• By Denis Richard, Jesse Cameron
  • InTech

Control the temperature of the fluids and air servicing a building.

The costs of ignoring temperature accuracy in energy measurement (British Thermal Unit) can be worse than expected. Inaccuracy occurs when poor techniques and equipment are used in the system, which can result in wasting thousands of dollars every year. This waste can be avoided by paying attention to the details. For instance, the temperature sensors used in the thermal energy system must be a matched pair, or uncertainties could instantly increase depending on the sensor’s performance. No matter how big or small the BTU measurement system is, it is important to comprehend that temperature sensors are a critical part of the system and should be procured, installed, and maintained correctly.

Maintaining proper measurement and control of the temperature of the fluids and air servicing the building is often thought of as a simple task. Still, it requires effort and diligence. Another major issue that is overlooked when caring for thermal BTU metering is the comfort of the people inside the building. It is important to maintain proper control, so the individuals can perform at optimal efficiencies. In this article, we will discuss:

  • fundamentals of BTU measurements
  • roles of relative temperature
  • techniques used to effectively measure temperature and reduce uncertainties
  • suitable temperature sensors.

Fundamentals of BTU measurements

BTU is a thermal energy unit used primarily in residential and commercial buildings to measure the energy used for heating, cooling, operations, and other applications involving the transfer of fluid for energy purposes. Systems typically contain a central heating or cooling unit and distributes energy throughout the facility. A BTU is defined as the amount of heat required to raise the temperature of one pound of water by one degree Fahrenheit. To give perspective on the unit, a gallon of gasoline contains about 125,000 BTU. It is an energy unit, like calories.

The governing equation for BTU is Q=mCp∆T. Q is the amount of energy, m is the mass, Cp is the specific heat of the fluid, and ∆T is the temperature difference between the two temperature sensors.

BTU metering applications use mass flow rate, , as many systems have circulating fluids instead of static reservoirs to circulate the energy. Thus, the equation becomes Q̇ =ṁCp∆T. This is the rate of heat transfer, which is multiplied by the time to get the total amount of energy that passed through the system.

Figure 1. A BTU metering system includes two temperature sensors, a flowmeter, and a BTU meter.

The instrumentation that is used in the system normally contains one set of matched-pair temperature sensors, one flowmeter, and one BTU meter with data logging and communication capabilities.

Resistance temperature detectors (RTDs) are the most common type of temperature sensor used in BTU metering. The measurement of the RTD resistance is made by passing a small current across the RTD and measuring the voltage drop across the RTD as its resistance varies with temperature. The change in voltage is then converted to a change in temperature electronically.

One temperature sensor is inserted into the inlet pipe, and one is inserted into the outlet pipe to capture the difference between the fluid temperatures after it travels through the system.

Roles of relative temperature

Figure 2. Matched-pair temperature sensors are important for increasing the accuracy of BTU metering.
The relative temperature between the input and output is more important to quantify BTUs than the absolute temperature at those locations. The equation for BTU is dependent on the difference between the two sensors, which is not relevant to their absolute temperature. For example, having the inlet temperature sensor at 140°F and the outlet temperature at 104°F does not add more energy into the system than if the values were much lower at 86°F and 50°F. They will input the same amount of energy, because they both have a temperature difference of 36°F.

High-quality resistance temperature detectors have accuracies of DIN Class A ± 0.15 + 0.002*|T|°C. This can result in the worst case to uncertainties of 0.27°F at 32°F and 0.63°F at 212°F, but this uncertainty is increased when they are not a matched pair. The relative difference between them can go as high as two times these values.

When temperature sensors are calibrated by manufacturers, they generally use a fluid bath, as it is the most stable medium, and create a two- or three-point resistance versus temperature (RvsT) calibration curve. This curve may change slightly between calibrations, as each RTD does contain its own manufacturing uncertainties. Once the curves are created, it is possible to match the RTDs that have the most similar curves. This is a time-consuming process and will increase the costs of the matched-pair sensors compared to buying the sensors without a matched pair. This purchase cost increase could represent substantial money over time.

Techniques to effectively measure temperature and reduce uncertainties

There are techniques that can save costs when implemented. First, place the temperature sensors to the extremities of the inlet and the outlet of the system. It is best to go as close as possible, but ensure space is left for servicing. The placement should be reachable from floor level when possible and as far away from any sources of electrical noise as possible.

Account for the additional resistance from the lead wires of the supplied temperature sensors. This is typically corrected from within the BTU meter and should always be verified on the specific unit. If the BTU metering unit comes already calibrated with dedicated temperature sensors, the wires cannot be cut or modified, or this would cause a significant reduction in the accuracy of the BTU metering. To prevent any tampering with lead wires, many manufacturers will prevent any access to the PCB port terminals. The use of shield cable is also required to avoid errors from interference.

Lead wires are also affected by the surrounding atmosphere and improper power usage, which will increase the drift of the sensor. This causes additional errors in the readings.

Ideally, temperature sensors will be inserted into the conduit used to circulate the water, steam, or other fluid. The best performance occurs when the tip is directly in the path of the fluid. This ensures that the temperature reading is coming directly from the fluid flow. If this is not possible in certain circumstances, then surface-mount temperature sensors should be used.
Control the temperature of the fluids and air servicing a building.

Figure 3. Proper mechanical installation requires the tip of the sensor to be in line with the fluid flow.

Circumstances that would lead to the need for surface-mount sensors could include a pressurized system, where shutdown is not an option, or higher costs to design for insertion.

When using surface-mount temperature sensors, using thermal paste to ensure a good transfer between the conduit and the sensor is recommended. It is also good to add some insulation over the temperature sensor to minimize the heat exchange with the environment. Modern BTU meters can account for the heat transfer losses through the piping and would include some form of offset constant.

Besides the temperature probe positioning, there are many other factors that can reduce the accuracy of the readings. Generic sources of error from RTDs include:

  • Interchangeability: Interchangeability refers to the “closeness of agreement” between an actual RTD RvsT relationship and a predefined RvsT relationship.
  • Insulation resistance: Current leaks into or out of the circuit from the body of the sensor or between the element leads.
  • Stability: Ability to maintain RvsT over time as a result of thermal exposure.
  • Repeatability: Ability to maintain RvsT under the same conditions after experiencing thermal cycling throughout a specified temperature range.
  • Hysteresis: Ability to maintain RvsT relationship when approaching temperatures from different directions and magnitude.
  • Stem/body conduction: Error that results from the RTD sheath/body conducting heat into or out of the process. Proper mechanical installation is critical to minimize stem or body conduction errors.
  • Calibration/interpolation: Errors that occur due to calibration uncertainty at the calibration points or between calibration points due to propagation of uncertainty or curve fit errors.
  • Lead wire: Discussed previously.
  • Self-heating: Since the RTD is a resistive device, it acts as a small heater. Self-heating errors can be higher on a 1000 ohm RTD. Proper electronics with lower measuring currents must be used on 1000 RTDs.
  • Time response: Errors are produced only during temperature transients, because the RTD cannot respond to change fast enough.
    Figure 4. Surface-mount temperature sensors are required in certain circumstances when in-line temperature sensors are not an option.

When uncertainties are not maintained, costs can quickly creep up. See below for the costs with simply 1.0°F of uncertainty at room temperature using water flowing at 100 gallons per minute. Assuming the system runs 24/7 for 365 days a year, then the total BTUs wasted would be 437,824,800 and would result in a cost of $15,393 at an energy consumption rate of 12¢/kWh.

Suitable temperature sensors

RTDs are the most suitable temperature sensors to use for BTU metering because of their excellent stability and repeatability over their full calibration curve.

There are some RTD errors that can be resolved by incorporating advanced sensor technologies, such as a microprocessor-based transmitter for an RTD. These transmitter technologies are capable of reducing RTD uncertainties, such as interchangeability, lead wire errors, self-heating, and thermal EMF. During their calibration procedure, they can be electronically replicated, creating a matched pair. When two temperature sensors with integrated transmitters are calibrated together, they essentially make an exactly matched calibration curve with each other.

Advanced circuitry, like the one found in Intempco’s MIST series, utilizes a three-wire configuration with matched current sources to deliver a constant current excitation to the RTD. The matched current sources are also used to generate the reference voltage that is used for the digital conversion of the RTD signal. That type of architecture improves the accuracy because of a radiometric reference voltage and also eliminates the error due to the RTD wires, making it the best solution for high-accuracy application like BTU measurement.

Figure 5. Microprocessor-based transmitter technology is enabling higher accuracy readings for RTDs. Intempco’s MIST probes are on the left, and Intempco’s DTG series is on the right.

Other features, such as a digital display, reprogrammability, and rescalability, can be extremely advantageous for the system. Digital displays allow technicians and inspectors to visually determine the temperatures at a point in the system. This is helpful when troubleshooting. Having the ability to reprogram the temperature sensor can also eliminate the need for a complete replacement if a system parameter has changed. For example, suppose the output was originally 0–5 Vdc, but after upgrading your BTU meter, now it only accepts 0–10 Vdc, then the ability to change output voltages without losing any accuracy is integrated into the sensor. The same principles apply to the temperature range. If at any time the temperature range changes in the system, then the scale can be adjusted in the sensor without the loss of accuracy.

Crucial aspect

Thus, increasing the accuracy of the temperature measurement is a crucial aspect of proper BTU metering. When temperature sensors are matched, they will perform much better than individual temperature sensors. It is important to reduce the errors by choosing high-quality RTDs that can be adjusted by having an integrated transmitter built into it or by the BTU meter itself. These upgrades could save tremendous costs over long periods of time.

Reader Feedback

We want to hear from you! Please send us your comments and questions about this topic to InTechmagazine@isa.org.

Like This Article?

Subscribe Now!

About The Authors

Denis Richard has been the senior electrical engineer with Intempco for more than 20 years. He works closely with clients to design custom PCB boards for instrumentation designs. He is responsible for managing all electronic assembly and design for all product lines. Richard also leads R&D projects and the development of new products.

Jesse Cameron has a degree in sustainable design engineering and has been a product engineer at Intempco for the past three years. He assists with product documentation and marketing materials. Cameron also has experience developing products involving thermodynamic principles.