Wireless Vibration Monitoring: Sampling Specs Buyers Must Ask For

Start the enquiry with the failure question

Identify the asset, speed range, operating modes and the decision that should follow a changed reading. Screening for a change in overall behaviour, investigating a bearing problem and comparing vibration during variable-speed operation can require different evidence. Include the person who will interpret the measurements, because the report needs to support that person's actual review method.

Download the vibration monitoring specification worksheet. It asks each supplier to connect the failure question with sample rate, frequency band, measurement duration, waveform access, operating context and power assumptions. The response and review fields are blank. Use one row per asset group or monitoring question, then retain the agreed configuration with the purchase specification.

Separate the four timing fields

Sample rate is the number of samples acquired per second within a measurement. Record length is the number of samples or the acquisition duration. Measurement interval is how often the device performs that acquisition. Reporting interval is how often results are transmitted or made available to the application. A data sheet may use terms such as heartbeat for the last of these, so ask for the vendor's definitions.

A high sample rate can apply to a short measurement taken periodically. It says little by itself about how much of the machine's operating time was observed. Conversely, a frequent dashboard update may display a previously calculated feature. The quotation should state which timestamp belongs to the measurement and which belongs to receipt, especially if decisions depend on catching a brief event or comparing a machine change with process conditions.

Read an actual data sheet beyond its headline frequency

The public Monnit ALTA Advanced Vibration Meter data sheet provides a useful specification-reading example. Page 3 gives a frequency range of 0.4–4,200 Hz for the X and Y axes and 0.4–2,900 Hz for Z, with configuration-dependent ranges and further response notes. The same page's measurement-range table varies with sample rate and the selected vibration quantity.

Page 5 separately describes measurements of 64 acceleration samples per axis and explains measurement interval versus heartbeat. Its example uses a six-second measurement interval and a one-minute heartbeat, giving up to ten measurements per heartbeat. These are distinct properties of that documented device. Use the example to structure questions for the equipment in your own proposal, and obtain its corresponding configuration, data sheet and output evidence before comparing capabilities.

Ask for the usable band and filtering together

The highest possible sample rate is only part of the acquisition specification. The sensor response, filters, mounting and selected measurement mode determine what signals are usefully represented. Ask the supplier to identify the usable frequency band and how signals outside the intended band are treated. Also request the measurement range, noise information and the conditions under which overload or clipping is indicated.

Keep units and calculated quantities explicit. Acceleration peak, acceleration RMS, velocity RMS and displacement answer different questions and can use different processing. Ask whether a value is directly measured or derived, which band it covers and which axis it represents. Two systems can both display a vibration number while applying different definitions, making a comparison unreliable until those definitions are aligned.

Check the record duration against the machine behaviour

Record duration determines how much of the machine's motion is present in each acquisition. A low-speed machine may need a longer observation to include the relevant rotations. A short burst that works for a faster repetitive signal may provide little context for a slower or intermittent event. Ask the reviewing specialist to connect the required observation period to the actual speed and failure question.

For variable-speed equipment, retain speed or an appropriate operating-state reference with the measurement. Explain whether the proposed analysis compares similar running conditions or uses a method that accounts for changing speed. Include startup, shutdown and normal steady operation where those states matter. A single baseline assembled from mixed operating conditions can hide the distinction between a process change and a developing mechanical issue.

Distinguish calculated features from inspectable evidence

Ask for a sample of the actual data that the customer can export. A device may calculate useful features internally without exposing the original waveform to the user. The procurement question is whether the available record supports the intended review and any follow-up investigation. Have the reviewer examine an example rather than relying on a feature list or a dashboard screenshot.

Where waveform or spectrum access is required, specify the axes, units, acquisition settings, timestamps and metadata that must accompany the record. Confirm how many records are retained, how they are requested and what happens when storage fills. Where a summary trend is sufficient, document the feature definition and the route for obtaining additional evidence when the trend prompts investigation.

Evaluate battery life at the proposed workload

Ask for the battery estimate at the measurement mode, acquisition interval, reporting interval and environmental conditions in the quotation. Include the expected alarm traffic and any requests for additional records. Record the battery type, replacement method and the low-battery indication. The maintenance team needs to know what site work will be required and how the configuration is preserved after replacement.

The Monnit data sheet itself states that reporting frequency and other variables affect battery life. Treat a headline lifetime as an estimate with conditions that must be supplied, then compare the proposed maintenance task across options. More frequent acquisition and transmission may change the power requirement; the supplier should document the effect for the selected device and configuration rather than leaving the customer to infer it.

Specify the communications and recovery behaviour

For a LoRaWAN and 4G monitoring project, identify which features or records cross each part of the data path. A radio link that is suitable for scheduled summary measurements may require a different arrangement for additional diagnostic records. Agree the intended workload, retained history and the way an operator requests information from the installed system.

Test missing data as well as normal reporting. Record what happens during an approved source or communications interruption, whether recovered records keep their measurement times and how an unresolved gap appears to the user. The maintenance workflow should distinguish a stable trend from a stopped machine or a device that is no longer reporting. Include this behaviour in the acceptance record alongside the measurement specification.

Compare proposals with an evidence-based review

Use the completed worksheet to record unresolved questions and the evidence required before acceptance. The pump condition-monitoring service applies these choices to pump trains and their hydraulic context. The asset-monitoring overview and manufacturing page place them within the wider plant requirement. For the operating workflow behind the purchase, see the vibration project-scoping guide.

Frequently asked questions

Source and worksheet scope

The manufacturer data sheet, particularly pages 1, 3 and 5, was checked on 5 November 2026. It is cited as a concrete example of separate acquisition, measurement and reporting specifications. The worksheet is an engineering procurement aid; device selection and acceptance criteria are agreed for the actual installation.