SmartCheck FAQs and HowTos: Analytics

Yes, SmartCheck can monitor multiple components. You can define multiple measurement jobs, which will be measured sequentially. Each job can monitor a different component with its own configuration. As long as SmartCheck can “see” the vibration of the component with its vibration sensor, it can monitor it. For example, if you have a machine with multiple bearings, you can place the sensor on the housing and configure multiple jobs to monitor each bearing individually.

There is no fixed limit on the number of configurations or components that SmartCheck can monitor. However, we advise to keep the number of configurations reasonable, say 5-10 per sensor position, since each configuration adds complexity to the setup and maintenance of the system. Also, the more configurations you have, the longer it will take for SmartCheck to cycle through all the measurements, which might affect the timeliness of the data.

The are several ways to analyze data with SmartCheck:

1. Check the measurement data in the web interface of SmartCheck. Here you can view the trends of all measured characteristic values. There is also is a basic viewer, in which you can view the time signals, both in the time domain and as spectrum. This view offers some basic tools like a cursor to measure times and frequencies, and the possibility to show the damage frequencies of components defined for the measurement configurations, like bearings.
2. Use the viewer in the SmartUtility software. This can also show trends and time signals, but offers more tools for analyzing the data, e.g. multiple cursors or a waterfall plot.
3. Use the ExpertViewer software. This offers the same functions as the SmartUtility viewer.

With SmartCheck, it is possible to monitor machines that operate at variable speeds, as long as the speed doesn’t change too much during the measurement. This would result in a smeared frequency spectrum, making analysis difficult. If you have such an application, e.g. a wind turbine, your should consider to use the ProLink device, as this supports the order analysis method, which is perfectly suited for such applications.

Monitoring machines with variable speed or load can be challenging, as traditional vibration analysis methods often rely on constant operating conditions. However, in SmartCheck you can use several mechanisms to effectively monitor such machines:

• Most measurement templates use speed dependent KPIs. For this a speed input has to used. The speed can be provided by a tachometer input (pulses), from an analogue speed input or from an external source like a fieldbus.
• A machine running at a higher speed or a higher load will usually have higher vibration levels. For this reason, the alarm levels should be adapted to the current operating conditions. In SmartCheck this can be done by using alarm maps. These can be configured in the measurement job configuration wizard in the “Set alarms” step. More information about configuring speed dependent alarm limits can be found in the manual

There are several ways to provide a speed signal to SmartCheck for its measurement jobs:

• When you know the speed of the machine, you can enter it manually in the input configuration as an “Input with a fixed value”. In the measurement jobs, this can then be used as any other speed input. Should the speed of the machine change (e.g. when there is a change in the setup of the machine), you can edit this fixed speed input value accordingly.
• Use a digital input when you have a pulse signal from a sensor like a tacho probe, e.g. one pulse per revolution or multiple pulses per revolution with a pulse wheel.
• Use an analogue input when you have a 4-20mA or 0-10V signal representing the speed of the machine, originating from an analogue speed sensor or an analogue output from a PLC.
• Use a communication channel, when the speed is provided by a PLC or another device via a network like SLMP, OPC/UA or PROFINET.

When you have multiple speeds depending on each other, e.g. in gearboxes or belt drives, you can add scaling factors to a speed input, directly in the input configuration. This way you can calculate the speed of a shaft based on the speed of another signal. These scaling factors can be used in the measurement configurations as if they were real speed inputs.

The alarm status of SmartCheck can be signaled in different ways:

• Via the digital outputs of the device. These can be configured to signal the pre-alarm and/or main alarm status.
• Via the status-LED on the front panel.
• Trigger an email notification.
• Signal via a communication channel, like SLMP or OPC/UA.

When a new measurement job is configured, default alarm limits are set. With the learning mode, the device can automatically set the alarm limits for the current measuring point and the current state of the machine. With the default settings for the learning mode, this will get good results for most cases.

What the learning mode does, after it is started for the measurement job, or for the entire device, is to collect trend values for each of the KPIs in the measurement job. By default, it needs 1000 values for this. When this number is reached, from these 1000 values a kind of average value is calculated, i.e. the normal value of the KPI in the current state of the machine. The alarm limits are then changed to be a bit higher than this value. By default, the pre-alarm is set to this value times two, the main alarm limit to this value times three.

It is possible for the user to change the settings for the learning mode. To do this, open the wizard for the measurement job, enable the expert settings and navigate to the fourth step. This shows the following settings:

Here you can change the following:

See the manual for more information on the learning mode.

The measurement time of a measurement job depends on several factors. In general, each recorded time signal has a duration which is defined by the following settings:

• Lowpass filter (e.g. 1kHz, 5kHz, 10kHz, …): The higher the lowpass filter, faster the signal is sampled and thus the faster the measurement is completed. The sampling rate is the lowpass frequency multiplied by 2.56.
• Number of FFT lines: this translates directly to the number of samples recorded for each time signal. In SmartCheck, the number of samples equals the number of lines multiplied by 2.56. For example, if you use 1600 lines, the measurement
  will record 4096 samples (1600*2.56) for each time signal.

The measurement time of a time signal is therefore calculated as follows:

Measurement time [s] = number of lines / lowpass frequency

The total measurement time of a measurement job is defined by the longest measurement time of a time signal. Since all time signals in a measurement job have the same setting for the number of FFT line, the job’s measurement time is defined by the time signal with the lowest lowpass filter. To this time, add 30 seconds distance between two measurement jobs.

Instead of calculating the measurement time manually, you can also see the measurement time directly in SmartWeb in the upper right corner of the status page for all measurement jobs:

A measurement condition can be used to enable measurement jobs only under certain conditions. For example, you can set a measurement condition to only enable a measurement job when the machine is running in a certain speed range. With a measurement trigger, you can start a measurement when a certain event occurs, for example when the speed exceeds a defined value. While this sounds straightforward, in practice the machine’s operating conditions can be instable. For example, the machine might be running in the defined speed range but with a lot of fluctuations.

When a measurement condition or trigger becomes instable, i.e. switching between valid and invalid very rapidly, this would start measurements which would be cancelled right afterward due to the condition becoming invalid. To the user, it might not be clear why no measurement is stored. Additionally, in the trend of the measurement condition and trigger, there would be many alarm state changes that would only be visible when zooming the time period as they are so close to each other. This could make it hard to read the trend and understand the machine’s operating conditions.

To prevent this, ProLink will detect an instable measurement condition or trigger and disable it until it becomes stable again. Both the beginning and ending of the detected instability will be written to the logbook and be visible with a gray background in the measurement condition’s or trigger’s trend.

When the turning speed is measured in a measurement job, the user can choose between Hz and orders in the SmartUtility Viewer and in the ExpertViewer. In SmartWeb, this is only possible for measurements with activated order analysis.

The minimum interval between two trend points depends on several factors, like the duration of the measurement jobs and the number of measurement jobs. The device will measure each measurement job after the other, with a waiting time of 30 seconds after each job. When there is only one measurement job, and the job is configured to store trend values as often as possible, you can get down to a minimum interval of around 7 seconds between two trend points when using the base measurement job.

The SmartCheck can monitor a wide range of roller bearings from any manufacturers. It has a bearing database which already includes all roller bearings from Schaeffler, including the brands FAG and INA. Unfortunately, Schaeffler is not allowed to provide bearing data of other manufacturers because of legal restrictions.

But bearings of any manufactured can be added manually, via Settings –> Bearings. Here you can enter a new bearing with its kinematic frequencies, which usually can be found in the datasheet provided by the manufacturer. Another way is to calculate these from the mechanical data of the bearing. A calculator for this is also included in the window when entering a new bearing.

SmartCheck provides templates for measurement jobs, which are defined to suit different types of machines and applications. Choosing the right measurement job type is crucial for accurate data collection and analysis. Here is the list of our measurement job templates and their recommended use cases:

With SmartCheck, you can monitor all sorts of mechanical components like bearings, gears, fans, pumps, belt drives, etc. The device comes with a configuration assistant, which already includes templates for most components. This makes it very easy to configure the measurement configuration for a specific component, since the wizard will ask you all the necessary questions and set up the measurement parameters accordingly.

The SmartCheck can monitor not only static machines. It can be set up to monitor machines with variable speeds and loads as well. Also special applications, for example machines which only run for a short time and then stay idle for a long time, are no problem. With the right input signals and configuration, SmartCheck can handle these applications as well.

This depends on the application and the specific use case. In general, SmartCheck can be used for a wide range of speeds, typically between 50 and 15,000 revolutions per minute (RPM).