The zero point checks out fine. But somewhere between 0% and 100% of the range, the reading has quietly stopped matching reality.
Ok, let me explain it properly. That's span drift, and it behaves very differently from a simple zero shift, which means it needs a different diagnosis and a different fix.
Pressure transmitter span drift is a condition where the slope of the transmitter's calibration curve changes over time, so the output error grows larger the further the process moves from zero, and it is typically caused by sensor material fatigue, temperature effects on the span coefficient, electronic component aging, static pressure effects, mechanical stress, or an overdue recalibration cycle.
Pressure Transmitter Span Drift: What It Means and Why It Matters
Pressure transmitter span drift shows up as a growing gap between what the instrument reads and what the process actually is, and that gap gets bigger as you move away from the zero point.
At 0% of range the transmitter might look perfectly fine. At 100% of range, though, it could already be off by a percent or more. Have you got it so far? That growing-with-range behavior is the signature that tells you it's span drift rather than something else.
Think of the calibration curve as a straight line drawn from the 4 mA point to the 20 mA point. A zero shift picks up that whole line and moves it up or down without changing its tilt, something we covered in detail in our article on pressure transmitter zero shift. Span drift is different. It pivots the line around one end, usually the zero point, so the far end swings away from where it should be while the near end barely moves.
This distinction matters a lot in the field, because the wrong diagnosis leads to the wrong fix. A zero trim will not correct this kind of gain error, and pushing ahead with one anyway just wastes a maintenance window while the actual problem keeps sitting there unresolved.
Span Drift vs Zero Shift: How They Differ
Before troubleshooting either one, it helps to be able to tell them apart quickly at the bench. A quick two-point check, zero and full scale, usually settles the question in under five minutes. If you'd like the full breakdown of terminology, our article on offset error, zero point error, and span error covers how these terms are formally defined.
| Characteristic | Zero Shift | Span Drift |
|---|---|---|
| Error at 0% of range | Large, this is where the error originates | Small or near zero |
| Error at 100% of range | Same as at 0%, error stays constant | Large, error grows with reading |
| Calibration curve behavior | Parallel shift, slope unchanged | Pivots or rotates, slope changes |
| Typical correction | Zero trim | Two-point recalibration with span adjustment |
| Common root causes | Mounting height, fill fluid, installation geometry | Sensor aging, temperature coefficient of span, electronics drift |
The Six Real Causes of Span Drift
Temperature Coefficient of Span
Most sensing elements don't just shift at zero with temperature, their gain changes too. Manufacturers publish this separately as a span temperature coefficient, and it's usually larger in magnitude than the zero coefficient covered in our pressure transmitter temperature effect guide.
Sensor Diaphragm Fatigue and Material Aging
Every pressure cycle flexes the sensing diaphragm a tiny amount. Over thousands of cycles across months or years, the material's elastic response changes slightly, and that shows up as a gradual gain change rather than a sudden fault.
Electronic Component Drift
Resistors, amplifiers, and reference voltage sources inside the transmitter's signal conditioning circuit age too. A slowly drifting gain-setting resistor is a classic, well-known cause of span error that has nothing to do with the sensor at all.
Static Pressure Effect on Span (DP Transmitters)
On differential pressure transmitters, high line pressure doesn't just shift zero, it can also affect the sensing element's response curve across the range, contributing a small but measurable span error alongside the zero error.
Overdue Recalibration Interval
Every sensing technology has a realistic drift rate per year, even under ideal conditions. Run a transmitter years past its recommended interval, and accumulated span error is simply expected, not a surprise.
Mechanical Stress From Installation
Over-torqued process connections, pipe strain transmitted into the transmitter body, or a poorly supported mounting bracket can all introduce mechanical stress that subtly changes how the sensing element responds across its full range.
Cause 1 in Detail: How Temperature Changes the Span
Have you got it so far? Good, because this is the cause worth understanding in real depth, since it's usually the largest contributor to this gain error in outdoor or unconditioned installations. Manufacturers specify a total effect that includes both a zero coefficient and a span coefficient, expressed as %URL per 100°F or per 100°C of ambient temperature change away from the reference calibration temperature.
The span portion of that error scales with how far up the range the transmitter is reading. At 10% of range, a given span coefficient produces a small absolute error. At 90% of range, the same coefficient produces a much larger absolute error, because the error is proportional to the reading itself, not a fixed offset. That's exactly why this error grows toward full scale while zero shift does not.
Visualizing the Pivot: Zero Shift vs Span Drift on the Calibration Curve
Try It: Span Drift Correction Calculator
Enter the transmitter's output at 0% and 100% of range from a two-point calibration check, and this calculator will work out the span error and the correction factor needed to bring it back on spec.
Let Us Take an Example
A technician performs a routine two-point calibration check on a transmitter that has been in service for six years. Venting to atmosphere and applying a certified full-scale test pressure, the readings come back at 4.12 mA at 0% and 19.30 mA at 100%, instead of the expected 4.00 mA and 20.00 mA.
Actual span = 19.30 − 4.12 = 15.18 mA. Ideal span = 16.00 mA. Span error % = (15.18 − 16) ÷ 16 × 100 = −5.13%. The correction factor needed is 16 ÷ 15.18 = 1.0540.
Ok, let me explain what that number is telling us. The zero error component here is only 4.12 − 4 = 0.12 mA, small and easily within a normal zero trim. But a −5.13% span error is well outside typical accuracy specs for most transmitters, meaning the gain has narrowed noticeably over six years. This transmitter needs a genuine two-point recalibration with a span adjustment, not just a quick zero trim. Have you got it? A zero trim alone would leave the reading close to correct near 0% but still off by roughly 5% near full scale, which is exactly the kind of error that goes unnoticed until it causes a real process problem.
Cause 2 in Detail: Sensor Aging and Long-Term Stability
Every sensing technology, whether it's a capacitive diaphragm, a piezoresistive strain gauge, or a resonant element, has a published long-term stability specification, typically expressed as a percentage of upper range limit per year. This single number is doing a lot of work: it's the manufacturer's honest estimate of how much span (and zero) error to expect purely from aging, assuming normal operating conditions and no external damage.
According to RealPars' explanation of analog signal drift, this kind of gradual electronic and sensor aging is one of the most common reasons a previously accurate instrument slowly falls out of spec without any single dramatic failure event. It's also why recalibration intervals exist as a scheduled activity rather than something triggered only by a complaint.
A good rule of thumb: if a transmitter's measured span error roughly matches what you'd expect from (years in service) × (published annual stability spec), aging alone likely explains it, and routine recalibration is the fix. If the measured error is significantly larger than that, look harder at temperature exposure, mechanical stress, or static pressure effects before assuming it's just normal aging.
A Step-by-Step Span Drift Correction Method
Perform a true two-point check
Vent to atmosphere for the zero reading, then apply a certified test pressure at or near full scale using a calibrated reference standard.
Separate the zero and span components
Use the calculator above to isolate how much of the total error is zero-related versus span-related before deciding on a correction method.
Check installation history and service life
Confirm how long the transmitter has been in service, what temperature extremes it has seen, and whether it's overdue for its recommended recalibration interval.
Rule out static pressure and mechanical stress
On DP transmitters, check the static pressure effect specification, and inspect the mounting for over-torqued fittings or unsupported pipe strain.
Apply a proper two-point recalibration
Adjust zero first at 0% of range, then adjust span at 100% of range, and repeat both checks once, since the two adjustments can slightly interact on some transmitter designs.
Log the drift rate, not just the correction
Recording how much span error accumulated since the last calibration lets you predict when this transmitter will next need attention, instead of waiting for it to fail a routine check again.
Why Tracking Drift Rate Over Time Beats a One-Off Fix
A single calibration check tells you today's error. A logged history of span error across several calibration cycles tells you the drift rate, and that's far more useful for planning maintenance before an instrument actually fails a compliance check.
Notice the rate isn't perfectly linear, it often accelerates in the later years as component tolerances and material fatigue compound each other. This is exactly the pattern that a simple annual calibration record can catch early, well before the error reaches a level that affects custody transfer accuracy or safety instrumented system performance.
Good Practices to Prevent Excessive Span Drift
✓ Do
- Perform a genuine two-point check at every calibration, not just a single zero verification
- Keep a logged calibration history so drift rate, not just current error, becomes visible over time
- Check both the zero and span temperature coefficients on the datasheet for outdoor or unconditioned installations
- Follow the manufacturer's recommended recalibration interval rather than extending it informally
✗ Don't
- Correct span drift with a zero trim, since it will only mask the error near 0% and leave full scale wrong
- Assume a large error near full scale is the same root cause as a zero shift issue
- Skip the full-scale check because venting to zero is faster and more convenient
- Ignore mechanical stress on the process connection as a possible contributor to span error
Worth Reading if You Want to Go Deeper
Questions Students and Technicians Often Ask
External References
- What Causes Zero and Span Offset in Pressure Transducers?, Ashcroft
- What Causes Analog Drift?, RealPars
What we learn today
- Pressure transmitter span drift is a change in calibration slope, so the error is small near zero and grows toward full scale, unlike zero shift where the error stays constant across the range.
- The six main causes are the temperature coefficient of span, sensor diaphragm fatigue, electronic component drift, static pressure effect on DP transmitters, overdue recalibration, and mechanical installation stress.
- A worked two-point example showed a transmitter reading 4.12 mA at 0% and 19.30 mA at 100%, giving a −5.13% span error and a correction factor of 1.0540 after six years in service.
- A zero trim cannot fix span drift, since it only corrects the 0% point, a proper two-point recalibration with a span adjustment is required instead.
- Logging drift rate across multiple calibration cycles, not just the current error, helps predict when a transmitter will next need attention before it fails a compliance check.