User Guide


Thank you for using the Particle, Activity and Context Monitoring Autonomous Node (PACMAN) version 5. We hope that your experience with the unit is pleasant and that your expectations are met.

The next sections briefly describe the different components of a PACMAN and what the operation procedures are. Finally a “datafile” sample is described and in the unlikely event that the unit malfunctions, there is a helpful troubleshoot section for your support.


The unit is divided in three “logic” systems:

Each of these systems consists of a few components and have clearly defined goals.



The main air quality sensor in the PACMAN is what gives it the P. The dust concentrations (no size selection) are measured by a Sharp Dust Sensor GP2Y1010AU0F shown in the picture below (image credi tSparkFun).

This unit is able to measure aerosol concentrations up to 1mg/m3 but the logged values are not calibrated nor corrected by temperature. You can see more details here


Carbon monoxide is a toxic gas that can, in high enough concentrations, cause severe acute health effects. The sensor used by the PACMAN is based on the MQ-7 CO Gas Sensor and it was developed by Parallax Inc. The image below shows what the unit looks like and more information can be found here


Carbon dioxide concentrations are a useful measure of human respiration and can give insights into how many people are in a room, as well as the level of ventilation in the microenvironment. The sensor used is based on the MG811 CO2 gas sensor and it was developed by Parallax Inc. The image below shows the unit and more information can be found here.

External temperature

Another crucial piece of information is the temperature of the surroundings of the unit. This has implications for the concentrations measured as well as giving indications about the nature of eventual pollution sources near the PACMAN. The sensor used is Texas Instruments’ LM335 Precision Temperature Sensor which is pictured below. More detailed information can be found here.

Internal temperature

In order to be able to correctly apply any calibration factor to the sensors it is necessary to have information about the temperature closer to where the sensors are. The solution implemented takes advantage of the choice of real time clock chip (see in the Control section) as it can output “instrument” temperature as required.


The AC in PACMAN are the most complex and loosely defined sections of the unit. The original aim was to obtain information about what was going on around the instrument in relationship with emitting activities (cooking, jumping, heating, smoking, spraying, running, shuffling, etc). It was decided that a motion sensor was a suitable starting point to identify if there was movement in the area where the PACMAN was deployed. The sensor selected was Hanseelect’s PIR Sensor Module SE-10 (see here for details) pictured below.


The second sensor aimed at obtaining context information is a range finder intended to aid the motion sensor in describing the surroundings of the unit. The sensor corresponds to Maxbotix’s LV-MaxSinar-EZ4 (details here).


Mains power adapter

The PACMAN internally requires at least 300mA@7.5vdc to operate correctly. The included AC adapter is rated 500mA@9vdc which should work for all cases. However, it is possible to use the unit with any DC voltage source between 7.5v and 12v.


Real Time Clock

In order to accurately sample and timestamp the records the Chronodot high precision RTC was used. The Chronodot is based on the DS3231SN real time clock chip which as a temperature compensated internal crystal which enable it to maintain an accuracy of ~1min per year (full documentation here). The supplied battery should keep the clock working for at least 8 years.

Data logging media

All data is logged in a single 𝜇SD card which should allow uninterrupted operation for 11 months.


The central controller of the PACMAN is an Arduino Pro Mini 5v (see here for details) which at its core has an ATmega328 chip running at 16MHz. This microcontroller coordinates the data acquisition and logging of the whole system. The choice of microcontroller was mainly driven by the availability of a large community driven code base (Arduino) and the needed analog-digital I/O hardware required for this system.

Normal operation

The PACMAN has been designed with ease of use in mind aiming to reduce the source of error between the chair and the screen.

First find a suitable location for the unit keeping in mind the purpose of the deployment. For example, if the purpose is to characterise a cooking source and environment, the preferred location would be as close to the stove as possible with its “activity” sensors (motion sensor and range finder) facing towards the front of the stove. On the other hand if the aim is to characterise a general occupancy area, a good idea might be to place the PACMAN ~1.5m above the ground facing the most occupied part of the room.

To turn on the unit first plug the power adapter and then switch the PACMAN on (see image below).

The PACMAN will start its operation by initialising the memory card, creating a new file and start saving data. If the system is operational then you will be able to see a red blinking light through the memory card slot (see image below).

That’s it. The PACMAN should be up and running and it will keep like that until power is disconnected or switched off.

Datafile sample

To give you an idea of the kind and format of information gathered by the PACMAN, below you can find a sample of a datafile as it is stored in the memory card.

Count    Distance    Temperature_mV    Temperature_IN_C    PM_mV    CO2_mV    CO_mV    Movement    COstatus

1    2012    7    11    19    39    20    53.71     24.78    23.50    270.40    3166.21    175.78    1    1

2    2012    7    11    19    39    21    53.71     25.19    23.50    212.80    3287.21    180.66    1    1

3    2012    7    11    19    39    22    64.36     25.19    23.50    240.00    1691.60    185.55    1    1

4    2012    7    11    19    39    23    1484.38     25.19    23.50    289.20    17.58    190.43    1    1

5    2012    7    11    19    39    24    1068.07     25.19    23.50    203.20    39.55    190.43    1    1

6    2012    7    11    19    39    25    1499.02     25.19    23.50    272.00    437.50    195.31    0    1

7    2012    7    11    19    39    26    1513.67     25.20    23.50    248.80    1010.45    195.31    1    1

8    2012    7    11    19    39    27    1513.67     25.19    23.50    377.60    1570.61    200.20    1    1

9    2012    7    11    19    39    28    1513.67     25.19    23.50    264.80    1938.28    200.20    1    1

10    2012    7    11    19    39    29    1513.87     25.19    23.50    253.20    2202.15    204.10    1    1

The first line of the file corresponds to the HEADERS which do not include the date fields. The next lines are the actual data being recorded by the unit every second. The explanation of the fields and their units is:


If when turning on the unit the memory card adapter does not show the red blinking LED, there are two options:

  1. Replace the memory card.
  1. Turn off the unit.
  2. Remove the memory card.
  3. Erase or re-format the card.
  4. Replace the memory card.
  5. Turn on the unit.
  6. Go back to a as many times as your sanity allows.
  1. Throw your hands in despair and send an angry email to NIWA (