to AC and supplied to the power grid.
DC dynamometers are not commonly used in racing engine development since they are generally configured for high torque applications at low shaft speeds. In this form, they have more inertia than other types of dynos and do not have the ability to turn high speeds. DC dynamometers can both absorb torque and power the engine. Eddy current dynamometers transfer the energy from the crankshaft into heat that must be dissipated by a cooling system. Eddy current dynos can run at high speed and can be very accurate but are generally not used in racing, due to their cost, complexity, and higher inertia. Eddy current dynos can only absorb, they cannot power the engine.
AC dynamometers are currently the state of the art. An AC dyno is simply a very large AC motor. AC dynos are used extensively in F1, CART, IRL, and a few Winston Cup engine shops. AC dynamometers can absorb and motor the engine, have low inertia, and have extremely fast response. AC dynamometers can truly simulate any on-track condition, including shifting. These dynamometers, when set for high dynamic simulation, are the most capable engine test facilities, but are extremely expensive, even by Winston Cup standards. There are only a handful of AC dyno facilities in the United States at this time that is configured for motorsports use. An AC test cell configured for motorsports will allow the engine builder to run simulated race miles in the engine shop at any time of the year and have the confidence that the results will correlate very well to on-track mileage.
The Value of Inertia dynamometer data
For all intent and purpose, this is about as close as you'll get to a racetrack without being there. While the "inertial" resistance of an engine and powertrain experience on an inertia dyno doesn't provide positively and negatively varying load conditions (as does a racetrack), allowing an engine to accelerate against a changing resistance comes close. Unlike so-called "acceleration" tests involving the controlled unloading of a power absorption unit (PAU), a "wheel" dyno's resistance can more closely approximate on-track loads, leading to a better environment for tuning a given engine combination for race performance.
In fact, the Saturday-night racer can approximate "wheel" dyno analysis by performing some on-track timing steps. For example, if you're either doing practice laps or tuning laps, select a point where straightaway acceleration begins and another where the throttle is consistently lifted. In a sense, this becomes your "dragstrip" or track section over which elapsed times can be measured and compared following each engine tuning change. While this approach does not compensate for weather conditions or chassis influence, it does provide an on-track comparison of changes in power level. Truth be known, this approach is not altogether that much different from performing tuning on a "wheel" dyno.
According to Dennis Wells of Wells Racing Engines, "Before I installed my wheel dyno, segmenting the track and recording elapsed times was a pretty good way of evaluating (engine) tuning changes. Even though the environment is less controlled, the approach is actually more helpful than complete lap times. Cornering conditions, where straightaway acceleration is more representative of rear-wheel power, tend to influence lap times. If you're only noting lap times and seeing changes, you won't know where these are occurring unless segmented times are being recorded."
In particular, inertia dynos enable evaluation of parts (gear ratios, wheel and tire size, clutch packages, crankshaft dampers, and related inertia-affecting components) that influence the overall resistance of a given powertrain to be accelerated under power. Of course, testing of power-producing engine parts can be easily measured on a wheel dyno, often revealing performance potential (or problems) that would be difficult or expensive to identify on the track.
The Goal The purpose of using a dynamometer is to get accurate and repeatable data on the engine being tested. If a dynamometer is not capable of returning consistently repeatable test results day in and day out, it's not effective. The level of competition in all forms of motorsports increases every year. Engine builders are challenged to continually find more and more power. In the past, gains of 5 hp were common. Today, the level of engine development is so high the only way you will find 5 hp is to add together five gains of 1 hp each. To do this, your dynamometer must be able to detect 1 hp consistently. Accurate and repeatable testing is the result of carefully controlling the variables that affect engine performance so that only the planned changes to the engine affect the torque and power.
Dynos do not have to be complicated or expensive to be a great tool. The dyno operator is often the key to the usefulness of the equipment. The aspects of the hardware that can affect the accuracy and the repeatability of the dyno are:
1. The type of testing 2. The control systems and data acquisition system 3. The combustion air supply 4. The oil and water temperature control system
Types of Tests The two most common type of tests used for engine development are the step test and the "acceleration" or sweep test.
For a step test, the engine is operated at constant engine speed until it is stable and then the data is taken. The engine speed is then changed to the next set point. The advantage of the step test is that with no acceleration, the inertia of the engine and the dynamometer do not impact the torque reading, and therefore the test can be more consistent. The step test requires the engine to be held at one rpm for a short period of time, but this is not a condition that engines see in operation and can impact the way the engine is tuned. (Note: Herein lies one of the fundamental reasons why optimal dyno tuning and optimal track tuning may differ.)
The sweep or acceleration test accelerates the engine at a constant rate: for example, 100 rpm/second or 300 rpm/second. (Sweep tests are typically the result of a controlled rate of unloading the power absorption unit.) Sweep tests simulate track conditions more closely than step tests, but control of the acceleration of the engine will impact the torque readings. It is extremely important that the rate of acceleration is consistent during the test and from test to test. Therefore, the dynamometer control system is a factor in producing consistency among sweep tests.
Control Systems and Data Acquisition System
Control System--The control of the dynamometer itself is one of the keys to getting accurate and repeatable data. Expensive computer controls are not required to make a dyno accurate and repeatable, but they can make the dyno easier to operate. Manual dynos require a highly skilled and experienced operator to be effective. The reliability of the dyno is impacted by factors such as the speed the engine is held prior to the sweep, how long it is held at that rpm, and how smoothly. A skilled operator running a manual dyno can repeat these conditions over and over again. The task is similar in requirements to driving--something that not everyone can master. A computer-controlled system, when properly tuned, can make this job easier. The operator can concentrate on controlling the other systems.
Data Acquisition--Most dynamometers today are equipped with a data acquisition system. This allows the operator to concentrate on running the dyno, not on reading gauges. Cost-effective sensors are available to measure just about any imaginable engine performance parameter. Data systems also allow a graphical presentation of the data that makes it easier to see trends rather than just looking at a tabular presentation of numbers.
Flexible data systems that permit the addition of any type of sensor allow dyno operators to gain greater understanding of what is going on in the engine. Some of the sensors that are key:
* Wide Band Air/Fuel Ratio The latest wide band sensors are both cost-effective and accurate. These sensors are the most accurate way to monitor air/fuel ratio and tune the engine.
* Fuel Flow Fuel flow meters are included with most dynos. Monitoring fuel flow from engine build to engine build is a good way of tracing the efficiency of the engine. Having one fuel meter for each float bowl is a good way to gain understanding of how each end of the carburetor is functioning.
* Blow-by Cost-effective blow-by meters are available and are a good tool for monitoring ring seal. With a dry-sump engine, the sensor should be connected to the vent on the tank. Remember, any crankcase leakage will show as an increase in blow-by.
* Combustion Air Supply (CAS) The air supplied to the carburetor is one of the keys to test-to-test repeatability. While the correction factor compensates for changes in atmospheric conditions, it can only do so on an average basis. Instantaneous changes are difficult to correct. It is best to have a large reservoir of stable air that the carburetor can draw from. For example, provide a duct from the largest room in your building directly to the carburetor. It would be best if the air were being drawn from a heated/air-conditioned portion of the building. It is important that the air in the room is stable from a temperature and humidity standpoint. If garage doors are opening during the dyno runs, the CAS can be severely impacted.
A fan that is capable of delivering significantly more air than the engine can use should supply the carburetor. For a 350-cid engine that makes less than 700 bhp (brake horsepower), a delivered pressure of 1,000 to 1,200 cfm should be sufficient. The air supply should be sealed to the carburetor, with a pressure relief door cut into the duct. Weights should be added to this door so that at maximum engine speed the carburetor is supplied with air at approximately 5 inches of water above ambient.
* Oil and Water Cooling It is important that the oil and water temperatures are constant throughout the test. Both the oil and water should be held to 3 degrees F from their set points.
Reading Dyno Sheets Many people who look at a dyno sheet look only at peak power. While peak power is clearly an important number, more important is the average power over the useful rpm range of the engine. The useful rpm range is the minimum rpm that the engine will see at wide-open throttle to the maximum engine speed that the engine will see.
Correction Factor The correction factor is used to compensate for how air density changes affect engine power. Brake horsepower (bhp) is the power the engine delivers from the crankshaft. Frictional horsepower (fhp) is the power taken by all of the friction in the engine. Air density changes do not affect the friction in the engine. The correction factor estimates the internal friction in the engine and corrects only the non-frictional component of the engine's output.
Pressure changes have the largest impact on the correction factor, but luckily the barometric pressure does not normally change quickly. Temperature has the next most significant impact, and temperature can change so rapidly that the correction factor will be ineffective.
Brake Specific Fuel Consumption Brake specific fuel consumption (b.s.f.c.) is observed engine power divided by fuel flow. This is a measure of fuel consumption that normalizes for power. These numbers should not be used for mixture tuning unless all factors affecting engine power are understood.
Consider this example: An engine is run on a dynamometer and achieves a 0.500 b.s.f.c. number. The mechanical water pump is removed from the engine and replaced by an electric one. When the engine is run, the power will increase and b.s.f.c. will decrease because of the improvement in frictional losses, but fuel flow will remain the same. Obviously, in this example, the mixture did not change, but the b.s.f.c. appears leaner.
Air/Fuel Ratio Engine tuners usually try to tune their engines for best power at an air/fuel ratio of 12.5:1 to 13.0:1. This is a comfortable average air/fuel ratio for a gasoline engine. It is common for carbureted engines to have a 1.5 air/fuel ratio difference from the leanest cylinder to the richest. Therefore, unless individual air/fuel ratios are available for each cylinder, it is best not to run any leaner than an average of 13.0:1.
Taking dyno information to the track Often heard among dyno operators is the phrase, "We aren't racing dynos." While this is true, both engine and chassis dynamometers have their place in optimizing on-track race car performance. Unraveling the piles of data each test can produce becomes the key to effective use of the information.
Recall some basics--In earlier parts of this series, a distinction was made between torque and horsepower. Notably, torque is an independent variable; horsepower is a dependent variable and computed by using torque values. Torque accelerates a race car, horsepower makes it fast.
We also spoke about brake specific fuel consumption (b.s.f.c.). To a large degree, this is a measure of how efficiently an engine is converting fuel into power (pounds of fuel per horsepower-hour). Generally, as b.s.f.c. numbers are lowered throughout the rpm range and without an attending loss of power, throttle response sharpens and acceleration times are typically quicker. Both ignition spark and fuel delivery can affect b.s.f.c. performance, as can mixture quality, contaminated combustion and parasitic losses on an engine.
When performing dynamometer tests to evaluate engine modifications, work toward minimizing b.s.f.c. numbers. Since peak torque and minimum b.s.f.c. typically occur at or near the same engine speed, here's a possible tuning protocol.
Baseline the engine through one full range of rpm, minimum to maximum. Select the peak torque rpm point. Adjust ignition spark and fuel to minimize b.s.f.c. at this without an attending power loss. Re-run the full curve. Deviations in b.s.f.c. values (usually seen as decreases) above and below peak torque b.s.f.c. can normally be attributed to a loss in combustion efficiency or improper air supply.
Absent any power losses, the lower the b.s.f.c., the sharper the throttle response. This method enables engine tuning on the dyno while minimizing the number of full runs or "pulls" required, thereby reducing wear and tear on engine parts while maintaining reasonable test-to-test repeatability.
Applying engine dynamometer torque and power data--Convert numerical data into graphs. By studying plotted information, rather than scanning columns of numbers, it is easier to evaluate an engine's power "picture." One reason is the value derived from looking at a "footprint" instead of making numerical comparisons. This leads us to the importance of how the shape of torque curves can be as important as peak values, and here's why: Suppose you are comparing two parts combinations (or engines), each of which delivers relatively the same peak torque and at comparable engine speeds (see illustration). In the graphic shown, note that one curve displays more "area under the curve" than the other, although both produce about the same peak torque and rpm points. All else being equal, the greater the torque curve area, the higher the probability for quicker vehicle acceleration. Plotting this type data provides a quick and graphic representation of potential on-track performance.
Additionally, engines tend to accelerate more quickly when being operated on the increasing-torque side of the curve rather than the decreasing-torque side. Attention to this helps in selection of gear/wheel/tire sizes and the most beneficial range of engine speed.
The message here is that peak horsepower and torque values paint only a partial picture of what can be expected of on-track performance. The rate of power development (the slope of the torque curve) is more descriptive, thereby underscoring the benefits derived from plotting data and studying the results. In other words, the steeper and broader a given torque curve, the greater the potential for race car acceleration.
Some concluding thoughts on this series Stated at its beginning, this series was intended to identify and explain areas of power development common between Cup and Saturday-night engine builders. While comprehensive to a point, space within its planned length prevented certain areas to be included and discussed. Both authors will visit these areas if Circle Track readers want that to happen. Your letters will dictate a decision to do so. Simply identify areas for which you want additional information. Send those letters to: Circle Track, 3816 Industry Blvd., Lakeland, FL 33811. CT