1   CONTENTS

2   External View & Dimensions

3   DESCRIPTION

4   LED - Sensor status

5   One or two channel sensor?

6   Measures to improve long-term stability

7   POWER REQUIREMENTS

8   Analogue outputs

9   Data Interface

10 Software

11 Gas Flow

12 CALIBRATION

2 External View & Dimensions

3 DESCRIPTION

1. Construction - the measuring method is based on the principle of light absorption in the infrared region, known as "non-dispersive infrared absorption" (NDIR). The broadband infrared radiation produced by the light source passes through a chamber filled with gas, generally methane or carbon dioxide. The gas absorbs radiation of a known wavelength and this absorption is a measure of the concentration of the gas. There is a narrow bandwidth optical filter at the end of the chamber to remove all other wavelengths before it is measured with a pyro-electric detector.
2. Function - The pyro-electric detector delivers a signal proportional to the gas concentration, which the microprocessor uses to calculate the correct volumetric concentration. The result is then converted into a linear output 0/4 - 20 mA and 0 - 10 V appropriate to the range of the sensor. The result is also available digitally at the RS232C interface. Here there is the madIR service software for communication and data retrieval.
3. Averaging of results - madIR 90 calculates a new result every second and keeps the last 60 results in the memory buffer for calculating a rolling average. So it is possible to ask the sensor at any moment for a result averaged for the last 1, 2, 3 - 60 seconds. The analogue outputs and external LCD display are updated every second. The result for the analogue outputs is calculated as a rolling average of the last 1..60 seconds. The averaging time for the analogue outputs and LCD display can be changed by the user using madIR PC software (15 seconds by default).
4. Calibration and settings memory - calibration data and all sensor settings are kept in the non-volatile EEPROM memory. Using the RS232C interface, madIR 90 PC software and standard 3-wire cable the user can change all sensor settings.
5. Programme memory  - sensor programme is written to flash memory. It can be reprogrammed by the user via RS232C. Reprogramming the Flash programme memory needs a special (non-standard) cable for communication. For reprogramming, special software from Atmel (freeware) is also necessary.

4 LED - Sensor status

There is a 3mm red LED in the sensor. The LED signalises the status of the sensor. Meaning of different LED signals:

5 One or two channel sensor?

The single channel sensors are used more widely than the two channel sensors for reasons of price. They are smaller, use less energy, which is most important in battery operation, and the detectors produce stronger signals, allowing more stable results to be achieved, especially at low concentrations. 

 The advantage of the two-channel sensor is the long-term stability.  The second channel (comparison channel) enables the sensor to compensate the following effects:

• Drift of the light source
• Temperature drift of the IR detector
• Changes to the measuring stretch (dirt etc.)

6 Measures to improve long-term stability

madur electronics has taken several steps to ensure a good long-term stability. These measures are especially effective with the one-channel sensors and have reached the point where they are only minimally exceeded by the two channel sensors. These are some of the measures used:

1. Light source - Use of special, sealed microbulbs as light source. The method of operation results in minimal wear and hence minimum change in the intensity of the source.
2. IR - Detector - the IR-detectors used are internally compensated for changes in temperature and the whole sensor is additionally temperature compensated from the microcontroller.
3. Protection against dirt - The gas chamber of the sensor is supplied with gas through a 20 µm porous plastic filter by diffusion. This avoids any entry of dirt particles into the sensor. Sensors used for forced flow are also diffusion sensors inside a housing filled with the gas. Since the filters are used only in diffusion mode there is very little tendency for them to become blocked. It is possible to change the filter at any time during the life of the sensor.
4. Protection against aggressive gases - The gas chamber of the sensor is constructed solely from chemically inert plastics and gold plated metals. For especially aggressive gases there are on special order fully gold plated gas chambers, closed off with a calcium fluoride window, so that the source and detector have no contact with the gas.

7POWER REQUIREMENTS

MadIR sensors are constructed as low-current, compact devices. The low current demand and hence minimal internal warming are a valuable improvement to the long-term stability. The power requirement is basically dependent on the voltage supply. The table below shows the four possibilities for the one-channel sensors. The power requirements do not include the power needed for the analogue outputs.

8Analogue outputs

MadIR 90 (A01) has two separate analogue outputs:

·         Current output Iout, maximum range 0..25mA

·         Voltage output Uout, maximum range (depending on power supply version): 0..5V or 0..10V

Both outputs are fully independent. User can set the minimum/maximum  voltage/current using RS232 and madir PC software as well as corresponding minimum/maximum of gas concentration separately for each analogue output.

  It is necessary to define a voltage range of more then 2 V and a current range of more than 4 mA to get an acceptable resolution of more than 2 % of the defined working range.

The analogue outputs operate from a 12-bit D/A converter. This allows a resolution of 4096 digits on 0…12 V for the voltage output and on 0…25 mA for the current output.

The current output can drive a maximum load of 500 ohm.

The voltage output requires a minimum load of 500 ohm.

The following settings are possible:

Setting A:

Setting B:

Voltage output in Schmitt mode

The voltage output can also be configured as a Schmitt output (instead of the normal linear mode). The Lo and Hi voltage as well as Lo-to-Hi and Hi-to-Lo trip points are definable by user via RS232.

For example following settings are possible for the voltage output working in Schmitt mode:

Setting C:

Setting D (for reversed logic of Schmitt output):

Settings E (for no-hysteresis Schmitt output):

9 Data Interface

MadIR Sensors are fitted with a RS232C serial interface as standard. This interface can be used for all communication between sensor and PC (programming the flash memory, calibration, sensor settings and reading out data).

The following options are also available:

·         SPI - solely for reading out data

·         I²C - can be used to drive an external 4.5 digit LCD display

10 Software

The sensors are supplied complete with the software for operation and configuration.

 The programme has the following functions:

·         Read data from sensor.

·         Processing the results:

·        Calculates averages

·        Stores the results continuously at a predetermined rate on the hard drive of the computer.

·        One-off storage of the results shown on the screen on the hard drive.

·         Zeroing of the sensor with gas.

·         Setting the analogue outputs.

·         Control: Shows the calibration curve and the parameters of the sensor.

·         Calibration of the sensor with span gas.

11 Gas Flow

For flow sensors the gas flow rate should not exceed 1.5 l/min. We recommend a flow rate of around 400 ml/min. There is no lower limit on the flow rate, but a reduced flow rate will slow the reaction of the sensor.

12 CALIBRATION

The calibration of a madIR sensor is carried out in the factory at ten separate points in the measuring range for methane or carbon dioxide. The user can view the calibration curve at any time using the software provided. It is also possible to recalibrate the sensor if necessary. This requires span gas of at least three different concentrations. It is, however, generally not necessary to recalibrate the sensor at any time. A zero correction may be sometimes required, and this can be carried out in four ways:

·     No Zeroing - No correction is made. The sensor retains the last zero point set.

·     Auto Zeroing - Every time the sensor is switched on the zero point is reset after 90 seconds.

·    Manual Zeroing - There is a zero button on the sensor. When this is pressed the sensor takes this value as the new zero.

·    Smart Zeroing - The sensor stores the averages for 15 minutes over a period of 14 days and takes the lowest value as being zero.

When zeroing the sensor be sure that the sensor is supplied with a gas of known concentration that can serve as a zero value. Fresh air is one example, assuming the area is relatively free of traffic!