Typical platform supply vessel.

The motor service company sent a technician on board with a CSI 2130 vibration data collector. This technician was trained to take measurements and provide basic analysis but not trained to do expert vibration analysis. The data collected on board in the coast of Africa was immediately transmitted to a specialized analysis team in Belgium to find out the source of the noise, and to determine if the severity of the problem would require a heavy and expensive maintenance action.

Figure 1. Propulsion system schematic.

Propulsion system consists of two air cooled DC motors with grease lubricated roller bearings. They are connected in series and coupled to a specific gearbox driving the propeller shaft (Fig 1). The gearbox is oil lubricated and is composed of 3 shafts with angular connections, the gearbox output shaft speed is approximately 180 RPM.

Before the arrival of the service technician, the ship owner had performed some preliminary trouble shooting tests and maintenance actions. The results indicated that the noise was present when the motors were connected to the gearbox, but it disappeared when motors were disconnected. The motors were completely checked and the bearings were changed without any significant changes in their performance. The coupling between the motors and gearbox was checked but no wear or fault was found.

Figure 2: Circular waveform of measured vibrations.

Analysis Process

The service technician used CSI 2130 A2 data collector with Machinery Health Manager 5.1 software for all on board vibration spectral measurements of the slow speed shafts. The shaft speed was measured with a Speedvue tachometer and all bearings were checked by spectral measurements and analysed remotely.

The vibration analysis of the DC motors did not indicate any significant defaults. There were no frequencies related to a bearing defect or lubrication failure. A frequency at 59.90 Hz was noticed on all collected data and an imbalance was noticed on the DC motor #2 cooling fan.

In the coupling between the motors and the gearbox no specific misalignment frequency was observed on measurements (no harmonics 1 and 2 of rotation speed). No coupling imbalance was noticed and the coupling had no significant looseness or wear.

The data collected from the gearbox showed a friction problem with a possible source from a lubrication problem, gear mesh friction or cavitation on the propeller (output shaft). The measurements also indicated a frequency at 59.90 Hz with sidebands.

The vibration analysis indicated that the motors and couplings were not generating the noise and vibrations. They were also visually checked and no signs of degradation were found.

Only two specific frequencies or phenomenon were noticed, friction and 59.90 Hz. The friction frequency may be caused by two root causes: problem in lubrication/gear mesh friction or cavitation of the propeller.

The frequency of 59.90 Hz, which was observed in all measurements, could be related to three root causes:

• Electrical frequency from motor or power drive.

The motor on vessels are powered by an exact 60Hz line frequency. The 59.90 Hz frequency is close to the motor drive frequency but not exact, and the conclusion was that the frequency was not generated by an electrical element.

• External vibration.

The measurements indicated imbalance in the cooling system of motor #2. The fan runs in theory at 3600 RPM (60 Hz) and the actual fan speed was measured to be 3562 RPM (59.38 Hz). This frequency was very different from 59.90 Hz and the conclusion was that the fan imbalance was not the source of this frequency.

• Gear mesh frequency in the gearbox.

Sidebands around the 59.90 Hz frequency were noted on some measurements. During spectral vibration analysis the sidebands are linked with the amplitude modulation of the vibration phenomenon as repetitions of the modulation frequency. In the vibration measurements the sideband had the same frequency as the rotation speed of the output shaft, and considering this, 59.90 Hz is the gear mesh frequency between the vertical and output shaft (2nd gear mesh), modulated by output shaft rotation speed.

The measured vibration amplitude was higher on the motor than on the gearbox. This was expected as the sensors were fixed on the motor bearings directly in contact with the shafts, whereas the sensors for gearbox measurements were fixed on the casing, far from gearbox shafts and bearings. Because of the long distance between the propeller shaft bearings and the sensors, the base frequency could not be detected as the amplitude was attenuated by the mechanical structure of the system.

Final Results

The vibration analysis confirmed that the motors were not deteriorated, the coupling between the motors and the gearbox was aligned correctly and was not the source of the vibration. Furthermore the input shaft of the gearbox did not show any sign of malfunctioning.

The following hypotheses were considered as the origin of the vibration and noise:

• Bearing problems on the output shaft of the gearbox including the propulsion system.
• Meshing problems between the output and the vertical shafts (second gear mesh in the gearbox).
• Deterioration of the propulsion system.

The recommendation was to check the bearings, the angle of the gearbox output shaft and the propulsion shaft at the next docking of the ship.

The results of the vibration analysis led to an expensive, one month maintenance work of the ship in a dry dock to replace the propulsion. After the maintenance work was completed, new measurements were done in the same conditions as before. The measurements indicated considerable decrease in the friction of the gearbox with no abnormal noise or vibrations.

The inspection of the replaced parts showed clearly a bearing defect on the propulsion shaft. Without the remote vibration analysis this defect could have led to a breakdown in the high sea with much more serious safety and financial consequences than the costs that incurred in this maintenance job.

Fabrice Brion

CEO, I-care Group

fabrice.brion@icareweb.com

The article was originally released in Maintworld 1/2013.