How to Calibrate a Differential Pressure Transmitter

Differential pressure transmitter calibration checks whether a transmitter produces the correct output for known differential pressure inputs. In a typical field or bench procedure, the technician isolates the transmitter, applies a traceable pressure input, measures the 4–20 mA loop output, records the as-found condition, adjusts only if required, and confirms the final as-left performance. The exact details depend on the transmitter model, manifold arrangement, process service, calibration range, and site procedure.

Step 1: Isolate the Transmitter and Operate the Manifold Safely

Before any differential pressure transmitter calibration work begins, the transmitter must be safely isolated from the pressurized process. This protects personnel from process media, prevents damage to the DP sensing capsule, and creates a stable pressure reference for the calibration check.

A differential pressure transmitter measures the pressure difference between its high-pressure side and low-pressure side. During normal operation, each side may be connected to a live process through impulse lines, remote seals, or a valve manifold. If the manifold is operated incorrectly, full line pressure can be applied to only one side of the DP capsule. That one-sided pressure shock can overload the sensing element, disturb the calibration, or damage the transmitter. For that reason, always follow the approved site procedure, the OEM manual, and the specific sequence for the installed three-valve or five-valve manifold.

In general, isolation involves closing the high-pressure and low-pressure block valves so process pressure no longer reaches the transmitter. The exact order may vary by installation and service conditions. Some procedures call for closing one block valve, opening the equalizing valve slowly, then closing the other block valve. Others require a different controlled sequence. The important principle is that the transmitter should not be exposed to an uncontrolled differential pressure during removal from service.

After the transmitter is isolated, the equalizing or balance valve is opened so both sensing chambers are connected together and held at the same pressure. With both sides equalized, the differential pressure across the measuring element is zero. This is the correct condition for a zero check, provided the reference condition is stable and appropriate for the calibration method.

If the transmitter chambers, impulse lines, or manifold cavities contain trapped liquid or gas, venting or draining may be required. Use the manifold vent, drain, or bleed points where provided, and discharge to a safe location according to the process hazard and site procedure. Liquid service may require draining trapped fluid; gas service may require controlled venting. Steam, corrosive, toxic, hot, cryogenic, or high-pressure services require additional precautions and should never be opened casually to atmosphere.

A true zero reference requires both sides of the DP element to be at the same reference condition. In many bench or field calibration setups, this means both sides are vented to atmospheric pressure. If only one side is vented while the other side remains pressurized, the transmitter is not at true zero. Likewise, if trapped head pressure, condensate, filled capillaries, or impulse-line effects remain unequal, the reading may not represent a clean zero condition.

For field checks on installed instruments, consider the actual application. A flow transmitter across an orifice plate, a level transmitter with a wet leg, and a remote-seal DP transmitter may all require different handling. Some applications cannot simply be vented to atmosphere without changing the reference condition or disturbing a fill system. The safe rule is to isolate, equalize, vent or drain only where applicable, and confirm that the reference condition used for calibration matches the approved procedure.

Step 2: Connect the Pressure Source, Loop Power, and Test Instruments

After the transmitter is isolated and the reference condition is established, connect the calibration equipment. The basic setup requires a controlled pressure source, a way to measure the transmitter output, and power for the transmitter loop.

A suitable precision pressure source is connected to the high-pressure side of the transmitter. The source may be a pressure calibrator, hand pump with reference gauge, deadweight tester, or automated pressure controller, depending on the required accuracy and pressure range. The pressure source must be appropriate for the transmitter span and process connection. It should also have sufficient accuracy for the calibration requirement; do not assume that any hand pump and gauge combination is acceptable. The pressure source accuracy should be checked against the transmitter specification, calibration procedure, or site metrology requirements.

When atmospheric pressure is used as the reference, the low-pressure side is normally left open to atmosphere. This means the applied pressure on the high side represents the differential pressure seen by the transmitter. For example, if the low side is open to atmosphere and the high side is pressurized to a known value, the transmitter measures that applied pressure as positive differential pressure. If the calibration involves negative differential pressure, compound ranges, suppressed ranges, or elevated zero applications, the connection method must match the transmitter configuration.

To measure the transmitter output, place a digital multimeter in series with the current loop. This is important: a loop current measurement is not made by placing the meter directly across the transmitter terminals in voltage mode. The loop must be opened at an appropriate point, and the meter must be inserted in the current path using the correct milliamp input terminals and range. Some process calibrators include a loop current measurement function, which can simplify the connection.

The transmitter also requires loop power. Many industrial two-wire transmitters operate from a DC loop supply, but the required voltage range depends on the specific device, loop resistance, intrinsic safety barriers, and installation. Verify the transmitter documentation before applying power. If the instrument is still connected to the plant control system, confirm whether the loop will be powered by the control system or by a temporary calibration supply. Follow lockout, bypass, alarm suppression, and control-room notification requirements before disturbing an active loop.

If parameter verification or trimming is required, connect a HART communicator or compatible modem according to the device and loop arrangement. The communicator is commonly connected across the transmitter terminals or across a loop resistor. HART communication requires sufficient loop impedance, but the exact requirement should be verified for the communicator, transmitter, and installation. If the loop has too little resistance, communication may fail even though the 4–20 mA signal is present.

Before applying pressure, verify the following:

• Transmitter tag number and service description
• Lower range value, upper range value, and engineering units
• Linear or square-root output configuration
• Damping, transfer function, and any output scaling
• Pressure source range and accuracy
• Loop supply voltage and total loop resistance
• Correct polarity and current measurement connection
• Required HART access and trim procedure
• Calibration tolerance from the approved procedure or specification

These checks prevent a common calibration error: testing the transmitter against the wrong range or output mode. For example, a DP transmitter configured for square-root flow output will not produce the same intermediate current values as a linear pressure transmitter unless the test is performed at the correct function level or the transmitter is temporarily placed in the appropriate calibration mode according to procedure.

Step 3: Verify Output with a Five-Point Calibration Check

A five-point verification checks transmitter output across the calibrated span rather than only at zero and full scale. This gives a better view of linearity, span error, and possible hysteresis. When required by the site procedure or calibration plan, use test points at 0%, 25%, 50%, 75%, and 100% of span.

For a linear 4–20 mA differential pressure transmitter, the expected outputs at those five points are:

These values apply to a linear output over the calibrated pressure span. They should not be blindly applied to square-root flow output, custom characterization, reverse-acting output, or special scaling without confirming the transmitter configuration.

Start at the lower range value and allow the transmitter output to stabilize. Record the applied pressure and measured loop current. Then increase pressure to the 25%, 50%, 75%, and 100% points, allowing stabilization at each point before recording the current. The applied pressure at each point is based on the configured span:

• 0% point = LRV
• 25% point = LRV + 0.25 × span
• 50% point = LRV + 0.50 × span
• 75% point = LRV + 0.75 × span
• 100% point = URV

Where span is the difference between the upper range value and lower range value.

For example, if a transmitter is ranged from 0 to 100 inH₂O and configured for linear output, the five pressure points would be 0, 25, 50, 75, and 100 inH₂O, with expected outputs of 4.00, 8.00, 12.00, 16.00, and 20.00 mA. If the transmitter is ranged from 20 to 120 kPa, the span is still 100 kPa, but the five pressure points become 20, 45, 70, 95, and 120 kPa. The output percentages are based on span, not simply on absolute pressure.

Compare the measured current at each point with the expected current. Do not invent a pass/fail tolerance. The allowable error must come from the transmitter specification, calibration procedure, quality system, or site requirement. Some applications require tight tolerances; others allow wider limits based on process needs, installed uncertainty, or safety classification. The technician’s task is to apply the specified acceptance criterion, not create one during the test.

For better diagnostic value, some procedures also require checking the points in both upscale and downscale directions. This can reveal hysteresis or mechanical effects that may not appear in a single upscale test. If the procedure requires both directions, record the readings separately rather than averaging them without authorization.

During the five-point check, avoid adjusting the transmitter after the first failed point unless the procedure instructs you to stop and correct immediately. The initial readings represent the as-found condition and are valuable maintenance data. Adjusting too early can erase evidence of drift, installation problems, impulse-line issues, or device malfunction.

Step 4: Record As-Found Results and Confirm As-Left Performance

The initial five-point test data are the transmitter’s as-found condition. These readings show how the instrument performed before any calibration adjustment, trim, repair, cleaning, or configuration change. As-found data are important because they indicate whether the device was operating within tolerance while in service.

For each test point, record the applied pressure, expected output, measured output, and error. The error may be expressed in mA, percent of span, pressure units, or another format specified by the calibration record system. Compare each measured value against the allowable tolerance from the applicable procedure, transmitter documentation, or site requirement.

A useful record normally includes:

• Instrument tag number and serial number
• Calibration range, LRV, URV, and units
• Output type and transfer function
• Applied pressure at each test point
• Expected mA at each test point
• Measured mA at each test point
• Calculated error
• Allowable tolerance
• As-found pass/fail result
• Adjustment or trim action, if performed
• As-left readings after adjustment
• Technician name or identifier
• Date and calibration equipment identification

If the as-found error is within tolerance, adjustment may not be necessary. In many calibration programs, unnecessary trimming is discouraged because it can introduce new error or mask long-term stability trends. If the instrument passes, document the results and continue to the return-to-service steps.

If the error is outside tolerance, investigate the cause before trimming. Confirm that the pressure source is connected correctly, the low side is at the intended reference, the range values are correct, the loop current measurement is wired properly, and the output mode is as expected. Also consider process-related causes such as plugged impulse lines, trapped gas in liquid service, liquid accumulation in gas service, leaking manifold valves, or incorrect wet-leg conditions. A transmitter should not be adjusted to compensate for an unresolved installation fault unless the procedure specifically addresses that condition.

When adjustment is required, use the model-specific HART or local configuration procedure. Modern smart transmitters may distinguish between sensor trim, output trim, range setting, and digital-to-analog trim. These are not the same operation. Changing the LRV and URV reranges the transmitter; trimming corrects measurement or output conversion errors. Follow the manufacturer’s instructions and the site calibration method.

Zero trim may be used at the low-pressure point where applicable, but only when the transmitter is at the correct zero reference. Span trim may be used at the full-scale point where supported by the transmitter and required by the calibration method. Some devices use two-point sensor trim, while others may use different calibration functions or restrict trim operations. Applying the wrong trim can make the transmitter appear correct at one point while increasing error elsewhere.

After any adjustment, repeat the five-point verification. These final readings are the as-left condition. The as-left data confirm whether the transmitter meets the required tolerance after adjustment. Record the final values clearly and keep them separate from the as-found values. If the transmitter cannot be brought within tolerance, it may require further troubleshooting, repair, replacement, or engineering review.

Step 5: Remove Test Gear and Return the Transmitter to Service

Remove the pressure source, multimeter, HART communicator, and temporary loop connections only after the post-adjustment verification is complete and the as-left data have been recorded. If the loop was opened for current measurement, restore the wiring exactly as required by the installation drawing or site standard. Check terminal tightness, polarity, cable glands, covers, and any temporary jumpers or barriers that were disturbed.

Before returning the transmitter to the process, make sure all vent, drain, bleed, and exhaust plugs are secured. Any open bleed point can create a leak path when the manifold is repressurized. On hazardous, hot, or high-pressure services, this step is critical. Use the correct plugs, gaskets, caps, and torque practices required by the site procedure and equipment design.

Return the valve manifold to service in a controlled sequence. The correct sequence depends on whether the installation uses a three-valve manifold, five-valve manifold, block-and-bleed arrangement, remote seals, or another configuration. Follow the approved procedure for the specific installation. The general principle is the same as during isolation: avoid applying full process pressure to only one side of the DP capsule.

In many manifold procedures, the equalizing path remains open during the initial controlled repressurization step. This allows both sides of the transmitter to rise together and reduces the risk of a sudden one-sided pressure shock. The high-pressure block valve is commonly opened slowly first while checking for leakage, then the equalizing valve is closed, and the low-pressure block valve is opened to establish the actual process differential pressure. However, the exact sequence must come from the site or OEM procedure, not from memory or generic practice.

As the transmitter returns to service, observe the output. The mA signal or digital process variable should respond to the actual process differential pressure. A reading near zero may be correct for a no-flow or balanced condition, but it may also indicate an equalizing valve left open, blocked impulse lines, or incorrect valve position. Compare the restored reading with expected process conditions and, where applicable, with the control system indication.

Perform a leak check around the manifold connections, vent plugs, drain plugs, transmitter process connections, and any fittings disturbed during calibration. Use the site-approved leak-check method for the process fluid and pressure class. Do not rely only on visual inspection if the service requires a formal leak test. For gas service, an approved leak detection solution or instrument may be required; for liquid service, visible seepage checks may be part of the procedure.

Finally, remove any loop bypasses, alarm inhibits, maintenance overrides, or control system suppressions that were applied for the calibration activity. Notify operations that the transmitter is back in service, and confirm that the final indication is reasonable. A differential pressure transmitter calibration is not complete until the instrument is safely restored, leak checked, documented, and returned to normal process monitoring or control.