|
|
|
FRICTION EXCITED OSCILLATIONS LINK BRAKE TECHNOLOGY REPORT—FEV1 Written by Arnold E. Anderson BASICS ABOUT FRICTION EXCITED OSCILLATIONS Definition and description We are all familiar with the squeak of a door hinge, the pure tone of a violin, and the chatter of chalk rubbing on a blackboard. These are all friction excited oscillations. Brake squeals and some violin solos can be long-lasting, seemingly without end. The stick-slip and lurch when a ski tow rope stops sliding in your glove are single events, always of short duration. All friction-excited oscillations have one common feature—the source of oscillation is an energy-dissipating frictional event. Friction-excited oscillations are so different from resonant and forced oscillations that they require special understanding and sometimes different techniques to measure, understand, and remedy. Friction excited oscillation problems have been around for a long time. Brake squeal is the oldest known engineering vibration problem. The first brake squeal on record was from a Sumerian quarry cart, 3,500 years ago. A bronze chain was wrapped around the axle and applied by a long wooden brake lever to control speed down steep grades. While highly energized and effective, they were so noisy as to get recorded for posterity. Until they are understood and can be designed out of a brake system, brake NVH (Noise, Harshness and Vibration) workers can count on job security—and frustration. The reason they are so frustrating is that friction excited oscillations produce vibrations in whatever combination that maximizes the conversion of vehicle kinetic energy into oscillations. Attempts at damping or suppressing one vibrational mode either provides a minimal vibration reduction or it causes another vibrational mode to get excited. There is only one way to end friction excited oscillations—and that is to stabilize the system. This can be done by reducing the frictional excitation, perhaps by changing the friction material. It also can be accomplished by providing damping at the excitation site. Damping introduced away from the excitation site is of little value. Some disc brake squeal problems have been resolved (but not ended) by forcing the squeal frequency beyond the human hearing range. Before attempting to resolve friction excited oscillations, it definitely helps to understand them and their excitation mechanisms. Three Excitation Mechanisms Stick Slip, Negative Damping, Positive Mechanical Feedback Stick Slip An example of stick slip is a squeaky door hinge. Slick-slip is a low speed source of noise generation that is caused by a non-linear drop of friction as motion starts or a buildup of friction as motion stops. Stick-slip also may occur if the static friction changes with time. As suggested by the name, stick-slip consists of two phases, stick and slip. During the stick phase, the brake lining and cast iron move together, with no slippage at the interface. Eventually, the windup of the brake assembly causes a momentary slip phase. The stick time period is variable, depending on speed, load, and system stiffness. When slip begins, a noise burst occurs. This involves a half-cycle of motion at the rubbing surface. The sudden energy burst often produces a more sustained audible oscillation. Most stick slip oscillations are of the rigid body form, so have low frequencies. Then noise bursts always start motion in one direction, and have peak amplitudes then. Many frequencies are excited at one time, so the time-domain wave form is complex. Diagnostics of stick-slip vibration problems require observations in the time domain—with sensing as close to the noise source as possible. Stick-slip in brakes generally is confined to creep-groan and similar low speed rigid body vibrations. Vehicle speeds are normally under 2 mph for brake stick slip. Many geometry, stiffness, and damping terms are involved in a rigorous modeling of disc brake creep groan. However, even the low frequency brake noise systems are similarly difficult to instrument for diagnostic studies. Some engineers, perhaps in frustration with creep-groan problems, have suggested that creep-groan provides a safety benefit and should not be eliminated. The argument here is that a driver at a stoplight may use creep-groan vibrations as a warning that the vehicle has started to move. Most drivers and all vehicle manufacturers appear to prefer brakes without creep-groan, even at traffic lights. It is possible for rounded particles within brake linings (such as glass beads) to undergo stick-slip within the brake lining, when excited by the rubbing surface. This stick-slip may occur at any vehicle speed, but still has very low rubbing speeds between the rounded particle and the brake lining matrix, where the stick-slip occurs. Negative Damping Dry clutch judder is an automotive example of negative damping oscillation. Judder, also called shudder, can occur when the clutch is slipped at startup. Like all friction-excited oscillations, it attempts to maximize conversion of energy into vibration. In a drivetrain, this normally is the lowest driveline torsional oscillation (typically around 15 hertz). Damping in the clutch hub is used to control this oscillation. Driveline damping is least in the lowest numerical gear ratios (reverse or first gear), when there is some vehicle motion. Consequently, if a clutch judders it will do so at startup. Oil contamination of the clutch facing is the most common cause for judder. Such oil contamination requires an oil film about 0.15 microns thick on the clutch surface. This produces a negative friction-velocity slope for rubbing speeds below 7 meters/second, with greatest negative slopes below 2 meters/second. Negative damping in brakes normally is a contributor to noise, but not the only cause. It generally is of consequence from 1.5 mph to about 12 mph vehicle speed, but effects have been seen to 20 mph. This is one reason why brake squeals are more commonly heard at the lower vehicle speeds. The other is that brake effectiveness rises at lower speeds, aiding positive mechanical feedback. Positive Mechanical Feedback This is the dominant and primary cause of most brake noise, excepting creep-groan. Sources of positive mechanical feedback can be found in the fibers of the brake lining matrix, the overhanging leading edge of the lining (especially with rivets or fractured bond lines), and poorly supported brake shoes (especially on outboard disc brake shoes). Rigid body oscillations, such as brake groans, also may involve positive mechanical feedback. Multiple positive mechanical feedback sites on a brake lining are known to generate wire brush and ‘gravel-gurgle’ sounds. These multiple excitation sites are the reason for many, and varied, noise frequencies and amplitudes. Two Vibration Types, Rigid Body and Continuum Oscillations Rigid Body Oscillations As the name implies, rigid body oscillations are those in which the vibrating masses move with little or no deflections. These are usually at vibrating frequencies well below those of brake squeal, and most are below 600 hertz. Rigid body oscillations can be of very low frequency, so they can be seen or felt, but not heard. Brake roughness, judder, and shudder are examples of these tactile forms of friction excited oscillations. They may have mechanical and thermal causes. Mechanical causes include machining errors (thickness variation, ovality, runout) and component wear (rotor thickness variation). Uneven heating of brake rotors can temporarily cause, or increase, thickness variation, and sometimes can produce a primary thermal buckling that warps the rotor into a washboard with three (sometimes five) high spots per revolution. Brake roughness is felt at wheel rotation speed, one torque pulse per wheel revolution. Some brake roughness may be felt at a low multiple of wheel speed. Disc brake thickness variation and drum brake eccentricity induce pulses at one per wheel revolution. Disc brake runout and drum brake ovality produce two vibrations per revolution. Lug bolt and other wheel distortions can cause 4, 5, or 6 pulses per wheel revolution. Groan and creep groan are also rigid body oscillations, but at higher frequencies, so they can be heard as well as felt. These usually fall in the range of 100 to 400 hertz. Groan vibrations occur at suspension and brake assembly natural frequencies, often involving 'windup' motions of the caliper and/or suspension arms. Creep groan is the name given to a groan which occurs at very low vehicle speeds (under 2 mph) that is caused by brake lining stick slip. Groan may occur at almost any speed, but is most often noticed between 5 and 20 mph with moderate braking. A pinchout groan is one that occurs as the vehicle comes to rest, usually with hard braking, and often following brake usage that got the brakes hot. Hum and moan are rigid body vibrations which are mostly heard, in the 150 to 400 hertz range. These often occur in disc brake assemblies as the result of a dragging shoe that excites a caliper mount windup resonance. Hum and moan can occur at highway speeds, with rotor thermal distortion as a cause. They also can happen at speeds under 20 mph, due mostly to negative damping effects. Continuum Oscillations Brake squeals make up these oscillations where, as the name implies, the vibrating components flex to provide both spring and mass for the vibrations. There are many different names used for brake squeals, unfortunately with no commonly accepted definitions. For this report, squeal is the generic name for a friction excited oscillation above 1000 hertz. Low frequency squeals range from 1000 to 4000 hertz, a frequency range that almost anyone can hear. These involve radial modes of the brake drum or rotor, and may include backing plate bending modes on drum brakes. They are caused by shoe-lining positive mechanical feedback and/or negative damping from the brake lining (under 20 mph). High frequency squeals are those which range from 5000 hertz to the limit of hearing—about 18000 hertz for women and young men (With age, most men progressively lose their high frequency hearing ability). High frequency squeals involve higher order diametral modes of rotors or higher order radial modes of drums. They are caused by positive mechanical feedback excitations from brake linings. Squeaks are brief squeals, differing only due to their short-term nature. A squeak may occur due to an apply transient, often seemingly random. They may occur regularly, such as one of the first brake applications in the morning. Squeaks may occur due to a repeating position-sensitive event, perhaps when one particular position on the rotor is contacted by the brake lining. Wire brush squeals tend to have several high frequency components, and result from multiple excitation sites on the brake lining. The only known cause of wire brush squeals is from positive mechanical feedback. Chirp squeals are short terms, like squeaks, but have a progressive variation of squeal frequencies. Squeaks often result from the change of brake apply pressure, which first excites on vibrational mode, then another as the brake linings change their contact patterns against the brake drum or disc. Ultrasonic squeals are not heard by humans, only sensed by instruments. Many audible brake squeals are preceded by ultrasonic squeals. For this reason, sensing ultrasonic squeals can be useful to indicate the onset of an audible squeal during testing. An otherwise quiet brake that exhibits ultrasonic squeal during development testing is one that could be prone to audible squeal, when many thousands of brakes are in customer service usage. Gravel gurgle is one of several names given to a warbling form of squeal (somewhat like a parakeet sound) that can occur at vehicle speeds under 6 mph, with low brake apply pressures. It can occur solely due to brake dragging action, again at very low speeds. LINK ENGINEERING COMPANY © 1995, All rights reserved reproduced on ifriction.com with permission of Link Engineering For more information visit: Link Engineering Or Contact: Link Engineering Email |