What Are Span and Zero?

Back to Blog

What Zero and Span Actually Set

Every analog instrument maps a physical input to a proportional output over a defined range. Zero is the lower range value - the input at which the output should sit at its minimum, typically 4 mA in a 4-20 mA loop. Span is the difference between the upper and lower range values, in other words the measurement width. A transmitter ranged 0 to 300 psi has a zero at 0 psi and a span of 300 psi.

Adjusting zero shifts the whole output line up or down without changing its slope. Adjusting span changes the slope - how much output change results from a given input change. Both must be correct for readings across the entire range to be accurate.

Zero Shift vs Span Error

The two failure modes look different on a calibration sheet. A zero shift adds a constant offset to every point - if the low end reads high by 2 percent, so does the middle and the top. A span error scales with the reading - the error is small near zero and grows toward the high end, because the slope of the output is wrong.

Recognizing which is which speeds up correction. A pure offset is fixed by adjusting zero alone; a growing error usually means span needs correction, often followed by a zero re-check because the two adjustments can interact on some instruments. Modern smart transmitters separate these as a digital zero trim and span trim.

Ranging and Re-Ranging in the Field

Because span and zero define the calibrated range, changing them is how an instrument is re-ranged to suit a new process condition - for example narrowing a level transmitter's span to gain resolution on a smaller tank. On a HART or smart transmitter this can be done from a handheld communicator without applying physical pressure, though a full loop check afterward confirms the result.

Span and zero live entirely in the field instrument and controller. A cloud SCADA like Merobix reads the scaled engineering value the controller reports; it does not set the transmitter's zero or span. The field ranging determines the meaning of the number; the SCADA displays and trends it.

The 4-20 mA Math, Worked Through

The whole relationship fits in one line: output equals 4 mA plus 16 mA times the fraction of span applied. Take the page's 0 to 300 psi transmitter. At 150 psi the input is halfway up the span, so the output is 4 plus 16 times one half, which is 12 mA. At 75 psi, one quarter of span, the output is 8 mA. Going the other way, a reading of 16 mA means 16 minus 4 over 16, or three quarters of span - 225 psi. Every scaling block in the controller is just this arithmetic run in reverse.

Now put errors into the same formula. A zero error adds a constant to the output line: every point along the range reads high or low by the same milliamp amount, whether the process is at the bottom or the top. A span error multiplies instead of adds - the output slope is wrong, so the error is negligible near the lower range value and largest at the upper one. This is why a calibration sheet with readings at several points tells you immediately which adjustment is at fault: a parallel shift is zero, a fan-shaped divergence is span.

Trim, Range, and Which Adjustment You Actually Need

On a smart transmitter, three different operations get lumped under the word calibration, and choosing the wrong one wastes a trip to the field. Sensor trim aligns the transmitter's internal digital measurement against a trusted reference standard. Output trim aligns the digital-to-analog stage so the mA on the wire matches what the transmitter thinks it is sending. Re-ranging just changes which input values map to 4 and 20 mA and touches neither measurement nor output accuracy. The distinction between sensor trim and output trim matters because re-ranging a transmitter whose sensor reads wrong simply relocates the error.

The decision is mechanical once you frame it. If a reference standard disagrees with the transmitter's digital reading, the sensor needs trimming. If the digital reading is right but the measured loop current is wrong, trim the output. If both are healthy and the process window has changed - a vessel now runs at different pressures, a tank was re-purposed - then range and re-range the transmitter to put the 4-20 mA window where the resolution is needed. Record as-found values before touching anything, so drift history stays intact.

Catching Zero and Span Errors Early

Routine verification is where these errors get caught before they cost money. A 5-point calibration check - rising and falling through the range - separates the failure modes cleanly, because each one leaves a distinct signature across the points. A single-point check at zero, by contrast, can pass a transmitter whose span is badly off, since span error hides at the bottom of the range.

The signatures read like this:

Signature on the sheet	Error	Correction
Same offset at every point	Zero shift	Zero adjustment or zero trim
Error grows toward the top of range	Span error	Span adjustment, then re-check zero
Ends good, middle off	Linearity	Sensor trim, or manufacturer service if it persists
Rising and falling readings disagree	Hysteresis	Investigate mechanical causes per the datasheet

How often to run the check is site-specific: criticality of the measurement, the instrument's drift history, and any regulatory or custody requirements set the interval, and a transmitter that keeps showing the same drift direction between checks is telling you to shorten it.

Frequently Asked Questions