              |
 Three primary roadway elements cause mountain grades to be hazardous. These hazardous elements include grade steepness, grade length, and curve severity. In their extremes, the combination of any two of these elements on one grade creates a difficult situation for a truck. The combination of all three of these elements on one grade can be extremely hazardous.  Truck drivers commonly descend grades too aggressively. These aggressive descents are too often believed by drivers to be safe since they may be relatively slow, brake failure does not occur, and the truck makes it to the bottom of the grade without incident. Drivers often believe this because what has been published on safe descent speed in driver training materials is either too liberal or too ambiguous. These aggressive descents require frequent use of the truck's brakes, heating them up to a point where they are incapable of absorbing the additional heat from an unexpected emergency braking event or even a normal stop. If this added demand on the braking system occurs, the brakes will overheat, fade, and fail. An example of one of these over-aggressive descents was witnessed during a research study of truck brake temperatures on the Grapevine Hill of Tejon pass. The event occurred when a truck driver recruited for the study at the top of the Grapevine was unable to stop at the base of the grade where he was to have his brake temperatures checked. Before the descent, the driver was only instructed to drive the grade as he normally would. The event was reported as follows:
the truck was not able to stop in the flat parking area just beyond the foot of the hill because the brakes had overheated and faded. It continued to coast along the highway for about a quarter of a mile, and finally pulled off the road onto the shoulder, allowing measurement of the brake drum temperatures. While five of the hottest brakes were over 800 °F (the hottest being 951 °F), the average brake temperature was "only" 588 °F… The driver commented, "I guess 30 miles an hour was too fast for that hill." (He was 53 years old with considerable experience on the Grapevine!).
This driver, as many do, believed that his 30-mph descent speed was not an aggressive descent. He only discovered that 30 mph was too aggressive on this grade because he was asked to stop in a place where he normally would not have to stop.To explain this trucking grade descent problem better the following will include a discussion of energy and brake theory, brake fade, downhill braking driver training, safe speed driver training, and safe speed research.  Energy and Brake TheoryThe basic principle of how brakes work is the same for most vehicles. This principle is the Conservation of Energy, which tells us that energy is neither created nor destroyed, but can only be converted from one form to another. Brakes are energy converters; they convert kinetic energy (motion) into heat energy through friction between either a brake lining and brake drum surface or a brake lining and brake rotor surface (friction surface). The amount of heat produced by a brake system is directly related to the mass (weight) of a vehicle and the driver's desired velocity (speed) reduction. Heat energy is generally measured in British Thermal Units (BTU's), and the amount of BTU's produced by the brake system is the result of the amount of kinetic energy that is being converted. Since Kinetic Energy = ½ m·V² (kinetic energy equals one half mass times velocity squared) the amount of energy that the brake must convert to heat is doubled by doubling a vehicle's weight and quadrupled by doubling its speed.  When sitting atop a grade in a truck, the truck weight and the vertical distance to the base cannot be changed, so the gravitational potential energy is constant at that point. Therefore, descent speed does not determine the amount of slowing required to keep the truck under control. However, since the transmission multiplies engine torque to slow the truck just as it does to get it moving, speed, and more importantly gear selection, determines how much of the potential energy will be converted by engine drag. Increasing engine drag reduces the need to use the truck's brakes, thereby reducing the truck's brake temperatures from the descent. The driver's choice of speed is also important, since slowing down increases the time it takes to descend the grade allowing more time for the gravitational potential energy to be converted and more time to move the heat off the brakes so they will not overheat. The effect of grade length on brake heat is also important to understand. When a truck brake is converting kinetic energy into heat faster than that heat can be removed, the temperature of the brakes will continually increase due to this thermal dynamic (movement of heat) imbalance. A truck brake can only absorb a limited amount of heat before it begins to experience brake to fade. With a thermal dynamic imbalance on a short grade, the brakes may never get hot enough to create a problem. However, on a long grade, even a small thermodynamic imbalance could cause the brake temperatures to progressively increase to the point of failure before reaching the base of the grade. So a 5 percent, one-mile grade is more forgiving of errors in judgment and aggressive descents than a 40-mile 5 percent grade. Since brake heat energy is only related to the amount of kinetic energy being converted, it doesn't matter whether the brakes are applied hard for a short time or light for a long time, the resulting heat will be the same if the desired speed reduction is the same. If all the brakes on a vehicle are generating the same amount of brake torque, the heat created by converting kinetic energy will be evenly distributed through all the brakes. However, imbalance caused by poor maintenance, poor load distribution, or light brake applications may cause an uneven distribution of brake heat with some brakes possibly overheating. The typical generic "normal driving" temperature range for well balanced vehicle brakes is 100 to 200 degrees. A controlled mountain grade descent can produce brake temperatures between 200 and 400 degrees. Carlisle reports that a brake resin odor is produced at about 550 degrees and visible smoke is produced at 850 degrees. Brake FadeBrake system heat is dissipated through radiation, conduction, and convection due to a temperature gradient. Radiation is the transfer of heat through space.  Friction Fade is a reduction in the friction at the friction surface. As previously discussed, friction is the mechanism used to convert kinetic energy into heat in a brake system. Friction is the resistance of motion between two objects that are in contact with each other. If friction at the friction surface is reduced to an unacceptable level, the ability to convert kinetic energy into heat will also be reduced. When friction fade occurs in a hydraulic brake system, the pedal will still feel hard to the driver, but he will notice a difference in the braking response of the vehicle. For air-braked vehicles, when friction fade occurs, the driver may report the pedal going to the floor. Brake friction is affected by the temperature at the friction surface. The heat/friction profile is different from lining to lining and can be linear or curvilinear. Either brake lining friction can gradually decline (linear) as heat builds in the brake, or alternatively, lining friction can build-up until it reaches a peak, then quickly begins to decrease (curvilinear). Generally a linear heat friction relationship is more desirable because its fade is gradual and predictable. Brake linings are made of many different types of materials that are bound together with resins ("glue") during production. These resins are often blamed for reducing the friction at the friction surface. A commonly disputed claim is that "gaseous bearing fade" occurs when the resin gasses escaping from an overheated lining create a layer of gas at the friction surface. This layer of gas is blamed for reducing the friction at the brake's friction surface much like a layer of water reduces the friction between a tire and the road during hydroplaning. Gaseous bearing fade is also often associated with the propensity of fade experienced within new brake linings (Green Fade). New linings are believed to produce a greater amount of resin gasses and, during normal driving, new linings do produce an odor that is associated with these escaping resin gasses. Gaseous bearing fade is also commonly claimed to be associated with melting metal and friction material creating this lubricating gas. Opponents to these claims believe that these gases cannot be produced at a rate that can keep up with the speed of the spinning brake and do not believe gaseous bearing fade to be a true cause of brake fade. Resins can also reduce the friction at the friction surface when the brake linings become glazed. Glazed brake linings can contribute to cause brake fade, particularly domino fade. Reports are that glazing of a brake lining occurs when the resins in the lining are "melted" and reharden to form a hard "crust" on the surface of the lining usually due to short braking cycles. This hard surface on the lining reduces friction at the brake's friction surface. Glazed linings are easily identified by an abnormally shiny surface.  Incomplete friction surface contact is also a contributing cause of friction fade. A brake's friction surface that has incomplete contact will not distribute the heat evenly across the friction surface and the portions of the friction surface that are in contact will get hotter than they normally would. Brakes in this condition are easily heated up to the point of friction fade. Evidence of brakes that fail due to incomplete friction surface contact is seen as localized discoloration (i.e. bluing) of the metal friction surface. Incomplete friction surface contact is very common. It can be caused by the following:
Brake rotor and drum thickness is also an element of friction fade. If the drum or rotor is excessively thin, its ability to store heat will be reduced. This also can cause the friction surfaces to get hotter than they would under the same circumstances. Brakes with excessively thin drums and/or rotors can easily heat up to the point of friction fade. Additionally, insufficient drum thickness can cause heat expansion of the drum to be excessive. These effects on brake fade are discussed in the following section.  Brake drum expansion is probably the most concerning aspect of brake fade in S-cam braked heavy trucks. In an S-cam brake, the application device is the brake chamber. A typical brake chamber has a maximum pushrod stroke capability of about 2.5 inches. Cold pushrod stroke measurements on a truck with Automatic Brake Adjusters (ABA's) are commonly found to be around 1.75 inches. So, a typical brake chamber will have a reserve stroke of around 0.75 inches. As a brake drum expands away from the brake linings, the cold pushrod stroke will increase. A brake drum temperature of 400 degrees can increase pushrod stroke by as much as 0.5 inches. The force output of a brake chamber is not linear, and a typical brake chamber reaches its knee point where force will begin to drop at about 1.75 inches. If the brake drum becomes hot enough, the brake chamber can bottom out and the brake chamber force output at that point will be 0. Drum brakes are also very susceptible to brake fade because they are self-energizing. In a self-energizing brake, when the lining is applied to the drum, the drum rotation tries to pull the lining and shoe along with it. This self-energizing action increases the force applied to the lining friction surface. When fade reduces friction at the friction surface in a self-energizing brake, a compounded reduction in applied force at the friction surface results in a compounded reduction in braking. The two common types of self-energizing drum brakes used in automotive applications are the Leading-Trailing drum brake and Duo-Servo drum brake. The Leading-Trailing drum brake has a fixed anchor on one side of the shoes. The Duo-Servo drum brake has "floating" anchorage. In a Leading-Trailing brake, only the leading shoe is self-energized. In a Duo-Servo brake, both shoes are self-energized. Therefore, a Duo-Servo drum brake can have a greater reduction in applied force at the friction surface than a Leading-Trailing drum brake and is more susceptible to fade.  The brake fluid used in cars and trucks typically has a boiling point of around 401degrees at sea level. Brake fluid is also hygroscopic, a characteristic which allows it to absorb moisture. Over time, brake fluid will become contaminated with moisture. As this happens, the boiling point of the brake fluid will be lowered, since the boiling point of water at sea level is only 212 degrees. Wet brake fluid has only a 3.5% water content and the wet boiling point of brake fluid drops to 284 degrees. I have tested the boiling point of brake fluid many times and have tested samples that boiled at temperatures as low as 260 degrees. For this reason, brake fluid maintenance recommendations are that it should be flushed about every 4 years. The boiling point of a fluid is also reduced as altitude is increased. For example, the boiling point of water is reduced by approximately 1% for every 1,000 feet above sea level. So driving mountain passes, which often exceed 11,000 feet, driving Pikes Peak highway that tops out at 14,110 feet, or Mount Evans Road, that tops out at 14,130 feet can easily cause brake fade, by producing a significant drop in the brake fluid boiling point. These locations also put extreme demands on a brake system that cause brakes to get hot and are the most common places brake fade is experienced. Domino Fade occurs when some brakes in the system have more brake torque than the others. This imbalance may be the result of poor maintenance, poor load distribution, or light brake applications. The brakes producing more torque will heat up much quicker than they should, which could cause them to fade. If the high torque brakes fail, the other brakes will then receive a disproportionate amount of heat. These now overloaded brakes will also likely fail, hence the domino effect. Domino fade is a characteristic that is common in heavy trucks that often have imbalances between the tractor and trailer's brakes. Light, steady brake applications may not activate all the brakes on a heavy truck. Hard, short brake applications promote more even braking and better distribute heat throughout the brake system. For this reason, it is recommended in the CDL manual that truck drivers use the snub braking technique when operating on steep grades. Downhill Braking Driver Training The following is the incorrect recommendation published in the 1989 CDL manual:
Some people believe that letting up on the brakes from time to time will allow them to cool enough so they don't become overheated. Tests have proven this is not true. Brake drums cool very slowly, so the amount of cooling between applications is not enough to prevent overheating. This type of braking requires heavier brake pressures than steady application does. Heavy pressure on the brakes from time to time builds up more heat than light continuous pressure does. Therefore, select the right gear, go slow enough, and maintain a lighter, steadier use of the brakes.
After the previous discussion on conservation of energy and its relationship to automotive brake systems, it should be clear that this old truckers' tale and the first edition of the CDL manual are obviously wrong. Right after the first CDL manual was published, a scientific study proved that the controlled braking method was wrong. This study resulted in downhill braking recommendations that are effectively the exact opposite of what was first recommended by the CDL manual. As a result, changes were made to the CDL manual to mirror the recommendations of this study and published in the 1993, second edition. Yet, the controlled braking method that was taught to truck drivers for years is today still taught to and practiced by many stubborn truck drivers.Snub braking became the recommended method of downhill braking as a result of the study done by the National Highway Traffic Safety Administration (NHTSA) and University of Michigan Transportation Research Institute (UMTRI). The following is the correct recommendation published in the current CDL manual:
Apply the brakes just hard enough to feel a definite slowdown. When your speed has been reduced to approximately five mph below your "safe" speed, release the brakes. (This brake application should last for about three seconds.) When your speed has increased to your "safe" speed, repeat steps 1 and 2.
As you recall, truck brakes are energy converters, converting kinetic energy of the truck into heat energy through the friction between the brake shoes and drums. The heat energy is then removed from the brake primarily by convection. Since the brakes are converting kinetic energy, then the amount of heat energy produced depends only on the weight of the truck and the desired speed reduction. If the weight and the desired speed reduction remain constant, the way the brakes are applied, either hard for a short time or lightly for a long time, will not change the amount of heat energy the brakes output.For example, if your "safe" speed is 40 mph, you would not apply the brakes until your speed reaches 40 mph. You now apply the brakes hard enough to gradually reduce your speed to 35 mph and then release the brakes. Repeat this as often as necessary until you have reached the end of the downgrade.  Pneumatic balance is created by having equal air pressure at all wheel ends. When a truck has a pneumatic imbalance, some of the brakes will work harder than others. Pneumatic balance is largely affected by the relay valves that control the application and release of the air brakes. A standard truck-trailer usually has one relay valve for the tractor drive axles and one for the trailer axles. These relay valves are controlled by air pressure from the foot valve (brake pedal). This control side pressure opens the relay valves allowing the desired amount of supply side pressure from the air tanks to pass through the valves and supply air pressure to the brakes. Pneumatic imbalance is a result of these valves opening at different pressures. For example, a tractor may be setup with a relay valve that opens at 15psi (15psi crack pressure relay valve) and the trailer may have a relay valve with a 3psi crack pressure. A truck set up this way will only apply the trailer brakes during a controlled brake application, which typically has an application pressure of less than 10psi. However, a "snub" brake application of 20 to 30psi will open all tractor and trailer valves. This type of imbalance can also result from contaminants and alcohol in the air system that cause these valves to hang-up and have higher than normal crack pressures. The UMTRI study found that trucks with properly balanced brake systems had basically the same average brake temperature when using either controlled or snub braking. However, trucks with poor brake balance were found to have more uniform brake temperatures when the snub method was used. Unless pneumatic testing is performed on a truck to ensure that proper brake balance is maintained, there is no way to know if a truck has good brake balance. This type of testing is difficult to perform in most trucking operations since a tractor is usually hooked to several different trailers over relatively short time periods. Therefore, because it is virtually impossible to determine if a truck has good brake balance, snub braking is the recommended method.  Since the snub braking method cannot compensate for torque imbalance, trucks should always be inspected and repaired with the following in mind. A truck's brake system should have matched mechanical components such as the same size brake chambers and the same length slack adjusters on both sides of an axle and, most of the time, on all brakes in a group of axles (i.e., tractor drive axles). When inspecting the condition of the brakes, any isolated premature wear found is an indication of a torque imbalance. If one brake wears faster than the rest, there is an imbalance and that brake is doing more work than the rest. If one brake wears much slower than the rest, then that brake is not working as hard as the rest. When brakes are repaired, the cause of an identified torque imbalance need to be found before repairs are made. Repairs made without correcting the torque imbalance could amplify the problem causing the overworked brake to work even harder and overheat. Equally important is to ensure that the same repairs are done on both sides of an axle. If the brake hardware is replaced on the right side of an axle, it should also be replaced on the left side. If the s-cam bushings are replaced on the right side, they should be replaced on the left side. Safe Speed Driver TrainingA long-standing recommendation for safe descent speed says you should never go down a grade faster than you can climb it. This Crawl Gear Method suggests that the Crawl Gear that a driver would use when climbing the same grade at its Crawl Speed is the correct gear and speed to use when descending the grade. While descending the grade slower than you can climb it is probably good advice, descending it at the same speed is most likely not good advice. Another recommendation claims that the safest speed for a mountain descent is what is known as the Control Speed. Control Speed is reportedly when the "uphill" and "downhill forces" are equal and the driver doesn't need to use his service brakes at all to maintain a steady speed. The truck's engine is the primary force that is used to balance the downhill force. So Control Speed is achieved by selecting the correct gear for the grade. This recommendation also reports that the Crawl Gear is the correct gear needed to achieve control speed. Ideally, Control Speed, if achieved, would be the safest speed for a mountain descent since the brakes would not need to be used to maintain a steady descent.  Trucks equipped with a retarder, known as an engine brake, reportedly have braking horsepower equal to a hundred percent or more of the engine's drive horsepower. The engine brake works by simply opening the engine's exhaust valves at the top of the compression stroke when the accelerator is released. When the engine brake is on, the intake air is still compressed and the truck's kinetic energy is still converted into the potential energy of the compressed air. The difference is that opening the exhaust valves allows the compressed air to exit the engine's exhaust rather than push the engine's pistons down on the power stroke. This escaping compressed air produces the distinct popping sound associated with an engine brake. Drivers of trucks with engine brakes are instructed to use the Crawl Gear method for determining the decent gear. However, some claim that, with an engine brake, you can descend grades one or even two gears higher than the Crawl Gear! Other common claims are that not only can an engine brake reduce trip time by allowing the driver to drive faster in hilly conditions while still maintaining control of his truck, but also that the engine brake can help prevent crashes by minimizing the chances of a runaway. While the engine brake is a very beneficial device, these marketing claims are more likely to encourage over-aggressive descents than to prevent runaways. Modern truck engines commonly make around 500 horsepower and can climb grades in gears and at speeds that are likely much greater than what is safe for descent. For this reason, more recently published recommendations suggest that modern trucks should use descent gears that are lower than the Crawl Gear. Some sources recommend one gear lower than the Crawl Gear. The recommendation in the CDL manual is very ambiguous but does recommend using lower gears. The current version of the CDL manual says:
With older trucks, a rule for choosing gears is to use the same gear going down a hill that you would need to climb the hill. However, new trucks have low friction parts and streamlined shapes for fuel economy. They may also have more powerful engines. This means they can go up hills in higher gears and have less friction and air drag to hold them back going down hills. For that reason, drivers of modern trucks may have to use lower gears going down a hill than would be required to go up the hill. You should know what is right for your vehicle.
This statement remains unchanged from the first publication of the CDL in 1989.In summary, all the published "safe" speed recommendations that I am aware of use some form of the Crawl Gear method. However, the specific recommendations as to the correct gear to use, either the crawl gear, one gear higher, two gears higher, or some unknown lower gear, are all over the map. Safe Speed ResearchBetween 1975 and 1989 the American Trucking Associations (ATA) and the Federal Highway Administration (FHWA) sponsored five or more research projects on safe descent speed. Initially the research was exploring the feasibility of a rating system for grade severity that would post a number 1 to 10 at the top of a grade in an attempt to communicate to truck drivers how they should drive that grade. The idea was that, regardless of the specific grade characteristics, a driver would drive all grades with the same severity number the same way. Since grade severity rating using the abstract rating system proved to be problematic, later research explored the use of Weight Specific Speed (WSS) signs posted on mountain downgrades (as discussed later).  The American Association of State Highway and Transportation Officials (AASHTO) publishes standards for the geometric design of highways. In their geometric design manual, they recommend the use of the Grade Severity Rating System User Manual and computer model for determining operating characteristics of trucks on downgrades. The geometric design manual also recommends that the computer model be used to determine the need and placement of Runaway Truck ramps. Where to place the runaway truck ramp is determined as the location where a truck is most likely to reach brake temperatures of 500 degrees or more. The Grade Severity Rating System User Manual and truck braking model are intended for use by highway departments to determine safe descent speeds on specific grades for trucks according to their weight. Those speeds can then be posted on a WSS sign at the top of the grade. The WSS sign has truck weights listed on the left side in 5,000 pound increments descending from 80,000 and has the safe speeds for the listed weights on the right. To my knowledge, the WSS sign has not been widely used. However, Oregon has a WSS sign on Deadman's Pass (I-84, aka, Cabbage Hill) like the proposed sign. The sign is a yellow and black advisory with the following information:
The information on this sign appears to have been developed based on the Grade Severity Rating System User Manual and truck braking model. In addition to the WSS sign, dynamic truck speed warning systems have been used on Sandstone Mountain and Straight Creek Pass (aka the Eisenhower). These systems use weigh-in-motion sensors to determine a truck's weight and calculate a WSS using the Grade Severity Rating System. The calculated WSS is then posted on a changeable message sign for the truck driver to see. To my knowledge, other than these three locations, the WSS information has never been communicated to truck drivers. Tables 1 and 2 is data output from the WSS Braking Model for the westbound descents from the Eisenhower and Vail Pass. The grades were figured as an average from the summit elevation and the base elevation. Data was obtained from a Colorado DOT road profile as well as physical inspections. TABLE 1 — RECOMMENDED SPEEDS FOR THE WESTBOUND EISENHOWER FROM THE GRADE SEVERITY RATING SYSTEM
TABLE 2 — RECOMMENDED SPEEDS FOR THE WESTBOUND VAIL PASS FROM THE GRADE SEVERITY RATING SYSTEM
This Grade Model and the above data are for trucks that are not using retarders. To account for retarders, the research indicates that the use of a retarder has the same effect as reducing the weight of the truck. The research suggests that trucks equipped with a retarder can descend the grade at the speed recommended for 1 weight classes (5,000 lbs) below their actual weight. Additionally this model assumes that the truck has a fully functioning and balanced brake system. Defective and unbalanced brake systems may still generate average temperatures under 500 degrees but some brakes will be hotter and others will be cooler than the average. Therefore, a truck with brake defects or imbalances could still experience some brake fade or domino brake fade even at the recommended speeds. Crash Forensics Field Study: To further explore the safe speed issue, a two-day field study of the operating characteristics of actual trucks (truck-tractor semi-trailers) on the Eisenhower was performed. This study included using radar equipment to determine both the uphill crawl speeds and descent speeds of a large sample of trucks on the west side of the Eisenhower. This study also included following and observing truck speeds and braking activity on the west side of the Eisenhower. For the uphill portion of the speed study, I was positioned about 1.5 miles west of the Eisenhower summit. The crawl speeds of trucks measured at this location ranged from 23 mph to 36 mph with the average being 28 mph. My downhill speed study measured truck speeds in the range of 23 mph to 50 mph with an average speed of 34 mph and almost half the sample was operating very close to 35 mph. Many of the trucks observed operating in this range were producing an identifiable brake odor and the truck operating at 50 mph was producing visible brake smoke (brake temperature over 850 degrees produce smoke). Note that there is a 30-mph downhill speed limit for trucks over 15T and, according to CSP officers, the speed limit is regularly enforced. My belief is that the speed limit and enforcement are the most likely reason that most of the trucks observed were operating at speeds around 35 mph. Most of the trucks followed while descending this grade were also operating at speeds around 35 mph. Of these trucks, some where using their brakes almost continuously while others never used their brakes at all. The Eisenhower data generated using the Braking Model in Table 1 shows that an unloaded truck should be able to descend this grade at 35 mph but that 35 mph is way too fast for a loaded truck. If the field study observations are compared to the Table 1 data, then the conclusion can be drawn that the trucks observed using their brakes continuously were more likely than not fully loaded while the trucks not using their brakes at all were more likely than not lightly loaded (under 30T gross). As discussed in this paper, there is an almost uniform agreement that truck drivers should never descend a grade faster than their uphill crawl speed on the same grade. Most of the sources indicate that the descent speed should be much slower than the uphill crawl speed. However, our spot-speed study showed that the average descent speeds of trucks on the Eisenhower grade exceeded uphill crawl speeds by six mile per hour and it is likely that these speeds would be even faster without the downhill speed limit for trucks. Many of these trucks were producing brake odor during the first two miles of this nine-mile descent. Brake odor is indicative of brake temperature in excess of 550 degrees. This brake odor is clear evidence that these trucks were exceeding the safe descent speed. |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
