Troy, Mich. — This is the second article in a series of four by Narcís Molina, Project Manager, Braking Systems in Applus IDIADA, about an investigation of the relationship between brake creep groan and brake squeal in disc brakes.

Relationship of Brake Creep Groan and Squeal in Disc Brakes (Part 1 of 4)

This article investigates how two distinct brake phenomena such as squeal and creep groan can be related to each other during the operational use of a disc brake, as they appear concatenated in time.

The first part of this series introduced the problem and the methodology, while this second article presents the main results associated to creep groan.

3. Results

3.1. Creep Groan

The mechanism to generate creep groan is commonly thought to result from a stick-slip motion in the friction interface. Under certain low pressure braking applications and low vehicle speeds, the brakes grab. This process might repeat itself and might be accompanied by a low frequency noise.

Severe creep groan problems are normally associated to a dominant intervention of the vehicle structure. The vibration at the friction interface starts the process; the brake by itself cannot generate enough noise to be an issue. In general, this means that there must be a resonance in one or more components that respond to the initial brake vibration frequency. The path is purely structural: the energy travels from the brake into the suspension, then to the vehicle structure, the steering system, etc. up to the vehicle interior panels, which boost the noise radiation.

3.1.1. Vibration Characterization

Figure 11 shows the spectrogram of the FL caliper at three instants in time, allowing the different groan phases to be clearly identified:

  • Triggering groan: a transient, broadband and incipient stick-slip motion.
  • Charging groan: also transient, the noise is the sum of the broadband signal with clusters of pure frequencies.
  • Unstable groan: a tonal response of exponentially-increased amplitude that is caused by a self-sustaining excitation of vehicle resonance.

The nomenclature used in previous findings is maintained.

Figure 11: Spectrogram of vibration of the FL caliper and the auto-powers at 3 instants in time, thus showing the evolution of the groan: triggering (67 s.), charging (69 s.) and unstable (71 s.).

Incipient groan generates relatively low levels of vibration energy during short periods of time. The frequency contents of the derived variations in brake torque are broadband and, thus, a wide range of vehicle resonances can be excited. This scenario tends to be confusing, particularly when a final steady-state vibration (the unstable phase) is reached, in which the vibration energy increases logarithmically with respect to the initial condition.

In this context, the concept unstable denotes the behavior associated to a high noise and vibration level that occurs with a controlled and constant brake pressure input.

The analysis of the spectrum suggests that the noise has two sets of harmonics. The first set —11.5, 23.5, 35, 47, 70 Hz— is excited during the charging phase and has quite a constant level, until the second set becomes over the unstable phase —whose base frequency is clearly identified around 63 Hz.

Note that the groan phases previously described appear in all the measurement points with different intensities —depending on the brake vibrations’ transmissibility over the vehicle structure.

3.1.2. Vibration Energy Evolution

The vibration energy per unit of mass is calculated as the spatial average mean square of the acceleration levels. It allows investigating how the vibration energy grows over the time domain, thus differentiating the already-presented creep groan development phases.

Figure 12 depicts the logarithmic increase in specific vibration energy, as measured in the pads (green; x-direction; point 3), the caliper housing (blue; x-direction; point 7) and the caliper anchor (red; z-direction; point 14) throughout the development of groan.

Figure 12: Specific vibration energy evolution throughout the development of creep groan. x and z refer to the local system of reference depicted on the caliper.

All measurement points refer to the number scheme of the FL caliper (Figure 2; Table 5), whereas x and z denote the direction with more vibration energy for each component (the local coordinate system is detailed in Figure 12).

3.1.3. EMA (Brake-Suspension Assembly, Static)

The in-vehicle modal analysis of the front axle and the FL brake caliper system allows  defining the vibration shapes associated with the base resonant frequencies occurring during the charging and unstable creep groan —i.e. 11.5 and 63 Hz, respectively.

Figure 13 depicts two different frames from the set-up, as seen from the top. They prove the deformation of the system at 11.5 Hz: indeed, a rotation of the wheel around the pivot point responsible for the steering movement (z-axis) occurs, the steering assembly following. Many other harmonics are found, but they are omitted for the sake of simplicity.


Figure 13: Modal shape at 11.5 Hz. Colors nomenclature: leaf spring, axle, track rod (black), steering column, steering rod, steering arm, knuckle. The brake caliper is hidden. The dashed, red line illustrates the wheel axis.

Figure 14 shows the same view of the modal shape at 63.0 Hz. At this frequency, observe the bending condition of the track rod.


Figure 14: Modal shape at 63.0 Hz. Colors nomenclature: leaf spring, axle, track rod [black], steering column, steering rod, steering arm, knuckle, brake caliper (pads, housing and anchor). The dashed, red line illustrates the track rod axis.

3.1.4. Operational Deflection Shapes

The analysis to obtain the operational modal shapes is equivalent to that undertaken for the impulse-response measurements. Indeed, the main difference is the excitation source: in the ODS, the initial vibration is coming from the stick-slip motion found in the friction interface, as the vehicle starts rolling on a 20% slope.

Figure 15 shows the operational modal shape of the front axle and the FL brake at 11.5 Hz. At this frequency, the brake pads tend to move longitudinally —forwards and backwards— within the caliper housing: it is the stick-slip motion. The wheel hub and all connecting parts do follow this oscillation by rotating around the steering pivot point (z-axis). This behavior is similar to the modal shape at 11.5 Hz (Figure 13). Many other harmonics (17, 23.5, 35, 47, 70 Hz) are found during the charging phase, exhibiting similar deformations; again, for the sake of simplicity, they are omitted.

Figure 15: Operational modal shape at 11.5 Hz. Colors nomenclature: leaf spring, axle, track rod [black], steering column, steering rod, steering arm, knuckle, brake caliper (pads, housing and anchor). The dashed, red line illustrates the wheel axis.

Figure 16 shows the operational modal shape at 63 Hz (unstable groan). At this frequency, the hub —and connecting parts— are not rotating around the z-axis, but the y-axis: this is the actual rolling rotation of the wheel, although such a movement may be also induced by the vibration resulting from the stick-slip motion.

The track rod, on the other hand, seems to exhibit a high relative displacement. This observation agrees with the modal shape seen at 63 Hz (Figure 14), where the first bending mode of the mentioned component was very active.

Figure 16: Operational modal shape at 63.0 Hz. Colors nomenclature: leaf spring, axle, track rod [black], steering column, steering rod, steering arm, knuckle. The brake caliper is hidden.

Therefore, it is concluded that the harmonics of the creep groan that are close to 63 Hz —base frequency of the unstable phase—  excite the track rod. In return, the hub and other parts —including the leaf spring— do follow the track rod.

3.1.5. EMA (Brake Assembly, Free-Free)

Figure 17 depicts the sum of FRF’s for the modal analysis conducted on the suspended brake assembly; two test set-ups are compared: with (2.6 bar) and without air pressure. The frequency range of the study is between 0 and 500 Hz. Note that the effect of the applied braking pressure at the lower frequency range is to stiffen the brake assembly structure —as seen with the increase of the low resonance frequencies by about 10 Hz.

Figure 17: Sum of FRF’s. Comparative results obtained with a braking pressure of 2.6 bar (red) and without pressure (green).

Similar modes shapes are identified for both test configurations. The first elastic mode is detected around 144 Hz, although the mode with the highest participation is close to 450 Hz.

Another important conclusion is the fact that, although the mode shapes of the isolated caliper show similarities with those found for the modal analysis of the front axle (including the FL caliper, as reported in section 3.1.3), the frequencies are very different. The in-vehicle modes are below 100 Hz, whereas those of the isolated brake assembly are above 100 Hz. This big difference is due to the differences in the level and frequency band of the excitation forces applied in the two types of modal analysis: in the in-vehicle case, loads on the brake are generated at the pad-disc interface, while the front axle suspension generates additional loads by shaking the brake system; in the isolated brake, under free-free conditions, the load scenario is completely different. Thus, the brake modes participating in the generation of groan can be considered as slave modes in comparison with those found in the free-free modal analysis, which could be seen as genuine.

About Applus IDIADA

With more than 25 years’ experience and 2,450 engineers specializing in vehicle development, Applus IDIADA is a leading engineering company providing design, testing, engineering, and homologation services to the automotive industry worldwide.

Applus IDIADA has locations in California and Michigan, with further presence in 25 other countries, mainly in Europe and Asia.