How to Verify Lift-Station Level Control Band Settings

Back to Blog

Confirm What You Need Before You Start

Pull the approved control narrative or the station drawing that lists every level setpoint, because the whole verification is a comparison against that document rather than a judgment call. You want the pump-off elevation, the lead pump-on, the lag pump-on, the high-level alarm, and any redundant float elevations, all expressed against the same datum. If the drawing gives depths from a top-of-slab reference while the level transmitter reads distance from the bottom, you have to reconcile the two before any of the numbers mean anything.

Have the level sensor identified and its span confirmed. Most wet wells use a submersible pressure transducer or an ultrasonic sensor, and the setpoints only make sense once you know what physical level the sensor reports as zero and as full scale. If the station uses an ultrasonic sensor, the reference on the drawing needs to survive the sign flip that comes with a sensor that measures distance down from the top rather than head from the bottom. It is worth reviewing how the site commissioned that instrument, since the level band inherits its span and datum directly from the sensor, as covered in the guide on commissioning an ultrasonic level transmitter.

Check the Setpoints Are Stacked in the Right Order

Read each configured setpoint out of the controller and lay them out from lowest elevation to highest. The correct order is pump-off at the bottom, then lead pump-on, then lag pump-on, then high-level alarm, with the redundant high-level float set above the analog high-level alarm so it only acts if the analog path has already failed. Any setpoint out of this order is a defect: a lag-on set below the lead-on means both pumps start together, and a pump-off set above a pump-on is a logical impossibility the controller may accept silently while the station never pumps down.

Confirm the deadband between pump-off and lead-on is wide enough to hold a reasonable working volume. That vertical separation, multiplied by the wet-well plan area, is the volume the pump moves per start, and it is what keeps the motor from short-cycling. A band that is too narrow makes the pump start and stop rapidly, which stresses the motor and the starter; the exact minimum separation is site-specific and follows from the wet-well geometry and the pump's allowable starts per hour on its datasheet. The relationship between these numbers is exactly what the site's chosen pump-off setpoint is meant to protect.

Verify the lag pump-on sits far enough above the lead pump-on that the lag pump is only recruited when the lead pump genuinely cannot keep up with inflow. If the two are too close, ordinary flow variation trips the lag pump on and off needlessly. Check that the high-level alarm sits below the lowest incoming pipe invert so the alarm annunciates before sewage backs up into the collection system, which is the entire point of the alarm existing.

Inject a Simulated Level and Watch Each Action Fire

With the pumps in a safe state, use the controller's simulation or force function to drive the level input from the bottom of the range slowly upward, and record the exact level at which the lead pump call energizes, then the lag call, then the high-level alarm. Each event should fire at the number on the drawing within the resolution of the sensor. Then drive the level back down and confirm the pumps drop out at the pump-off elevation, not before and not after.

Watch for the two classic faults this sweep exposes. If a pump call activates at a level different from the configured setpoint, either the engineering-units scaling of the input is wrong or the setpoint was entered in raw counts rather than engineering units. If a pump call flickers on and off right at its setpoint as you hold the level there, the setpoint has no deadband or hysteresis and will chatter in service. Both are fixed in configuration before the station goes live, and the fix should be re-verified with the same rising-and-falling sweep.

Verifying the Result and Common Mistakes

A verified level band reads as a clean, monotonic staircase: level rises, lead starts, level keeps rising under heavy inflow, lag starts, both pumps draw the well down, lag drops out, lead drops out at pump-off, and the high-level alarm never annunciated because the pumps handled the flow. Capture the as-left setpoint list and file it against the drawing so the next technician has a verified baseline rather than a guess. On a platform like Merobix the same setpoints appear as tags on the wet-well trend, so a later drift or an accidental edit shows up as a change in the recorded control behavior.

The most common mistake is verifying the numbers on the screen against the drawing but never injecting a level to confirm the logic acts on them, which misses scaling and hysteresis faults entirely. The second is forgetting the datum reconciliation, so every setpoint is internally consistent but shifted from the physical wet well by the sensor offset. The third is setting the redundant float at or below the analog high-level alarm, which defeats the backup because both trip together instead of the float acting as a true last line of defense, a point covered in the guide on the wet-well float backup control.

Frequently Asked Questions