GB2580634A - Automatic bump test device - Google Patents

Abstract

An automatic gas sensor functional bump test device transfers a calibrated volume and pressure of gas to a gas sensor 15, holding a portion of the gas around sensing surfaces, for a predetermined period. It measures the sensor response to the gas and response to recovery following test gas exposure. From the response to the test gas and the release of the gas, it is computed whether the sensor 15 is operating normally and has a gas path open to the environment. A single actuator provides three valve functions, firstly filling a reference volume 7 via an inlet valve 5 to a predetermined pressure with a test gas or span gas, the second valve 9 venting gas onto the sensors 15 and the third valve function closing and opening the sensor gas path to the environment, such that the first and second valve functions operate in series and not concurrently. The third valve function is synchronised to the first two to effect a discharge of gas over the sensors which is substantially retained over the sensors 15 while the second valve function is open.

Description

TITLE OF INVENTION

Automatic Bump Test Device

Technical Field

The present invention relates to the automatic "bump" testing or calibration Sof gas sensors in situ, in equipment such as gas monitors, gas controllers, breathing equipment and rebreathers. In particular, the present invention relates to a method for exposing the sensors to a known change in the gas periodically and validating that the sensors are functioning correctly in response to gases in their sensing environment.

Background of the Invention

Gas sensing is critical to the provision of a safe breathable gas in many environments. For example, confined spaces in industrial plant or on large vessels may expose workers to poisonous gases such as.0O2, CO, NO2, HCN, SO2 or H2S, or to explosive gases such as H2, VOCs and LELs, or may 15presenting a risk of hypoxia from depleted oxygen levels. Personal, transportable and fixed gas monitors are used to control those hazards and ensure a safe working environment. In other applications, such as in re-breathing equipment, the gas sensors are the eyes and ears of an active feedback loop that maintains the oxygen and carbon dioxide levels within a safe range.

Gas sensory systems have fault modes that can prevent a hazard being signalled when the hazard is present. These fault modes may include faults in the gas sensor itself, faults of sensitivity loss through ageing or contamination, faults in the temperature compensation of the sensor, or obstruction of the membrane separating the sensor from the environment by vapour, water, dust or paint, or 25faults in the control system.

To manage the risks posed by those failure modes, it is common practice to perform a functional test of the sensors periodically, e.g. daily. The test frequency is stipulated by some international standards and in by workplace regulations, therefore traceability that it has been met is desirable.

The periodic test verifies that the gas sensors do respond correctly to changes in the ambient gas environment and will trigger an alarm when required.

The most basic functional test is known as a "bump test". It involves the sensor to a different gas concentration than in the ambient environment for a time and at a concentration to activate all of the alarms to at least the lower alarm lOsettings. The concentration of the gas of interest, is chosen typically to be 5% or 10% more than the alarm level for that gas, or it may test the 25% FSD, 50% FSD or the Full Scale of the sensor (i.e. 100% FSD). A bump test confirms that the gas is capable of reaching the sensors, that when they are exposed to gas the sensors respond, the response time (time to alarm) after gas is applied is within normal 15Iimits, and that the alarms are activated and function properly. However, this basic test does not necessarily verify the accuracy of the readings or output of the sensors when exposed to gas.

A "calibration check" is a more rigorous quantitative test using a traceable source of known concentration test gas to verify that the response of the sensors 20is within the manufacturer's acceptable limits. For instance, a manufacturer might specify that readings in a properly calibrated instrument should be within ±10% of the value of the gas applied. If this is the pass / fail criterion, when 20 ppm H2S is applied to the instrument, the readings must stabilise between 18 ppm and 22 ppm in order to pass the test. It should be stressed that these pass / fail criteria are 25manufacturer guidelines. Different manufacturers are publish different requirements. It can be seen that a calibration check is a more rigorous test than a basic bump test.

Both types of test, bump test and calibration tests, may be performed manually or using bump test or calibration stations, such as that described in 30US6442639B1. Using a docking station involves taking the gas monitor to a test station, and running the test. For larger systems or for fixed monitors, the test station is taken to the monitor using a cap over the sensor, such as described in US20060081033A1.

The test gas, or "span gas" used to in bump testing or in sensor calibration may contain a mixture of gases to enable systems with a combination of gas 5sensors to be tested in a single step, or in a sequence of tests each with a dedictated single span gases.

Some sensors respond much more quickly than others. For example an oxygen sensor may have a 90% response time under 7 seconds, but some CO2 sensors or LEL sensors may take 2 minutes. The duration of the test is 10determined by the slowest sensor.

In gas control applications, gas sensor testing may not be able to create the alarm conditions for a true bump test. For example, in an electronically controlled rebreather used by divers, the maximum oxygen level on the surface is limited to 1 atm, but in use the equipment may be expected to indicate alarms at a Partial 15Pressure of Oxygen (PPO2) of 1.6 atm. Partial pressures of gas above 1 atm can occur when the equipment is used under water with more than 1.6 atm ambient pressure, but would require a pressure chamber to create on the surface. To compensate for this limitation, rebreather gas sensors are typically tested in a pressure vessel every month or so, and before each use are tested to the lower 20partial pressures that can be created at surface ambient pressure conditions. This procedure does not guarantee that the sensor is not failing, so the general practice is to fit multiple oxygen sensors, preferably from different batches or of different types, and check during the dive that the reading from each match within a narrow band.

Tests that involve correlating the output of multiple sensors can be difficult to perform under some environmental conditions because gas sensors often have long thermal time constants, typically 30 minutes, and many types of sensor are highly temperature sensitive. This means that either the environment should be stable or the temperature compensation of the sensor must perform very well if 30sensors are to track each other within any predefined band. This temperature tracking causes problems when one sensor is checked against another sensor.

For this reason it is better to test the sensor by exposing it to a known span gas, rather than by comparing it with the output of a reference sensor.

When an oxygen sensor fails, it typically becomes unable to measure partial pressures of oxygen (PPO2) above a drooping ceiling concentration, with 5the result that if the gas controller tries to establish a PPO2 above that ceiling then the environment becomes filled with oxygen. For example, if the oxygen sensor has a PPO2 ceiling fault of 1.2 bar, it cannot produce an output corresponding to more than 1.2 bar regardless of the PPO2 of the gas it is reading. If the controller tries to establish a PPO2 in a breathing loop of 1.3 bar, with a sensor having such 10a fault, then the controller will generally keep injecting oxygen until the oxygen is exhausted; at which point the PPO2 may be far higher than the intended 1.3 bar level and pose a serious risk to anyone breathing from that system. To remove that risk, it is preferable to test the sensors to the highest alarm levels, in service or immediately prior to service.

During testing, alarms are usually localised by showing the indication of the alarm on the device being tested but any telemetry of that alarm is suppressed.

The detailed execution of the bump test varies considerably in detail between different sensors and different applications. To tell the equipment that it is being bump tested, either a switch or an RF protocol is used. RF protocols can 20include Near Field RF or BlueTool. Where switches are used they may be a Hall sensor, reed switch, piezo button, or a contact pad, that detects the bump test equipment, or the switch activation may be fully manual, such as a reed switch activated by a wand used by the technician performing the test, or a keypad or touchscreen. Once the device is in bump test mode, a display may take the 25technican through the steps involved in the bump test, or the technician may use a Work Key Point sheet, checklist, or the test may be fully automatic interacting with bump test equipment. In any event, the sensor is usually exposed to either a zero or ambient gas level and then to a gas level high enough to trip the alarm.

Many docking station systems verify not only the final stable reading of the sensor, but the time it takes to reach the desired output level, as well as the shape of the sensor response curve, which can provide important diagnostic information on the health of the sensor.

Sensor bump test stations are usually large relative to the sensor, and involve gas cylinders, typically of 1 litre to 3 litres in size. The cylinders contain toxic gas at high pressure and are hazardous themselves. For example, if a cylinder leaks, the operator may breathe in a high concentration of the poisonous gas, or the gas may inject to the operator directly through his skin, or the leak may 10force liquids through the skin.

The reason for the span gas cylinder being so large is that the central station does a large number of bump tests and in each test the gas must pass through valves, tubing and the coupling space into the sensor. When the test is complete some of that gas is released in the vicinity of the technician performing 15the test. In particular, the prior art applies a flow of gas over the sensors under test, resulting in a much larger volume of gas being consumed per test, than is the case with the present invention.

There is a significant amount of other prior-art, which suffers from one or problems preventing them being integrated within the sensory instrument except 20under highly specialised conditions. One such condition is in rebreathers, where gas is injected onto a reference sensor periodically.

Object of the present invention.

It is a primary objective of the present invention to eliminate the need for frequent manual operations in the bump testing of gas sensors.

It is a further objective to minimise the amount of gas used for each bump or calibration test.

It is a further objective of the present invention to ensure that sensors are tested on a prescribed schedule.

It is a further objective of the present invention to enable sensors to be tested in situ.

It is a further objective of the present invention to minimise the amount of gas used in each test, to enable a small cylinder to provide span gas for at least 5the service interval of the equipment.

It is a further objective of the present invention to predict sensor failure.

If is a further objective of the present invention to provide maximum availability of the system.

BRIEF SUMMARY OF THE INVENTION

The present invention relates to devices, techniques and methods to minimise the exposed gas volume of the sensor and apply a predetermined dose of calibration gas to the remaining volume, measure the response of the sensor to that exposure, and thereby either validate the sensor or inidicate a fault. The invention then releases the test gas around the sensor and checks that the sensor 15output returns to the output expected from the environment within a predetermined time period.

The key aspects of the present invention is the dose (in Mols, controlled by applying a predetermined volume of gas at a predefined pressure), which is released onto a confined space around the sensor's sensing surfaces, held there 20for a predertermined time, and then released. The extra pressure created by the dose is relieved by features in a gas flow restrictor. The response of the sensors to the predetermined dose, and the recovery of the ambient reading when the dose is released, are used to validate the operation of the sensor and that the gas path to the environment to be measured, is not blocked.

BRIEF DESCRIPTION OF THE DRAWINGS

For a better understanding of the present invention and the advantages thereof and to show how the same may be carried into effect, reference will now be made, by way of example, without loss of generality to the accompanying drawings in which: Figure 1 shows an a block diagram of the present invention where solenoid valves are used to charge and release a fixed volume of gas into a space around or over the gas sensor.

Figure 2 show an embodiment of the present invention in which the valves are 5connected to a single actuator such that the dosing, release and sensor closure and opening functions all occur synchronously in a single mechanism.

Figure 3 shows a cross section through an embodiment of the type shown in Figure 2 and Figure 4 Figure 4 shows an example embodiment of a mechanism operating synchronously lOwith the valve in Figures 2 and 3, to close and open the sensors to the environment for the purpose of the test.

DETAILED DESCRIPTION OF THE INVENTION

The invention will now be described without loss of generality by reference to the aforementioned figures and by use of example embodiments.

l5Figure 1 shows an a block diagram of an example embodiment of the present invention where a cylinder containing a span gas under pressure (1) is connected via a pressure reducer (3) to a normally open inlet valve (5) passes gas to a fixed reference volume (7), and then later the inlet valve (5) is closed and a normally closed outlet valve (9) is opened, which releases the gas in the gas reference 20volume (7) into a space (11) around or over the gas sensor (15) displacing the gas there previously. A gas flow restrictor (13) such as a gas permeable membrane that slows down the gas flow or a moving mechanism that closes off the sensor gas path to the its normal sensing environment (17), or combination thereof. The gas flow restrictor (13) allows excess pressure around the gas sensor (15) to be 25flushed. The output of the gas sensor (15) is read by a processor such as microcontroller or an FPGA (19), which manages the test by driving the inlet valve (5) and outlet valve (9).

If a membrane is used as the gas flow restrictor (13), the permeability is chosen to let excess pressure through easily, whilst allowing gas at a pressure near ambient to diffuse sufficiently slowly that the gas sensor respond before so much gas is lost that the sensor fails to change by an amount that is the bump test condition, e.g. 55% above the alarm threshold.

Gas permeable membranes have a deleterious effect on gas sensing in that they increase the time constant of the sensors gas response. Where the membrane that would retain the gas would have too deleterious an effect on the time constant, then a more permeable membrane can be used in conjunction with a lOpneumatic or electrically driven closure device. An example of such a closure device described with the aid of Figure 2.

Figure 2 describes of a novel means to of performing the functions of the gas restrictor (13), and the inlet valve (5) and outlet valve (9) such they operate synchronously with the a type of gas restrictor (13) that rotates to close off the 15sensors to the external ambient gas. The valve comprises four disks (21, 23, 25, 27) through which passes a shaft (30). The two outermost disks (21, 27) are fixed to a housing, while the inner disks (23, 25) rotate with the shaft. The disks are pressed apart in this embodiment by a spring (29), that enables 0-rings (31, 35) to maintain a seal when when there is no gas inside the volume. The space between 20the inner two disks act as the gas reference volume (7). Rotating the shaft (30) first connects the reference volume (7) to the gas inlet, then closes the volume, before releasing the gas to the outlet (37) via the top 0-ring (35). The shaft and the moving disks are sealed by use of further 0-rings, for example 0-ring (21).

Figure 3 is a cross section through an example of a mechanism implementing the 25type of embodiment illustrated in Figure 2.

Figure 4 The single test mechanism can supports the simultaneous functional test of multiple gas sensors (15), by use of a mixture of gases in the span gas, with an inert carrier gas such as nitrogen or helium. In this embodiment, six sensors are located in a sensor housing (41), over which a sector plate (42) is placed, on top of 30which is a gas director plate (43). The director plate (43) can be rotated by connection to the shaft (30) of the valve assembly shown in Figure 2, such the the cutout sectors align with those in the plate below, or are between the sectors in the plate below. The director plate is fitted to the shaft such that when the cutouts in the plate are aligned with the cutouts in the plate below, the reference volume (7) is disconnected from the sensors but connected to the gas inlet (33) from the cylinder (1) or pressure regulator (3). When the shaft (30) rotates to the position 5where the cutouts in the director plate (43) are between the cutouts in the sector plate (42) below, then the reference volume (7) is disconnected from the gas inlet (33) but connected to the gas outlet (33), causing gas to flow along the routing channels in the director plate (43), over the sensors (15), trapping the gas there. The director plate (43) does not have to be sealed well, as it is desirable that gas 10above the ambient pressure is released, allowing the measurement of the sensors to be made at the same pressure as that of the environment. If the director plate seals firmly, then it is necessary to determine the pressure of the gas over the sensors (15) and compensate the output of the sensors (15) for that pressure: this is an unnecessary extra complexity that can be avoided by allowing the director 15plate (43) to seal sufficiently to contain the gas over the sensors but not so well as to allow additional pressure. It is desirable that the gas directed onto the sensor mixes with the gas in the dead space around the sensor and the excess gas volume is vented. One means to increase the time the test gas dwells over the sensor is to bond a gas permeable membrane over the housing (41) or director 20plate (43).

The carrier gas in the gas cylinder (1) usually comprises the major part of the span gas; in the case of highly toxic gases, almost all of the span gas is the inert carrier gas. As the carrier gas is the majority of the gas mixture, the flow dynamics to multiple sensors can be uniform: that is, the device does not require different flow 25geometries for different gas types as a small percentage of a heavy gas in a carrier such as nitrogen will have almost identical flow dynamics as a small amount of light gas in the same carrier.

The service interval for the gas cylinder (1) may be affected by how the gases settle out within the span gas mixture. Several methods can be used to minimise 30gas settling, in particular the gas cylinder (1) should preferably be on its side within the device or equipment using the test device, and turbulence can be used to agitate the gas in the cylinder (1). Preferably the gas inlet valve (5) is chosen have a sudden action which tends to cause turbulence in the inlet gas supply: a solenoid valve or other quick acting valve. Where gases are particularly prone to separate and the gas is stored for long periods, the pressure reducer (3) can be eliminated thereby increasing the turbulent effect when the inlet valve (5) opens. 5Solenoid valves tend to have a narrow pressure range over which they operate reliably, but the valve type shown in Figure 2 can operate over a very wide temperature range.

The use of a pressure regulator on the gas source is desirable but not necessary. Without a pressure regulator (3), the pressure of gas, and hence amount of gas in 10the reference volume (7) will vary. The device can be calibrated for that variation, as it should be known based on the initial charge pressure of the cylinder and the number of tests that were carried out, or from a pressure sensor.

An example embodiment of a novel means to combined the valves (5), (9) and gas restrictor (13), is shown in cross-section diagrams forming Figures 2, 3 15and Figure 4. In each embodiment of the device, the automated bump testing of gas sensors is performed by closing the gas path of one or more sensors to the environment, then dosing a predetermined volume of test gas of a known constitution over one or more sensors for a predetermined period, measuring the level read by the sensor, then releasing the test gas while opening the sensor to 20the environment, before compute from the response of the sensor to the test gas and the release of the gas, whether the sensor is operating normally and whether the sensor has a gas path open to the environment.

The embodiment in figure 3 and figure 4 show device with a single actuator which acts to provide the three valve functions, i.e. the first valve fills a gas volume at a 25predetermined pressure, the second valve function vents the space onto the sensors under test and the third valve function closes and opens the sensor gas path to the environment, such that the first and second valve functions operate in series and not concurrently, and the third valve function is synchronised to the first and second functions to effect a discharge of gas over the sensors which is largely 30retained over the sensors while the second valve function is open.

In each case, the test gas is of a known constitution and it mixes with a known volume of gas that is in the dead space around the sensor. It is usually necessary to performs a computation that compensates for diluting effect of the dead space in determining whether the sensor is functioning correctly. For example, if the volume of the chamber(7) is 2cc, filled at 10 bar, then the volume of test gas dosed is 20cc at normal atmospheric pressure. If the dead space around the 5sensors is also 20cc in total, then the 50ppm of the test gas, e.g. Carbon Monoxide, would need to be used to provide 25ppm at the sensor.

Each device perform a computation to compensates for the diluting effect of the dead space in determining whether the sensor is functioning correctly including adjustment for the reading measured from the sensor prior to the test. Preferably, 10the computation that compensates for diluting effect of the dead space in determining whether the sensor is functioning correctly includes a compensation for the reading measured from the sensor prior to the test. That is, the test gas will cause a predetermined change to the sensor output, if the sensor is operating correctly, and that change is additional to any gas the sensor may be sensing and l5which persists around the sensor during the test due to the dead space around the sensor.

The rate at which the sensor changes in response to either the test case or the venting of the test gas, should preferably be used to determine whether the sensor is functional, as well as the absolute readings obtained.

20The device preferably contains a control means to carry out the test at predetermined intervals. During the test, the normal alarm function of the sensors would preferably be suppressed, to avoid false alarms.

The device would typically annunciate an audible or visual alarm when the sensor response is determined to be outside predetermined limits as a result of the test of 25either the test gas presence or the release of the test gas.

Claims (11)

1. WE CLAIMA device for automated bump testing of gas sensors with the ability to close one or more sensors to the environment, then dose a predetermined volume of test gas of a known constitution over one or more sensors for a predetermined period, measure the level read by the sensor, then release the test gas while opening the sensor to the environment, and compute from the response of the sensor to the test gas and the release of the gas, whether the sensor is operating normally and whether the sensor has a gas path open to the environment.
2. 2. A device according to claim 1 wherein the device comprises a single actuator which acts to provide three valve functions in a synchronous manner, the first function of which fills a gas reference volume to a predetermined pressure with a test gas or span gas, the second valve function vents gas in the reference space onto the sensors under test and the third valve function closes and opens the sensor gas path to the environment, such that the first and second valve functions operate in series and not concurrently, and the third valve function is synchronised to the first and second functions to effect a discharge of gas over the sensors which is largely retained over the sensors while the second valve function is open.
3. 3. A device according to claim 1 in which the test gas is of a known constitution mixes with a known volume of gas that is in the dead space around the sensor, and performs a computation compensating for the diluting effect of the dead space when determining whether the sensor is functioning correctly.
4. 4. A device according to claim 1 in which the test gas is of a known constitution mixes with a known volume of gas that is in the dead space around the sensor, and performs a computation that compensates for diluting effect of the dead space in determining whether the sensor is functioning correctly including a compensation for the reading measured from the sensor prior to the test.
5. 5. A device according to claim 1 in which the rate at which the sensor changes in response to either the test case or the venting of the test gas, is used to determine whether the sensor is functional.
6. 6. A device according to claim 1 where the test is carried out at predetermined intervals.
7. 7. A device according to claim 1 in which the normal alarms associated with the sensor are suppressed until the sensor has had time to recover from the test.
8. 8. A device according to claim 1 that annunciates an audible or visual alarm when the sensor response is determined to be outside predetermined limits as a result of the test of either the test gas presence or the release of the test gas.
9. 9. A device according to claim 1 which signals to a remote location data on sensors that are deemed not to be functional as a result of the test.
10. 10. A device according to claim 1 in which the dwell time of the test gas over the sensors is prolonged by use of a gas permeable membrane over the overall mechanism.
11. 11. A device according to claim 1 in which two plates with holes that align during normal sensor operation and which are opposed during the test function, are connected to the valve assembly.