Intermittent Left-Turn Loop Detection After Rain
During a post-installation inspection of a new left-turn lane on a four-lane arterial, you find that the left-turn call is intermittent. The lane is controlled by a 6-foot by 30-foot quadrupole loop saw-cut into the pavement, with a 450-foot lead-in to the cabinet. The lead-in is 14 AWG unshielded THHN, spliced in a handhole with twist-on wire connectors, and for 120 feet it shares a 4-inch conduit with the 120 VAC signal conductors. The loop detector is a rack-mounted two-channel unit with sensitivity set at 4. During dry-weather testing, the detector calls reliably; after a 30-minute rain, the call fails about 40 percent of the time, and the detector sometimes holds a call with no vehicle present. A loop tester shows 95 microhenries when wet and 140 microhenries when dry. The cabinet ground electrode conductor measures 38 ohms to earth; the project specification requires 5 ohms or less. The contractor proposes raising detector sensitivity to 8 and moving on.
1. Evaluate the contractor's proposal to increase sensitivity. What would it fix and what would it not fix?
Read the worked answer
Increasing sensitivity from 4 to 8 is not an acceptable correction. A higher sensitivity setting can make a weak loop appear to work under dry conditions, but it does not repair a wet splice, damaged insulation, shared-conduit noise, or a poor ground. It will also make the detector more likely to respond to adjacent-lane vehicles, to hold a call when no vehicle is present, and to produce false calls during rain - the opposite of the reported failure mode.
The change in inductance from 140 microhenries dry to 95 microhenries wet is significant. Moisture in the splice or a nick in the lead-in insulation can change loop-circuit characteristics and create leakage to ground; that is a wiring defect, not a sensitivity-tuning problem. The detector's inability to reliably drop a call also points to a circuit or grounding problem. Raising sensitivity masks the symptom and may leave the intersection with phantom left-turn calls, which can cause unnecessary phase changes and delay.
2. Identify the wiring and grounding deficiencies and the state the accepted practice for each.
Read the worked answer
Several deficiencies are visible or measurable:
- Twist-on connectors in a handhole - deficient. Loop lead-in splices should be permanent, waterproof, and mechanically protected. Use an encapsulated direct-burial splice kit or a sealed splice enclosure; do not rely on wire nuts in a wet location.
- Unshielded 14 AWG lead-in - deficient for this environment. Loop lead-in should be twisted shielded pair, with the shield grounded at the cabinet end only, to reduce electrical noise and interference.
- Sharing conduit with 120 VAC signal conductors for 120 feet - deficient. Loop lead-in should be run in a separate conduit from power conductors, or at least maintain the required separation. Shared conduit couples noise into the detector circuit.
- 38-ohm ground resistance - deficient. The specification requires 5 ohms or less. Inspect the ground rod, clamp, and bonding jumper; drive additional rods or improve the electrode until the measured resistance meets the specification.
- Wet/dry inductance shift - deficient. The splice and lead-in insulation must be tested and repaired; the loop itself should be checked for nicks or damage at the saw-cut edges.
Each deficiency should be corrected at the source rather than compensated for at the detector.
3. Describe the test sequence you would require after corrections before accepting the detection system.
Read the worked answer
After the splice, lead-in, conduit separation, and grounding are corrected, require a documented sequence:
- Insulation test on the loop and lead-in with the detector disconnected, using an appropriate insulation tester at the manufacturer's recommended voltage; record resistance to ground.
- Loop-circuit measurements: inductance, resistance, and turns ratio or equivalent, compared with the design values; confirm the wet/dry shift is gone.
- Grounding verification: measure the cabinet ground resistance and confirm 5 ohms or less; check the neutral and ground bonding and the surge-protection ground.
- Detector calibration and test with the manufacturer's test set or a test vehicle: verify a call on entry, a hold through the loop, and a drop-out after exit, with no false hold when the lane is empty.
- Wet-weather simulation if practical, such as applying water to the handhole and splice area while monitoring for false calls.
- Operational test through the controller: confirm that the left-turn call is registered and that the phase terminates correctly, and observe several cycles.
Record all readings and test results; acceptance should be based on the corrected wiring and measured performance, not on a dry-weather call test alone.