How accurate are multi-gas calibrations?
When you switch from flowing one gas to flowing a different gas in the same flow meter, are the new measurements with the new gas just as accurate as the original calibration gas? As the originator of multi-gas flow meter calibrations some 20 years ago, Alicat is asked this question often. In this post, we’re going to look at Alicat’s approach to gas selection and why you experience no change in accuracy when you change gases on an Alicat mass flow instrument.
A review of non-ideal gas laws
Alicat’s laminar, differential pressure-based mass flow meters determine flow rates by measuring pressure drops across the laminar flow element (LFE). Higher flow rates generate proportionally higher pressure drops. Sounds easy, right? It gets more complicated when we look at real-world gas behavior, which factors in a fluid’s viscosity (its resistance to flow). According to Poiseuille, the pressure drop of a flowing fluid is directly proportional to its absolute viscosity at a given flow rate under laminar conditions. For example, honey is more viscous than water, so it has a much greater resistance to flow than water does and generates a higher pressure drop at the same flow rate. (But don’t put honey through your Alicat flow meter.)
What affects fluid viscosity? Every fluid, whether gas or liquid, has an inherent absolute viscosity that speaks to its interactions at the molecular level. Viscosity is unrelated to molecular weight. Both hydrogen and helium have far lower molecular weights compared to air, but whereas hydrogen has about half the viscosity of air (it flows more easily), helium has a slightly higher viscosity than air does. Viscosity also increases with temperature and, much less so, pressure. For gases, absolute viscosity varies with the square root of absolute temperature. Pressure, on the other hand, has minimal effects upon viscosity, increasing the viscosity of air at 25°C only 0.8% from 1 to 10 bar.
Another complication is gas compressibility (z), which is a measure of how greatly the behavior of a gas departs from ideal gas laws (PV=nRT). Given that a z value of 1 indicates ideal gas behavior, typically, gases with more complex molecules exhibit lower compressibility factors (z in the non-ideal gas equation PV=znRT), which also indicates greater densities of these gases. Compressibility affects flow readings more when line pressures increase, or when temperatures decrease. Keeping tabs on all these variables calls for some advanced flow measurement technology.
Alicat’s approach to changing gases ensures that your flow control or flow measurement process does not suffer any loss of accuracy as your gas medium changes.
Alicat’s Gas Select: No K-factors allowed
If you’ve ever used a thermal mass flow controller, you’re probably familiar with the concept of K-factors. Let’s say your flow controller was calibrated on nitrogen, but you want to use it to flow argon. You find argon in the manufacturer’s table of K-factors and multiply this conversion factor against your nitrogen-calibrated flow reading to get the equivalent flow reading for argon. Depending on your accuracy requirements and operating conditions, this could be a suitable approach.
If you are concerned about maintaining a high degree of accuracy, however, you may be disappointed. The problem is that the thermal characteristics of gases used to run a thermal flow controller are also highly dependent upon pressure and temperature, just like viscosity and compressibility. If your operating pressure or temperature differs significantly from those defined in the K-factor chart, your K-factor will not provide an accurate flow correction. For this reason, many manufacturers of thermal mass flow meters provide much less stringent accuracy specifications when using K-factors. Recognizing that a single-point K-factor cannot account for these changing temperature and pressure conditions, Alicat uses a very different approach.
When you enter the Gas Select menu of an Alicat mass flow controller and switch from air to argon, the Alicat replaces its calibration data for air with data for argon, using a mathematical function derived from 3-dimensional gas properties data for the non-ideal gas properties of argon. This data surface charts NIST-traceable gas properties data from Refprop 10 for compressibility and viscosity against pressure and temperature across the entire usable range of the Alicat mass flow controller. The data is then stored as a unique function on-board the Alicat. What this means is that the Alicat accesses different data points when it is flowing argon at 2 bar and 23°C than when it is flowing argon at 6 bar and 37°C. The mass flow controller’s own sensors are reporting changes in pressure and temperature a thousand times every second, so the Alicat is always referencing the correct point on the calibration surface. Because of this 2-pronged approach—unique functions derived from 3-dimensional gas properties data combined with real-time sensing of pressure and temperature—you experience no change in accuracy when you change gases on an Alicat mass flow instrument.
Recent testing by an independent calibration house revealed a total error of just 0.15% of the reading at full scale when an air-calibrated Alicat was set (via the Gas Select menu) to measure a NIST-traceable, calibrated 3-gas mixture of 12% CO2 + 7% O2 + 81% N2. This measured error was well within the published accuracy specifications for the mass flow controller.
Read the fine print
These days, more and more thermal mass flow controller manufacturers are coming out with multi-gas calibration for some of their mass flow controllers. Most of these manufacturers increase their accuracy bands in some way when you use a gas other than the gas on which the instrument was physically calibrated. This may take the form of converting percent of reading accuracy specifications to percent of full scale specifications, sometimes while also widening the accuracy band itself (e.g., from 1% to 3%). For a discussion of how full-scale specifications differ from percent of reading specifications, please see Alicat’s previous post on Flow Meter Accuracy: Demystifying Spec Sheets. No matter how the change is applied, the presence of any specification change when you select a different gas indicates that not all relevant factors are being taken into account in the manufacturer’s gas model when you select a different gas, and this can be reason for concern.
Alicat’s approach to changing gases ensures that your flow control or flow measurement process does not suffer any loss of accuracy as your gas medium changes. Our lightning-fast speed ensures that your flow readings—whatever the temperature or pressure—are the most accurate possible, a thousand times every second.