Every measurement device has some degree of imprecision. A ruler may tell you that two pencils are the same length, but that deduction is only sound within the smallest measurement possible by the ruler. The pencils might match up to the centimeter or millimeter, but there is no guarantee they’re not different lengths when considering micrometers or nanometers.
No matter the tool, there will always be a limit to its ability to capture the true measurement of the measurand, i.e., the object the tool is measuring. There will always be a range of values that the true measurement of the measurand can lie within. We call that phenomenon measurement uncertainty. Just like every other measurement device, pressure gauges are susceptible to measurement uncertainty, and it's important to know how different variables affect a device’s uncertainty.
In this article, we’ll walk through the general considerations for pressure measurement uncertainty, the types of uncertainty at play, and briefly cover how that uncertainty is calculated.
General Considerations for Pressure Measurement Uncertainty
There are two primary types of uncertainty: Type A and Type B uncertainty. We’ll discuss each of these in more detail later in this article, but in simple terms, Type A uncertainty is based on statistical analysis and Type B uncertainty is not. (And together, they make up the overall measurement uncertainty.)
However, Type A and Type B are very broad classifications. When it comes to pressure gauges, some of the more specific potential sources of measurement uncertainty include:
- Type of measurement device
- Operating mode of the measurement device
- Fluid media
- Environment
- Orientation of the device
- Reference uncertainty
- Calibration frequency
- Dwell time at setpoint
- Zeroing
Let’s review each of these in more detail.
Type of Measurement Device
This is the simplest consideration for pressure gauge uncertainty: What type of device are you using? Different pressure gauges operate differently, have different calibration processes, and can react differently depending on the other considerations we’ve mentioned, like fluid media, operating mode, and the environment in which someone uses a pressure gauge. Pressure gauges also have different levels of precision and resolution. Many brands and models of gauges come with a list of uncertainties and details about how different situations and considerations will affect the final measurement.
Operating Mode of the Measurement Device
The exact operating mode of the pressure-measuring device will, to a degree, affect its measurement uncertainty. Some operating modes include absolute pressure and gauge pressure. The operating mode generally does not have as big of an influence on uncertainty as some other variables, but it’s still a factor you’ll need to consider.
Fluid Media
The fluid inside the gauge that relays the pressure of whatever it’s measuring can cause some uncertainty, too, since the type and density of the fluid can change. Many pressure gauges use liquid glycerin or silicon oils, while others might actually use a gas — either air or nitrogen. The larger differences between these media are usually taken into account in the measurement, but smaller differences in density between measurements can lead to uncertainty.
Environment
The environment in which you take your measurement is one of the most important aspects to consider when determining the validity of your results. In fact, environment is one of the most important influences to consider when determining the validity of your results. Environmental factors not only affect the pressure of whatever you’re measuring but the actual measurement device you’re using.
For instance, fluctuations in temperature can change the physical properties of the pressure gauge in small but meaningful ways, such as changing how the fluid media inside the gauge behaves and therefore changing the readings. Temperature can also subtly affect the different components of a pressure gauge, such as a piston-cylinder, via thermal expansion and contraction, leading to uncertainty in the measurement. Additionally, adiabatic effects that change the temperature of the media in a system can alter pressure gauge readings, especially in gas-based systems or systems that experience rapid pressure changes.
Most advanced pressure gauges will have some level of temperature compensation, but there will always be some level of uncertainty no matter how controlled the conditions are.
Another environmental consideration is barometric pressure, or atmospheric pressure. Barometric pressure changes primarily depending on your elevation: Higher elevations lead to lower barometric pressure and lower elevations lead to higher barometric pressure. Because this atmospheric pressure largely depends on the elevation of your location, you need to account for it when measuring pressure.
Local gravity is another environmental factor you’ll need to consider when determining uncertainty for your pressure measurements. Local gravity depends on several things, including latitude, nearby topographical features like mountains, and elevation. While small differences in local gravity will not be meaningfully noticeable for a human, they are critical to account for when you’re taking extremely precise measurements. These tiny changes in gravitational force can mean big differences in exact pressure measurement and lead to some uncertainty. Particularly with deadweight testers that calibrate pressure measurement tools and need to reproduce a known pressure point based on mass, gravity, and area, local gravity is a crucial input.
Yet another, perhaps less obvious, environmental factor to consider is the presence of vibration. Vibrations from the surrounding environment can affect a pressure gauge and cause potential errors. The vibrations can be natural, but they’re more likely to come from machines or devices that are operating nearby. Vibrations often aren’t consistent, so if you fail to account for them in an uncertainty analysis, your pressure measurement data can be unpredictable. To lessen the uncertainty from vibrations, ensure that pressure gauges are properly and securely mounted to whatever surface is appropriate for the use case.
Orientation of Measurement Device
Some pressure sensors operate better in certain orientations or mounting positions and can be sensitive to how they’re set up prior to measurement. (In the case of deadweight testers, it’s not just orientation that matters: Users need to account for the difference in fluid column height of the reference and the device under test (DUT). Not accounting for this difference can significantly affect results, especially at low pressures.) Ensuring you orient the gauge according to the manufacturer’s recommendations helps minimize any orientation sensitivity.
Uncertainty of Reference Device for Calibration
Uncertainty affects more than just the pressure gauge you use to take measurements for day-to-day work. The reference standards that metrologists use to calibrate your pressure gauge are also subject to measurement uncertainty. If an ISO/IEC 17025-accredited laboratory calibrated your pressure gauge, you should be able to find the reference device’s uncertainty listed on your pressure gauge’s calibration certificate. Because the reference standard is subject to uncertainty, it can impact the calibration — and thus the measurement performance and uncertainty — of your pressure gauge.
Frequency of Calibration
Over time, measurement devices will naturally have slight changes in the accuracy and reliability of their results. This measurement drift is why it’s critical to calibrate your measurement devices regularly. (In the past, organizations and calibration labs relied on the OIML D10 from the International Organization of Legal Metrology (OIML) for guidance on calibration intervals, but that guidance has been withdrawn. Now, organizations can look to the ILAC G24 from the International Laboratory Accreditation Cooperation (ILAC) or the EURAMET cg-19 from European Association of National Metrology Institutes (EURAMET) for guidance on calibration intervals.)
Most uncertainty analyses will assume a device is recalibrated once a year, but the real frequency of calibration for the gauge itself will help determine uncertainty. Longer periods without calibration will naturally lead to greater ranges of potential uncertainty, while more frequent recalibrations should tamp down on uncertainty.
Dwell
All pressure measuring systems require time to stabilize and reach equilibrium. They must hold, or dwell, at a specific point before you can take an accurate reading. That stabilization time can sometimes be 30 seconds or longer, and measurements will fluctuate before the device reaches its dwell time. If you try to take a measurement before your pressure gauge stabilizes, you risk introducing uncertainty into your measurement. Any fluctuations during that time period can lead to a level of uncertainty in the measurement, so it’s important to know and take into consideration your pressure gauge’s dwell time when you calculate uncertainty.
Zeroing
Just as a measurement device’s ability to accurately take measurements can drift over time, so can its zero point. This drift of the zero point can stem from mechanical hysteresis, residual stress, lack of regular calibration, surrounding ambient factors, or simply from the equipment aging. Regardless of the cause, zero drift causes slight uncertainty in exact measurements. For digital pressure gauges, zeroing is relatively easy and can be done by pushing a button or finding the zeroing function in the right menu. For analog gauges, zeroing can be trickier and will require physically venting the pressure. (The venting method depends on your pressure gauge and how its built.)
Now that you’re familiar with some of the specific factors that can contribute to pressure measurement uncertainty, let’s look at the broader classifications of Type A and Type B uncertainty.
Determine Sources of Type A Uncertainty
Type A uncertainties are uncertainties that you can attribute to quantitative measurements and statistical analysis — both of which must come from a series of repeatable observations. For instance, any sort of uncertainty that you can calculate averages, standard deviations, or degrees of freedom for would fall under Type A. Because of these strict classification parameters, there are usually fewer Type A uncertainties in an uncertainty analysis.
Determine Sources of Type B Uncertainty
Type B uncertainty, on the other hand, is any uncertainty that you can’t classify as a Type A uncertainty. Sometimes, these are qualitative uncertainties such as using your judgement as an expert in your field to make certain assumptions about the performance or characteristics of a pressure gauge. However, there are quantitative Type B uncertainties, like data points that come from the gauge’s manufacturer specifications or calibration certificates.
Many of the considerations we’ve already covered are Type B uncertainties. Drift, calibration frequency, reference uncertainty, and the model of measurement device are all Type B uncertainties.
Calculate Your Uncertainties
Once you’ve determined the Type A and Type B uncertainties for your pressure gauge, you’re in a good place to actually calculate the overall uncertainty of your pressure measurement.
Note that there’s no one-size-fits all when it comes to uncertainty. All the considerations outlined above will affect different pressure gauges in different ways. And every pressure gauge will have slightly different uncertainties you’ll need to check for, especially Type B uncertainties. For any given device, you can find much of the information you need in manufacturer’s manuals and calibration reports.
Of course, the process of calculating uncertainty can be complex and time-consuming — so much so that there are entire courses and webinars dedicated to calculating measurement uncertainty. That said, generally speaking, the steps to calculate uncertainty are as follows:
- Identify and eliminate external sources of uncertainty
- Estimate and confirm standard uncertainties (Type A and Type B)
- Determine the probability distribution of standard uncertainties
- Convert standard uncertainties to standard deviations
- Combine the Uncertainties
- Calculate Expanded Uncertainty
If you’d like to learn more, take a look at our guide on how to calculate measurement uncertainty.
Reduce Uncertainty in Your Pressure Measurement With the Right Tools
Due to the precise nature of pressure measurement, there will always be some level of uncertainty. However, you can take steps to account for and reduce it. Investing in quality pressure gauges and pressure standards with a high level of precision and reliability, that are traceable to national and international standards, and have a clearly documented uncertainty — like those from Fluke Calibration — is a great place to start. Always remember that even the best tools are never truly infallible and should be routinely calibrated to maintain their accuracy and reliability. Uncertainty may affect every measurement we take, but as long as we know that and prepare for it, measurements will always remain precise and reliable.