Continuous vs. Periodic Vibration Monitoring at Power Plants: A Comparison | Bently Nevada — KEG TRK

A power plant's chief mechanic's office rarely has the budget for continuous vibration monitoring on every pump and fan in the unit. The realistic choice is a combination of continuous monitoring on critical units and a periodic route on the rest of the fleet. The question isn't "which is better," but where each method is economically and technically justified.

Bently Nevada offers the full spectrum: the 3500 rack and System 1 for continuous protection, Orbit DCM for the mid tier, SCOUT for route-based collection. Below is a comparison of these approaches as applied to auxiliary equipment at thermal power plants (TPPs, state district power plants, and large industrial boiler houses).

Definitions

Approach	Essence	Typical Bently tool
Continuous (online)	Sensors 24/7, alarms, relays	3500, Orbit DCM, 3500/42M
Periodic (offline)	Measurement on a route schedule	SCOUT, Ranger Pro
Hybrid	Continuous on Class A, route on B/C	All of the above + System 1

Criteria for choosing a method

1. Criticality to unit load

• A shutdown immediately reduces or stops the unit → continuous monitoring.
• A backup unit with automatic switchover exists → periodic monitoring with regular rounds is acceptable.

Example: a feedwater pump with no backup → 3500/42M; a circulating pump with backup → a monthly SCOUT route.

2. Speed of typical defect development

Defect	Time to failure	Minimum offline interval
Pump bearing	2–8 weeks	2 weeks
Fan imbalance	1–3 months	4 weeks
Gearbox (GMF)	2–6 months	4 weeks
Loose anchor bolts	days–weeks	Online only

Fast-developing mechanisms (loosening fasteners after a post-repair restart) are not caught by a monthly route.

3. Downtime cost vs. monitoring cost

A simplified decision formula:

If (downtime cost × annual failure probability) > (online monitoring cost × 3) → install continuous monitoring.

For a 200 kW TPP auxiliary pump, a 4-hour unit shutdown from its failure often outweighs 3–5 years of Orbit DCM operation on that pump.

4. DCS and regulatory requirements

Some sites require relay-based shutdown on vibration through the DCS — this is only possible with continuous monitoring and a dry contact (3500, Orbit DCM). A SCOUT route provides no automatic protection.

5. Existing Bently infrastructure

If a turbine generator already has 3500 and System 1, adding 3500/42M channels for auxiliary machines is cheaper than building a separate ecosystem.

Comparison table

Parameter	3500 / 3500/42M	Orbit DCM	SCOUT (route)
Capital cost	High	Medium	Low
Operating effort	Low	Low	Round-based labor
Reaction to failure	Minutes	Minutes	Days–weeks
Relay protection	Yes	Yes	No
History in System 1	Native	Integrated	Trendmaster / import
Coverage of 100 points	Expensive	Selective	Realistic
Spectrum on alarm	Yes	Yes	Every reading

Recommended model for a TPP

Class A — continuous 3500/42M or Orbit DCM

• Feedwater pumps with no backup;
• ID fans with no backup;
• Turbine lube-oil pumps;
• Any unit with a history of causing an unplanned unit shutdown.

Class B — Orbit DCM or a 2-week route

• Pumps with backup but a long switchover time;
• High-pressure heater fans;
• Coal mill drives.

Class C — SCOUT route every 4–8 weeks

• Cooling-tower auxiliary pumps;
• Small room ventilation fans;
• Units with full redundancy and fast switchover.

Hybrid scenario: a typical 150–300 MW unit

1. Already in place: 3500 on the turbine — add 12–20 channels of 3500/42M on pumps and ID fans.
2. Orbit DCM on a pump station of 6 circulating pumps (one panel).
3. SCOUT — a 180-point route, twice-monthly rounds, a one-engineer crew.
4. System 1 — a single point for trending continuous channels; reporting for the chief mechanic.

On a SCOUT alarm → an in-depth spectrum; on a repeat alarm → a decision to move the point to Orbit DCM.

Common mistakes when choosing

1. "Route only" across the entire fleet — missing fast failures on Class A.
2. "3500 on every pump" — unjustified CAPEX on redundant equipment.
3. No unified trend database — continuous and route data live in separate systems with no comparison.
4. ISO thresholds without a baseline — false alarms on fans with high background vibration.

Training and competencies

Periodic monitoring requires qualified staff: reading an FFT spectrum, knowing BPFO/BPFI, understanding bearing failure stages. Continuous monitoring reduces how often manual analysis is needed, but doesn't eliminate the need to investigate an alarm.

KEG TRK provides on-site training and helps develop a "unit — method — interval" matrix.

Scope of this article

This article covers thermal power plants and industrial generation within KEG TRK's supply scope. Nuclear power, oil refineries, and petrochemicals follow separate regulations and equipment lists and are not covered here.

Example matrix for a 240 MW unit (fragment)

Unit	Backup	Method	Interval / channels
Feedwater pump P-1	None	3500/42M	4 channels, 24/7
Feedwater pump P-2	Backup for P-1	3500/42M	4 channels, 24/7
ID fan D-1	None	3500 + 3300 XL	8+ channels
Circulating pump	Yes	Orbit DCM	4 channels
HP heater fan	Yes	SCOUT	2 weeks
Cooling-tower pump	Yes	SCOUT	1 month

The matrix is approved by the station's chief engineer and reviewed every 3 years or after a major failure.

Total cost of ownership (TCO) over 10 years

A simplified comparison per measurement point:

Method	CAPEX	OPEX/year	10-year total*
SCOUT (route)	Low	Round labor	Medium
Orbit DCM	Medium	Low	Medium–high
3500/42M	High	Low	High

* Excluding the cost of a single prevented unit shutdown, which can outweigh the difference by an order of magnitude.

The decision point isn't minimum TCO but minimum risk for units with no backup.

System 1's role as a single pane of glass

Even when part of the fleet is on SCOUT and part is on 3500/Orbit DCM, importing route data into System 1 (or a unified report) spares the chief mechanic from juggling "two worlds." KEG TRK configures Trendmaster → System 1 import at sites with mixed architectures.

Regulatory and insurance requirements

At a number of TPPs in Kazakhstan, insurers and auditors require documented vibration monitoring of critical equipment. A SCOUT route log signed by the engineer is the bare minimum; for an ID fan or feedwater pump with no backup, an insurance claim often requires continuous protection with a relay — an argument in favor of 3500/42M.

18-month implementation roadmap

1. Months 1–3 — inventory, SCOUT route across the entire fleet, baselines.
2. Months 4–6 — Orbit DCM on 2 pump groups (pilot).
3. Months 7–12 — 3500/42M on feedwater pumps and the ID fan (if a rack already exists).
4. Months 13–18 — scaling, System 1 integration, KPIs and reporting.

Auditing an existing program: 10 questions for the chief mechanic

If the TPP already has both a route and separate vibration-protection panels, check:

1. Is there a single list of units with an assigned method (online/offline)?
2. Has the matrix been updated since the last major failure?
3. Do the thresholds match the actual baseline, or were they copied from ISO?
4. Is RPM recorded during route readings on VFD-driven equipment?
5. Is there a "failures prevented" KPI, not just a "readings taken" count?
6. Is the standby shift trained to respond to a 3500/42M alarm?
7. Is SCOUT data imported into the same system as 3500 data?
8. Are backup pumps monitored, not just the duty units?
9. Is the relay logic coordinated with the DCS without conflicting rules?
10. Is a spare-parts budget allocated for Velomitor sensors and 3300 XL probes?

Positive answers to 7+ items indicate a mature hybrid program; fewer than 5 is a reason for an audit with KEG TRK.

Connection to the broader downtime-reduction strategy

Hybrid vibration monitoring at a TPP is part of a broader program for reducing unplanned downtime and maintenance strategies. The economic effect compounds: fewer unplanned load reductions, fewer emergency bearing purchases, less overtime for repair staff. The station's chief engineer rolls the vibration program's KPIs into the annual report to shareholders or the regulator — "number of auxiliary-equipment failures prevented" becomes a measurable maturity indicator for maintenance, rather than a declaration that "predictive maintenance has been deployed."

Frequently asked questions from TPP staff

Can we get by with just SCOUT if the budget is tight? Yes, for Class B/C. For feedwater pumps and ID fans with no backup (Class A) — no; a route interval doesn't cover the risk.

Do we need 3500 if Orbit DCM is already in place? Not necessarily. Orbit DCM provides relays and trends. 3500/42M makes sense when a turbine rack already exists and a unified API 670 standard is required.

How often should the matrix be reviewed? Every 3 years, and after every major auxiliary-equipment failure.

Who should analyze the spectrum on an alarm? A vibration analyst or a trained chief mechanic; the shift operator's role is limited to confirmation and escalation.

The bottom line: don't choose a single method for the entire fleet. Use 3500/42M where a rack already exists and criticality is highest; Orbit DCM for pump groups; SCOUT for the rest. KEG TRK will calculate the right balance for your unit within 1–2 weeks after a site survey.

The same selection criteria apply to own-needs units at large metallurgical plants (boiler houses, shop-level TPPs) as at a municipal TPP — see also 3500/42M auxiliary units for a consistent monitoring architecture across an entire holding company's sites.

An annual review of the hybrid program takes 1 working day given an up-to-date matrix: reconciling the unit list with the DCS, analyzing false alarms over the year, proposing 3–5 points to move between SCOUT and Orbit DCM. KEG TRK includes such a review in the System 1 support contract — cheaper than one-off consultations after every major failure.

Summary

The optimal strategy at a TPP is a hybrid: 3500 and Orbit DCM on critical auxiliary equipment, SCOUT on the rest of the fleet, unified analytics in System 1. KEG TRK helps calculate the right balance of continuous and periodic monitoring for your unit.

KEG TRK contacts | Power-generation solutions.