Dual Absolute®: the new frontier of leak testing
In the field of leak measurement, two types of pressure drop equipment are currently known and widely used: those with absolute, or relative, drop, and those with differential drop with reference part.
Although the quality of measurement for both types of equipment is described and catalogued in standards, the application field of one or the other method is not always clear and well-defined, so much so that in the majority of practical cases the two systems overlap with each other. This is also because over time, technologies and components have evolved; the basic schematics have been enriched, becoming increasingly more performant, both in management software and in terms of variants, options and pneumatic modules not foreseen in the original circuitries. Sample volumes, electronic regulators, capacitance meters, vacuum generators, manifolds, isobaric or coaxial housings, have been added to the equipment in order to improve their effectiveness and reliability.
One of the circuit variants that is now beginning to spread is dual technology.
This new system, also called Dual Absolute Leak Tester, is not positioned in the list of optional features nor in a middle ground between the two previous types of measurement, but rather opens new horizons, improving the quality of measurement and simplifying the two previous existing basic types.
While historically differential systems were developed with the dual purpose of increasing the resolution of pressure drop and compensating for the thermal behavior of the part under measurement, it is also true that over time technology has improved exponentially in terms of strain gauge acquisition and electronic transducers, simultaneously increasing the quality of absolute decay systems.
A technological parallel can be drawn in the weight measurement sector, comparing spring or dynamometric scales with beam-type scales using standard weights. Although comparison scales may seem infallibly precise, over time these systems have given way to strain gauge measurement, which, thanks to electronics, has for several decades now surpassed all previous mechanical systems in both precision and ease of use, reducing costs, maintenance, mechanical parts while increasing efficiency and reliability.
The resolution of the Δp/Δt
Similarly to a gravimetric dosing measurement system, in a pressure drop leak measurement system the measurement resolution factor plays a key role in application quality, perhaps even more so than the overall long-term accuracy characteristic. In numerical terms, referring to the technical data available on the web declared by various leak tester manufacturers, it emerges that the standard widely adopted today is to guarantee a resolution of 0.1 Pa up to 16 bar for differential systems, a figure that reduces to 3 bar for absolute drop meters.
This means that by installing a transducer, for example with a +/- 50 millibar scale, in a differential system and comparing it with the direct pressure of an absolute decay meter having a maximum scale of 3000 millibar, the two measurements in terms of Δp/Δt resolution function in the same way. If instead we installed a 6 bar transducer in the absolute system, the resolution would worsen by half, that is, doubling to approximately 0.2 Pa per division.
Given the limited benefit in terms of acquisition, it is clear that differential systems, in light of their greater mechanical complexity, contribute to providing benefits mainly in measurement compensation in terms of temperature and mechanical stress of the part, rather than in terms of pressure drop precision.
Considerazioni tecniche
More specifically, it is necessary to dwell on this aspect related to resolution for some purely technical considerations.
The first is that, in the practice of the thousands of applications solved by ForTest, approximately 90% are within the pressure of 6 bar, while 60% are within 3 bar of testing. Therefore, although the difference in benefits is already minimal, in more than one case out of two there is no advantage, not even theoretical, in terms of resolution when using a differential gauge.
Furthermore, in applications above 6 bar it is inadvisable to set a leak threshold Δp/Δt close to the resolution limit of the measuring equipment. Indeed, it is evident that the higher the test pressure, the proportionally greater the drop to be measured becomes. In reality, at pressures greater than 8 bar it is strongly discouraged to set Δp values lower than 100 Pa; in such operating conditions, having a resolution of 0.1 or 0.2 Pa makes no difference, except from an advertising standpoint.
The last consideration is that these resolution data are typically measured at zero, i.e., compared with primary standards at ambient pressure. In practice, little is known, except through laboratory trials conducted with leaks and sample volumes at actual test pressure, about the effective behavior in terms of resolution hysteresis. That is to say, any diaphragm, whether absolute or differential, subjected to an offset pressure (or common mode pressure in the case of differentials) is inevitably subjected to mechanical “noise” and stress that are not considered under certification conditions.
Obviously, the quality of the differential transducer, being the “heart” of these systems, largely defines the metrological quality of the devices on the market as well as the reliability in terms of robustness to pressure spikes and compatibility with moisture or contaminants in the test air or present in the parts under test.
Temperature compensation
If over the years the technology of signal transduction and digitization of force and pressure signals has significantly brought together the characteristics of various measurement systems, at least in terms of resolution, it is equally true that the “physical” issues of compensating for thermal and mechanical variations have remained the same. It is in this scenario that differential systems still play a leading role.
Analyzing symmetrical differential pressure drop measurements, i.e. compared to a reference sample, we still have two opposite cases where absolute leak testers are at a disadvantage. These are cases of testing on very small volume parts with very high productivity (tire valves, fittings, biomedical components, etc.), where cycle time is paramount and measurement speed is a predominant parameter, and cases of large test volumes, where elastic and temperature drifts affect too significantly not to be compensated.
In reality, in both cases, alternative measurement systems have highlighted more suitable solutions compared to differential meters. For example, compliance recovery systems or bell jar interception for small volumes and mass-flow for large-sized parts.
The current research is being conducted within the scope of improving the application of differential systems. In this vision, balancing through sample parts is what we are focusing on. Also because the concepts presented of absolute, differential and dual measurement are actually transversal to the various typologies and applicable in different ways to the physical principles of the transducers used.
Measurements on small-volume parts
Regarding measurements on small volume parts, with a view to containing test times (for example 1.2s total start-to-end cycle, 1cc test, leakage = 10 cc/h @ 2bar) and while absolute measurement enjoys a very high dynamic range and does not require the long stabilization times conversely required by differentials, it should be noted that the “mechanical” balancing of the differential transducer is actually even more immediate and faster than dual systems.
This means that with such restricted times (generally 100/200 ms) for acquiring the Δp/Δt drop, even small phase shifts in the signal from the two membranes, or their resonances, result in significant overall measurement errors. Such errors are actually non-existent when the passband is kept below 100 Hz on the signal slopes, that is, with Δp/Δt times exceeding half a second. In any case, in these “ultra fast” cases for micro-volume applications, traditional differential pneumatic circuits are certainly preferred, albeit redesigned with micro-valves, transducers and tubes with dimensions reduced as much as possible.
Considering that in these particular micro-volume conditions the pressure drop that occurs in case of leakage is always of great magnitude, the application of transducers with low dead volume such as MEMS or solid-state bridges, instead of capacitive transducers, simplifies breakdown and reliability issues while ensuring an extremely high dynamic range of the elevated measurement scale, albeit with more limited resolution.
This is the opposite case to measuring large volume parts, where it is necessary to have measurements that are as stable as possible and immune to any type of noise and drift, even at the expense of bandwidth. Here, resolution and stability in measurements up to 60/120 seconds are the peculiar characteristics. In all cases of “direct” measurement, it is good to remember that the relationship with the leak is always inversely proportional to the pressure drop Δp/Δt. It is in these cases that it is advisable to have designed in all possible conversion bits that AD components offer, as well as greater filters and EMC immunity.
Measurements on large-volume parts
Compared to the conditions of small-sized parts, the physical and pneumatic scenario of measuring parts with large volumes is quite different, namely in cases where high sensitivity is required, starting from sizes above 250 cc. It is in this field that all manufacturers of measuring equipment, including ForTest, have researched systems to support the use of reference sample parts. A large part of the technology based on software algorithms involves the characterization of tests considered “good,” that is, within an extremely safe band, so as to recreate in an “anti-transient” way a dynamic offset and be able to continuously adjust a Dynamic Offset Compensation on the measurement (DOC). These are all systems already widely used as auto-zero algorithms in the most common weighing systems, which actually only partially adapt to the broader issues of complex leak testing processes.
The drawback of these systems and alternative solutions
The main disadvantage of these offset correction systems is related to their inability to separate and correct the various errors one by one. However effective, these automatic compensations can only provide assistance when used in small percentages of the set point, as they serve the sole purpose of tracking slow/very slow error variations. Generally, instead, the overall measurement is corrupted by various spurious phenomena due to the overlap of multiple factors such as mechanical movements, material stress, elasticity of connection fittings to parts, and only partially by ambient temperature variations.
Other widely used systems allow sampling through temperature probes to monitor the progression of environmental factors, creating offset compensation in terms of Pa/Degree Celsius (DOCT). In this mode, after a period of analysis of practical tests in production, i.e., acquisition in Excel format of measurements correlated with measured temperatures, a correction factor is introduced to the measurement in order to compensate for temperature oscillations. Although more laborious during the setup phase, these algorithms have the advantage of limiting themselves to compensating only the thermal phenomenon and therefore avoiding excessive accumulation of phenomena to be corrected.
In all cases, balancing through a sample piece or reference emulator greatly helps measurement stability and repeatability, if nothing else in terms of acquiring ambient thermal conditions.
Differential meters and repeatability
It should be kept in mind that differential leak meters are commonly used in three practical configurations, which can generally be summarized as:
- Asymmetric differential, with the reference side blocked by a plug. This is a simplification during installation that makes it equivalent to an absolute system.
- Zero-centered differential, designed to measure two pieces at a time.
- Symmetric differential, the true balanced comparator, where the reference side is connected to a hermetic reference part.
Let’s now analyze the benefits of using reference sample parts in the various modes.
Of these three usage layouts, the symmetrical one with sample piece proves to be the method that provides the best results in terms of accuracy, repeatability and especially rejection of noise generated by temperature and mechanical stress.
Applications at micro volumes
In micro-volume applications, where the predominant thermal mass and expanding elements are essentially the connection tubes to the part, using a reference circuit as similar as possible to the measurement side allows perfect system balancing and correction of not only temperature but also the expansion of both circuit sides (Test and Reference), since parts under test of such small size are generally rigid. In these cases, a simple identical sealed tube on the reference side with the same length as the Test connection tube is more than sufficient to achieve both excellent repeatability and drastic reduction of settling times. In the case of metallic parts, a blind fitting such as a plug at the end of the reference tube ensures a “temperature capture” function, further improving the application.
Applications at larger volumes
This reasoning is no longer valid in the second case of using a differential gauge, namely in the most frequent applications for testing parts with volumes that are already larger than the dead volumes of the connecting tubes. To complicate the scenario, which is already complex in itself, problems arise related to the mechanical stress of the parts and the endogenous generation of parasitic temperatures when tests are repeated on the same part.
In practical uses of sample parts, and contrary to the desired measurement compensation, it is observed that the volume variance due to the expansion of the two parts under test introduces measurement errors in turn. It should be considered that in a differential pressure drop system, commonly used for high-rate industrial production operations, the mechanical expansion of the part under test will be limited to the measurement operation only, while the mechanical stress on the reference sample part will accumulate throughout the entire usage time of the equipment for an indefinite number of times, effectively leading to a continuous drift in the behavior of the two parts after just 15/30 minutes of steady-state operation.
In these cases, the dilation of tubes or internal circuits within the equipment is no longer predominant, as in micro-volume applications, but the parts themselves create the repeatability error.
Similarly, due to the continuous pressurization and emptying of the reference sample piece alone, there is an increasing thermal accumulation that triggers endogenous phenomena which largely nullify the measurement compensation, creating unwanted drift. In practice, empirical findings have shown that a metallic part with a volume of 300cc subjected to a pressure of 2bar gauge requires at least 20 minutes to restore elastic and quiet temperature conditions, that is, to return within a repeatability margin of 10% compared to the first test performed.
For this reason, the concept of apparent repeatability in the use of differential pressure drop leak testers has been introduced over time, namely that phenomenon of good repeatability when performing repeated measurements on the same part, measurement stability that is however not maintained during practical use in production.
The birth of dual absolute systems
To overcome all these problems of drift and stress in reference parts, dual absolute systems were developed. In an initial version, or rather in the experimental phases, these systems were presented as simple expansion and modification kits for normal equipment, both absolute and differential. Through a three-way pneumatic valve, a sampling procedure was then introduced, namely automatic “self-learning” of DOC, with time frequencies fast enough to follow ambient temperature evolution, but allowing sufficient settling time for the reference side to return to the initial elasticity condition, i.e., the real elasticity condition to be compared with production parts under test. The same systems are sporadically used by various manufacturers to sample environmental factors (Tamb and Pamb) through sample nozzles and practically compensate volumetric flow test measurements.
In the case of pressure measurement, over time it has been found that, thanks to a fortunate combination of positive factors all working in the same direction of product improvement and economy, the creation of two symmetrical branches of absolute measurement that are independent of each other but governed by different software modes has led to an unparalleled improvement in all types of measurement. As can be intuited, in addition to improving symmetrical measurement, the possibility has indeed been discovered, thanks to different test management modes, to significantly improve both zero-centered measurement and asymmetric type measurement.
Absolute decay meters
Always considered the “poorest” system, thanks to the acquisition and transduction improvements already outlined, absolute decay meters have gained increasing popularity, now commonly standing alongside both differential and mass flow systems. This success is largely due not only to the actual quality of measurement, but also to enormous simplicity, robustness and reliability in maintenance and use compared to any other leak tester present in the industrial field. Now far removed from the basic concept of PLC, valve and pressure transducer, through methodical development over time of hardware and firmware, it has been possible to obtain precise and versatile machines, with a more immediate approach to the leak testing procedure.
It is indeed necessary to always remember the application environment of this equipment (which is generally not an ideal laboratory with sterile conditions) where even simple things very often become complicated with enormous ease.
Although apparently less sensitive on small scales compared to other systems, the high dynamics during the settling and absolute decay measurement phases and the absence of limits at high pressures have established their application in fields not recommended for differential gauges and mass-flow systems. For example, in the biomedical field where, in addition to the reliability of pneumatics and the need for sterility and non-contamination of the parts under test, the high oscillations of elastic materials used such as bags or transfusion sets have defined these systems as standard to the disadvantage of others.
Obviously, being able to rely on a complete range of technological solutions and different measurement methods, spanning from tracer gases to micro flows, from recovery systems to pressure drops, the approach to the application always involves the most suitable solution, primarily in terms of purpose and scope of use, then sensitivity, and finally the required cycle time.
Advantages of absolute-type measuring equipment
It remains a fact that the application of an absolute decay system, where possible, always has the appeal of “install and forget” while every other dual sensor method requires some additional attention in the metrological field, due to the dual measurement. Periodically, more careful verification and drift control are indeed required, as well as, as in every case, dual certification. For example, in cases of mass-flow flowmeters (which have nonetheless reduced and simplified the intervention scenarios related to capillary systems) it is always necessary to control the quality of the air used and the cleanliness condition or degradation of the measurement sensors. In particular, in differential decay systems, wear and dirt in the equalization valves is inevitable due to the discharge necessary to preserve the life of the measurement transducer, while the pneumatics are much more sensitive and sophisticated than any other system in comparison.
Although both pneumatic and mechanical engineering and the periodic verification and calibration procedures of all systems have evolved drastically over time, it is evident even at a simple glance that all these technologies are more complicated when compared to absolute type meters.
In this type of flow meters, the only transducer used is of excellent quality and covers the entire measurement range. It is therefore very robust, does not necessarily require discharge at the end of the test and can withstand water hammer caused by non-synchronized discharges from outside the equipment, is not particularly affected by dirt and is insensitive to the dielectric capacity of the gas used and, within certain limits, to its humidity.
Furthermore, the simple pneumatics involves the use of mostly commercial components, oil and silicon free, if necessary supplied with certifications for food, packaging and pharmaceutical applications. The pneumatics is therefore easy to maintain and, if properly designed, intrinsically safe, i.e. always leaking in case of malfunction. All characteristics that are difficult to obtain in pneumatics for differential systems, whether with symmetrical scheme, master less axis, or isobaric cavities. For this reason, this second type of device requires more frequent maintenance and more accurate periodic checks.
In our T8960 differential models, for example, we have studied the opportunity to use commercial valves in order to take advantage of the interchangeability and versatility benefits of absolute decay models, delegating the equalization and transducer protection functions no longer to mechanical components, but to software procedures and high-speed PWM signals.
In practice, however, it is difficult to define which system is more practical to use. For example, how do we understand whether diesel or gasoline is better? Could the future lie in hybrid technology?
Dual technology
As already mentioned, the new dual systems do not arise from the assumption of positioning themselves as a middle ground between the currently known measuring equipment, but rather to complement and improve them where possible. Drawing from the characteristics of both currently known types, they fundamentally aim to merge their functionalities, simplifying and enriching the measurement cycles. On one side, the reliability and safety of absolute systems, on the other, the “leak amplification effect” of differential decay systems.
The main distinguishing features
Although it is still early to define standards, as this is largely still within the research and development scope of software in various modes, it is nevertheless possible to already draft a brief description of dual absolute systems.
The most evident distinguishing element lies in the comparison to a symmetric differential use with a sample piece. In this case, the strategy is to sample the reference piece during the Test measurement phase, just as in a differential, but only at time intervals that allow for correct comparison to the piece under test, while not distorting the elastic and thermal characteristics of the reference piece. In turn, these samplings are stored and compared in vector mode to the ongoing tests, effectively creating a virtual comparison until a new sampling occurs.
The evidence of improvement is even stronger when used in zero-centered symmetric differential mode, where current differential systems are now completely abandoned, being considered unreliable due to measurement uncertainty in case of leakage from both sides. In this mode, the full power of the dual system is expressed, being able to exploit the benefits of symmetric compensation while making the system safe. In practice, the measurement cycle in this mode involves extending the test time only in case of deviation of absolute values detecting a low differential factor. In other words, this way it is possible to benefit from both the high immunity to environmental noise from mechanical stress and temperature drift provided by true symmetric balancing, and the reliable simplicity of absolute decay.
In asymmetric differential mode, the software instead focuses on the ability to discharge air only when necessary. Due to the lack of need to protect the transducer, it is no longer necessary to generate a discharge phase at the end of the test, as required for differential meters. This allows maintaining pressure as much as possible on both measurement sides, settling them and avoiding complicated isobaric mechanics, coaxial tubes and other anti-expansion devices designed to reduce internal elasticity phenomena in the equipment. In practice, where possible, the discharge phase occurs at the beginning of the test, no longer at the end, and control is achieved by the software intercepting when the operator or the test bench is about to empty the part under test.
In conclusione
These are, in summary, the most evident peculiarities of the new technology described here. Beyond these aspects, measurement certification is always and only related to a relative measurement, and in practice, all the simplicity and reliability of an absolute decay system is respected. In practice, while losing some decimal places of Pascal resolution and with operating pressures above 6 bar test, an incredible simplification of the most well-known differential systems is achieved. With this technology, nothing necessarily “revolves” around the differential transducer anymore; instead, the hardware is reduced to minimal terms while the software is continuously evolving.
We therefore encourage both industry technicians and equipment manufacturers to contact ForTest for evidence and further details and not to hesitate in testing this new promising technology.