by Frank Spevak
On August 27, The Energy Conservatory (TEC) participated in a webinar called “Meeting the RESNET Equipment Calibration Requirements. Learn from the Experts,” hosted by the Residential Energy Services Network (RESNET).
While preparing for this webinar, we started thinking about the most important issues related to measurement and calibration. To ensure our gauge users are getting the most accurate readings, we have listed four factors every building tester should know about measurements and calibration of gauges.
The energy rater world relies on many types of measurements. These are just some of the many items that help energy raters gather:
- Air tightness tests of the building
- Air tightness tests of the ductwork
- Pressure measurements around the house and around the combustion appliance zone
- Flow measurements at the air handler and at register supplies and returns
- Temperature and humidity measurements
- Physical measurements of walls, windows and insulation
In the case of physical measurements, just about anyone can approximate an inch or a foot. I can pace out large measurements just by walking and can be pretty darn close to the actual measurement. Some people, especially those in the engineering professions, are very good at how much one thousandth of an inch can fix or hinder a new product. Many people have built-in thermometers and can judge air or surface temperatures.
But when we get into the realms of pressure, flow and velocity, a lot of our visual cues of the world can’t help us. We need to rely on instruments to translate a pressure value into something we can understand. Same for flow and velocity.
Many years ago an engineer at the National Institute of Standards and Technology (NIST) made a comment regarding the difficulty of measurements and accuracy that resonated with me. He said that NIST can measure time to around 20 decimal places and distance (feet or inches) to around 12 decimal places, but when you put the two together, as in cubic feet per minute, everything goes to pot.
With this information as a backdrop I thought that I would highlight the four most important issues related to measurement and calibration.
1. The real world is different than the laboratory world
Calibration: To alter the raw characteristics of the device to be in alignment with a known measurement value
As a manufacturer of instruments, TEC calibrates thousands of instruments each year. The location of our measurements is always the same; there is no wind; the temperature and humidity vary slightly, but we adjust our readings to take those values into account when we produce the final calibration. With a sense of controlling many of the variables, we are able to say that our gauge is accurate to plus or minus 1% or that our fans and gauges as a system are able to measure flow to plus or minus 3%.
The conditions in the lab are very different from those at a house, say in January, in northern Minnesota, and the outside temperature is minus 10 F. Oh, and the wind is blowing the snow around at 10 to 15 mph. Then you have to retest the house in August, in northern Minnesota, with the temperature in the 90s and dew points in the 70s or even higher.
Just as we account for temperature and humidity in the lab, it can also be done out in the real world, but no matter what, conditions change from hour to hour, month to month and season to season. So while the instruments will still be accurate to 1% or 3%, the conditions may have changed such that there can be very different readings even when no work was performed to change the house.
2. Accuracy is a range of acceptable values
Instrument manufacturers such as TEC make thousands of products each year. To make those instruments, we buy thousands of parts from a wide variety of other manufacturers. Each of those parts has a range of tolerances and accuracies, so we choose parts that have high accuracy and low tolerances.
Even with our quality part selection process and our ability to incorporate extensive computational algorithms, some gauges will be slightly high, some slightly low, and some right in the middle. And even when they are right in the middle at 1000 Pascals (Pa) the values can be very different at say 500, 100 or 25 Pa. The result of the variation in parts, conditions and range is the manufacturer’s specification called accuracy. This is stated as plus or minus a given value. Most of the time that value is stated as a percentage, sometimes stated as a fixed value and sometimes as a combination of percentage and fixed value.
With an accuracy of plus or minus 1%, the measurement of 1000 Pa can actually result in a variety of values anywhere from 990 Pa to 1010 Pa. So while one gauge could be reading 990 Pa and another reading 1010 Pa, given that the standard is 1000 Pa, they are both accurate. The same is true for flow measurements. In this case it is plus or minus 3%. So when someone says they are measuring what is supposed to be 3000 cubic feet per minute (CFM), the range of acceptable values is 2910 CFM and 3090 CFM. Even though they are 190 CFM different, they are both accurate within the specified level of accuracy. Take care when trying to determine which instrument is more accurate than the other.
3. Factory calibration is different from a field calibration check
When we perform a factory calibration on an instrument, we are able to go inside the gauge and make adjustments so it will display values that correspond to the standard values. This requires computer software to simultaneously read values of the gauge and the standard, as well as enter in the various formula changes to the gauge. Our pressure standard devices only leave our building to go back to the manufacturer for their regularly scheduled calibration interval.
A field calibration check, on the other hand, is a way for anyone, at any time to check if their gauge needs to be returned for factory calibration. This is done by comparing one gauge to another. We recommend when a gauge is returned to you after a factory calibration, you use that gauge to check all of your other gauges to see if any of them are out of calibration. Many organizations allow their members to perform field calibration checks in lieu of a factory calibration. While we recommend a two year factory calibration interval, periodic field calibration checks ensure your process as well as your instruments are in good working order.
4. Accuracy of the instrument is based on each individual value compared to the standard, not an average of individual accuracy values
When a gauge receives a factory calibration, each point above the range of the instrument is compared to a standard value. Each point has the accuracy calculated. If an instrument has an accuracy specification of +/- 1% then each of the individual points must be less than 1%. If any single point is greater than the manufacturer’s specification then the entire instrument is deemed out of calibration.
The results of a factory calibration are reported on a calibration certificate that accompanies each instrument when it is returned to the customer. Some of your customers will require you to provide a copy of the certificate with your report documentation, so we recommend keeping the certificate with the instrument or having a copy with the instrument while the original is filed in the office. Your customers are looking for how old the certificate is as well as reviewing the values and accuracy of each point of the instrument.
Accuracy and calibration are different parts of the same issue – how do you know if your instrument is measuring what it is supposed to? When using or purchasing a pressure gauge, consider the accuracy of what you need, the values you are measuring, the ease of factory calibration, and the actual accuracy of the gauge.