Efficient Plant April 2021 - 15

feature | condition monitoring
protocols. Each condition-monitoring module has four ports
for connecting IEPE vibration sensors and these modules
support characteristic-based and frequency-selective vibration analysis, which can be streamed or exported for further
analysis using outside tools or cloud services.
Characteristic-based analysis is, in essence, machine health
monitoring using root mean square analysis of mechanical
velocity and acceleration data. Such analysis is easily configured and monitored for bearings and other components, and
alerts can be generated when movements exceed manufacturer recommendations. While this is a great starting point
for effective predictive maintenance, characteristic-based
analysis alone does not reveal what type of vibration damage
is occurring.
To discover the type of damage, frequency-selective analysis can be applied by breaking up the net vibration frequency
recognized in characteristic-based analysis into its contained
collection of vibration signals. These signals are superimposed over their respective frequency spectrum using a Fast
Fourier Transformation (FFT) algorithm, with the CMS
performing this mathematical operation internally.
The result of the FFT calculations reveals the following
types of bearing damage:
 unbalance eccentricity
 misalignment soft foot
 rotating looseness
 rotor rubbing
 vane faults
 electrical field faults
 gear meshing faults
 roller bearing noise.
The FFT analysis can also be configured and visualized through
integrated webservers.

RESULTS
For a manufacturer using a CMS on a milling machine, early warning
of bearing failure empowered them to minimize downtime. As later
snapshots revealed, the machine was experiencing increasing vibrations over time, but problems were undetectable by human sensory.
When vibration reached the preconfigured maximum, the CMS
generated an alert to notify staff.
Lead time on a replacement bearing was several months, but the
early warning provided time to continue production while waiting.
Once the replacement part arrived and was installed, they found
considerable damage on the old bearing. The bearing would have
probably failed completely if left unmitigated, so early detection may
have spared the manufacturer from months of unplanned downtime.

APRIL 2021

EP2104fsiemens.indd 15

Above. An integrated CMS, such as a
Siemens SIPLUS CMS 1200, allows modular expansion and simple integration
into automation systems.
Left. The Siemens SM 1281 condition-monitoring module has four ports
for connected vibration sensors. Data is
accessible through a web interface using
a browser from any networked device.

FUTURE OF PREDICTIVE MAINTENANCE
Early detection of mechanical and electrical damage is one of the
most effective means for reducing unplanned downtime and an
integrated CMS can provide transparency into machine health to
make this possible. Furthermore, these notifications assist operations,
management, maintenance, resource planning, and engineering staff
to perform their jobs more effectively by forecasting machine failure.
The future of condition monitoring will expand upon predictive
maintenance by performing prescriptive maintenance to provide
additional insight into the cause of problems, along with suggested
mitigation remedies. Edge- and cloud-based solutions will likely
also play larger roles as machine-learning algorithms become more
refined. EP
Luis Narvaez is the S7-1200, S7-1200F, and SIPLUS CMS product
manager for Siemens Industry, Washington, DC (siemens.com).
Narvaez has more than ten years of industry experience in various
roles within the construction, entertainment/theme park, manufacturing, and process industries.
EFFICIENTPLANTMAG.COM |

3/26/21 8:39 AM

http://www.siemens.com http://www.EFFICIENTPLANTMAG.COM

Efficient Plant April 2021

Table of Contents for the Digital Edition of Efficient Plant April 2021