Hydrostatic Tank Level: Separate Density Error from Pressure Zero Offset
Updated: Nov 12

AI-generated illustrative scene, not a product screenshot or a completed-project photograph. The worked calculation below uses explicit example inputs.
A tank-level trend can look smooth and believable even when its conversion is wrong. A pressure sensor does not directly measure cubic metres or percentage full. It measures pressure at its sensing point; the monitoring system converts that signal using assumptions about the liquid and the measurement reference. Two incorrect assumptions can even partly cancel at one operating level, making a single spot check reassuring for the wrong reason.
This example concerns a quiet, vented tank containing uniform-density water. It is an offline calculation review, not an instruction to recalibrate a working instrument, open a tank or change pump controls. Pressurized vessels, wet legs and remote seals require their own pressure-reference treatment.
Name the three references first
For an atmospheric tank with an appropriate gauge-pressure reference, liquid depth above the sensor is h = p / (ρg). WIKA's technical explanation identifies density as a required input and discusses its temperature dependence. The sensor's elevation is a separate issue: depth above a side-mounted diaphragm is not automatically level above the tank floor.
Record the pressure unit, pressure type, configured density and diaphragm elevation before comparing the displayed level with a drawing. If pressure is reported in kPa, multiply by 1,000 before using density in kg/m³ and g in m/s². An absolute-pressure reading cannot be inserted unchanged into the gauge-pressure equation.
Work through a known-depth example
Assume an actual water depth of 2.400 m above the sensor, uniform density 998 kg/m³ and g = 9.81 m/s². The density is an explicit illustrative input, not a claim about the temperature or composition of a particular tank. First calculate the pressure that would exist without sensor error.
True gauge pressure = 998 × 9.81 × 2.400 = 23,496.912 Pa = 23.496912 kPa.
Suppose the monitoring conversion instead assumes 1,000 kg/m³. With a perfect pressure reading, its displayed depth would be slightly low.
Density-only indicated depth = 23,496.912 / (1,000 × 9.81) = 2.3952 m; error = −4.8 mm.
Now add an assumed positive pressure zero offset of 0.500 kPa. The indicated pressure becomes 23.996912 kPa. Keep this deliberately separate from the density mismatch.
Combined indicated depth = 23,996.912 / (1,000 × 9.81) = 2.44617 m; error relative to 2.400 m = +46.17 mm.
The density mismatch did not cause that positive error. It slightly reduced the error introduced by the pressure offset. Correcting only the configured density would leave a significant offset problem and could make the single-point disagreement larger. That is why unexplained agreement at one level is not adequate evidence that both settings are right.

Original assumed vented-water-tank example; not a calibration instruction.
Compare slope and offset rather than one percentage
For this simplified case, indicated depth equals actual depth multiplied by the actual-to-configured density ratio, plus the pressure offset divided by configured density and gravity. A density mismatch changes the slope of the level relationship. A constant pressure offset adds a fixed depth error. Their relative importance therefore changes as the tank level changes.
At a true depth of 0.600 m, the same assumptions give 0.64977 m indicated: approximately +49.77 mm. The absolute error is similar, but its percentage of the true depth is much larger. Do not report one full-scale percentage as though it were the percentage error at every operating level.
Carry level into volume only after geometry is established
If the diaphragm is 0.150 m above a flat tank floor, the true surface elevation above that floor is 2.550 m in the first example. Adding the sensor elevation does not correct pressure or density errors; it changes the datum. In a constant-area tank, a level error can be multiplied by plan area to estimate its volume effect. For a horizontal cylinder or shaped bottom, use the applicable level-volume relationship instead.
A useful review record keeps raw pressure, timestamp, reference type, water condition, conversion density, sensor elevation and an independently obtained level observation together. Label uncertain inputs rather than tuning the calculation until it matches a preferred answer. Any instrument adjustment or control change belongs to the authorized commissioning process, with the instrument manufacturer's procedure and competent personnel.
Keep the next design task separate
For the next stage of a tank-and-booster review, use the verified level and pressure references to define the water-system boundary before checking the associated piping. The MEP Pipe Design and Sizing Handbook is a related reference for plumbing, booster pumps and pipe-sizing calculations; it is not an instrument calibration manual and this article's example is separate.
Continue the calculation review
Sources and calculation basis
Primary references checked on 27 October 2026. The worked examples and diagrams are original illustrative calculations, not manufacturer selections or project approvals. Apply the stated limitations and the requirements governing the actual project.

Comments