Analyzing Gearbox Degradation Using Time-Frequency Signature Analysis
Analyzing Gearbox Degradation Using Time-Frequency Signature Analysis
Analyzing Gearbox Degradation Using Time-Frequency Signature Analysis
Signature Analysis
SpectraQuest Inc.
8205 Hermitage Road
Richmond, VA 23228
(804)261-3300
www.spectraquest.com
March, 2006
Abstract: In this work, vibration signature caused by gear tooth seeded fault is studied
in the time and frequency domains. The experiment was conducted on a Machinery Fault
Simulator TM (MFS). The pinion gear in the gearbox was intentionally faulted with
increasing severities, and a vibration signal was collected for each case. The signal was
analyzed in a time and frequency domain. The results indicated that the vibration
signature of a faulted bevel gear tooth is a pulse in time domain. Because of this impulse
signal, strong sidebands arise in the spectrum around the mesh frequency.
Test Setup
The experiment is conducted on a Machinery Fault Simulator TM (MFS). The MFS and
the position of the gearbox in the MFS are illustrated in Fig. 1.
The gearbox is shown in Fig. 2(a). The transmission ratio is 1.5:1. There are two straight
bevel gears in the gearbox and they are shown in Fig. 2(b). The number of teeth on the
pinion is 18. The gearbox input shaft is connected to a sheave and driven by a “V” belt
drive. The gearbox output shaft is connected with a variable brake loader. The tachometer
is installed on the rotor deck with a bracket and used to measure the speed of the belt
sheave connected to the gearbox input shaft as illustrated in Fig. 3.
1
(a) gearbox (b) pinion and gear
Figure 2. Gearbox
The vibration data was collected by a tri-axial accelerometer installed on the top of the
gearbox. The accelerometer is also shown in Fig.3. The frequency limit was set to 5 KHz
during data acquisition. During the test, the motor speed was kept at 1000 RPM and the
brake loader was set to 2.5. First, the baseline of the gearbox vibration data is collected.
Then the gearbox is disassembled and the pinion gear is taken out. The surface of one of
the pinion’s teeth on the meshing side is milled using a drill mill. After the damage is
done to the tooth, the pinion gear is put back into the gearbox. Vibration data is collected
2
again under the same running conditions (speed and load) as before. This process is
repeated four more times with additional damage applied to the same pinion tooth for
each case scenario. The five damage severities are denoted as Severity Level 1 to Level
5, from the least to the most. The final pinion tooth damage is illustrated in Fig. 4. The
main body of the faulted tooth has been milled out and the damage is referred to as the
“missing of a tooth”.
Test Results
The waveforms for the damaged pinion with Damage Severity Level 1 are illustrated in
Fig. 6. Comparing Fig. 6(b) with Fig. 5(b), notice that there are impulses appearing in
Fig. 6(b). By a careful study of the position of the faulted pinion tooth, the positions of
the pulses indicated by arrows in Fig. 6(b) correspond to the instances when the faulted
pinion tooth enters meshing.
4
The waveforms for the damaged pinion with Damage Severity Levels 2 through 5 are
illustrated in Fig. 7 through Fig.10. Notice that the impulses caused by the faulted tooth
in meshing are consistently clear in each of the four faulted case scenarios. Comparing
the signal amplitudes in Fig. 10(b) and Fig. 7(b), it can be concluded that impulse
amplitude increases with damage.
6
(a) Tachometer signal
7
(a) Baseband Spectrum
The spectrum for the damaged pinion with Damage Severity Level 1 is illustrated in Fig.
12. Comparing Fig. 12(a) with Fig.11(a), notice that although the mesh frequency and its
harmonics are still clearly visible, there are more spectrum lines appearing. Most of the
spectrum lines can be attributed to the pinion speed sidebands around the mesh
frequencies and its harmonics which are illustrated by Fig. 12(b). By comparing Fig.
12(b) with Fig. 11(b), it is easy to see that there are more pinion sidebands appearing.
8
(a) Baseband Spectrum
The spectrums for the damaged pinion with Damage Severity Levels 2 through 5 are
illustrated in Fig. 13 through Fig.16. A careful study of these figures indicates that the
pinion sidebands become more and more dominant as the damage severity levels
increase. In particular, Damage Severity Levels 4 and 5 reveal that the mesh frequency
components are almost immersed by the pinion sidebands as illustrated in Fig. 15(a) and
16(a). And, Fig. 15(b) and 16(b) indicate the sidebands have higher amplitude than the
mesh frequency component under these same two Damage Severity Level examples.
9
(a) Baseband Spectrum
10
(a) Baseband Spectrum
11
(a) Baseband Spectrum
12
(a) Baseband Spectrum
13