Field Case Study: Troubleshooting a Differential-Pressure Flow Meter Loop - Just Measure it

Field Case Study: Troubleshooting a Differential-Pressure Flow Meter Loop

1. Executive Summary

A differential-pressure (DP) flow loop on a process line suddenly reported PV=0 and lost local display. The instrumentation team followed a “electrical chain first, then process side” approach. Measurements isolated the fault to a non-standard cable splice ~4–5 m upstream of the transmitter that had oxidized and opened the loop. After corrective re-termination (cold-pressed ferrules + dual-wall adhesive heat-shrink + IP67 gland), the loop recovered and remained stable. This case highlights the risk of outdoor in-tray splices and formalizes a preventive inspection plan.

2. Background and System Description

  • Tag/Loop: 42FIT-302B (AI-302B-01)

  • Measured Variable: Volumetric flow via primary element (DeltaBar/DP primary device) + secondary DP transmitter

  • Medium/Service: Process liquid (clean to mildly fouling)

  • Design Range: Sized for normal operating flow; DP transmitter span configured per orifice/DeltaBar calculation (details in P&ID)

  • Power/Signal: 24 VDC two-wire, 4–20 mA + HART, single-ended shield grounding

  • Impulse Lines & Manifold: Two SS impulse lines; standard three-valve manifold (H/L isolate + equalize)

  • Environment: Outdoor tray routing, UV/rain exposure; junction box (JB) and transmitter at grade

  • Recent Work: No transmitter change; suspected historical field splice during minor routing change

3. Failure Symptoms

  • Timestamp (UTC+8): 2025-08-10 09:33–09:35 incipient fluctuation; 09:35 PV dropped to zero

  • DCS Alarm: BAD PV / Lo-signal

  • Local Indicator: Transmitter display off

  • Operator Observation: No apparent process upset; control valve steady, upstream/downstream pressures within normal band

4. Safety & Work Permits

  • Work under valid hot/cold work permit and LOTO as needed.

  • Verify loop service is non-hazardous; if hazardous, use SIS/MOC protocol and gas test.

  • Ladder/height PPE, cut-resistant gloves, eye protection.

  • Confirm control strategy in MANUAL before equalizing/closing manifold valves.

5. Diagnostic Methodology and Evidence

5.1 Electrical Chain (Power/Signal)

Objective: Determine where 24 VDC is lost.

CheckpointTestObserved ValueInterpretation
Control room isolatorLoop VDC24.0–24.1 VOK (supply healthy)
JB#42-2 (X1 +/-)Loop VDC23.8 V (stable)Power present at JB
Transmitter terminalsLoop VDC0.6 V (unstable)Voltage collapse between JB and TX
Transmitter outputLoop mA≈0 mAConsistent with PV=0 / no power

Finding: Voltage drop exists between JB and transmitter.

5.2 Cable/Continuity & Shielding

  • Visual on tray section #C-7: found non-standard inline splice (~4.3 m from TX). Copper strands blackened/oxidized, heat-shrink brittle, no sealing gland, no drip loop.

  • Continuity test across splice showed intermittent open under light bending.

5.3 Process Side (Impulse & Manifold)

  • Prior to electrical repair, manifold left isolated; no attempt to force PV.

  • After restoration (see §6), manifold operations conducted (see Appendix B). ΔP and PV corresponded to expected flow, confirming no impulse blockage.

6. Corrective Actions (As-Built)

  1. Remove corroded splice and recover clean conductor to bright metal.

  2. Re-termination: cold-pressed ferrules; dual-wall adhesive heat-shrink (2× overlap) over joint.

  3. Ingress protection: added IP67 inline gland; formed drip loop; secured to tray with UV-rated ties.

  4. Mechanical QC: Pull test ≥ 30 N per conductor; visual check for full barrel fill, no stray strands.

  5. Shield & Ground: Shield continuity preserved; single-point ground at JB maintained.

  6. Documentation: Marked tray section C-7 and JB X1; updated loop drawing and cable schedule.

7. Verification & Restoration

  • Post-repair readings (power off process equalized): 24.0 V at TX; loop current ~4.0 mA (LRV).

  • Return to service (manifold per Appendix B):

    • Equalize closed → H/L isolates opened: PV climbed smoothly with process.

    • DCS trend stabilized; loop current ~12.1 mA at observed flow, PV = expected ±0.3%.

  • 24-hour soak: No alarms, no drift; no moisture ingress observed at new joint.

8. Root Cause

Outdoor, non-standard inline splice in cable segment (JB→TX) suffered environmental ingress (rain/condensation) and conductor oxidation, leading to an intermittent open circuit and loss of loop power.

Contributing factors

  • No IP-rated enclosure or sealing; no adhesive heat-shrink.

  • No drip loop; water path along cable sheath.

  • Splice location unmarked; not recorded in drawings.

9. Preventive Actions & Lessons Learned

9.1 Wiring & Materials Standard

  • Prohibit open splices in tray. Use IP67 junction box + terminal strip when extension is unavoidable.

  • Conductors: tinned, fine-strand copper; cold-pressed ferrules; torque per terminal spec; dual-wall heat-shrink for all field joints.

  • Routing: enforce drip loops, UV-rated ties, bend radius ≥ 8× OD.

9.2 Inspection & Testing

  • Quarterly sampling of outdoor analog loops (≥20%):

    • Insulation ≥ 20 MΩ @ 500 V (de-energized)

    • Loop resistance baseline vs. design

    • Shield continuity & single-point ground check

  • Add pull-test and photo record to closeout checklist.

9.3 Process/Operations

  • Maintain a laminated Three-Valve Manifold Card at the transmitter (Appendix B).

  • Capture before/after trends (PV/mA/status) and attach to the work order.

9.4 Documentation & Training

  • Update loop drawings, cable schedules, tray section IDs.

  • Toolbox talk: “Why open splices fail outdoors” (moisture + capillary action + galvanic/oxide growth).

10. Appendices

Appendix A — DP Primary Element Quick Notes

  • Straight-run: follow vendor/ISO guidance (typical: ≥10D upstream, ≥5D downstream unless straightener used).

  • Tappings: high-pressure tapping upstream, low downstream; keep impulse lines short, level, and heat-traced if needed.

  • Mounting: orient transmitter to minimize gas pockets (liquid service) or condensate pools (gas/steam with seals).

Appendix B — Three-Valve Manifold Operation Card

To return to service after maintenance

  1. Verify downstream isolation closed; transmitter power healthy.

  2. Ensure equalize OPEN, H & L isolates CLOSED.

  3. Slowly OPEN H isolate, then SLOWLY OPEN L isolate (monitor PV rise).

  4. CLOSE equalize.

  5. Check leaks; confirm PV stable and matches expected.

  6. Hand back to operations; restore control mode.

Reverse sequence for safe isolation: H & L CLOSE → equalize OPEN → depressurize via drain as permitted.

Appendix C — Static Pressure Effect (Heads-Up)

  • High static pressure can bias DP readings via transmitter mechanics/fluid properties.

  • Mitigation: live-zero checks, apply vendor static pressure compensation, verify span under operating pressure where applicable.

Appendix D — Maintenance Record (Template)

  • Work Order / Date / Techs: ______ / ______ / ______

  • Process Conditions (medium / T / P / valve): __________________________

  • Electrical Data:

    • Control-room supply: __ VDC

    • JB terminal X1 +/-: __ VDC

    • TX terminal +/-: __ VDC; Output: __ mA

  • Cable/Shield Checks: continuity __ / insulation __ MΩ @ __ V; shield ground (□ single-point / □ issue)

  • Manifold Actions: followed card (□ yes / □ no); ΔP before/after equalization: __ → __ kPa

  • Findings (with photo IDs): __________________________________________

  • Corrective Actions/Materials (ferrule, heat-shrink, gland, torque, pull-test): ___________________

  • Verification: PV __ ; mA __ ; 1-h trend stable (□) / notes: ____________

  • Closeout: drawings updated (□), training note filed (□)

11. Takeaways (One-Page for Toolbox Talk)

  • Symptom: PV=0, display off → suspect power path first.

  • Fast triage: control room → JB → TX voltage ladder; never skip measurements.

  • Outdoor splices fail; use IP67+ practices or proper JB.

  • Document everything: data tables, photos, trend captures—future you will thank present you.

Share This Story, Choose Your Platform!

Contact Us

    Please prove you are human by selecting the star.
    Translate »