GB2429921A

GB2429921A - CO2 scrubber monitor

Info

Publication number: GB2429921A
Application number: GB0512509A
Authority: GB
Inventors: Alex Deas
Original assignee: Individual
Current assignee: Individual
Priority date: 2005-06-18
Filing date: 2005-06-18
Publication date: 2007-03-14
Also published as: GB0512509D0

Abstract

A scrubber life indicator for a rebreather including a plurality of temperature sensors 29 from which a proportion of the integral temperature rise is measured to indicate scrubber life or life remaining. Preferably the temperature gradient is measured and analysed to indicate scrubber health. The scrubber operation may be divided into a multiple phases to track the position of the maximum temperature, take the average temperature of selected sensors and monitor the gradient. The temperature readings may be scaled as a function of depth, starting temperature, thermal capacity of the gases, flow rate, ambient temperature, humidity and moisture. The sensors near the output may be closer together and the sensors on the inlet may comprise an array. The sensors may be automatically calibrated. Compensation may be made for the thermal mass of oxygen or helium. The results may be displayed 19, computed 53 and stored in memory 52.

Description

C02 Scrubber Monitor

Background

Carbon Dioxide (C02) scrubbers have been in use since 1726 when Stephen Hale developed the first scrubber: a flannel liner, soaked in sea salt and tartar, used in a helmet for mine rescue.

Rebreathers are recirculating gas systems to enable a person to breath underwater or in hazardous environments and rely on a scrubber. The rebreather uses the scrubber to remove the C02 that is exhaled, and then injects 02 to maintain a breathable gas mixture. Rebreathers have been in use since 1876 when Henry Fleuss demonstrated a mine rescue rebreather.

A modem scrubber is a strong alkaline immobilised in either granules or a porous plastic extrusion. The active chemicals are usually Calcium and Sodium Hydroxide, such as in the brand Sofnolime (RTM), though others are also used, including Potassium Hydroxide and Lithium Hydroxide.

Sonfolime (RTM) is the most widely used scrubber material and comprises Calcium Hydroxide - Ca(OH)2 (75%), Water H20 (20%) and Sodium Hydroxide NaOH (3%).

A general description of the reaction of sodium and calcium hydroxide with C02 is as follows: 1. The gaseous C02 reacts with water to form carbonic acid - H2C03.

2. The NaOH reacts with the carbonic acid to produce Na2CO2 and H20.

3. The Na2CO2 reacts with the Ca(OH)2 which has been disassociated into Calcium and Hydroxide Ions, Ca-H- and OH-, to produce CaCO2 (calcite, the main component in limestone).

4. There is a net production of one extra H20 molecule for every molecule of C02 which is taken in.

The chemical reactions are: C02 (g) + H20 (I) ≤> H2C03 (aq) H2C03 (aq) + 2 NaOH (aq) -> Na2CO3 (aq) +2 H20 (I) + Heat 2NaCO3 (aq) + Ca(OH)2 (aq)-> 2NaOH (I) CaCO3 (s) + Heat For every molecule of C02 one additional molecule of water is liberated. The dissolution of C02 in water to produce carbonic acid is reversible but is the start of the subsequent two reactions.

Without water the first reaction would not take place. Water to prime the reaction is from the water vapour exhaled by the user, after which the reaction produces more and more water. The sodium hydroxide is regenerated.

Some Ca(OH)2 and NaOH based scrubber products also include Potassium Hydroxide - KOH, typically (1%). The waste product in every one of these cases is calcite (CaCO3), the primary component of limestone.

A similar cycle exists for potassium hydroxide and lithium hydroxide based scrubbers.

The C02 absorption reaction sequence is highly exothermic. As the scrubber nears depletion, C02 levels rise. It is useful for the diver to know what proportion of the scrubber has been exhausted, and what the level of C02 is. The latter is important in bail-out situations, where the rebreather may be operated manually in a semi-closed mode with flushes to keep the C02 within safe limits.

The absorption process is influenced by * Heat Balance (Chemical Reaction <> Convection / Conduction) * Water Balance (Chemical Reaction <> Evaporation) * Mass Balance (Chemical Reaction <> Flow / Diffusion / ppCO2 I Absorbent concentrations) The speed of the chemical reaction is dependant on the temperature. 10 C warmer can doubling the speed, and so decreasing the needed mass transfer zone. Colder absorbant means less evaporation and so too water condenses instead of passing as vapour, the water covers the surface of the absorber and reduces the active area. Deeper dives mean that the scrubber is cooled more by the compressed gas mix which results in a less efficient absorber: typically a 40% loss in efficiency at a depth of lOOm. Excess heat causes too much evaporation, resulting in a scrubber fIll lasting a shorter period in warm shallow water because the water vapour is required to initiate the reaction.

The first published attempt at mathematical modelling of the kinetics of a C02 absorber was done in 1993 using an Atari-ST and a Basic Program by Kramer using finite element methods driven by the basic chemical formula. Kramer concluded that the ppCO2 will travel as a wave into the scrubber, decreasing the further in amplitude the further it travels. Kramer concluded that if the scrubber is "too small" because it is depleted / was "drowned" I is too cold I the diver is working too hard I has a bad breathing-pattern, the "tail" of the PPCO2 wave reaches the end of the scrubber and the ppCO2 in the breathing loop increases. This is true for very large scrubbers, but rebreathers attempt to minimise both breathing resistance, mass and swept volume, so this model is does not hold for a modem rebreather.

It is being proposed by some health and safety organisations that both the scrubber life and the level of C02 be monitored, for example, in the rules proposed by the OSHA in Document, FR Doc 03-372 Federal Register: January 10, 2003 (Volume 68, Number 7). The OSHA ignored the issue that at the time of their publication there was no known rebreather that provided a C02 monitor that could withstand the moist operating conditions of a rebreather, nor was there any known method of providing a voice alarm also proposed so a diver that would be heard only by them rather than by all in the neighbourhood because sound travels very well underwater but divers lose the sense of which direction the sound is coming from. There was also no known method for predicting when failure of a scrubber would occur.

Scrubber manufacturers have for many years routinely measured the heat distribution in the scrubber as a means of assessing the scrubber life in any particular design. Manufacturers of scrubber materials, such as Extend Air mc, perform characterization tests on scrubber materials including among these being tests on how the heat is distributed.

The paper "Computer Modelling of the Kinetics of C02 Absorption in Rebreather Scrubber Canisters" by J. R. Clarke, Navy Experimental Diving Unit, 321 Bultfinch Rd, Panama City, Florida 32407-7015, published in OCEANS, 2001. MTS/IEEE Conference and Exhibition Publication Date: 2001 Volume: 3, vol.3, pp 1740 -1744 shows that the position of the temperature peak is not correlated significantly with scrubber depletion, nor with the peak temperature but moves within the scrubber primarily due to gas flow and work rate.

Sensors have been developed in an attempt to measure the life of the scrubber and indicate this to the rebreather diver, mostly by measuring the temperature differentials using an array of thermistors in the scrubber. Such sensors were in common knowledge by 1985, disclosed by lain Middleb rook of HSM Engineering in the early 1 990s and for that reason were included in Readey et. Al. in US Patent 6,003,513 as a matter of fact type item rather than a claim. Readey measures the temperature of the scrubber in eight positions.

Figures 2a, b, and c shows a similar system developed by Peter Steggle disclosed first in 2002, also using 8 temperature sensors.

Despite all of this background, US Patent 6618687 was issued in 2003 to Warkander on a basic temperature stick, that is a linear array of temperature sensors, which attempted to forecast scrubber life by the temperature difference normalised to the maximum temperature. As well as the issue of prior art, from the data in Figure 5 it can be seen that there is no direct correlation between temperature differentials and scrubber life.

The Evolution rebreather produced by APD, Heston, Cornwall, also uses what the manufacturer describes as a "scrubber temperature-measuring stick, which indicates on the handset which part of the scrubber is being used and so helps the user understand its depletion. It measures temperatures throughout the profile of the stack and is claimed to compensate automatically for depth, work-rate, water temperature and part- used scrubber material." The Frog rebreather also has a maximum scrubber temperature based performance monitor, claimed by the manufacturer to monitor the scrubber "during use by temperature probe to warn of exhaustion or operation outside of desired parameters." Some systems have been proposed on public internet forums which measure the position of the peak of the heat in the scrubber, and purport to give an indication of what portion of the scrubber remains, or has been used up. Again, on many scrubbers in rebreathers there is no movement of the temperature peak, though this does happen on very large scrubbers.

A Matlab (RTM) Simulink (RTM) model of a C02 scrubber by Vladimir Davidov disclosed in part via the internet in 2005 shows that modem scrubbers do not exhibit a distinct PPCO2 wave behaviour.

The actual temperature of any section of the scrubber depends on very many factors, including the ambient temperature, the storage temperature of the scrubber, the work rate of the diver - hence amount of C02, the packing of the scrubber, the air flow through the scrubber and the thermal mass of the gas mix - helium mixes tending to cool the scrubber very differently to air.

In practice, the reading from scrubber life indicators that track the position of the heat front in the scrubber, can be very misleading. The fundamental problem is that these systems assume the scrubber converts the C02 to heat instantly, and the scrubber can be considered as a series of stages as shown in Figure 6. This is a false assumption, as the mathematical results of the model of the assumed behaviour shown in Figures 6 to 16, and the actual behaviour as shown in Figure 5: The mathematical models of scrubber behaviour are developed and considered in detail in describing the present invention, but at this juncture, a summary that they do not show a correlation between either C02 level or scrubber depletion and either the maximum temperature or position of peak temperature. This is the first problem with the contemporary scrubber life sensors: the assumption which forms the fundamental basis on they operate is false.

All of the thermal differential scrubber sensors in the prior art fail to indicate the actual value of the C02, nor do they indicate reliably the scrubber life remaining, but simply that the scrubber is operating.

The OHSA proposals require that a rebreather gives a measurement of the C02, which is different from simply measuring scrubber life. The reason for this requirement is that there are various conditions which can cause inaccurate measurements from a scrubber sensor, including errors in packing the scrubber and water flooding the scrubber. In the case of a partial flood, the scrubber sensors will show a heat differential at the level of the water, and the scaling electronics will tend to scale the real C02 heat front such that it will not be apparent to the user. The user will believe the scrubber is working normally, when it is in fact exhausted. Warning signs such as a gurgling sound from the rebreather when breathing generally occurs only in the case of gross flooding.

Another problem with simple scrubber life indicators based on the position of the thermal peak, is that the C02 absorption depends on the Vital Capacity of the lungs of the user and their Tidal Volume. These volumes determine the actual dwell time, which varies considerably as a function of the amount of gas exchanged during the breathing cycle. This can vary from 0.6 litres at rest, to 4.5 litres for a deep breath with effort to maximise capacity: this assumes the diver has a Vital Capacity of up to 6 litres. A low dwell time from either poor scrubber design or rapid deep breathing, can reduce the scrubber efficacy by a factor of 4 compared to that with a good scrubber with shallow slow breathing. With slow breathing, the thermal front can be distributed, or even double peak depending on the scrubber packing. This can cause the sensor to give misleading readings.

Another problem with thermal front scrubber life measurements is that during the first part of a dive, the diver will be working hard, generating a lot of C02, which means the scrubber will be used up very quickly. During the ascent portion of a dive, which is the great majority of the time spent in the dive, the diver is hanging or performing very little work, so the amount of C02 generated is low and dwell time is long: this means the scrubber will be used up very slowly and previously "used" regions of scrubber material will come back into use, foiling the sensor.

Yet another problem is radial scrubbers which are used commonly, for example in the Prism and Cis-Luna rebreathers. The Cis-Luna uses the outside of the scrubber as the inlet, moving to the centre. As the design is radial, the volume of scrubber material reduces as the square of the distance from the centre. This can cause "flash" failure indication of the scrubber unless the thermal sensors are spaced on a square law from the centre.

These various problems can both mislead the diver into thinking their scrubber is working, when it is not, and gives no indication of a C02 spike that can quickly incapacitate a diver. When C02 is excessive under the conditions of elevated 02, the diver hallucinates and loses the ability to think normally: there are usually no warning signs such as headaches that occur when PPO2 is not elevated, that is when it is around 0.2. This means that these problems are very dangerous.

Methods well known to the art to measure C02 include infra-red absorption, particulaily mutti- wavelength and multi-channel absorption. The problem with infra-red methods is the very high level of humidity in the rebreather tends to cause condensation which affects the sensor deleteriously. Another method known to the art is based on a chemical reaction involving the C02, not unlike that in the scrubber, and measurement of the chemical byproducts. That method is effective in a laboratory, but is not suitable for use in a rebreather as it involves wet chemical reaction vessels.

Another method of C02 sensing is parametric measurement of an ionic reaction, such as the paper "Potentiometric C02 gas sensor with lithium phosphorous oxynitiide electrolyte", C. Lee, S Akbar and C. Park, Sensors and Actuators B 80(2001), pp 234 to 242, and "C02 Sensorn fur die Uberwachung Der Raumlauftqua(itaV', Christian Vott, Article 743 Published in Electronic Industry 11, 2000. Other descriptions of a similar C02 sensor can be found in J. Maier et. Al, J. Chem Thermodyn V18, 1986, pp309, J. Maier, Solid State lonics V62 1993, pplO5, and M. Holzinger, J Maier and W. Sitte in Solid State tonics, V86-88 (1996) pp 1055 - 1062. The main drawback of parametric sensors is they generally require a high operating temperature, often around 400C, which is not possible in a rebreather due to limits of power consumption and safety concerns.

The present invention can be used in conjunction with one or more of these parametric C02 sensors or infra red C02 sensors.

Object It is a primary object of the present invention to provides an accurate indication of the scrubber life and a further object to monitor scrubber health. A third objective is to determine the general C02 level in a rebreather.

Statement of Invention

The present invention determines the scrubber life by modelling the scrubber as a battery with a known capacity to remove C02, which as it is expended causes the temperature rise in the scrubber, hence the scrubber life is the inverse of the integral of that temperature rise after correcting for the losses that occur as a function of depth, start temperature, operating temperatures and the gas cooling the scrubber.

The health of the scrubber is determined in the present invention from the relative temperature rise of different sections of the scrubber, and gradients thereof, divided into multiple phases.

The indication of C02 levels in the present invention is computed from on how much work the scrubber has done and gradient of the temperature differential, with correction factors based calculation of the total heat generated by the scrubber, the rate at which 02 is being consumed, the effects of depth, temperature, start temperature and gas cooling.

Brief Description of the Invention and Figures

The invention will now be described by way of example, without limitation to the generality of the invention, and with reference to the following figures: Figure 1 is a block diagram of the present invention. In its most basic form, the invention includes a distributed set of temperature sensors 29, referred to herein as a "Temperature Stick", pressure sensor 42, ambient temperature sensors 44, a multiplexer 50 taking the output of the sensors to an Analogue to Digitial conversion circuit 51 which samples the data and converts them to a set of digital values, which are processed using finite state machines or microcontroller 53 to drive an information display 19 and a warning annunciation system shown here within 19.

Values particular to the design of the rebreather and scrubber are held in a memory 52 part of which at least is non-volatile.

The Temperature Stick 29 has array of temperature sensors distributed axially through the core of the scrubber in the case of an axial scrubber, and across the scrubber in the case of a radial scrubber. A logarithmic distribution of the sensors has advantages of greater resolution during the phase where the scrubber enters the depleted region, and is preferred over a linear distribution. A distribution of the sensors favouring the centre of the scrubber is also advantageous.

Sensors which show a reading less than 5 degrees above the temperature of the gas into the scrubber should preferably be removed from the calculation of the temperature, as the sensor could be in water collecting due to a slow flood or simply the water condensed from the by product of the C02 absorption.

Embodiments which indication C02 value include a second pressure sensor 46 to determine the difference belween breathing loop pressure and ambient pressure, gas flow rate 45 and full gas monitoring to determine the composition of the breathing gas: in the case of 02 and nitrox only rebreathers this can be just three 02 sensors 47,48 and 49 which are converted to a single 02 value using voting logic or suchlike in the processor 53 or by a dedicated circuit, but otherwise includes a helium sensor 54 such as a platinum resistance helium sensor operating in a bridge circuit.

A preferable embodiment includes an independent C02 sensor 55 for calibration and gas monitoring, such as an infra-red C02 sensor protected from moisture by location, using hydrophobic gas permeable membranes 56 such as GE PTFE Laminated Membrane or GE Osmonics OEM Nylon Hydrophobic membrane, moisture absorbent material around or near the C02 sensor, a metal condensation point 57 cooled to the outside ambient on which condensation settles rather than the C02 sensor, a design avoiding condensation by use of heat and low thermal conductivity materials, or combination thereof.

A preferred embodiment of the present invention includes a scrubber which has uneven gas flow, such that it breaks down gradually rather than very suddenly, with one Temperature Stick for each scrubber or each major region of the scrubber where the regions have markedly dissimilar gas flow, hence multiple instances of the Temperature Stick 29 are shown. In the case of a single scrubber operating as a single device, only one Temperature Stick 29 is essential.

Figure 2 (Prior Art) shows (2a) a scrubber life sensor display, (2b) a scrubber life sensor and (2c) a sensor fitted to a Buddy Inspiration rebreather, all these pictures being Prior Art of a scrubber life sensor developed by Peter Steggle in 2002.

Figure 3 is a photograph of a rebreather display which uses both C02 sensing and scrubber life sensing according to the present invention Figure 4 a dive handset showing an information display with a separate text display, an on-off control concentric with the inlet cable and shown in the on position, a menu select control under the text display.

Figure 5 shows the results of empirical measurements of the temperature distribution in a perfectly uniformly packed scrubber, the measurement being taken using a 8.5 digit UI 1906 Computing Bench Reference Multimeter froml 8 temperature sensors distributed linearly along the axis of the centre of the scrubber. Sensor I is nearest the inlet to the scrubber and sensor 18 nearest the outlet. The axis are time in minutes in the X axis, and temperature on the left Y axis and C02 percentage on the right Y axis. The bottom and rightmost trace is the C02 level, all others are of temperature. The measurements were taken using a 4 inch diameter, 150mm long prepacked scrubber from Extend Air Inc. Similar readings were observed using granular scrubber material.

Figure 6 (Prior Art) shows in diagram form a model of scrubber behaviour assumed by known scrubber life sensors. The scrubber comprises n layers, shown in Figure 6 as 10 LI to LI 0 by way of an example, with parameters using the equivalent electrical model. The temperature resistance between layers is R112, R_2/3, ..., R_9/10. The resistance between the layers and outside space is Rout/I and R_1 0/out. The chemical reaction of each layer absorption defined as a souse of energy is CR1, CR2, CR3, ..., CR10. The "GND" potential corresponds to the Absorbent outside temperature.

Figure 7 (Prior Art) shows an example of the absorption model for a single layer of the scrubber, with the concentration of C02 at the output of the scrubber layer normalized to 1.

Figure 8 (Prior Art) shows the Simulink (RTM) model of a single layer of the scnbber.

Figure 9 (Prior Art) shows the C02 output from a single stage using the model in Figure 8.

Figure 10 shows an example of the combined Simulink (RTM) model of a scrubber modeled as layers Figure 11 shows the temperature distribution through a scrubber modeled as a series of layers.

Figure 12 shows the C02 absorption of each layer with time.

Figure 13 shows the change in temperature along the axis of the scrubber, as a function of the model parameters.

Figure 14 shows the position of the peak temperature differential over time for a scrubber that can be modeled as a series of layers.

Figure 15 shoes the peak temperature of the scrubber relative to ambient. That is ambient temperature is show as zero.

Figure 16 shows the gradient of temperature when the scrubber layers are modeled using equal thermal resistances.

Figure 17 shows a table of temperature gradients as a function of the thermal resistance.

Figure 18 shows the display of a scrubber monitor based on temperature differentials when the scrubber can be modeled by a series of layers.

Figure 19 is a block diagram of a rebreather incorporating the present invention. This shows a rebreather with a high pressure oxygen cylinder (1) and a tank pressure sensor, with the pressure reduced and regulated to an intermediate pressure using a first stage regulator (2), feeding an 02 injector (4), a breathing loop comprising breathing tube (10) and (11), with a mouthpiece 12 which may also incorporate warning lights or buzzers, with two one way valves (5a), (5b), a breathing bag (17) with over-pressure relief valve (18), and a scrubber housing (27).

Dilutent is added from a second high pressure cylinder (6), reduced to an intermediate pressure using a regulator (7) preferably with a tank pressure sensor (8), feeding a dilutent adding device which is in this case an auto-dilutent second stage valve (9) preferably with an autoshutoff valve (23) to maintain the life critical section of the breathing loop to the minimum, power supply and sensing electronics (13), driving a handset display and annunciation system (19). The Temperate Stick(s) 29 for the scrubber life monitoring and for the C02 sensor that is the subject of the present invention, are in this case co-mounted and traverses the length of the scrubber (15), while the C02 sensor is placed in the inlet air stream (31) and on the control electronics on the exhaust air stream along with the power supplies (13). The scrubber may be one or more scrubber units connected in parallel, or very short sections connected in series.

Operation of the Present Invention The operation of the invention will be described, by reference to example embodiments without limit to the generality of the invention, by starting first with the description of models of the scrubber and experimental data of the thermal effects seen as the scrubber is depleted.

The actual scrubber behaviour is central to this invention and the principle on which the present invention is based, runs contrary to the widely held view of the operation of a scrubber and without this description, the reader is at risk of not correctly comprehending the invention. In particular, the scrubber is widely held to display a heat front which moves through the scrubber progressively as the scrubber is depleted. This is an incorrect assumption, and as it is so widely held, and is fundamental to the present invention, the correct model will be developed as follows.

The aforementioned models of scrubber behaviour that was developed by Vladimir Davydov and published previously in part makes the following assumptions to build a model of the temperature distribution along the Absorber axis.

1. The scrubber can be modeled as a series of layers, shown for example in Figure 6 and Figure 10 as 10 layers.

2. Temperature resistance between any two nearest layers (R_1/2, ... R9/10) is the same.

3. Each layer absorption is proportional to the C02 concentration 4. For each layer the speed of absorption cannot be higher the maximum level.

5. Each absorption process has saturation.

6. For gas flow with constant concentration each layer's absorption consist of two processes. The first stage is the constant absorption and the second one is decreasing of absorption along positive feedback curve as it shown in Figure 7.

7. Energy of the chemical reaction is proportional to the absorption.

The model of the absorption was built using the Simulink (RTM) facility within MatLAB (RTM), a widely used tool for modeling the properties of systems. One such overall model, layer model and the total model are shown in Figures 6, 8 and 10 respectively.

The active capacity of a stage depends on the point at which the state is saturated, defined by 1n2. The curve of the second stage absorption is defined via the parameters of the Transfer Fcn" block. The gain of 0.187 is taken to provide the following data: the input C02 is 4%, and the Absorber output C02 is 0.5 % - measured empirically. The formula for gain if the scrubber is modeled as 10 stages is therefore (1-gain)'iO = 0.5/4.

The absorber output: "scope 1" is shown in Figure 9, in which the output is normalised to the 4% and 0.5% levels.

Each layer has an absorption model shown in Figure 7. The temperature of the Absorber layer depends on the heat generated by chemical reaction and the Absorbers temperature resistances between the layers and outside space as it is shown in Figure 13.

The table in figure 13 shows the temperature curves as a function of the Absorber temperature resistance between the edge layers and the outside space. The outside temperature resistance between the first/last layer and space are R_outll, and R_1 0_out correspondingly. Theirs values is 1/4 and 4 times of the internal resistance between nearest to layers named "R".

Layers having the max temperature and its temperature value is a function of time and the Absorber temperature resistance between the edges layers and the outside space, and is tabulated below, as well as plotted in Figures 13 and 14.

Normalised Rout/I =0.25; R_oufll =0.25; Rout/I =0.25; R_outIl =4; R_outll =4; time R_out/1 0=0.25 R_outll 0=0.25 R_outll 0-4 R_out/1 0=0.25 R_outfl 04 _________ Layer Temp Layer Temp Layer Temp Layer Temp Layer Temp 1.0 5 13.9 5 8.8 6 17.2 3 26.3 4 44.9 1.5 5 13.9 5 8.8 6 17.2 3 26.3 4 44.9 2.0 5 13.9 5 8.8 6 17.2 3 26.3 4 44.9 2.5 5 14.0 5 8.9 6 17.3 3 26.3 4 45.0 3.0 5 14.1 5 9.0 6 17.6 3 26.2 4 45.0 3.6 5 14.4 5 9.3 6 18.4 3 25.6 4 44.9 4.1 5 14.8 5 9.8 6 19.6 4 24.8 5 44.6 4.5 5 15.3 5 10.3 6 21.1 4 23.85 44.5 5.0 5 15.3 5 10.5 6 22.8 5 21.6 6 43.4 5.5 6 14.8 6 10.1 7 23.3 5 19.3 6 41.4 6.1 6 13.0 6 8.8 7 22.3 6 15.5 7 37.3 6.5 7 11.3 7 7.4 8 20.7 6 12.7 7 33.6 7.0 7 8.8 7 5.5 8 17.9 7 9.0 8 27.7 7.5 8 6.1 8 3.5 9 13.8 7 5.4 8 21.0 8.0 8 3.5 8 1.8 9 9.6 8 2.7 9 14.0 8.5 9 1.6 9 0.6 10 5.0 9 0.9 9 7.2 9.0 10 0.6 10 0.1 10 2.1 10 0.2 10 2.9 Temperature gradient between layers and the outside space for the same outside and internal temperature resistances is shown in Figure 17. The relationship of the gradient of the outside absorber temperature resistances is in Figure 17.

The multistage model is an accurate and useful model for large scrubbers, such as is used for saturation dive chambers, but for rebreathers, it can be seen that the experimental results correspond to a scrubber with just a single layer. This is not unexpected, as issues of dwell time have been central to rebreather design from inception: that is the volume of the scrubber is so small, that the time the gas is in the scrubber is often the minimum needed to ensure there is an efficient reaction initiated by the water vapour and C02.

It is apparent that using this model and empirical data on a particular scrubber, such as that shown in Figure 5, that there are three main phases to scrubber operation: 1. During the light off phase, seen in the first 19 minutes of the test result shown in Figure 5, the temperature front moves progressively from the start of the scrubber to a stable position nearer the scrubber centre, and during this phase successive thermal sensors in the Temperature Stick(s) 29 show the peak temperature.

2. During the plateau period, lasting from after the first 22 minutes until 130 minutes in figure 5, the core of the scrubber heats up, but all parts of the scrubber are operating.

3. During the stage where the scrubber is beginning to show signs of being exhausted, the scrubber indicator still shows the temperature peak to be in the centre. Later on, the entire scrubber become exhausted so the temperature peak does move, but movement of the peak has a logarithmic relationship to scrubber life.

The scrubber life sensor sums and integrates the output from all sensors on the Temperature Stick 29, and shows the norrnalised inverse of this as a measure of scrubber life remaining. The normalisation is performed relative to a stored value, which is the capacity of the scrubber under fixed conditions, then modifies the value derived from the stored capacity value of the scrubber as follows: 1. Depth reduces the capacity by a factor which is stored. That factor is typically 40% compound per I OOm, but can vary from 20% to 80% depending on the scrubber type and design.

2. Initial Start Temperature. When the scrubber starts the dive cold, then it appears to never recover its full capacity even after being warmed up. This adjustment can be applied to the scrubber capacity, for example, reducing the capacity by a further 30% when cold, this factor being stored and determined by empirical data obtained from tests on the scrubber by the manufacturer or designer.

3. Gas Flow Rate. The porlon of loss of temperature due to cooling by the breathing gas should be determined by the designer or manufacturer and a correction factor relating to the breathing rate and the gas composition applied. The loss due to flow rate is a simple scaling, or scaling using a look up table of scale values as a function of gas flow rate.

4. Gas Composition.

If breathing gases other than nitrogen and oxygen are used, a further correction factor is necessary. This can be achieved by selecting a worst case factor when for example helium is selected by the diver, or preferably, is based on actual measurements of the helium content. Preferably, full gas monitoring is provided.

The conction factors for gases take the form of stored values or a table based on the heat change = gas mass * specific heat capacity * temperature change. The gas mass is function of depth and should have already been accounted for. Helium has a specific heat of 5193 J/(kg.K), compared to Nitrogen 1042 J/(kg.K), Oxygen 920 J/(kg.K).

Helium can be detected using the difference of the speed of sound (965m/s, compared to 317m1s for 02 and 334m/s for N2), using ultrasonic impulses travelling between two points, or by the difference in thermal absorption using a Wheatstone Bridge with a platinum resistance element exposed to the gas. Where the rebreather uses different ratios of helium and nitrogen as dilutent, it is necessary to measure the portion of helium in the gas mixture, and based on the pressure and gas capacity of the scrubber, to multiply the temperature differential across the scrubber by the portion of helium in the gas mix times 4.98 (the ratio of the specific heat capacity of helium and nitrogen respectively). It is preferable for reasons of accuracy, also to scale the measured temperature differential by the portion of oxygen times 0.88 (the ratio of the specific heat capacity of nitrogen and oxygen). Oxygen PPO2 and pressure are sensed on almost all rebreathers, using voting logic and other high reliability and failure tolerant techniques.

Similar compensation should be applied preferably for the water vapour in the breathing loop, and the mass of gas due to temperature changes.

5. Loss to the environment.

The loss to the environment can be determined by the manufacturer by heating the scrubber to a calibration temperature when the scrubber housing is immersed in water, or air in the case of rebreathers designed for non-diving applications, and measuring the amount of energy needed to maintain the temperature as the water temperature is changed. This should produce a table, interpolation of which, allows an environmental loss to be added to the scrubber temperature to produce an overall integral which more accurately represents the total heat generated by the scrubber.

6. Moisture and Humidity Water is needed for the reaction to operate, but excess moisture reduces the effective surface area of the scrubber. The reduction can be estimated by the amount of moisture in the system or on a sensor. The temperature values can be scaled by the humidity level, such that low humidity levels indicate low scrubber efficiency, and high moisture levels indicate a reduction in scrubber activity. The moisture and humidity sensors are normally an integral part of the rebreather control system, which is assumed in all drawings and descriptions, to be integrated within the processor 53, memory system 52 and display and annunciation system 19. Humidity is the amount of water vapour, and moisture is the amount of water that is condensing.

The controller in the present invention in the example embodiment performs the foHowing compensation steps: 1. Determine the readings from temperate sensors.

2. To measure the gas pressure and scale dissipation correction factor Tdis by the absolute pressure in bar, or indicator thereof.

3. To measure the portion of 02 in the gas, and divide Tdis by that portion times 0.88.

4. Preferably, to measure the portion of He in the gas, and multiply Tdis by that portion times 4.88 5. Adjust for the gas flow, using a table or a scaling factor determined experimentally or by calculation and scale the correction factor by that amount.

6. Apply the each of the compensation factors in a similar manner.

7. Scale the temperature from each sensor by the combined scaled compensation factor and integrate the result.

8. Sum the value of all the integrated results, to produce a parameter related to the amount of the absorbent capacity of the scrubber that has been consumed, and use the inverse of this to show scrubber life imaining.

AD scaling factors should ideally be applied prior to summation, as some factors are a non-linear function of temperature and involve either several parameters or are stored in a look up table as a function of temperature.

The compensated integral of the temperature rises can be stored regularly in non-volatile memory 52 in case the rebreather is switched off, or on the rebreather sensing a power down.

It will be appreciated by anyone skilled in the art that one or more compensation factors can be omitted, with a consequence of a larger error margin on the resulting indicator.

It will also be appreciated by anyone skilled in the art, that the above factors can be compensated for in a variety of ways: linear mathematical operations can be rearranged in order, and inverse measurements can be applied to produce an inverse indicator as is most convenient to the designer.

In an extremely simple case, the scrubber life sensors in the Temperature Stick 29 can be reduced to a sensor on the gas inlet and a sensor on the gas outlet, as the temperature rise in the scrubber is conducted to the gas stream. This extreme case highlights the principle on which the present invention works, in contrast to the prior art, in that it measures the total thermal output of the scrubber or a parameter related thereto. However, preferably the sensors are in contact with the scrubber material and measure the rise in temperature of the scrubber mass.

In extending the indicator of scrubber life to provide a scrubber health indicator, the same sensors and same data stream can be used. The difference between the two applications of the data is that scrubber health is indicated in three phases: 1. During the first phase, the temperature should rise on each indicator in turn, until the stable operating point of the scrubber is reached. In the data in Figure 5, sensors Ito 8 show a peak temperature in turn during the startup phase, the minimum temperature reached by the first sensor is 35 degrees and the minimum peak temperature is 45 degrees. Figures less than these absolute numbers during the startup phase indicate a faulty scrubber.

The startup phase is initiated as soon as the user breaths the rebreather, indicated by a temperature rise of the second and third sensors successively. That is, the rebreather monitors the temperature sensors in the Temperature Stick 29, and when the third temperature sensor is higher than the second, which is higher than the first, then the unit can indicate it is ready to dive. A fixed amount by which one sensor is higher than another can also be used, for example, that the second sensor is 3 degrees warmer than the first sensor.

The processor can monitor easily the march of the peak temperature through the sensors and show that the scrubber has capacity which reduces as the integral of the temperature rises.

2. During the second phase, the temperature is stable and the scrubber operates on a plateau. The peak temperature will vary in position between known positions on the Temperature Stick, and the compensated gradient of those temperature changes will match a table that should be stored. The gradient of the temperature differences in the scrubber is the primary indicator of scrubber health during this plateau phase.

The minimum temperature for all sensors during the plateau phase, except the first 2 or 3, and in some cases, the last Iwo or three, should be maintained at not less than 43 degrees, this number depending on the design of the scrubber and is a stored value for any particular design.

3. The depletion phase of the scrubber is indicated by either the scrubber life indicator showing the scrubber is exhausted, or by all temperatures in the scrubber dropping below prescribed limits. In the case of the scrubber with the data shown in Figure 5, there is a drop of temperature of all sensors below the previously recorded peak and the first three sensors drop below 40 degrees, this being a stored value. Different scrubber designs and Temperature Sticks with different number or spacing of the sensors will show a different number of sensors failing to reach a minimum temperature. During depletion phase, the scrubber monitor can indicate scrubber depletion by a factor which is a fUnction of how large the temperature drops are on specific sensors.

A "rough" but cheap scrubber health sensor according to the present invention needs only the values of the temperature gradients. The output of such sensor is shown in Figure 18 (R_outll = R_1 0/out = R). The "1" of the table correspond to the state when the gradient is more than the half of the max gradient. A row of the table shows sensor's outputs along Absorber's axis. A column contains a one sensor output during the Absorber work. A more sophisticated and more accurate scrubber health monitor will include the correction factors that have already been listed for the scrubber life indicator.

In the extension to the present invention to indicate C02 levels, the method by which is done shall now be described in more detail.

Respiration uses oxygen and produces carbon dioxide. The ratios are: GAS INHALED -vs- EXHALED 02 20.71% 14.6% C02 0.04% 4.0% H20 1.25% 5.9% This means that if no gas is lost from the system, then for a given 02 injection into the system, 65% becomes C02. The scrubber absorbs some of the C02, generating heat. The amount of heat depends on the scrubber design, the gas pressure, the thermal capacity of the gas mixture, the amount of C02 being absorbed by the scrubber and the other factors that have already been described.

If the embodiment of the invention uses the gas temperature differential across the scrubber, then the temperature sensors are preferably an array of sensors, such as resistors or a track on a circuit board, such that they measure the average temperature across the gas and not any specific spot in the gas flow. That is, the temperature sensors should preferably be distributed.

As it is the temperature differential that is being measured, the two sensors (in the gas flow prior and after the scrubber respectively), can be connected into the same bridge circuit, to form two different arms, such that a differential is measured. However, it is preferable that the sensors are in contact with the scrubber and integration and summation of the individual sensor readings is performed.

The heat given off by the absorption and conversion of C02 into calcite, is dissipated in the scrubber, into the environment and into the air. The scrubber mass being constant, the loss to the scrubber is also constant. The discharge to the environment varies with water temperature, but is reasonably constant and as the scrubber is a poor thermal conductor, environmental dissipation is low. The primary cooling route is via the gas passing through the scrubber and loss to the environment. The heat differential for a default gas mix is measured, stored in a table and the table value is used to convert the C02 into heat.

For C02 sensing in particular, it is necessary that the volume of gas flowing through the scrubber is sensed. The gas flow data can be used to determine dwell time in the scrubber, and also used as a correction factor to produce a more accurate forecast of scrubber life. Measurement of the gas flow rate can be achieved using any one of many different methods, preferably using one of the many micromachined gas flow sensors that are available, but other methods may be used including the volume or frequency of the sound in a gas vortex, a micro vane on a spring, or solid state method using either the transit time of uttrasonic impulses in one, two or more directions, or by using thermal methods, or by a pressure differential using a venturi such as a Pitot Dracy tube.

The amount of C02 in the system is 0.65 times the amount of oxygen injected, minus the amount of C02 removed, minus the amount of oxygen and C02 vented from the system.

The amount of oxygen being injected is known from the on time of the 02 solenoids, or the motor position of the 02 injector. The volume of 02 injected per pulse into the 02 solenoid can be determined by the system automatically during pre-dive checks, by ensuring the system is closed, lightly pressurised and the controller injecting 02 and measuring the increase in pressure. For any given rebreather design the volume of the rebreather will be known and constant, so the amount of 02 injected per pulse can be deduced. This injection volume remains constant until the 02 cylinder is nearly exhausted.

The amount of 02 vented from the system can be determined by the drop in pressure that is not accounted for by the removal of C02. Most rebreathers measure the loop pressure, and this sensor can be used to sense gas loss from the system, attematively a differential pressure sensor 46 can be used. The volume of the breathing loop can be regarded as constant, as the variable volume of gas in the rebreather is simply exchanged with the gas in the lung of the diver, and it is the total amount of gas that is relevant.

When the diver ascends, there is excess gas in the rebreather, which is normally vented. The amount of gas lost is proportional to the drop in ambient pressure.

The other complication is the addition of dilutent. In autodilutent systems, dilutent gas is added automatically as the diver descends. In manual dilutent systems, the diver must add the gas. In either case, the diver experiences difficulty in breathing if the difference between ambient pressure and the breathing loop is more than about 0.5 bar. This means that as the diver descends, gas is added to the system, such that the total volume of gas being the dead volume of the rebreather and breathing bag, times the depth in bar absolute. Upon the gas pressure increasing in the rebreather, the calculation is performed to reduce the C02 concentration by the proportion of the added gas.

To avoid errors accumulating, when there is high temperature differential in the scrubber, such as all sensors except for example the first 3 and last 1, are at a temperature greater than 20C above ambient, it can be assumed in a particular embodiment that the scrubber is working and the C02 is very low. When the temperature drops below this level, it can be advantageous for the system to start calculating C02 as described here, from that start point and ignore prior history. 02 injection rates can be calibrated by loop pressure rises during 02 injection in predive checks.

An algorithm or flow chart describing the overall function can be written easily from the description of its function above, to produce an absolute C02 reading. Such an algorithm can be encoded into a micro-controller in the rebreather or into a state machine with data path section for the computation.

Various alarms can be annunciated if the C02 level is above limits, the scrubber life falls below other limits, the scrubber health monitor indicates the scrubber behaviour is outside limits, such as during the initial phase the successive temperature sensors in the Temperature Stick 29 do not progress in phase or reach minimum temperatures within a prespecified time.

Claims

Claims 1. A scrubber life indicator for application in a rebreather, with

a plurality of temperature sensors from which a measurement proportional of the integral temperature nse is used to indicate scrubber life or scrubber life remaining.
2. A device according to claim I where the temperature gradient is measured and parameters derived therefrom are used to indicate or provide warnings on scrubber health.
3. A device according to claims I or 2, where the scrubber operation is divided into a plurality of phases, which are tracked, the first phase of which the relative position of the maximum temperature is tracked, the second phase of which the average temperature of selected sensors and the temperature gradient across those sensors is monitored to indicated or warn the user on scrubber heatth.
4. A device according to any of claims 1 to 3 where a depletion phase is monitored that is triggered by the readings from particular temperature sensors or groups of temperature sensors falling in value below stored values and the maximum temperature also falling below a predetermined value.
5. A device according to any of claims 1 to 4 where the readings from the temperature sensors or a function derived therefrom are scaled as a function of depth.
6. A device according to any of claims 1 to 5 where the readings from the temperature sensors or a function derived therefrom are scaled as a function of initial start temperature of the scrubber.
7. A device according to any of claims I to 6 where the readings from the temperature sensors or a function derived therefrom are scaled as a function of the thermal capacity of one or more of the gases.
8. A device according to any of claims 1 to 7 where the readings from the temperature sensors or a function derived therefrom are scaled as a function of flow rate.
9. A device according to any of claims 1 to 8 where the readings from the temperature sensors or a function derived therefrom are scaled as a function of temperature differential between the scrubber and ambient environment.
10. A device according to any of claims 1 to9 where the readings from the temperature sensors or a function derived therefrom are scaled as a function of humidity.
11. A device according to any of claims I to 10 where the readings from the temperature sensors or a function derived therefrom are scaled as a function of moisture.
12. A device according to any of claims 1 to 11 where the temperature sensors near the output of the scrubber are spaced more closely than those near the input.
13. A device according to any of claims I to 12 where the temperature sensors near the output of the scrubber are spaced more closely in the centre region than those near the input.
14. A device according to claims 1 to 13, where an indication of C02 level is computed the difference between the integral of the amount of 02 injected into the breathing loop, data from sensors on the gas lost from the breathing loop and computation of the C02 removed by the scrubber based on the integral of the thermal data obtained through a device according to any of claims 1 to 13.
15. A device according to claims ito 14 where the temperature sensors measure the temperature of the input and output gas stream from the scrubber instead of the temperature of the scrubber itself and compute a function proportional to scrubber temperature from the differential gas temperature across the scrubber with correction factors that include any combination of data on depth, flow rate, ambient temperature, initial start temperature, humidity, moisture and gas composition.
16. A device according to any of claims I to 15, that includes a scrubber life sensor measuring the position of the thermal front in the scrubber as an indication of scrubber life.
17. A device according to any of claims I to 16 that includes the automatic calibration of the sensors during pre-dive tests.
18. A device according to any of claims I to 17 that includes the calibration of the thermal C02 sensor with an infra-red absorption C02 sensor during pre-dive checks.
19. A device according to any of claims I to 18 with compensation for compensation for the specific thermal mass of the portion of oxygen in the gas mixture.
20. A device according to any of claims I to 19 with compensation for compensation for the specific thermal mass of the portion of helium in the gas mixture.
21. A device according to any of claims I to 20 including compensation for the dwell time of the scrubber using a sensor to measure gas flow.
22. A device according to any of claims I to 21 including a display of both scrubber life, scrubber health and a C02 reading.
23. A device according to any of claims I to 22 where the temperature sensor on the inlet and exhaust of the scrubber is distributed, comprising an array of sensors or a pcb track.
24. A device according to claims I to 23 which stores the value of the scrubber life or parameters related thereto in non-volatile memory and initialisation using those values in the event the system is powered down and then powered up without replacement of the scrubber.