ISA CCST Level III study materials
ISA CCST Level III study resources

Loop-Check Sign-Off Under Schedule Pressure

ISA CCST Level III · Put your knowledge to work in context.

All case studies

Loop-Check Sign-Off Under Schedule Pressure

A greenfield 25 MGD wastewater plant is in commissioning. The loop-check package has 80 loops; 15 analog loops and 6 control valves cannot be stroked because the plant air compressor has not been commissioned. Eight loops have instrument range mismatches; one flow transmitter is configured for square-root extraction in the transmitter while the DCS also extracts the square root. The EPC contractor’s commissioning manager wants your signature on the loop-check sheets for all 80 loops so the owner can begin biological seeding; he says the remaining items can be handled under a “punch list after start-up.” The project schedule has a $40,000/day liquidated-damages clause, and your supervisor has asked you to “be practical.” Two of the unfinished loops are on the influent pumping station, which must run to seed the plant.

1. Should you sign the loop-check sheets for all 80 loops? Explain your decision.

Read the worked answer

No. A loop-check signature is a technical record that the loop was verified, not a promise to fix it later. Signing all 80 would misrepresent the status of 23 loops and would remove the visibility that the owner needs before start-up. The double square-root configuration is not a minor punch item: if the transmitter outputs 50% at 25% flow and the DCS extracts the square root again, the indicated flow is about 70.7% instead of 25%, which could cause overfeeding, pump damage, or permit exceedance. A range mismatch such as 0–250 kPa in the field versus 0–200 kPa in the DCS produces a 25% error at full scale.

I would sign only the loops that have been fully verified, and I would issue the rest as “not verified” with clear exceptions. I would support a partial mechanical completion or a controlled start-up for the influent pumps using local/manual control and independent protection, but not a blanket loop-check acceptance. The commercial pressure is real, but the record must match the physical condition.

2. What immediate actions would you take for the 21 unstroked loops and the eight range/configuration problems?

Read the worked answer

For the valves, I would rent a portable compressor or use nitrogen if the actuator and safety requirements allow it, and stroke each valve at 0, 25, 50, 75, and 100% in both directions. I would verify fail-safe action, travel limits, position feedback, and air leakage. For the analog loops, I would use a portable pressure source or loop calibrator to inject 4, 8, 12, 16, and 20 mA and confirm the DCS indication, alarms, and interlocks.

For the eight configuration problems, I would correct the range in the DCS or transmitter to a single agreed range, and remove the duplicate square-root extraction from exactly one location. I would verify the result at several flow points, not just at 100%. Each correction needs a retest record and a revised loop-check sheet. The two influent pump loops should be prioritized because they are needed for seeding, but priority does not justify skipping the checks.

3. How would you handle the schedule and management pressure while protecting the commissioning record?

Read the worked answer

I would put the exceptions in writing: a short memo or dashboard listing each unverified loop, the reason, the risk, the owner, and the target date. I would ask for a management-of-change or readiness review that explicitly accepts the remaining risk for start-up, rather than allowing a blanket signature to imply completion. I would separate “ready for seeding” from “loop check complete” so the project can move forward without falsifying the record.

With my supervisor, I would focus on the cost of a wrong start-up: a double square-root error or an unstroked valve can cause a trip, a spill, or a permit violation that costs more than the liquidated damages. I would offer a practical path: verify the minimum set needed for seeding, use manual control where appropriate, and schedule the rest before handover. The goal is not to block the project; it is to make the residual risk visible and owned at the right level.