A differential temperature source methodology for the design of a single beam NDIR gas sensor is advanced. This methodology uses a low and a high amplitude voltage cycle to drive a closely approximated Blackbody source for generating at different times two distinct detector outputs obtained from the same detector equipped the same narrow band pass filter but strategically designed for the detection of only a particular portion of the absorption band for the gas of interest. The ratio of the high amplitude cycle detector output over the low amplitude cycle detector output is used to calibrate such an NDIR gas sensor after it is normalized by a similar ratio when there is no target gas present in the sample chamber.

FIELD OF THE INVENTION
The present invention is in the field of non-dispersive infrared (NDIR) gas sensors of a type used to measure the concentrations of unwanted or combustible gases so that an alarm can be enunciated when their concentration approaches or exceeds a harmful or dangerous level. More specifically, the present invention relates to a comparatively small, simple and low cost apparatus having no moving parts and capable of measuring the concentration of most common gases in the atmosphere.
BACKGROUND OF THE INVENTION
The NDIR technique utilizing the characteristic absorption bands of gases in the infrared has been widely used for decades in the gas analyzer industry for the detection of these gases. Such gas analyzers utilize the principle that various gases exhibit substantial absorption at specific wavelengths in the infrared radiation spectrum. The term “non-dispersive” as used herein refers to the apparatus used, typically a narrow-band optical or infrared transmission filter instead of a dispersive element such as a prism or diffraction grating, for isolating for the purpose of measurement the radiation in a particular wavelength band that coincides with a strong absorption band of a gas to be measured. The NDIR technique has long been considered as one of the best methods for gas measurement. In addition to being highly specific, NDIR gas sensors are also very sensitive, relatively stable and easy to operate and maintain. In contrast to NDIR gas sensors, the majority of other types of gas sensors today are in principle interactive. Interactive gas sensors are less reliable, short-lived and generally nonspecific, and in some cases can be poisoned or saturated into a nonfunctional or irrecoverable state.
Despite the fact that interactive gas sensors are mostly unreliable and that the NDIR gas measurement technique is one of the best there is NDIR gas sensors still have not enjoyed widespread high volume usage to date. There are three main reasons for this. First, there are several applications in existence today that would require a very large number of gas sensors typically running into millions of units per annum. One very prominent example of these is the long overdue smart fire detector that needs the assistance of gas sensors for detecting specific effluent gases from a fire such as Carbon Monoxide and Carbon Dioxide. Detection of these effluent gases when a fire first breaks out would greatly help the conventional smoke detector not only to eliminate most of its nuisance false alarms but also to detect fires like smoldering or even fast-moving ones in a much shorter time. But gas sensors to be deployed in such an application must be extraordinarily reliable and just about all gas sensors ever designed and manufactured to date, irrespective of what technology is being employed, invariably have significant output drifts over time.
Another high volume usage example in the millions of units per annum range is the so-called “harmful or dangerous gas level fuse.” Many gas heaters, inclusive of kerosene heaters and gas water heaters, are required by law to have a safety device equipped with the heater in order to warn users of poor ventilation and hence low oxygen levels in the heater's enclosed space. Either an NDIR high Carbon Dioxide fuse (for detecting CO2 levels>5,000 ppm) or an NDIR high Hydrocarbon fuse (for detecting lower explosion limit [LEL]>2.5%) would be a much better candidate for use than an expensive, short lifespan and unreliable electrochemical oxygen sensor. However, such NDIR gas level fuses must once again be extraordinarily reliable and should not require frequent re-calibration in order to assure their output accuracy over time.
The second reason why today's NDIR gas sensors do not enjoy widespread high volume usage has to do with their size. They are typically several inches in length, width and height dimensions. Like in the application cases mentioned above with regard to their potential use as an augmented smart smoke detector or as a “harmful or dangerous gas level fuse,” their sizes are generally considered to be too big. Even if they have overcome their output drift reliability problem, their physical dimensions remain a significant impediment to their utilization and must be drastically reduced to gain usefulness. Although the size of NDIR gas sensors has indeed been greatly reduced to just a couple of inches in all three dimensions during the past couple of years, they still have to be further reduced, preferably to just thumb-sized scales, in order to remove their size hindrance in a number of high volume usage applications.
Recently the present author in issued U.S. Pat. No. 8,003,944 (“Saturation filtering NDIR gas sensing methodology”), Aug. 23, 2011, U.S. Pat. No. 8,143,581 (“Absorption biased NDIR gas sensing methodology”), Mar. 27, 2012 and U.S. Pat. No. 8,217,355 (“Self-commissioning NDIR gas sensors”), Jul. 10, 2012 disclosed teachings which are capable of eliminating substantially all NDIR gas sensor output drifts over time. These methodologies represent for the first time an NDIR gas sensor that can now be designed and manufactured to overcome this performance deficiency. Furthermore, these methodologies, when appropriately implemented, are capable of reducing the size of NDIR gas sensors to thumb-sized dimensions thereby removing for the first time any size hindrances affronting them in many high volume usage applications.
The third reason why NDIR gas sensors do not enjoy widespread high volume usages is their unit sensor cost which has been too high for almost all such applications. Recalling about four decades ago, an NDIR medical CO2 sensor was sold for more than $10,000.00 each. By the early 1990's, the unit price for an NDIR CO2 sensor dropped to less than $500.00. Today the unit price of an NDIR CO2 sensor goes for about $200.00, reflecting the fact that the unit production cost for such a sensor has to be just around $50.00 or less. But even this unit production cost today is considered to be too high for many applications including the two examples mentioned above, namely the augmented smart smoke detector and the “harmful or dangerous gas level fuse”. For both of these applications, the unit production cost for an NDIR gas sensor has to be well under $10.00.
Since the first two out of three main reasons why NDIR gas sensors do not enjoy widespread high volume usages today appear to be under control for elimination as noted above, the object of the present invention is to reduce the unit production cost for NDIR gas sensors to an absolute minimum possibly just a few dollars. This unit production cost is likely to be the ultimate bottom price for future non-interactive NDIR gas sensors. As it turns out, when comparing the difficulty to overcome this third reason as versus overcoming the first two, it is indeed the toughest.
The current invention reduces unit cost by reducing component cost while at the same time rendering the implemented NDIR gas sensor with significantly reduced output drifts over time and also with thumb-sized dimensions. As a result, the current invention not only eliminates the first two reasons why NDIR gas sensors have not enjoyed to date widespread usages as discussed above, but also allows an NDIR gas sensor to be designed and manufactured for the first time with volume unit production cost well under $10.00.
ISweek（http://www.isweek.com/）- Industry sourcing & Wholesale industrial products