How to Zero-Trim a DP Transmitter With a Wet Leg
DP level transmitters on condensing services - boiler drums, columns, closed vessels - carry a wet reference leg, and that leg changes what "zero" means. The most damaging mistake a technician can make on one of these loops is to equalize the manifold and trim the transmitter until the level reads zero, which quietly erases the elevation the wet leg builds into the calibration. This page explains the distinction between zero differential and zero level, and the sequence that gets the trim right with the loop in service.
Wet-Leg DP Zero Trim in one line: To zero-trim a DP transmitter with a wet leg, first confirm the wet leg is actually full, because a partly evaporated leg is the usual cause of an apparent zero shift. Then equalize the transmitter per site procedures and check that it reads zero differential pressure - not zero level. If the equalized reading is off zero, that is a genuine sensor zero shift and a zero trim corrects it. The head of the wet leg itself belongs in the range values as an elevated or suppressed zero, and must never be trimmed away.
Why a Wet Leg Changes What Zero Means
A wet leg is a reference line kept full of condensate or fill fluid, applying a constant hydrostatic head to one side of the DP cell. Because that head is always present, the transmitter sees a nonzero differential even when the vessel level is at the bottom of its range, so the calibrated range is elevated or suppressed: the lower range value is not zero DP but the DP that corresponds to empty with the leg full. Zero level and zero differential are two different points on these loops, and every step of the check has to respect that.
This is exactly why the naive procedure fails. Equalizing the manifold puts the same pressure on both sides, which drives the true differential to zero. A healthy transmitter must then read zero DP, which corresponds to some level reading well away from zero on an elevated range. A technician who trims until the level display reads zero at that moment has shifted the sensor by the entire wet-leg head, and the loop will read wrong across the whole range afterward.
What You Need
You need the loop documentation showing the calibrated range - the lower and upper range values with their elevation or suppression - the manifold arrangement, and which side the wet leg feeds. You also need a configuration tool or HART communicator to read the sensor in pressure units rather than percent of range, and the fill-fluid details if the leg uses something other than process condensate. As-found and as-left readings should be recorded like any calibration.
Because the transmitter is on live process, every isolation, equalization, venting, or refilling step is governed by site procedures and the permit system. The sequence of manifold operations that protects the cell from overrange during isolation is covered separately in the guide to the 3-valve manifold blowdown and zero; this page assumes you follow the site's version of that sequence.
Confirm the Wet Leg Is Full First
Most "zero drift" complaints on wet-leg loops are not transmitter drift at all; they are a leg that has partially evaporated, leaked, or been drained and never refilled. When the reference head shrinks, the measured differential shifts and the indicated level steps in whichever direction the lost head dictates for that piping arrangement. The tell is a step change after maintenance or a hot spell, or an indicated level that disagrees with an independent reference such as a sight glass while the transmitter itself checks out.
So before touching trim, establish that the leg is full. If the site's procedure calls for refilling, do that first, let the leg reach a stable temperature, and re-check the reading against the independent reference. Trimming a transmitter to compensate for a half-empty leg builds a lie into the calibration: the moment someone properly refills the leg, the loop is wrong by the amount of the earlier trim.
Check True Zero, Then Trim or Re-Range
With the leg confirmed full, equalize per site procedure and read the sensor in pressure units. A healthy cell reads zero differential, within the tolerance the site applies. If it does not, that offset is a genuine sensor zero shift - from overpressure history, mounting stress, or ordinary drift - and a zero trim at this equalized condition is the correct fix. Record the as-found offset, trim, confirm the reading sits at zero DP, and record the as-left value.
Keep the two adjustments in their lanes. Zero trim corrects the sensor's reading of an actual zero-DP condition. The wet-leg head, by contrast, is a known physical offset that belongs in the range values as elevation or suppression. If the leg geometry or fill fluid changed - a rebuilt leg, different fill, relocated transmitter - the correction is a re-range with newly calculated range values, not a trim. Mixing these up is how loops end up with a trim fighting a wrong range, which no amount of further adjustment untangles.
Verifying the Result and Common Mistakes
Return the loop to service per procedure, making sure the equalizer is fully closed and the process valves are back in their operating lineup, then compare the indicated level against an independent reference: a sight glass, a bridle gauge, or a second transmitter. Watch the trend afterward; a correctly trimmed loop holds its agreement, while a loop with a slowly leaking or re-evaporating leg starts walking again, and that trend is the evidence for a piping repair rather than another trim.
The classic mistakes: trimming the level display to zero while equalized, trimming out a lost wet leg instead of refilling it, leaving the equalizer cracked open so the loop reads a fixed low level, ignoring trapped vapor in a freshly filled leg, and failing to record as-found values so nobody can later reconstruct what was changed. Each of these produces a loop that looks calibrated on the bench sheet and lies in service.
Frequently Asked Questions
Should an equalized wet-leg DP transmitter read zero level or zero DP?
Zero differential pressure, never zero level. Equalizing removes the real process differential, so the sensor should report zero DP; on an elevated or suppressed range that corresponds to a level reading well away from zero. Trimming until the level reads zero at the equalized condition shifts the sensor by the entire wet-leg head and corrupts the whole range. Always judge the equalized check in pressure units, not in percent of level.
Why did the indicated level jump after someone zeroed the transmitter?
Almost always because the zero was performed as if the loop were a plain DP transmitter: equalized, then trimmed to make the level display read zero. That erases the elevation the wet leg builds into the range. The fix is to restore the correct range values from the loop documentation, confirm the wet leg is full, and then do a proper zero-DP trim at the equalized condition if the sensor actually needs one.
How do I know a wet leg has partially lost its fill?
Look for a step change in indicated level after maintenance or hot weather, a growing disagreement with a sight glass or redundant transmitter, or drift that correlates with ambient temperature. Confirm by comparing the reading against an independent reference before and after the leg is refilled per site procedure. If refilling moves the reading back into agreement, the leg was the problem and the transmitter never needed a trim.
Automation services
Need help turning this into a working system?
Merobix integrates SCADA, programs Allen-Bradley and Siemens PLCs, and designs and fabricates industrial control panels.
Meeting requests are reviewed before confirmation.