How to Measure Oil Tank Temperature with a Transmitter
A temperature reading is only useful when it matches the process, appears where operators need it, and fits the accuracy required by the job. In an oil tank, that means the instrument must handle the expected liquid temperature range, send a reliable signal to a remote system, and still give a clear local indication at the tank.
This example walks through a simple process measurement case: an oil tank with continuous temperature monitoring from -10 °C to 90 °C, with an accuracy requirement of 1% of span or better.
The practical answer is to use a temperature transmitter, configured so the process range matches the standard 4 to 20 mA output signal.

What the measurement system must do
The process requirement is straightforward:
Requirement | Value |
Minimum temperature | -10 °C |
Maximum temperature | 90 °C |
Required accuracy | 1% of span or better |
Local indication | Required |
Remote indication | Required |
The temperature must be measured continuously. A handheld thermometer or periodic manual check would not meet that need. The system must send a live signal to another location, such as a control room, PLC, recorder, or monitoring panel.
Because the value must be shown both locally and remotely, a temperature transmitter with a local display is a suitable choice. The transmitter accepts the sensing element signal, converts it into a standard output, and can show the measured value at the tank.
In a typical installation, the measurement system includes:
A temperature sensor, often an RTD or thermocouple
A thermowell, when the sensor needs process protection
A temperature transmitter
A local display, either built into the transmitter or mounted nearby
A remote receiving device, such as a PLC, DCS input, indicator, or recorder
A 4 to 20 mA current loop or other approved signal connection
For this example, the key point is the transmitter setup. The process range and transmitter output must match correctly.
Define the range before selecting the transmitter
The range is the part of the process variable that the measurement system must cover. In this case, the oil temperature can go as low as -10 °C and as high as 90 °C.
So the required measuring range is:
```text
Range = -10 °C to 90 °C
```
The transmitter must be configurable for that range. It should not merely be capable of measuring temperature in a broad catalog sense. It must support this specific calibrated range and still meet the accuracy requirement.
For example, a transmitter that can be configured from -50 °C to 150 °C may be suitable, but only if its performance over the configured span meets the required accuracy. A broad measuring capability does not automatically mean the installed measurement is accurate enough.
A good selection should answer these questions:
Can the transmitter be ranged from -10 °C to 90 °C?
Can the sensor operate correctly in oil service over that range?
Does the transmitter provide a local display?
Can it send a remote standard signal?
Does the total measurement meet or beat ±1 °C for this application?
That last question comes from the span calculation.

Calculate the span of the measurement
The span is the difference between the upper range value and the lower range value.
For this oil tank:
```text
Upper range value = 90 °C
Lower range value = -10 °C
Span = Upper range value - Lower range value
Span = 90 °C - (-10 °C)
Span = 100 °C
```
So the transmitter span is:
```text
Span = 100 °C
```
This is one of the most important numbers in the setup because the accuracy requirement is stated as a percentage of span.
When accuracy is expressed as a percentage of span, it does not refer to the full physical capability of the transmitter. It refers to the configured measurement interval. Here, that interval is 100 °C.
Convert the accuracy requirement into degrees Celsius
The process requires accuracy of 1% of span or better.
Since the span is 100 °C:
```text
Required accuracy = 1% × 100 °C
Required accuracy = 1 °C
```
That means the measurement error should not exceed:
```text
±1 °C
```
If the actual oil temperature is 40 °C, the indicated value should remain within the allowed error band, assuming normal operating conditions and proper installation:
```text
Acceptable indication = 39 °C to 41 °C
```
This does not mean the reading will always move in full 1 °C steps. It means the maximum allowed error is ±1 °C based on the stated process requirement.
For this example, a transmitter accuracy of 1% of span means each temperature reading should stay within ±1 °C of the actual process temperature.
In real projects, the total error can include more than the transmitter. Sensor tolerance, thermowell placement, wiring, input card accuracy, display resolution, and calibration practice can all affect the final reading. For a basic process measurement example, though, the transmitter must be selected and calibrated so the specified range meets the ±1 °C target.
Understand depression in this range
This example includes a lower range value below zero:
```text
Lower range value = -10 °C
```
The range does not start at 0 °C. It starts 10 °C below zero. This is often described as depression in instrument ranging.
The amount of depression is:
```text
Depression = 10 °C
```
As a percentage of span:
```text
Depression percentage = 10 °C ÷ 100 °C × 100
Depression percentage = 10%
```
So the measurement has:
```text
Depression = 10 °C, or 10% of span
```
This matters because the transmitter output must still be tied to the actual lower and upper range values, not to 0 °C. The lower output signal represents -10 °C, not 0 °C.
If this relationship is misunderstood, the remote display can show the wrong value even though the current loop appears to be working.

Match the process range to the 4 to 20 mA signal
A common analog transmitter output is 4 to 20 mA. The signal does not begin at 0 mA. The live zero of 4 mA helps distinguish a valid low reading from a broken loop or loss of power in many systems.
For this application, the transmitter should be calibrated as follows:
Process temperature | Transmitter output |
-10 °C | 4 mA |
90 °C | 20 mA |
In instrument terms:
Term | Value |
Lower range value | -10 °C |
Upper range value | 90 °C |
Lower output value | 4 mA |
Upper output value | 20 mA |
Output span | 16 mA |
The current output changes from 4 mA to 20 mA across a temperature span of 100 °C. That means the 16 mA output span represents 100 °C of process temperature.
A simple linear relationship applies:
```text
Output current = 4 mA + [(Measured temperature - LRV) ÷ Span] × 16 mA
```
For this example:
```text
Output current = 4 mA + [(Measured temperature - (-10 °C)) ÷ 100 °C] × 16 mA
```
At the lower range value:
```text
Measured temperature = -10 °C
Output = 4 mA + [(-10 - (-10)) ÷ 100] × 16
Output = 4 mA
```
At the upper range value:
```text
Measured temperature = 90 °C
Output = 4 mA + [(90 - (-10)) ÷ 100] × 16
Output = 20 mA
```
At the midpoint, 40 °C:
```text
Measured temperature = 40 °C
Output = 4 mA + [(40 - (-10)) ÷ 100] × 16
Output = 4 mA + (50 ÷ 100) × 16
Output = 12 mA
```
So a remote device receiving 12 mA should interpret it as 40 °C, assuming the receiver is scaled with the same -10 °C to 90 °C range.
Configure the remote display with the same scaling
The transmitter is only half of the measurement path. The remote device must use the same scaling.
If the transmitter sends:
```text
4 mA = -10 °C
20 mA = 90 °C
```
then the remote display, PLC, or control system input must be configured the same way.
A common mistake is to scale the receiving device as:
```text
4 mA = 0 °C
20 mA = 100 °C
```
That looks close because the span is still 100 °C, but it is wrong for this process. Every displayed value would be shifted by 10 °C.
For example, if the tank is actually at 40 °C, the transmitter sends 12 mA. With correct scaling, the remote display shows 40 °C. With incorrect 0 to 100 °C scaling, the same 12 mA would show 50 °C.
That error is far larger than the allowed ±1 °C accuracy.
Correct scaling is simple, but it must be applied at every point where the signal is interpreted:
Transmitter configuration
Local display configuration, if separate
Remote indicator configuration
PLC or DCS analog input scaling
Historian or monitoring software tag scaling
A measurement loop is only accurate when the whole loop agrees on what the signal means.
Check the installation details that affect accuracy
The calculation proves what the transmitter needs to do. The field installation determines whether it can do it well.
For oil tank temperature measurement, pay attention to these details.
Place the sensor where it represents the liquid
The sensor should measure the oil temperature, not the air above the liquid or the temperature of the tank wall. If the tank has stratification, where warmer and cooler layers form, one measurement point may not represent the whole tank. The correct location depends on the process purpose.
For inventory, quality, or process control, the measuring point should represent the temperature that matters to that operation.
Use a suitable thermowell when needed
A thermowell protects the sensor from the process and allows sensor removal without opening the tank in many installations. It can also slow the response. The thermowell material, length, insertion depth, and fit all affect measurement performance.
For a tank with slow temperature changes, response time may not be critical. For faster processes, it can matter.
Keep wiring and grounding clean
RTD and thermocouple signals are small before they reach the transmitter. Poor wiring, loose terminals, electrical noise, or grounding issues can cause unstable readings. Mounting the transmitter close to the sensor can reduce signal problems, especially before conversion to 4 to 20 mA.
Verify both local and remote readings
After calibration, compare the transmitter display with the remote display at known test points. If both are scaled correctly, they should agree within the expected tolerance of the system.
This step catches scaling errors, display setup mistakes, and wiring issues before the measurement is trusted in operation.

A complete setup for the example oil tank
For the stated process, the selected temperature transmitter should be configured with these values:
Item | Required value |
Measured variable | Oil tank temperature |
Lower range value | -10 °C |
Upper range value | 90 °C |
Span | 100 °C |
Depression | 10 °C |
Depression as span percentage | 10% |
Required accuracy | ±1 °C or better |
Local indication | Transmitter display or local indicator |
Remote signal | 4 to 20 mA |
4 mA value | -10 °C |
20 mA value | 90 °C |
This is the core of how to measure oil tank temperature with a transmitter. Select a transmitter that supports the required range, confirm that its accuracy meets the process requirement, and scale its output so the minimum and maximum temperatures match 4 mA and 20 mA.
The final success check is simple: when the tank temperature is at the lower range value, the transmitter sends 4 mA and the displays show -10 °C. When the temperature is at the upper range value, it sends 20 mA and the displays show 90 °C. Across the full span, the measurement stays within ±1 °C.
A reliable measurement starts with clear numbers. In this case, those numbers are a -10 °C to 90 °C range, a 100 °C span, 10 °C of depression, ±1 °C allowable error, and a correctly scaled 4 to 20 mA output.
Need help with your application?
ProSense Instruments — explore instruments, sensors, data loggers and test equipment.
ProSense Solutions — discuss complete monitoring and automation systems.
ProSight Software — explore cloud dashboards, alarms and reporting. Device integration must be confirmed for your application.
Instrument enquiries: sales@prosenseinstruments.com.au | 1800 560 854
Project enquiries: sales@prosense.com.au. Software enquiries: info@prosightsoftware.com.


Comments