Fail-Safe by Design: Applying the Open-Circuit Rule to Bently Nevada Vibration Trips
News

Fail-Safe by Design: Applying the Open-Circuit Rule to Bently Nevada Vibration Trips

A broken wire is a foreseeable fault, not a rare edge case. A protection circuit should detect it and move to the state required by the machine's hazard analysis. This review draws on InstrumentationTools' “Basics of Fail-safe Circuits” and applies its open-circuit lesson to Bently Nevada 3500 machinery monitoring. The required response is not automatically an immediate trip in every installation.

The principle behind open-circuit detection

In a simple energize-to-run, de-energize-to-trip circuit, loss of power or an open series path can release a relay and initiate a protective action. That arrangement can make certain wiring failures visible instead of silently preventing a demand. But short circuits, stuck contacts, common power failures, incorrect wiring, and bypasses require separate analysis. The safest state must be defined for the particular machine and process, then verified throughout the complete sensor-to-final-element path.

From probe to monitor to trip relay

A proximity measurement passes through the probe, extension cable, signal-conditioning hardware, monitor channel, configured alarm logic, relay output, and downstream shutdown system. A Bently Nevada 3300 XL 8 mm proximity probe, 3500/42M monitor, and 3500/33 relay module illustrate different parts of that chain. Their catalog listings do not establish how a particular rack handles Not OK, relay energization, voting, bypasses, or loss of rack power.

Check the installed configuration: what declares a channel Not OK, how it is annunciated, whether it is inhibited from voting, and what the output relay and downstream circuit do. A probe fault must not be mistaken for a valid low-vibration reading. Equally, it is unsafe to claim that every cable break must trip the machine without checking the approved cause-and-effect design.

Reduce nuisance trips without masking danger

Investigate each unwanted trip using sensor gap or bias data, wiring condition, connector integrity, startup transients, alarm setpoints, diagnostics, and event timestamps. Voting may reduce certain single-channel false trips, but it also changes dangerous-failure probability, common-cause exposure, and degraded-mode behavior. Select voting only through a documented protection study; a universal 2oo3 recommendation is not appropriate for every machine.

Proof-test the actual failure response

Safety note: Never unplug a live probe or break a running machine's trip circuit solely to follow a generic test guide. Use an approved procedure, permit, operations coordination, and suitable simulated inputs or isolated test facilities.

  1. Review the approved cause-and-effect, drawings, monitor configuration, relay logic, bypass status, and permitted test boundary.
  2. Record baseline channel health, vibration values, relay states, and operator indications.
  3. Under the approved test method, simulate the specified sensor fault and overrange conditions separately.
  4. Verify channel status, annunciation, voting, relay output, and downstream response against the design. Where a full shutdown cannot safely be exercised, document the tested boundary and remaining test scope.
  5. Restore all test connections, clear authorized bypasses, and confirm live channel readings and protective readiness.
  6. Record response times, settings, discrepancies, witnesses, and corrective actions.

Govern every bypass

A bypass used for a test must have an owner, reason, authorization, visible status, expiry, compensating measure, and restoration verification. Do not assume same-shift clearance is possible for every repair, but require any extension to be reviewed and approved. Keep drawings current so technicians can tell a normally energized healthy circuit from an abnormal energized fault, and do not infer state from a single voltage measurement.

Conclusion

Open-circuit-aware design makes certain failures detectable and can drive the intended protective response. In a Bently Nevada 3500 installation, the outcome depends on the configured channel diagnostics, voting, relay behavior, and downstream shutdown scheme. Prove that chain with controlled tests and manage every bypass so protection is neither silently lost nor needlessly tripped.

Link copied