Carbon Monoxide Health Hazard Monitor

The invention relates to a carbon monoxide monitor for use on marine vehicles and more particularly to an apparatus and method for monitoring the health hazard to the occupants of a marine vehicle occasioned by breathing the carbon monoxide fumes generated by the internal combustion engines of the marine vehicle.

Carbon monoxide is an odorless, colorless, and tasteless gas that is lethal if present in sufficient concentration and if at a particular concentration it is breathed over a sufficient period of time.

Carbon monoxide detectors are available for use on marine vehicles, however, these instruments function only to detect a specific selected level of carbon monoxide and to sound an alarm and/or to provide a visual indication when this level has been exceeded. It is perfectly safe to breathe a relatively low concentration of carbon monoxide e.g., 50 PPM, over an extended period of time, whereas it would be lethal to breathe a high concentration of carbon monoxide e.g., 450 PPM, over even a relatively short period of time. Thus, prior art detectors that are set at the lower level will give an alarm even though no health hazard exists. On the other hand, prior art detectors that are set to the higher level will not alarm even though carbon monoxide concentration below the alarm set point is present that may be lethal if breathed over a sufficient period of time, as for example, the amount of carbon monoxide that may be breathed by a person sleeping on a marine vehicle while internal combustion engines of the vehicle, or other sources of carbon monoxide, are operating.

It has also been determined that the lethality of carbon monoxide is not only a function of concentration and the length of time a person breathes carbon monoxide at that concentration, but also a function of such factors as the rate of diffusion of carbon monoxide into the lungs, barometric pressure, minute respiratory volume and level of exertion or work which together establish a level of carbon monoxhemoglobin (COHb), in the blood. It is the level of COHb in the blood that constitutes a health hazard rather than carbon monoxide concentration or the length of time that carbon monoxide at that concentration is breathed per se.

Embodiments of the invention function to detect and alert to the presence of toxic concentrations of CO (carbon monoxide) in two ways--both in terms of the present CO level (i.e., the CO concentration, in PPM, at the most recent update, or "measuring period") and in terms of exceeding an "exposure limit" which is based on %COHb (percent carbon monoxhemoglobin) in the blood. The present CO level is visually displayed on a seven segment bar LED whch indicates the relative CO concentration at the last "measuring period" (MP). The existence of a "Health Hazard", i.e., when the exposure limit (EL) has exceeded a present %COHb level, is indicated visually by both a "Health Hazard" (HH) light and one of three "condition lights" that advises if the HH was caused by a short term exposure to a high CO concentration, long term exposure to a low CO concentration, or somewhere between those extremes. A HH condition is also accompanied by an audible alarm (buzzer) where the sound rate varies corresponding to one of the three HH conditions.

Three self-testing schemes are provided: Manual Test Mode (activated by depressing a TEST switch), Automatic Test Mode (periodic self-test during normal operation), and Production Test Mode (where the unit is commanded to self-test various circuit areas and transmit the resultant test data to a production test fixture for analysis). The Manual Test Mode is also used to activate three different test sequences: Self-test, Self-test plus memory reset, and Self-test/memory reset/"super sensitivity for one MP."

If at any time a fault is recognized by any of the self-tests, a visual FAULT light and audible alarm will be activated.

Provision is also made for an optional interface (relay) that will control external apparatus, e.g., close solenoid valves or shut off engine ignition systems when a HH condition exists.

The sensor element utilized in embodiments of this invention requires a precisely timed variable power supply which cycles the sensor current at a 60 second high current/90 second low current rate. A measurement period takes place at the end of each cycle (i.e. at the end of the 90 second low current period). The power control/MP timing is provided by a programmed microprocessor (uP).

A sensor/load resistor circuit provides a varying output voltage during the MP which is directly proportional to the CO concentration level. The output voltage of the sensor is fed to an A/D input of the uP. Three other uP A/D inputs are used for various self-test tasks. The sensor input voltage to the uP is then compared to reference data in the uP's ROM, the CO concentration level (in PPM) determined, and the appropriate LED segment activated. The CO concentration level (in PPM) is then applied to a mathematical formula to calculate %COHb and the HH and condition alarms triggered when the preselected EL is exceeded. The "condition" is determined by analyzing the rate at which COHb increases.

It is therefore the object of the present invention to provide a safe and reliable carbon monoxide monitor unit for use on a marine vehicle which not only detects and indicates the concentration of carbon monoxide present in the vehicle but also determines and displays the health hazard, resulting from the %COHb in the blood of the occupant (and the rate of increase of %OCHb) to an occupant of the vehicle, occasioned by breathing this concentration of carbon monoxide for a given period of time.

FIG. 1 is a schematic circuit diagram of a preferred embodiment of the invention;

FIG. 2 illustrates the carbon monoxide sensor power and measurement timing cycle;

FIG. 3 is a graph showing R/R.sub.s as a function of COC;

FIGS. 4A-4C illustrates three manual test modes implemented by the preferred embodiment;

FIG. 5 shows, in block diagram form, the microprocessor portion of the preferred embodiment along with associated components;

FIG. 6 illustrates the initial power up sequence of the preferred embodiment;

FIGS. 7A-7N are a flow diagram of the steps formed by the preferred embodiment; and

FIG. 8 is a block diagram of a production test fixture according to the invention.

The preferred embodiment of the invention comprises four main functional blocks as follows:

Sensor and Sensor Power Control

Computation

Output (Visual Display, Audible Alarm, and Relay)

Power Supply.

Each of these main functional blocks will be described below, referring to FIG. 1 and other figures of the drawings.

FIGS. 7A-7N constitute a flow diagram of the steps performed by the preferred embodiment of the invention. These steps will be referred to in the following description.

Referring now to FIG. 1, the carbon monoxide sensor RS-4 may comprise a Figaro Model 203. When heated to various operating temperatures, the sensor responds selectively to numerous fumes (including carbon monoxide). However the temperature at which the sensor is selective to carbon monoxide is relatively low, and at that low temperature the sensor collects contaminants which impairs its performance. For these reasons, during each operation cycle (OC) of 150 seconds (21/2 minutes) the sensor drive power is cycled high current, I.sub.H, for 60 seconds which raises the temperature to a point high enough to purge the sensor of any contaminants which may have collected on it (see FIG. 7B), followed by low current, I.sub.L, for 90 seconds which allows the temperature to cool down to the point where it is selective to carbon monoxide (see FIGS. 7C-7E). After the 90 second period has been concluded, a measuring period (MP) of 500 MS (1/2 second) takes place. During this time, power is removed from the heater elements of the sensor. The voltage, V.sub.S, appearing across the residual resistance, R.sub.S, of the sensor is then proportional to the concentration of carbon monoxide to which the sensor is exposed. In addition, during each OC, two mini measuring periods of 25 MS take place in order to guard occupants of the marine vehicle against being exposed to concentrations of carbon monoxide as would be an extreme health hazard to breathe for the 21/2 minute OC, as well as for other purposes which will be later discussed. The sensor power and measurement cycle for an OC is shown in FIG. 2 where the sensor current (in ma) is shown plotted as a function of time. The voltage, V.sub.RL, appearing across the sensor load resistance, R.sub.L (FIG. 1) is also shown in FIG. 2.

Continuing to refer to FIG. 1, after a MP, the next OC sensor power cycle is started and simultaneously the information from that just completed MP is analyzed, processed, and appropriate alarm action is initiated if necessary, as will be more fully discussed below. Periodically (as frequently as the microprocessor (uP) IC-2 operation and software will allow) the unit self-tests and certain internal voltages are checked for correct levels. At any time after the power up sequence is concluded (i.e., during a normal OC) the instrument may be manually put into one of three test modes by depression of a test switch as will be more fully discussed below. Normal OC's resume after such tests. These later functions will also be more fully discussed below.

Referring more specifically to FIG. 1, the specific operation of the sensor and sensor power control section is as follows. During the 60 second 369 ma period, microprocessor pin 14 (PC5) is at logic 1 (approximately 5 V) which causes the open collector output section of comparator section IC-1A to be off. Thus the voltage at pin 5 (+input) of comparator section IC-1D is 0.55 v, as determined by the voltage divider network consisting of R2 and R4. The current flow through R1, Q1, Q2, and the sensor elements determines the voltage across Resistor R1, which is fed to pin 4 (-input) of IC-1D. For the current to be 369 ma, the voltage across R1 must be 0.55 V. If the current is higher than 369 ma, the voltage across R1 (and thus the -input of the comparator) would be greater than 0.55 V. In that case, that output of IC-1D would turn on, which would turn Q2 off reducing the sensor current back toward 369 ma. Similarly if the current is lower than 369 ma, the voltage across R1 (- comparator input) would be less than 0.55 V. This would cause IC-1D output to turn off, allowing Q2 to conduct more current increasing it back toward 369 ma. A gentle oscillation takes place, softened by the RC network consisting of RS2-3, 2-4, and 2-5 plus C3 that maintains the current flow within less than .+-.1% of 369 ma.

During the 90 second 133 ma, period, microprocessor pin 14 (PC5) switches to logic 0 (approximately 0 volts) which causes the output of comparator IC-1A to turn on, placing R3 in parallel with R4, reducing the voltage at pin 5 (+input) of comparator section IC-1D from 0.55 V to 0.20 V. Following the same analysis as above, 0.20 V across R1 (-input to IC-1D) yields 133 ma sensor current.

At the end of the two sensor drive periods (60 seconds+90 seconds), microprocessor pin 13 (PC4) changes from logic 1 to logic 0 for 0.5 seconds. This, in turn, causes comparator section IC-1C output stage to turn on, shutting Q2 off. Simultaneously comparator section IC-1B output turns off thus shutting Q1 off. Therefore, during this 500 ms measuring period a voltage divider is established consisting of the residual resistance, R.sub.S, between the primary and secondary elements of the sensor RS-4, (not shown) and the calibrated load resistor R.sub.L. The voltage, Vs, at the junction R.sub.S and R.sub.L is proportional to the concentration of carbon monoxide present wherein the sensor is located. As CO increases, R.sub.S decreases thus raising V.sub.S. Voltage V.sub.S is then sent to microprocessor analog-to-digital conversion input ANO (pin 24, PDO). Resistors RS2-6 and 2-7 establish a reference voltage for comparator sections IC-1A, 1B, and 1C while resistors RS2-1 and -2 bias Q2.

The response of the sensor is somewhat a function of ambient temperature, in that the sensor exhibits a negative temperature coefficient. Temperature compensation is provided indirectly by measuring temperature with a voltage divider consisting of resistor RS2-10 and a negative temperature coefficient thermistor RS-3. The voltage (V.sub.T) at the junction of RS2-10 and RS-3 is fed to another A/D input (pin 23, AN1-PD1) of the microprocessor.

The concentration (COC) is derived from a chart of COC vs (R/R.sub.S), FIG. 3, where (R/R.sub.S) is calculated by the uP as follows: ##EQU1## where Vsc is the temperature compensated sensor voltage.

This formula is valid when R.sub.L =R.sub.S (200 PPM) and V.sub.cc =5 V, where:

R.sub.L =sensor load resistor

R.sub.S =sensor resistance when the sensor is exposed to 200 PPM CO

R=sensor resistance at any CO concentration.

As a function of incoming quality control inspection, all sensors received are operated in a 200 PPM CO environment and its R.sub.S determined. The sensors are then identified such that the R.sub.L will equal R.sub.S in the unit in which that particular sensor is installed. Thus, for purposes of the invention, (and uP software), the conditions for the (R/R.sub.S) formula are always valid.

As noted above, the value of V.sub.S must be factored to compensate for the variation in sensor response vs. temperature. The sensor does exhibit a negative temperature coefficient (TC). Therefore, the temperature compensation factor (F) also has a negative TC. The Vsc (V.sub.S - temperature conpensated) is calculated as follows:

where F is determined from a temperature compensation voltage (TCV) as shown in the table below:

TCV is generated by thermistor/resistor network RS-3, RS2-10 and is fed into the uP via another A/D input port (PDI, ANI). Thus, the complete formula for (R/R.sub.S) is: ##EQU2##

Referring to FIG. 1 of the drawings, resistor R2S-8 (in parallel with an internal pull-up resistor in the microprocessor) and capacitor C4 set up an RC time constant charging circuit to delay activating the microprocessor until initial power-up of the unit has been completed per standard practice. Diode D3 provides rapid discharge of C4 upon power-down. Crystal X1 and capacitor C6 set the operating frequency of the microprocessor and capacitor C5 is a power supply filter capacitor per standard practice.

Resistors R9 and R10 form a voltage divider that establishes the voltage reference (high) for the analog/digital conversion in the microprocessor. Resistors RS2-8, RS2-9, RS2-10, and RS1-1 through 1-7 function as pull-up resistors.

The output section consists of two types of visual displays, an audible alarm, and (optionally) a relay that can be used to control external apparatus, such as a solenoid valve or the ignition system of the internal combustion engine.

The first visual display is a seven segment bar graph type light emitting diode assembly wired with common anode and individual cathodes connected to the collectors of buffer/driver/switches (IC-3). Since there is only one LED activated at a time, a single current limiting resistor (RS2-12) from the common anode connection to +5 V is sufficient. The seven microprocessor outputs (PB0-6, Pins 25-31) drive the inputs of the B/D/S devices in IC-3 with the logic 1=LED on, 0=LED off. The voltage at the junction of RS2-12 and the common anode line is fed to the AN2 A/D input (Pin 22, PD2) for self-test analysis.

The second visual display consists of 6 incandescent lamps, one side of each connected to +12 V and the other side of each connected to the collectors of B/D/S devices in IC-4. Operation and logic is the same as noted above and is obvious.

Relay RY-1 is connected across lamp LA-1 and operates with LA-1. The relay coil may be replaced with a fixed resistor to simulate the same current drain for self-test purposes. Diode D2 is used to prevent damage to the B/D/S device due to inductive transients caused by the relay coil when power is removed.

The audible alarm (Buzzer, X2) is similarly driven by the microprocessor through a B/D/S device in IC-4. Resistor R6 and Capacitor C7 form a low pass filter to prevent feedback of the buzzer's oscillations into the B/D/S device or microprocessor.

The emitters of all B/D/S devices in IC-4 are common (Pin 8 of IC-4) and grounded through R5. The voltage at the junction of R5 and IC-4 Pin 8 is fed to the AN3 A/D input (Pin 21, PD 3) for self-test analysis.

The audible tones are produced by piezoceramic transducer/oscillator circuit ("Buzzer") that is activated by a transistor driver (switch) which in turn is driven from uP I/O port PA1 with the following logic:

There are four (4) different sound pulsing rates as follows:

Finally the power supply simply consists of a half wave rectifier/reverse polarity protection diode, (D1), filtering (C1, and C2), and 5V regulator (IC-5). Provision is made for the addition of a surge protector (D5) and a high performance regulator with internal microprocessor reset signal generation controlled by capacitor C8.

During the MP, the voltage (Vs) from the sensor load resistor network (which is proportional to the CO concentration) is fed into an A/D input to the (PDO, ANO), factored for temperature compensation, analyzed, and the resultant data used to cause an LED to light per the following:

For reference, COC vs (R/R.sub.s) vs Vsc at the key points for LED operation is as follows:

A health hazard (HH) exists when a preset exposure limit (EL), in terms of %COHb, has been exceeded. The %COHb is determined by applying the COC to the following formula:

where:

"%COHb.sub.2.5 " is the present %COHb level, or the %COHb over the last 2.5 minute period,

"%COHb.sub..phi. " is the initial %COHb level, i.e. the %COHb level at the beginning of the last 2.5 minute period,

"COC" is the measured CO concentration in PPM.

A more general statement of the equation would be:

The calculated %COHb.sub.2.5 from the last 2.5 minute measuring period becomes the %COHb.sub..phi. for the next 2.5 minute measuring period.

The initial power-up %COHb (i.e., the %COHb.sub..phi. for the first %COHb.sub.2.5 calculation after power-up) is 0%. The RESET function (FIG. 4b and 4c) clears the %COHb.sub..phi. in memory to 0%.

The EL (i.e. the preset %COHb.sub.2.5 point) at which a HH alarm is to take place is selectable via the PA7, PB7, PC7, and PD7 uP I/O ports as follows:

where

DIGITAL 0--Ground potential

DIGITAL 1--+5VDC.

There are three HH conditions (HHC) or causes for a HH alarm as follows:

L/L or long term exposure to a lower COC level,

M/M or moderate term exposure to a moderate COC level, and

S/H or short term exposure to a high COC level.

The HHC is determined by comparing the %COHb.sub.2.5 to the %COHb.sub..phi. and calculating the rate of rise in %COHb in terms of %COHb per OC, as follows:

where

The HHC (visual and audible) and uP port status are as follows:

Since the drivers are of the inverting type, for an "OFF" condition the uP port is a DIGITAL 0 and an "ON" condition occurs when the uP port is at DIGITAL 1.

The HH lamp and appropriate HHC lamp stay ON as long as the %COHb.sub.2.5 level is above a selected EL alarm set point. However, the audible alarm (S1, S2, or S3) sounds only when the present COC is 100 PPM plus. When the present COC drops below 100 PPM, so that either LED G1 or G2 is ON, the audible alarm stops sounding. If at any point where the present COC level again goes above 100 PPM, and the %COHb.sub.2.5 level is still above the selected EL alarm set point, the appropriate audible alarm resumes sounding.

As shown in FIGS. 4A, 4B, 4C, there are three manual test functions that are activated by depression(s) of the TEST SWITCH (FIG. 1) as follows:

In all cases, the test switch depressions noted above are to be made within a 2.5 second period.

Test mode "A" cycles the unit's audible and visual alarm indicators and sensor power control operation as shown in FIG. 4A.

Test mode "B" is similar to "A" plus the COHb.sub.o memory is reset to 0%, as shown in FIG. 4B.

Test mode "C" is similar to "B" plus the action of increasing the sensitivity of the by a factor of 10 for the next MP ONLY, then resetting the %COHb.sub.o RAM to 0 and returning to normal. As an example, during this one "super sensitive" MP, 20 PPM actual COC would be measured as 200 PPM. This feature allows checking of the sensor's response to CO without the need for using highly toxic high levels of CO. Cigarette smoke contains 50 PPM CO. Therefore, cigarette smoke could be used to check the sensor and will cause the LED "present COC indicators" to respond. Refer to FIG. 4C for details of test moade "C".

In addition, during each of the three test modes the uP measures the LED currents (uP port PD2, AN2) for each LED and the lamp/buzzer currents (uP port PD3, AN3), the thermistor voltage, and the sensor's voltages.

Correct voltages at all uP A/D input ports vs operating conditions are detailed in the table below:

If at any time (during any test sequence or during normal operation) an incorrect voltage is detected at any uP A/D input port, the "FAULT" lamp lights and the audible alarm (S4) sounds as detailed above.

During any test sequence (i.e., initial power-up or during any of the three manual tests) additional depressions of the test switch are ignored.

Periodically during normal operation, the uP checks all A/D input ports and determines if the voltages present are for the current operating condition. During every OC, the automatic self-test checks for the following:

High Level Sensor Current;

Low Level Sensor Current;

LED Current;

Lamp(s)/Buzzer/Relay Current; and

Temperature compensation network voltage (for abnormal voltage or ambient temperature outside the rated operating temperature range of 0.degree. C. to +to.degree.C.).

In the event that an abnormal current or voltage is detected, the "FAULT" lamp (LA-6, FIG. 1) lights and the audible alarm (S4) sounds as detailed above.

At the time that power is initially applied to the unit (and at all subsequent "power-ups"), the %COHb.sub.o and %COHb.sub.2.5 memory locations and all other data RAM locations (FIG. 5) are reset to 0. Then the unit cycles through the initial power-up sequence detailed in FIG. 6 and self-test.

During the sensor power control timing cycle (FIG. 2), there are two short duration periods, Mini Update Period (MUP), where power is momentarily removed from the sensor in addition to the normal MP. These periods occur at 90 seconds and 120 seconds after the start of the OC. During these MUP's (MUP--Mini Update Period), the unit measures CO and alarms (HH+SH+S3) if COC is in excess of the ceiling value specified below. It should be noted that absolute sensitivity may be somewhat below normal during the MUP, therefore, requiring slightly higher COC than the ceiling value to alarm.

At any time that the unit records a COC of 1600 PPM or higher (i.e., at any MP or MUP), the alarm is immediately sounded, HH+SH+S3, regardless of whether the %COHb.sub.2.5 has dropped below the EL set point. Thus, 1600 PPM is the ceiling value.

Referring now again to FIG. 5,, micro- processor, IC-2, may be Motorola part 6805 R3, or the like, programmed and utilized as herein disclosed.

FIG. 7A-7N, is a flow diagram of the steps performed by the preferred embodiment of the invention described above.

An illustrative program which may be utilized to program the microprocessor and associated compliments is appended hereto as Appendix A.

A table showing the PIN connections of the microprocessor, IC-2, is given below:

Referring now again to FIG. 1, the alarm set, e.g., 15% is accomplished by connecting micro-processor, IC-2 ports, PA 7, PB 7, PC 7, and PD 7, through a pad to +5 V or to ground. Port, PC 7, has coupled thereto the most significant bit (MSB) while port PA 7 has coupled thereto the least significant bit (LSB).

Referring now to FIG. 8, during production testing, the unit under test will be prompted by the production test fixture (IBM P.C. XT or the like) to check certain internal voltages and transmit the resultant data back to the test fixture/P.C. for analysis. The production test sequence will generally be as follows:

a. Insert units to be tested into test fixture in test environmental chamber.

b. Close chamber and activate test sequence by appropriate key entry on the P.C.

c. The test sequence begins with a purging cycle where the chamber's atmosphere is replaced with "clean" air.

d. All units under test are powered-up and allowed to operate for 10-15 minutes in the "clean" air to stabilize.

e. After a stabilized condition is reached, the %COHb.sub.o memory of each unit is reset to 0 and then each unit is prompted to step through various operating conditions and measure and record certain voltages for later transmission back to the P.C. as follows:

f. The uP is prompted to transmit the data recorded in Step "e" to the P.C. and stored in the P.C.'s memory for later analysis.

g. At this point the Test Fixture P.C. causes a valve to open which allows Carbon Monoxide to be injected into the chamber. The COC level is monitored by a calibrated COC meter, the output of which is fed to the P.C. When the COC level reaches 50 PPM, the valve closes. The COC meter continues to monitor the COC level in the chamber and causes the P.C. to open the valve as necessary to maintain a level of 50 PPM.

h. After the chamber and the units under test have stabilized in the 50 PPM COC environment, each unit is again prompted to step through the various operating conditions noted in Step "e", and transmit this data to the P.C. as noted in Step "F".

i. Steps g and h are repeated at 200 PPM and 800 PPM COC.

j. At the conclusion of the operational test, the P.C. causes the chamber to purge and replace its 800 PPM CO atmosphere with "clean" atmosphere (COC less than 15 PPM). When the chamber atmosphere is clean, a "test cycle completed" indicator will alert the operator to open the chamber and remove the units under test.

k. After the test sequence is completed, the P.C. system prints out the test data for each unit, to indicate pass or fail condition, and, if a failure, the cause of the failure.

For purposes of digital readout of COC in PPM, micro-processor port PCO (Pin No. 9) serves as a clock out for production test and port PC2 (Pin No. 11) similarly serves as Data Out (and port PC1, Pin No. 10, a strobe input). During periods when the uP is not being prompted (via strobe input) to down load data to the test fixture, the Data and Clock output lines are to drive a P/N IC-D0024 33 segment LCD display driver and display COC in PPM up to four digits to 10 PPM resolution. For Digital Display of COC (in PPM), the uP interrupt (INT) line (FIGS. 1, 5), Pin No. 3, is held at logic 0. The PPM display is updated at each MP.

The drive ports, etc., readout %COHb.sub.2.5 in percent to 0.01% resolution with leading zero blanking above 1%. For %COHb.sub.2.5, the INT line is held at logic 1.

Switching the display back and forth between COC (PPM) and %COHb.sub.2.5 occurs at the time the INT line is switched from logic 0 to 1 or 1 to 0.

As noted above, if at any time during any test sequence (i.e., Manual, Automatic, or Production), or during normal operation, an incorrect internal voltage/current is detected at any uP A/D input port, the "FAULT" lamp lights and the audible alarm sounds (S4) as detailed above, and all other indicators (LED, lamps, buzzer, etc.) are inhibited, until such time that the fault is corrected.

During the production test, the nature of the fault is printed out at the conclusion of the test indicating when the fault occurred, at what A/D input port the fault was observed, the incorrect voltage measured, and what it should have been at that point.

Upon acknowledging that a fault has occurred, an error code corresponding to the cause of the fault is displayed per the following:

The foregoing relates to the preferred embodiments of the invention, it being understood that other variants and embodiments thereof are possible within the spirit and scope of the invention, the latter being defined by the appended claims. ##SPC1## ##SPC2##