ISA CCST Level I common mistakes
ISA CCST Level I · Understand the traps and how to avoid them.
Leaving the loop calibrator in 'source/simulate mA' mode and connecting it across the transmitter terminals of a still-powered 2-wire loop, then recording the simulated value as the transmitter's output.
The calibrator's source and measure modes look similar on the display and the leads go to the same two terminals; candidates conflate 'simulating a signal into the receiver' with 'measuring what the transmitter produces'.
What to do instead
Before every hookup, state out loud what is being driven and what is being read. To test a 2-wire transmitter, break the loop and put the calibrator in mA-measure (or use a dedicated read-back channel) with the loop supply in series; to test the receiver or DCS card, use mA-source into the input with the transmitter disconnected. Verify by observing a change when the process or test pressure changes.
Calibrating a differential-pressure flow transmitter linearly in milliamps across 0-100% and declaring it good, when the flow indication is the square root of the differential pressure.
The 4-20 mA output really is linear with dP, so a linear check passes; the error only appears when the candidate converts mA to flow percent and forgets that flow% = sqrt(dP%).
What to do instead
Determine from the loop sheet/transmitter configuration where the square root is extracted. If the transmitter outputs linear dP, apply the square root when converting to flow: a 25% dP equals 50% flow. Check at 25/50/75% of dP and confirm the flow values land at 50/70.7/86.6%.
Zeroing a differential-pressure level transmitter with both legs vented to atmosphere, or with a wet leg not yet filled, and accepting the resulting 4 mA as the true 0% level.
The static-head offset from a wet leg (or elevation/suppression from a sealed or remote-seal installation) shifts the true zero; with equal pressure on both sides the transmitter reads zero dP, not zero level.
What to do instead
Confirm the installation type from the P&ID and datasheet first: dry leg, wet leg, or remote seal. Fill wet legs fully and let them stabilise, apply the calculated elevation/suppression, and verify with a known level or by calculating the head (h x SG x 0.0361 psi/in for water) and comparing to the transmitter's LRV/URV.
Breaking a loop or opening an impulse line for maintenance without first isolating and de-energising it - pulling a transmitter out of a live 24 VDC loop, or cracking a fitting on a line that is still under process pressure.
In a training lab the loop is often already dead, so the habit of 'just unplug it' carries over; in the field the loop supply and the process are still live and the candidate is focused on the instrument, not the energy sources.
What to do instead
Use a written isolation sequence every time: identify all energy sources (loop power, separate 24 V, instrument air, process pressure/temperature), close and tag the block valve, vent/bleed the impulse line, verify zero energy with a meter, and apply lockout/tagout before disconnecting anything. Restore in reverse order and leak-check before returning to service.
Swapping out a suspect transmitter, card, or fuse and closing the work order as soon as the indication comes back, without establishing why the original failed or checking loop supply voltage and burden first.
A restored reading feels like a diagnosis, and parts are readily available; the underlying cause (a sagging supply, a corroded terminal, a shorted shield, a wet junction box) stays in place and fails again.
What to do instead
Before replacing anything, record the evidence: supply voltage measured at the transmitter terminals while the loop is loaded, measured loop current, insulation resistance to ground, and terminal torque/condition. Replace only after the measurements point at the component, then re-measure after repair to confirm the root cause is gone, and document the readings.
Starting the diagnosis at the DCS screen and working outward one component at a time, or replacing parts by guess, instead of measuring at a midpoint and checking the basics (supply, fuse, breaker, air pressure) first.
The HMI is the most visible symptom and the schematic is not memorised, so the candidate chases the reading rather than the signal path; without a baseline measurement there is no way to know which half of the loop is at fault.
What to do instead
Use half-split: measure the signal at the midpoint of the loop, decide which half is bad, then split again. Before that, verify the obvious inputs - 24 VDC at the transmitter with the loop loaded, instrument air at the filter regulator, fuse/breaker status, and the DCS card's channel status. State the expected value at each test point before measuring.
Reading 3.6 mA, 21.5 mA, or 22 mA on a transmitter and interpreting it as a legitimate low or high process condition rather than a fault or saturation signal.
The candidate reads the number as a process value because the arithmetic works, and may never have been taught that modern transmitters use out-of-range current levels to annunciate failure.
What to do instead
Learn and apply the standard fault bands: below about 3.6 mA or above about 21.0 mA is outside the normal 4-20 mA range and typically indicates a saturated or failed device (NAMUR NE43 uses <3.6 mA and >21.0 mA as fault). When a reading sits in these bands, treat it as a device/wiring fault and check supply, sensor, and diagnostics rather than reporting a process excursion.
Energising a new or modified loop before comparing the field wiring to the loop drawing, checking polarity, ground/shield landing, fuse rating, and the transmitter's tag and address against the instrument index.
Start-up pressure pushes toward 'power it up and see', and the as-built wiring may differ from the design drawing; reversed polarity and mis-landed shields cause damage or intermittent faults that surface later and look like instrument failures.
What to do instead
Work a written pre-energisation checklist for every loop: point-to-point ring-out with power off, polarity verified at both ends, shield grounded at one end only per the grounding standard, correct fuse/breaker rating, correct tag and range on the transmitter and in the DCS database, and as-built markup of any deviation. Sign off the checklist before the first energisation.
Commissioning a control valve by confirming it strokes in both directions, without verifying the fail-safe action (air-to-open vs air-to-close, loss of air, loss of signal) and without checking positioner calibration, limit switches, and travel against the P&ID and cause-and-effect.
Full stroke proves the actuator moves, which looks like success; the fail direction only reveals itself when air or signal is removed, which the candidate does not test because it feels like creating a trip.
What to do instead
Before handover, verify the specified fail action against the P&ID/spec sheet, then physically remove instrument air and separately remove the electrical signal and observe the valve go to its safe position. Check 0/25/50/75/100% travel against the positioner, confirm limit switch and position feedback setpoints, and record the results with the tag number.
Performing a loop check as a continuity or 'wire is good' test only - buzzing out the pair - and not driving an actual signal from the field device through to the DCS/PLC and confirming the displayed value in engineering units.
Continuity is fast and requires no communication with the control system, so it gets substituted for the real check; a loop can be continuous and still be landed on the wrong channel, wrong polarity, or wrongly scaled.
What to do instead
For each tag, inject a known signal at the field end (or simulate the sensor) at 0/50/100% and read the value at the operator station, verifying tag, engineering units, scaling, and alarm limits match the loop sheet. Record the as-found/as-left values, initial and date the loop check sheet, and confirm the control room and field technician are on the same tag before and after each injection.