Experiences on How to Work With Sensors (specifically PASCO Environmental Sensors)

As an intern at SUNY ESF - Office of Sustainable Facilities and Operations, my duty is to help the Energy Manager maintain the internal environment of the campus (Temperature, Humidity, Lightning, CO2 levels,...). Because all of the factors I mentioned above are parts of HVAC systems. We don't want to use too much energy on the HVAC systems.

One way to effectively maintain the internal environment is to set up the sensor system. My boss assigned me to figure out how to set up and utilize the PASCO Sensors.

This Instructable is like an experienced guidebook on how to work with sensors in general and some potential possible ideas for these specific PASCO sensors.

For my circumstances, I was assigned to monitor the temperature, humidity, wind speed and CO2 levels in SUNY ESF buildings so at a minimum I only need:

• PASCO weather sensor (in the 1st picture)
• PASCO CO2 sensor (in the 1st picture)

Also to assist with the use and setup of the sensor, I also need documents from the manufacturer which are:

• Sensor user guide (in the 2nd picture): The reason I need this is because every product specification (like battery, number of data points stored, how to set up, and data format) is included in the sensor user guide.

As you can see in pic 3 and 4, this is the structure of PASCO weather sensors I obtained from the PASCO user guide. This PASCO weather sensor is a big "container" with several small sensors inside:

• Temperature, Pressure, and Humidity sensor are coded as position 2
• Light sensor are coded as position 11

From this structure, I think the way the PASCO sensor works is: The whole PASCO sensor will have a circuit that has different small sensors connected to it. And based on number and the types of parameters we want to collect, it can disable or turnoff the current to particular sensors except the one we want to collect data from.

Based on the experience I had when I worked with Arduino sensors in the past and PASCO sensors currently, I realize every sensor will have this same issues:

• Product specifications
• Setup and utilize the sensors
• Sharing data

For every sensor you work with, the first thing first is always to try to understand your product specifications. In my point of view, every sensor's specifications will have these 4 aspects:

• Battery: Including the usage time, how to charge the battery and how to tell if the battery is run out
• Data storage: How many data points can this sensor store for each time of measurement and what is their data types
• Measurable parameters: What parameters can this sensor measure, what are the units and is it possible to change the units of measurement
• Measurement accuracy and precision: This includes the accuracy and precision of sensor measurement and if possible finding out about the Limits of detection (LOD) and Limits of quantification (LOQ).

We will discuss all of those things in further detail below.

The battery is definitely the most important aspect of every sensor. Because based on the maximum usage time of the battery, it is easy to determine the suitable data collection time.

• We must know that for environmental parameters such as Temperature, CO2 levels, and Humidity,... the values usually fluctuate over the course of the day. Thus, usually, you will see a lot of people set up the sensors for a day, or even a week just to have the full picture of the data and avoid any biased conclusions. Because of that when you buy environmental sensors, you do want your sensors to have the ability to work for at least a day.
• Secondly, for some really advanced sensors like PASCO in this case (Arduino sensors do not have this feature), the sensor can both work in remote and non-remote mode. Non-remote mode is when you set up the sensor, collect the data and it displays on the software's user interface for you to see in real-time (like pic 1). Remote mode is when you set up the sensor, and let it sit at 1 place to collect data by itself for sometimes and after that you download the data the sensor has collected over that time into your software (like pic 2). Usually, the remote mode will work longer than the non-remote mode because in non-remote mode, not only do you have to spend some portion of your battery to collect the data, part of your battery will be used to transfer and display the data on the software's user interface for the user to see.
• Thirdly, you must know the relationship between the number of measurements and Battery life. As you can see in pic 3, this PASCO sensor can collect a lot of measurable parameters. However, the more parameters I want to collect, the more the sensors have to work. Because in this sensor, it has small separate sensors just to collect temperature, and humidity individually. The more parameters we collect, the more small sensors have to be turned on in these big PASCO sensors, increasing battery consumption (as you can recall from the supplies step I provided the detailed structure of the PASCO sensor).
• Finally, you must know when the sensor's battery is running low. Take a look at the 1st picture in this step, you see on the PASCO sensor, there is a Battery sign with a light next to it. So based on the user guide, it said: "The battery light will blink yellow while charging, blink green when full battery and blink red when running low." This is actually an application of Ergonomics to Product design.

In my case, this PASCO sensor can work for 44 hours in non-remote mode and for around 11 days with remote work. However, this only applies for all measurements enabled, meaning you measure temperature, humidity, wind speed, UV index, light intensity, GPS location,... (which is really unnecessary).

Ergonomics is the scientific discipline concerned with the understanding of interactions among humans and elements of a machine or system. Basically, it will focus on how humans interact with machines, from there designing better machines to enhance the experience of the users. There are several aspects of Ergonomics:

• Workplace Ergonomics: Workplace ergonomics is really about building a better workplace. For example in the 1st pic, you see 1 typical type of Workplace Ergonomics. By designing the setup of the computers, desks and chairs in a way that maximize workers comfort, the worker can work for longer hours without having back issues like before.

• Physical Ergonomics: This type is concerned with human anatomical and anthropometric as they relate to physical activity. Products with this type of Ergonomics will design in a way to reduce the harmful effects they may bring to users. For ex, in the 2nd pic, you see an Ergonomic mouse will have a shape that better fits human hands and therefore reduces the pain from working for long hours.

• Cognitive Ergonomics: This type is concerned with mental processes and how humans perceive information. Usually, systems or products with this type of Ergonomics will design the user interface, and buttons in a way that any user can immediately know the usage of those without much clarification or instructions. For example. take a look at the 3rd pic, if you see those types of buttons or knobs in real life, you will know how to use those without instructions like, "First, rotate the knob." or "First, slide the slider."...

Based on the thing we discussed above, the type of Ergonomics to be used in this sensor would be Cognitive Ergonomics. Because all the buttons on the software user interface and the LEDs blinking things conveyed a lot of information to us without the need to have numerous labelling or instructions on how to know or recognize them.

Some good books about Ergonomics to read for Mechanical Engineering and User Interface design:

• Ergonomics in Design Methods & Techniques (pic 3)
• Ergonomics for Beginners (pic 4)

The second most important factor after the battery is data storage. We must know that in the sensor, the units of storage people use is the Number of data points, not MB (Megabyte), GB (Gigabyte), or b (byte). Knowing the number of data points a particular sensor can collect and store is really important in determining the Data collection frequency.

• Frequency: Describe the number of data points collected in 1 second.
• Periodic: Describe the time it takes (in second) to collect 1 data point.

So for our PASCO sensor, it can collect 35000 data points. For ex, if you set the PASCO sensor for 1 week, your calculation will be like below:

• 7 days / week, 24 hours / day => 24 x 7 = 168 hours / week.
• 168 hours / week, 60 minutes / hour => 168 x 60 = 10080 minutes / week.
• 10080 minutes / week, 60 seconds / minute => 10080 x 60 = 604800 seconds / week.
• Because our sensor can only collect 35000 data points so take the number of seconds per week / number of data points to get the periodic cycle = 604800 / 35000 = 17.28 seconds.

So it means to collect data enough for 1 week, you should collect data for every 17.28 seconds or any number larger than 17.28 seconds.

Pasco sensor has 2 types of measurements:

• Primary measurements: This is the type of measurement that has the values directly provided to us without any calculations from previous measurements.

Some PASCO sensors' primary measurements are: Wind Speed, Temperature, Relative Humidity, and Barometric Pressure.

• Secondary measurements: Each secondary measurement is a calculation based on one or more of the primary measurements.

Some PASCO sensors' secondary measurements are: Absolute Humidity (calculated from Relative humidity), Dew Point, and Wind Chill,... An example of Absolute Humidity was provided above to show how Secondary measurements can be calculated from Primary measurements.

Speaking about measurements, we cannot forget to mention units of measurement. In different countries and regions, people use different units for each type of parameter. For ex, in the US people use Fahrenheit for Temperature, while in Asia, it is Celsius. To avoid any confusion in displaying and understanding sensor data, it is best to figure out how to change the units of measurement.

• As you can see in pic 2 of this step, the software user interface displays the data both in Graphs and Table format. So we should see how to change the units of measurement in Graphs and Table.

Graphs:

First, take a look at the graph, we see that on the y-axis of the graph, there is 1 button that is kind of rises up a little bit (kind of inviting us to click on it, once again, this is the concept of Ergonomics). That button also has units of measurement on it. So I suspect this button is used for changing units of measurement. As we see, it is correct, when I clicked on it, I can change the units of measurement displayed on my graph (pic 4).

Tables:

For the table, things are really similar. If you take a look at the table (pic 5), there is 1 similar button on the top of each column that is kind of rises up a little bit (still kind of inviting us to click on it, and once again, this is the concept of Ergonomics). So when I clicked on that button, I was able to change the units of measurement of that column (pic 6).

First, we see in these pictures (pic 1 and 2) that for every value that the sensors are collected, there always ranges. I really want to use this opportunity to introduce the concept Limit of Quantification.

• Limit of Quantification is the lowest value of a signal that can be quantified with acceptable precision and accuracy. Every sensor or method of measurement will have a limit of quantification. As you can see in pic 3 of this step, you will want your measurement to have values that fall inside the green region to have acceptable precision and accuracy.
• For example in our case, we see that the Barometric Pressure range is around 225 to 825 mmHg. So if you measure with your sensor and it gives back a value of 200 mmHg, you will not use that value because it is well below the Limit of Quantification, therefore, the accuracy of the value is not true anymore.

Also, I do want to introduce the concept of Accuracy and Precision.

• Accuracy: How close to a measured value is the true value.
• Precision: The reproducibility of a result.

Take a look at pic 1 and 2 again, for ex Barometric accuracy is 0.1 mmHg. This means that suppose I know exactly the Barometric value of a certain environment is 250.8 mmHg. When I use the PASCO sensor to measure that environment, the value will either be 250.9 mmHg or 250.7 mmHg. It means that there will be 0.1 variations around the true value of the measurement. For precision in this case they didn't provide but you could do it by yourself. Just simply measure something 5 or 7 times and record those values and calculation variations and you could have a conclusion of whether there is the reproducibility of the result.

To turn on PASCO sensor, press the ON/OFF button until the 3 status LEDs shine momentarily. The purpose of the 3 LEDs is actually from the concept of Ergonomics. It allows the user to know the status of the sensor.

Take a look at the 1st pic in this step of the PASCO Weather sensor, you see there are 3 LEDs for Bluetooth, Battery and GPS. So based on our common sense, when we turn this sensor on, we expect to see green light blinking at these 3 LEDs. If they are blinking red, it means something wrong is happening with them.

For Arduino sensors, they don't have the software interface but for this advanced PASCO sensor, we have the software interface (like in pic 2). As you can see there are 3 modes you can choose from:

• Manual entry: This is when you have the dataset in advance and just want to display it on the software.
• Sensor data: This is when you connect your sensor with the software user interface and display data to the user in real-time.
• Remote logging: This is when you connect your sensor with the software but your sensor will collect data independently and you can only see the data after you finish collecting it and download it back to the software.

For options 2 and 3, in this PASCO sensor, everything will be the same. You will have the data and you have the option to display the data in your way of preference:

• By tables and graphs
• By figures and graphs
• By graphs only
• By figures only

As you can see throughout this instructable, my favourite type of display was by Tables and Graphs because I feel that I can have more measurements being displayed and also it was easier for me to notice any trends in my data. This is really helpful because me and my boss had to make a quick conclusion on whether we should turn on the heat of the room after witnessing our data or we should keep everything the same (we don't have hours or days to analyze it).

For every sensor, because the number of data points they collect is large so it is really impractical to scroll through or look at it on the software user interface. So it is best to have the ability to share and view it. The 2 most preferred file types to be shared are either xlsx file or csv file (they are both Excel files).

For the PASCO sensor, when I click on the shared options, I see these 3 options:

• Share Current Page: This is basically share the current screenshot image of the data now, as you can see in pic 2. It is not so useful and not the kind of file type we want (pic 2)
• Share SPARKLab: Based on the name, I suspect it only shares the app data, and we can only open this file type if we either have the SPARKVUE or Capstone software from PASCO. And I was right, this only meant to open by PASCO software. It is somehow useful but not the kind of file type we want.
• Share Data: The data file to be shared here is .kml file. I actually look up on way to convert it to Excel file but not possible.

However, when I play around with the software and clicked on the 3 line icon on the top left corner, I see the option called Export Data. From there I was able to export the data into csv file. Actually the csv file will look like in pic 5.

• As you can see it include the date and time when the data was collected and all the values from all of the sensors I want to collected in PASCO Weather sensor.