Choosing the wrong sensor for a critical asset in a power plant does not produce a warning — it produces a gap in your data at the moment you need it most. Smart sensor selection is the foundation of any predictive maintenance program: the right vibration sensor on a turbine bearing shaft, the right temperature probe on a transformer winding, and the right ultrasonic detector on a pump seal can each cut unplanned downtime by 30–50% on that single asset. This guide walks maintenance engineers and reliability teams through sensor selection by equipment type — covering turbines, generators, pumps, transformers, conveyors, and boilers — so every monitoring point collects data that actually drives decisions. The framework maps directly to OxMaint's IoT sensor integration module, which connects sensor streams to automated work order triggers and real-time condition dashboards. Start your free OxMaint trial to connect your first sensor asset today, or book a live demo to see how sensor data drives predictive work orders.
IoT Sensor Integration · OxMaint
Smart Sensor Selection Guide
for Power Plant Predictive Maintenance
Turbines · Generators · Pumps · Transformers · Conveyors · Boilers
30–50%
Downtime reduction per asset with correct sensor selection
6 Types
Equipment categories covered in this guide
ISO 13373
Vibration monitoring standard referenced throughout
Real-Time
OxMaint work order triggers from sensor threshold alerts
Why Sensor Selection Determines PdM Program Success
A predictive maintenance program is only as good as the data flowing into it. Incorrect sensor placement, wrong measurement range, or mismatched technology for the failure mode being monitored creates blind spots that look like coverage on paper but deliver nothing in the field. The table below shows the failure mode mismatch risk by equipment type.
Wrong Frequency Range
An accelerometer rated to 1 kHz misses high-frequency bearing defects that appear at 5–20 kHz — the most critical early warning signal.
Wrong Mounting Location
A vibration sensor mounted 150mm from a bearing housing loses 60–80% of signal fidelity due to structural transmission loss.
Wrong Measurement Type
Using surface temperature instead of winding temperature on a transformer means detecting insulation failure 3–6 weeks later than necessary.
No Integration Path
Standalone data loggers with manual download cycles defeat the purpose of continuous monitoring — alerts arrive days after the failure mode develops.
Sensor Selection by Equipment Type
Equipment 01
Steam & Gas Turbines
High-speed rotating machinery · Most critical asset class in power generation
Failure Risk: Bearing Wear · Blade Erosion · Imbalance · Oil Degradation
PRIMARY
Eddy Current Proximity Probes
Shaft vibration — radial and axial displacement
Range:0–2 mm displacement
Standard:API 670 compliant
Mount Point:Bearing housing, X-Y orientation
Alert Trigger:>125 µm peak-to-peak (ISO 7919)
SECONDARY
High-Frequency Accelerometer
Bearing defect frequencies, blade pass frequency
Range:0.5 Hz – 20 kHz
Sensitivity:100 mV/g preferred
Mount Point:Bearing cap, stud-mount
Alert Trigger:ISO 10816-3 Zone C/D threshold
TEMPERATURE
RTD (Pt100) Bearing Temperature
Lube oil temperature, bearing metal temperature
Range:0–150°C
Accuracy:±0.5°C Class A
Mount Point:Bearing housing pocket, lube oil line
Alert Trigger:>85°C or 15°C rise from baseline
OIL HEALTH
Oil Particle Counter / Ferrography Sensor
Metallic wear particle count in lube oil circuit
Tech:Laser extinction or inductive
Output:ISO 4406 cleanliness code
Install:In-line on lube oil return line
Alert Trigger:ISO code >18/16/13
Equipment 02
Generators & Alternators
Electrical + mechanical failure modes · Requires both vibration and electrical sensing
Failure Risk: Winding Insulation · Rotor Eccentricity · Partial Discharge · Bearing Wear
PRIMARY
Partial Discharge Sensor (PD Coupler)
Stator winding insulation degradation detection
Tech:Capacitive coupler or HFCT
Sensitivity:1 pC detection threshold
Install:Cable termination / bus duct
Alert Trigger:Trending increase >20% month-on-month
VIBRATION
Piezoelectric Accelerometer
Rotor imbalance, misalignment, looseness
Range:1 Hz – 10 kHz
Sensitivity:50 mV/g
Mount Point:Drive-end and non-drive-end bearing
Standard:ISO 10816-1
THERMAL
Embedded Winding RTD (Pt100)
Stator winding temperature — per-phase monitoring
Range:0–200°C
Quantity:Minimum 6 per machine (2/phase)
Mount Point:Embedded between stator coils
Alert Trigger:>130°C Class F insulation limit
ELECTRICAL
Motor Current Signature Analyzer (MCSA)
Rotor bar cracks, air gap eccentricity
Tech:CT-based current measurement
Analysis:FFT of current spectrum
Install:On power supply cable — no shutdown needed
Alert Trigger:Sideband asymmetry >50 dB below fundamental
Equipment 03
Pumps (Feedwater, Circulating, BFW)
High-consequence failure · Seal failures cause environmental events
Failure Risk: Cavitation · Seal Failure · Bearing Wear · Impeller Erosion
PRIMARY
Ultrasonic Acoustic Emission Sensor
Cavitation detection — impeller and suction side
Range:20–100 kHz
Tech:Resonant AE sensor
Mount Point:Pump casing, suction-side
Alert Trigger:AE RMS rise >6 dB from quiet baseline
VIBRATION
Triaxial Accelerometer
Imbalance, misalignment, looseness — all axes
Range:2 Hz – 10 kHz
g Range:±50 g
Mount Point:Bearing housing, both ends
Alert Trigger:ISO 10816-7 Zone B/C boundary
SEAL HEALTH
Mechanical Seal Temperature Sensor
Seal face friction — early dry-run detection
Tech:Thermocouple Type K
Range:0–300°C
Mount Point:Seal flush piping / gland area
Alert Trigger:>80°C seal flush temperature
PROCESS
Differential Pressure Transmitter
Pump performance trending — head vs. flow curve deviation
Range:0–10 bar DP (application-specific)
Output:4–20 mA HART
Install:Suction and discharge tapping points
Alert Trigger:>5% deviation from duty curve
Equipment 04
Power Transformers
Long replacement lead times · Failure is a multi-week outage event
Failure Risk: Insulation Aging · Dissolved Gas · Cooling Failure · Bushing Degradation
PRIMARY
Dissolved Gas Analysis (DGA) Sensor
Incipient fault detection in transformer oil
Tech:Photoacoustic or electrochemical
Gases Detected:H2, CH4, C2H2, CO, CO2, C2H4
Install:Oil sampling valve — top of tank
Alert Trigger:C2H2 >1 ppm or Doernenburg ratio fault
THERMAL
Fiber Optic Winding Hot-Spot Sensor
Direct winding temperature — top oil and hot-spot
Range:0–200°C
Accuracy:±1°C
Install:Embedded in winding during manufacture or retrofit
Alert Trigger:>98°C top oil (IEC 60076-7)
BUSHING
Capacitive Bushing Monitor
Insulation power factor — bushing moisture ingress
Tech:Online capacitance and tan-delta
Frequency:Continuous, real-time
Install:Bushing tap — no outage required
Alert Trigger:tan-δ >0.5% or >20% change from baseline
COOLING
Oil Flow & Fan Current Monitor
Cooling bank performance — ONAF/OFAF systems
Tech:Magnetic flow meter + CT
Output:4–20 mA flow + digital fan status
Install:Oil cooler piping + fan MCBs
Alert Trigger:Flow <80% rated or fan current drop
Equipment 05
Coal Conveyors & Belt Systems
High availability requirement · Failure stops fuel feed to boiler
Failure Risk: Belt Slip · Idler Bearing Failure · Drive Gearbox Wear · Belt Misalignment
PRIMARY
Belt Speed & Slip Sensor
Conveyor throughput and belt slip detection
Tech:Encoder or proximity switch on head pulley
Output:Pulse count / 4–20 mA speed signal
Mount Point:Head and tail pulley comparison
Alert Trigger:>3% speed differential between head and tail
VIBRATION
Low-Frequency Accelerometer (Idler Survey)
Idler bearing defect frequencies — rolling element faults
Range:1 Hz – 5 kHz
Method:Walk-by or wireless IoT node per critical idler
Mount Point:Idler housing, radial direction
Alert Trigger:BPFI/BPFO spike >3x baseline RMS
ALIGNMENT
Belt Alignment Switch / Tilt Sensor
Belt edge tracking — prevent belt damage and spillage
Tech:Mechanical limit switch or MEMS tilt sensor
Install:Both sides of belt at 30m intervals on long runs
Output:Digital trip signal to DCS
Alert Trigger:Immediate — conveyor stop on sustained misalignment
GEARBOX
High-Sensitivity Accelerometer (Drive)
Gearbox mesh frequencies — tooth wear and lubrication failure
Range:1 Hz – 20 kHz
Analysis:TSA (Time Synchronous Averaging)
Mount Point:Gearbox housing — input and output shafts
Alert Trigger:Gear mesh frequency sideband >2x
Equipment 06
Boilers & Pressure Vessels
Safety-critical · Regulatory inspection requirement
Failure Risk: Tube Corrosion · Pressure Excursion · Refractory Failure · Drum Level Loss
PRIMARY
Guided Wave Ultrasonic (GWT) Sensor
Tube wall thickness — corrosion and erosion monitoring
Tech:Long-range guided wave UT
Coverage:Up to 50m of tube from single sensor
Install:Collar clamp on tube at header entry
Alert Trigger:>20% wall loss from nominal thickness
THERMAL
Tube Skin Thermocouple (Type K)
Superheater / reheater tube metal temperature
Range:0–1200°C
Mount:Weld-pad attachment or spring-loaded
Quantity:Per design — typically 20–40 per unit
Alert Trigger:>design metal temperature per tube spec
PRESSURE
Drum Pressure Transmitter (Smart HART)
Steam drum pressure — continuous recording per regulation
Range:0–250 bar (subcritical) / 0–300 bar (supercritical)
Accuracy:±0.075% of span
Standard:ASME PTC 4 compliant recording
Alert Trigger:>MAWP — direct safety interlock
LEVEL
Differential Pressure Level Transmitter
Drum water level — boiler safety critical
Tech:DP with compensated reference leg
Redundancy:2oo3 voting — 3 transmitters per drum
Install:Equalizing taps on drum — per ASME BPVC
Alert Trigger:High-high or low-low — direct boiler trip
Connect Every Sensor to Automated Work Orders in OxMaint
OxMaint's IoT sensor integration module receives threshold alerts from vibration, temperature, oil, and process sensors — and automatically generates prioritized work orders, assigns technicians, and logs every response with a timestamp. No manual data transfer, no missed alerts.
Sensor Selection Summary Table
| Equipment | Primary Sensor | Failure Mode Detected | Standard / Threshold | OxMaint Trigger |
|---|---|---|---|---|
| Steam / Gas Turbine | Eddy Current Proximity Probe | Shaft radial / axial displacement | API 670 / ISO 7919 | Work order on >125 µm |
| Generator | Partial Discharge Coupler | Stator insulation degradation | IEC 60034-27 | Alert on 20% PD trend rise |
| Feedwater Pump | Ultrasonic AE Sensor | Cavitation, impeller erosion | ISO 10816-7 | Work order on AE RMS >6 dB rise |
| Power Transformer | DGA Sensor (online) | Internal arcing, thermal fault | IEC 60599 / Doernenburg | Alert on C2H2 >1 ppm |
| Coal Conveyor | Belt Speed / Slip Sensor | Belt slip, throughput loss | Internal baseline | Alert on >3% head-tail differential |
| Boiler / Pressure Vessel | GWT Ultrasonic Sensor | Tube wall corrosion / erosion | ASME BPVC / PTC 4 | Work order on >20% wall loss |
IIoT Integration: Connecting Sensors to OxMaint
1
Sensor Installation & Calibration
Mount sensors per manufacturer specification and relevant standard. Record baseline readings for each asset at commissioning — stored in OxMaint asset profile.
2
Edge Gateway / IoT Hub
4–20 mA, HART, Modbus, or wireless (WirelessHART, ISA100) signals aggregated at plant-level IoT gateway. OxMaint connects via REST API or MQTT broker.
3
Threshold Configuration in OxMaint
Set alert thresholds per sensor and asset. Thresholds linked to work order templates — OxMaint auto-assigns technician and priority on alert fire.
4
Condition Trending & PdM Dashboard
Every sensor reading logged against asset history. OxMaint trend charts flag deteriorating assets before threshold breach — plan maintenance proactively.
Frequently Asked Questions
What vibration sensor standard applies to power plant turbines and pumps?
ISO 10816 covers vibration measurement on non-rotating parts (bearing housings), while ISO 7919 covers shaft-relative displacement. API 670 defines the machine protection standard for turbines and compressors. OxMaint alert thresholds map directly to ISO 10816 Zone C/D boundaries for pump and turbine assets, giving your PdM program a defensible engineering basis.
Can OxMaint receive data from existing plant DCS or SCADA sensors?
Yes. OxMaint integrates with plant historian systems and IoT gateways via REST API, MQTT, and OPC-UA connectors. Sensors already wired to your DCS can feed condition data into OxMaint without rewiring, enabling work order automation from existing infrastructure. Book a demo to discuss your plant's specific integration architecture.
How often should vibration baselines be re-established on rotating equipment?
Baselines should be re-recorded after any major maintenance event — bearing replacement, coupling change, impeller replacement, or rewind — since the mechanical signature changes. OxMaint work order completion prompts technicians to re-record baseline readings so trending always compares against the correct post-maintenance condition.
Is online DGA monitoring worth the investment for power transformers?
For any transformer above 20 MVA or feeding critical auxiliary busses, online DGA pays for itself after preventing a single unplanned failure. Transformer replacement lead times of 12–24 months mean that an undetected internal fault results in an outage measured in months, not days. Online DGA provides the earliest possible warning. Start tracking transformer DGA alerts in OxMaint free.
What communication protocol should I use for wireless sensors in a power plant?
WirelessHART (IEC 62591) is the most widely adopted standard for industrial wireless sensors and is suitable for most power plant monitoring points. ISA100.11a is an alternative for applications requiring lower latency. Both are supported by OxMaint's IoT integration module via compatible gateway hardware from major vendors.
Build Your Sensor-Driven PdM Program with OxMaint
From single asset monitoring to plant-wide IIoT integration — OxMaint connects sensor data to maintenance workflows, work orders, and compliance records. Most plants complete their first sensor-to-work-order integration within one week of sign-up.
