PurpleAir sensors measure airborne particulate matter (PM). Particulate matter describes solid particles suspended in air; this includes dust, smoke, and other organic and inorganic particles. PurpleAir sensors use laser particle counters to count the number of particles by particle sizes 0.3, 0.5, 1, 2.5, 5, and 10 μm, and use the count data to calculate mass concentrations of PM1.0, PM2.5, and PM10.

PurpleAir sensors use laser counters to measure particulate matter in real time. A laser counter uses a fan to draw a sample of air past a laser beam. Any particles in the air will reflect some light from the laser beam onto a detection plate, like dust shimmering in a sunbeam. The reflection is measured as a pulse by the detection plate, and the length of the pulse determines the size of the particle while the number of pulses determines the particle count. The PM1.0, PM2.5, and PM10 mass concentrations are calculated from these particle counts.

PurpleAir sensors are used by a wide variety of individuals and groups from government air districts, school districts, and universities to industrial and commercial organizations to home enthusiasts and concerned citizens.

There are two major differences between PurpleAir sensors and regulatory sensors: the method of measuring particulate matter and the averaging time of the data collected.

Methods:

PurpleAir sensors use a laser particle counter to count the number of airborne particles in the air. That count is used to calculate a mass concentration, assuming an average particle density in an algorithm developed by the laser counter manufacturer, Plantower. An average density must be used because not all PM of a particular size is made of the same stuff. For instance, PM2.5 from wildfire smoke will have a different density than PM2.5 from dust blowing off a gravel pit. This means that mass concentration reported by a PurpleAir sensor can vary depending on the specific composition of PM for a given area thus making the sensors appear to "read high." So far, two different research groups have completed studies for their areas and created conversion factors specific to the composition of particulates in their air: AQ&U and LRAPA.

Federal reference sensors typically measure mass concentration of PM by drawing air through a filter and weighing the filter. This method is expensive, difficult to install, requires a specialist to maintain the sensor, and reports on an hourly scale. Because of this, many cities have a limited number of these sensors (or none at all) and it's not feasible for the general public to have their own.

Averaging Time:

PurpleAir uses the AQI breakpoints established by the US EPA to convert the mass concentration into the AQI published on the PurpleAir map. However, most regulatory groups report AQI as a 24-hour average that gets updated every hour or so. If you look at particulate matter data on a website like AirNow, a PM2.5 AQI of 150 means the average AQI in the last 24 hours was 150.

PurpleAir sensors use laser particle counters to count the number of particles in sizes from 0.3um up to 10um. These get converted into a mass concentration (ug/m3) and reported every 120 seconds. Since air quality can fluctuate greatly throughout the day, the real time PurpleAir AQI reading may appear "high" when compared to 24-hour averaged AQI data.

If you want to compare 24-hour averaged AirNow data with 24-hour averaged PurpleAir data, you can look at longer term averages on the PurpleAir map by selecting the averaging period from the options in the bottom left, or by checking the "Averages as Rings" in the map legend in the lower left-hand corner of the map.

Each ring represents an average for a time range:

Center of the circle = Real time average
1st ring = Short-term average
2nd ring = 30-minute average
3rd ring = 1-hour average
4th ring = 6-hour average
5th ring = 24-hour average
6th ring = 1-week average

The PA-II and PA-II-SD both have WiFi capabilities and come with 2 laser counters. The difference is that the PA-II-SD comes with an internal microSD card reader and real time clock that logs the data and tracks time with or without a WiFi connection. 

Currently, you must remove the microSD card from the sensor and plug it into your computer to read and analyze the data. It is not possible to upload the SD card data to the PurpleAir map. The PA-II-SD is intended for places that do not have access to WiFi or have intermittent WiFi.

The sensor housing on the PA-II and PA-II-SD is designed to protect the device from the elements while allowing air to flow freely past the two laser counters. We recommend installing in a location high enough so that water splash or snow cannot get into the underside of the sensor.

All PurpleAir sensors can be used indoors. Our most popular sensor is the PA-II, and many people use the PA-II sensor as an indoor sensor. This sensor can be mounted on the wall and contains dual laser counters, so you can compare the data from each laser counter to confirm your sensor is reporting correctly. Additionally, we have an indoor specific sensor, the PA-I-Indoor. It is designed to sit on a flat surface like a countertop or desk, and contains an LED light on top that illuminates the color of the current AQI (green, yellow, orange, red, purple, and maroon).

We do not recommend a regular cleaning schedule for PurpleAir sensors. "If it ain't broke, don't fix it."

If one of the laser counters is not reporting as expected, we do recommend that you try to clean out any debris, insects, or spiders that may have made their way inside of the laser counter. You can use a shop vacuum or vacuum hose and place it on the bottom of the sensor for 2-3 minutes, or try clearing out the laser counter with compressed air. Focus on the four vertical holes on each laser counter.

The laser counters used in PurpleAir sensors come factory calibrated. It is not possible to calibrate the laser counters post-production. If you suspect your sensor might have a laser counter misreporting you can try cleaning the laser counters with a vacuum, shop vacuum, or canned compressed air.

Before shipment, we verify that every sensor is reporting data that are consistent from sensor to sensor during tests in a smoke chamber.

It is possible that some debris, an insect, or spider got into one of the laser counters in your sensor. To clean out the laser counters, you can use a shop vacuum or vacuum hose and place it on the bottom of the sensor for 2-3 minutes. Another option is to try clearing out the laser counter with canned compressed air. Focus on the four vertical holes on each laser counter. 

We do not recommend a regular cleaning schedule of the sensors.

We do not recommend covering the base of your sensor with any material. This will impact the air flow and ability for the sensor to properly measure particulate matter.

We have heard of a number of individuals running their sensors with solar power. However, we personally have not tested PurpleAir sensors using solar power in-house. That said, it seems reasonable that solar panels between 10W and 15W should be adequate to power a battery pack to run a PurpleAir sensor. How quickly that battery pack recharges is situationally dependent and might take some experimenting on your part to fit the seasonal sun conditions. 

Power draw: The sensor uses 180mA (milliamp), about 1W. We have experimented with a 20Ah battery pack and it kept the sensor running for about four days.

Every icon on the PurpleAir map represents a public PurpleAir sensor, and the color indicates the real time PM2.5 reading on the US EPA Air Quality Index scale. The sensors with no outline are registered as outdoor sensors and the sensors with black rings are registered as indoor sensors.

When clicking a sensor icon, two informative boxes display. One shows the sensor information and averages over the last week.

The other shows a graphical chart of readings from Channels A and B over the last week.

The default display on the PurpleAir map is the United States Environmental Protection Agency (EPA) Air Quality Index (AQI) scale for PM2.5. The AQI allows comparison for different pollutants with an easy-to-visualize color scheme. The scale goes from good, green air to more hazardous, maroon air. The graphs give further information on different counts, relative weights and sizes of particles in the air.

When you click on a sensor icon on the PurpleAir map the graph will populate with 10-minute averaged data for each laser counter for that sensor. You can add more sensors to the same graph by selecting additional sensor icons.

You can display or hide individual sensor channels by clicking on the name in the legend. Visible channels are black in the legen, while hidden channels are gray.

You can select the data type displayed on the PurpleAir map using the Map Data Layer drop-down menu in the map legend.

While we’re unable to say what factor would be best for a particular scenario and cannot make health recommendations, we’re glad to provide more information to help users better understand this option on the map.

We’ll start with why the conversion factors were created. Our PurpleAir sensors measure particulate matter (PM). Particulate matter is used to describe a mixture of inhalable particles - solid and liquid - suspended in the air and is grouped by size, such as PM2.5 and PM10. (The number refers to the diameter of the particle in microns.) These airborne particles can originate from dust, smoke, and other organic and inorganic sources. However, particles from these sources can be the same size but have different densities. For instance, PM2.5 from wildfire smoke will have a different density than PM2.5 from dust blowing off a gravel pit.

Secondly, PurpleAir sensors use laser particle counters that measure the particle count, before converting that count into a mass concentration. The purpose of this is to make it comparable to data reported by regulatory sensors. To do this, the laser counters must assume an average particle density. An average density must be used because, as mentioned above, not all particulate matter of a certain size has the same density. For example, if you weighed 1000 particles of wildfire smoke and 1000 particles generated from gravel dust, the wildfire smoke would be much lighter. In the case where the predominant source of PM2.5 is from wildfire smoke that has a lighter density than the assumed density used by the sensor, the sensor data will predictably overestimate the mass concentration and read higher than the regulatory monitor. Similarly, if the predominant source of PM2.5 was a more dense material like gravel dust, the sensor data would predictably underestimate the mass concentration and read lower than the regulatory monitors (such as those used to report to AirNow.gov, Weather Underground, etc.).

According to this recent study published by researchers in Australia, the AQ&U conversion factor has proven very effective in conditions involving wood smoke as a major pollution source. The LRAPA conversion factor is very effective as well; it was developed by Oregon’s Lane Regional Air Protection Agency for their particular area during a smoke event caused by wildfires. 

Additionally, the EPA has developed their own conversion factor for wildfire and woodsmoke conditions. (The study they used to create this factor can be viewed here.) On the PurpleAir map, each sensor’s map marker can only display real-time data with the EPA conversion factor applied. On the sensor’s graph window, however, you can view the data for a variety of averaging times using the EPA conversion factor. AirNow’s Fire and Smoke Map has a similar factor applied to its data.

In the map legend there are checkboxes for "Outside Sensors", "Inside Sensors", and "Show My Sensors". Check the boxes of the sensors you want to be visible on the PurpleAir map.

Outdoor sensor icons have no outline. Indoor sensor icons have a bold outline.

If you prefer to see the sensor averages to display on the map dots, check “Averages as Rings” on the map controller.   The rings give a visual display of the average historical readings, showing the same averages you see when clicking a specific sensor.

Each ring represents an average for a time range:

Center of the circle = Real time average

1st ring = Short-term average

2nd ring = 30-minute average

3rd ring = 1-hour average

4th ring = 6-hour average

5th ring = 24-hour average

6th ring = 1-week average

At this time, you can log in to the PurpleAir website with a Google account. If you set the Owner's Email to a Google account when registering the sensor, you can log in to the PurpleAir map. This enables you to use the "Show My Sensors" checkbox in the map legend to view your PurpleAir sensors, which can be especially useful if you have registered your sensor as private.

To log in:

1. Click Login in the upper right-hand corner of the PurpleAir map
2. Select the Google account set as the Owner's Email for your PurpleAir sensor(s)
3. Enter your Google account password

To change the Owner's Email on a PurpleAir sensor that was already registered, send a request here.

If you registered your sensor as private, there are two ways to view your private sensor on the PurpleAir map:

1. You can use the "view it here" link provided in the registration confirmation email sent to the Associated Email and Owner's Email indicated on the registration form. This link can be bookmarked or shared with others that you would like to see your private sensor, too.

2. You can login to www.purpleair.com/map in the top right corner of the PurpleAir map. At this time you can only login with a Google account. This means the Owner's Email must be set to a Google account. If you need to update the Owner's Email, please contact us.

At this time, we don't have an easy way to get sensors outside of the PurpleAir network reporting to the PurpleAir map.

PurpleAir PA-II and PA-II-SD sensors contain two laser particle counters that measure the real time particulate matter and report as Channel A and Channel B. If the values being reported for the two laser particle counters drift apart, our system may "downgrade" one of the channels. The downgraded channel will no longer be used in the data average and will be grayed out in the graph's legend. You can still view the downgraded channel on the graph by clicking on the downgraded name in the graph's legend.

1. Find the sensor you wish to order on the Air Quality Sensors page here
2. Enter the quantity in the box
3. Select Add to Cart
4. Select Checkout
5. Enter your email and shipping information and select Continue to Shipping
6. Select shipping method
7. Enter preferred payment
8. Select Complete Payment to finalize your order.

An order confirmation email will be sent to the email address entered for the order. A shipping confirmation email will be sent upon shipment of your order.

You can request an invoice or quote by sending us a message on our contact form

Yes, we do! Please contact us using our contact form if you need to complete a purchase using a purchase order.

At PurpleAir, we do our best to process orders in 1-2 business days of the purchase date. However, this time can vary due to stock availability, weekends (Saturday and Sunday), or national holidays.

For current estimated order fulfillment time, send us a message here.

Within USA:
USPS Priority Mail delivers in 1-3 business days.

International Shipping:
DHL International delivers in 4-12 business days. Tracking is available for most orders.*

FedEx International delivers in 4-12 business days. Tracking is available for most orders.*

Australia Only:
USPS Priority Mail International delivers in 6-24 business days. Tracking is not available for most countries.*

*Actual number of days may vary based on origin, destination, and customs delays. Please note: International shipping costs do not include duty, taxes or brokerage fees.

Contact us here with the order number and updated shipping address as soon as possible and we'll gladly update your order.

Yes! We ship with FedEx and DHL International. Shipping rates are calculated at checkout.

Tariffs/customs duties are paid by buyer and due at delivery. 

Firmware version 4.02 and later

1. On a WiFi enable device open the list of available WiFi networks.

2. Connect to the sensor's network "PurpleAir-xxxx". The xxxx is a 2-4 dight code specific to your sensor.

   Important: If your sensor network name is "AirMonitor_xxxx" you will need to go to www.purpleair.com/configure to configure WiFi on your sensor.

3. Depending on your operating system or device you may get a pop-up window or a message to sign into the network. If you get a message to sign into the network, press it to make the pop-up window appear.
4. The pop-up window will list available WiFi networks.
5. Choose the WiFi network you would like the sensor to use, enter the password, and click save. A hidden WiFi network can be entered by selecting the option at the bottom of the list.
6. The WiFi status bar will turn green once your PurpleAir sensor is connected to WiFi and transmitting data.

Firmware version 3.00 and earlier

1. On a computer, tablet, or smartphone open the PurpleAir configuration page in Google Chrome and leave it open

2. Go to the list of available WiFi networks and look for the PurpleAir sensor network, AirMonitor_xxxx

3. Connect to the AirMonitor_xxxx network

4. Go to the PurpleAir configuration page opened in step 1 and click on the "Connect to Sensor" button

5. Select the desired WiFi network from the list

6. Enter the WiFi network password  and click Save

7. Your sensor should be online

If it was successful the sensor data should become visible in the web browser and the AirMonitor_xxxx network should no longer be visible.

Cannot see the AirMonitor_xxxx network

- Make sure the sensor is receiving power by checking that the power outlet is working and the power supply is securely connected to the power outlet and sensor then check for a dim red glow underneath the sensor housing.
- If the sensor is in range of a WiFi network it was previously configured to you will need to turn off that WiFi network or take the sensor out of range of that WiFi network to make the AirMonitor_xxxx network visible again.

Cannot connect to AirMonitor_xxxx network

- Sometimes computers, tablets, or smartphones will not connect to the AirMonitor_xxxx because this network does not reach the internet. This is a security feature on some devices. If a pop-up alert appears you may need to select trust the network or "connect anyways to the network, or repeat attempts to connect to the AirMonitor_xxxx on the same device.

WiFi network name is incomplete or not on the list after clicking "Connect to Sensor"

- Make sure your PurpleAir sensor is near the wireless router to ensure the sensor is within range of the WiFi network.
- PurpleAir sensors can only be configured to 2.4GHz networks. Any other networks will not be visible.
- If the WiFi network name contains an apostrophe (Example: John's WiFi) you will need to manually configure your sensor to the WiFi network.

Sensor will not connect to the WiFi or is not showing up on the PurpleAir map

- PurpleAir sensors must be configured to a WiFi network and registered to be visible on the PurpleAir map. Make sure your sensor was successfully registered.
- PurpleAir sensors work on WPA2 Pre-Shared Key networks. If there is any kind of secondary authentication (like clicking on a button in a web page) to connect to the WiFi network, the sensor would need to be whitelisted on the router.
- Check the password for 2.4GHz WiFi network by logging in to the router to double check your password for your 2.4GHz WiFi network.

If the desired network requires any secondary authentication (such as captive portals where you must click on a button in a web page or login with a username and password), the devices MAC address would need to be whitelisted on the router in order to use that network.

The list of fully qualified domain (FQDN) names for PurpleAir sensors are:
www.purpleair.com
api.purpleair.com
update.purpleair.com
api.thingspeak.com
sensors-ingest.wunderground.com
pool.ntp.org
www.googleapis.com

Outgoing ports used by PurpleAir sensors:
Port 80 for HTTP
Port 443 for HTTPS
DNS: 53
NTP: 123

There is not a button to reset the WiFi network connection.  To make the sensor network, AirMonitor_xxxx, visible again you will need to turn off the WiFi network or take the sensor is out of range of the WiFi network the sensor is currently configured to.  Once sensor network is broadcasting again, you will be able to connect it to a different network. Just make sure the WiFi network it is currently installed on remains out of range/offline until it is successfully configured to the new network.

The registration form can be used to register a new PurpleAir sensor or update registration details on a previously registered PurpleAir sensor. A registration confirmation email will be sent to the Associated Email and Owner's Email indicated on the registration form.

Individual registration fields are described below.

Device-ID (MAC address): A unique code for every sensor the is printed on the sensor sticker or listed in the shipping confirmation email.

Associated Email: The email address used by the buyer to complete the purchase of the sensor. This email must be used to register the sensor. Look for a green "Looking good: Device-Id and Associated Email are related..." box on the registration form. Contact us to associate a different email address with a sensor.

Installed: Indicates whether this sensor will be installed indoors or outdoors

Location Name: The name of the sensor icon that will appear on the PurpleAir map

Visibility: Choose whether you want the sensor icon to be visible on the public PurpleAir map or only visible to the sensor owner.

Map Location: Drag and drop the red marker on the map to select a location for the sensor icon to appear on the PurpleAir map. If no location is set the sensor icon will be placed at the lat/long "0/0", which is off the west coast of Africa.

Data Processor #1: By default sensor data will be sent to Weather Underground and used in WU air quality report. Use the drop down menu to select another 3rd party service to receive sensor data or select None.

Data Processor #2: Use the drop down menu to select an additional 3rd party service to receive sensor data or select None.

Owner's Name: The name of the owner of the PurpleAir sensor

Owner's Email: References the physical owner of the sensor, which can be the same or different as the Associated Email. The Owner's Email can be any email address, but must be a Google account if you want to be able to login to the PurpleAir map and use the "Show My Sensors" checkbox. Once the Owner's Email is set you must contact us if you'd like to change the Owner's Email.

SMS Alert Phone Number: The phone number of the owner of the PurpleAir sensor.

You can update registration details by resubmitting the registration form. You must enter the same Associated Email and Owner's Email that was entered when original registration. Any other details entered on the registration form will be used to update the PurpleAir map.

Moving a PurpleAir sensor to a new location:

Before moving the sensor to a new location or changing it from an indoor to outdoor sensor, contact us to request the data gathered to date be archived to the physical location and the sensor reset to gather fresh data in the new location. Once the data archive is confirmed by our team, the Registration form can then be resubmitted with the new location information entered.

PurpleAir sensor owners have the option to select 2 additional locations to send their PurpleAir sensor data. By default PurpleAir sensors will send sensor data to Weather Underground's air quality report.

None: PurpleAir sensor data will not be sent to an additional location

Weather Underground:

Check "Create new Weather Underground sensor ID"

The default setting will check the checkbox for "Create new Weather Underground sensor ID" and will send data to them through their new API that is not part of the PWS system but they are working on having the two integrated together. This API feeds the algorithms that drive their Air Quality readings and forecasts.

Un-check "Create new Weather Underground sensor ID"

When you un-check the checkbox for "Create new Weather Underground sensor ID" the KEY3 and KEY4 fields will become visible. You can add the station id (KEY3) and station key (KEY4) to display the sensor data on a PWS, which will not share the data through the API used to create the Weather Underground air quality report. Sending sensor data to a PWS essentially make the air quality data private unless someone looks specifically at that PWS.

Custom: Any HTTP server of your choice

KEY3 - Should contain the full server address including “http://” or “https://” and any querystring
KEY4 (optional) - Should contain any header to be included in the request. For example “SomeKey: SomeValue”
METHOD - The Http method being GET, POST or PUT

On the registration form, you have the option to set the visibility of your sensor to public or private. Public sensors are visible on the public PurpleAir map and the data can be accessed by anyone. Private sensors are only visible to the individual with the unique link provided in the registration confirmation email or if you log in to the PurpleAir map with a Google account when the Owner's Email is set to that Google account.

If you'd like to update the Owner's Email to a Google account please contact us.

- Find a location with easy access to a power outlet and is within range of the WiFi network

- You can mount the sensor using either cable ties or a screw

- Find a location with easy access to a power outlet

- Mount the sensor away from any foliage that would increase the chances of inscets or spiders getting inside of the laser counters. 

- The sensor is designed to lay flat on a hard surface like a counter top or desk. 

The simplest way to download the data is using the PurpleAir Sensor List available at https://www.purpleair.com/sensorlist. This page provides an easy to use interface to download data based on a date range.  

***You will need to use a Chrome browser to download any data.***

- Find and mark the sensor/s in the list you want to download - use CTRL + F to quickly search by sensor name
- Return to the top of the page and enter the desired date range, then click Download Selected

If you want to download data from a private sensor the Owner's Email for the sensor must be a Google account and you must login with that Google account at the top of the Sensor List. To update the Owner's Email on a sensor that was previously registered contact us.

Make sure you are using Google Chrome as your web browser to download data and turn off any pop-up blockers and ad blockers. 

PurpleAir sensor data is stored in the ThingSpeak data cloud. ThingSpeak stores the data in "channels" that have a limit of 8 data fields per channel. PA-II and PA-II-SD sensors contain 2 laser counters that report independently, hence Channel A and Channel B. PA-I-Indoor sensors contain 1 laser counter, so only report on one channel, Channel A. Each channel is further divided into 2 data sets, Primary and Secondary, due to the data field limits per ThingSpeak channel. Below are the header descriptions for Channel A Primary, Channel A Secondary, Channel B Primary, and Channel B Secondary data files.

Channel A Primary:

created_at: Timestamp of data posting to ThingSpeak (created by ThingSpeak)
entry_id: row number relative to all data channel A primary data (created by ThingSpeak)
PM1.0_CF_ATM_ug/m3: mass concentration calculated from count data for particle sizes ~0.3um to ~1.0um for “atmospheric” particles (From Plantower 5003/1003)
PM2.5_CF_ATM_ug/m3: mass concentration calculated from count data for particle sizes ~0.3um to ~2.5um for “atmospheric” particles (From Plantower 5003/1003)
PM10.0_CF_ATM_ug/m3: mass concentration calculated from count data for particle sizes ~0.3um to ~10um for “atmospheric” particles (From Plantower 5003/1003)
UptimeMinutes: the amount time the unit CPU has been running since it was powered up (From ESP8266 WiFi chip)
ADC: Voltage (From ESP8266 WiFi chip)
Tempertautre_F: Temperature inside of the sensor housing in Fahrenheit. On average, this is 8F higher than ambient conditions. (From BME280)
Humidity_%: Relative humidity inside of the sensor housing as a percentage. On average, this is 4% lower than ambient conditions (From BME280)
PM2.5_CF_1_ug/m3: mass concentration calculated from count data for particle sizes ~0.3um to ~2.5um for “standard” particles (From Plantower 5003/1003)

Channel A Secondary:

created_at: Timestamp of data posting to ThingSpeak (created by ThingSpeak)
entry_id: row number relative to all data channel A primary data (created by ThingSpeak)
0.3um/dl: aerodynamic diameter of >0.3 micrometer particle counts per deciliter of air
0.5um/dl: aerodynamic diameter of >0.5 micrometer particle counts per deciliter of air (From Plantower 5003/1003)
1.0um/dl: aerodynamic diameter of >1.0 micrometer particle counts per deciliter of air (From Plantower 5003/1003)
2.5um/dl: aerodynamic diameter of >2.5 micrometer particle counts per deciliter of air (From Plantower 5003/1003)
5.0um/dl: aerodynamic diameter of >5.0 micrometer particle counts per deciliter of air  (From Plantower 5003/1003)
10.0um/dl: aerodynamic diameter of >10.0 micrometer particle counts per deciliter of air (From Plantower 5003/1003)
PM1.0_CF_1_ug/m3: mass concentration calculated from count data for particle sizes 0.3um to ~1.0um for “standard” particles (From Plantower 5003/1003)
PM10.0_CF_1_ug/m3: mass concentration calculated from count data for particle sizes ~0.3um to ~10um for “standard” particles (From Plantower 5003/1003)

Channel B Primary:

created_at: Timestamp of data posting to ThingSpeak (created by ThingSpeak)
entry_id: row number relative to all data channel A primary data (created by ThingSpeak)
PM1.0_CF_ATM_ug/m3: mass concentration calculated from count data for particle sizes ~0.3um to ~1.0um for “atmospheric” particles (From Plantower 5003/1003)
PM2.5_CF_ATM_ug/m3: mass concentration calculated from count data for particle sizes ~0.3um to ~2.5um for “atmospheric” particles (From Plantower 5003/1003)
PM10.0_CF_ATM_ug/m3: mass concentration calculated from count data for particle sizes ~0.3um to ~10um for “atmospheric” particles (From Plantower 5003/1003)
UptimeMinutes: the amount time the unit CPU has been running since it was powered up
RSSI_dbm: WiFi strength (From ESP8266 WiFi chip)
Pressure_hpa: Pressure inside of the sensor housing in hectopascal. (From BME280)
Blank: No Data
PM2.5_CF_1_ug/m3: mass concentration calculated from count data for particle sizes ~0.3um to ~2.5um for “standard” particles (From Plantower 5003/1003)

Channel B Secondary:

created_at: Timestamp of data posting to ThingSpeak (created by ThingSpeak)
entry_id: row number relative to all data channel A primary data (created by ThingSpeak)
0.3um/dl: aerodynamic diameter of >0.3 micrometer particle counts per deciliter of air
0.5um/dl: aerodynamic diameter of >0.5 micrometer particle counts per deciliter of air (From Plantower 5003/1003)
1.0um/dl: aerodynamic diameter of >1.0 micrometer particle counts per deciliter of air (From Plantower 5003/1003)
2.5um/dl: aerodynamic diameter of >2.5 micrometer particle counts per deciliter of air (From Plantower 5003/1003)
5.0um/dl: aerodynamic diameter of >5.0 micrometer particle counts per deciliter of air (From Plantower 5003/1003)
10.0um/dl: aerodynamic diameter of >10.0 micrometer particle counts per deciliter of air (From Plantower 5003/1003)
PM1.0_CF_1_ug/m3: mass concentration calculated from count data for particle sizes 0.3um to ~1.0um for “standard” particles (From Plantower 5003/1003)
PM10.0_CF_1_ug/m3: mass concentration calculated from count data for particle sizes ~0.3um to ~10um for “standard” particles (From Plantower 5003/1003)

PurpleAir sensor data is stored in the ThingSpeak data cloud. ThingSpeak stores the data in "channels" that have a limit of 8 data fields per channel. PA-II and PA-II-SD sensors contain 2 laser counters that report independently, hence Channel A and Channel B. PA-I-Indoor sensors contain 1 laser counter, so only report one channel, Channel A. Each channel is further divided into 2 data sets, Primary and Secondary, due to the data field limits per ThingSpeak channel. Below are the header descriptions for Channel A Primary, Channel A Secondary, Channel B Primary, and Channel B Secondary data files.

The CF_ATM and CF_1 values are calculated from the particle count data with a proprietary algorithm developed by the PMS5003 laser counter manufacturer, PlanTower. The specifics of the calculation are not available to the public (or us for that matter). However, to convert the particle count data (um/dl) to a mass concentration (ug/m3) they must use an average particle density. They do provide 2 different mass concentration conversion options; CF_1 uses the "average particle density" for indoor particulate matter and CF_ATM uses the "average particle density" for outdoor particulate matter. Depending on the density of the particles you are measuring the sensor could appear to read "high" or "low". Some groups have developed conversion factors to convert the data from the sensor to match the unique average particle density within their airshed. 

PurpleAir provides access to our real time data in a JSON format. This format allows developers to access current data for all sensors or a subset of sensors.

There are two ways to access the JSON data:

All sensors: https://www.purpleair.com/json
One entry: https://www.purpleair.com/json?show=<id> where ID is the “ID” of the sensor you want (in the case of dual laser where ParentID is “null”).

If you want to access data from a private sensor the Owner's Email for the sensor must be a Google account and you must login with that Google account on https://www.purpleair.com before navigating to https://www.purpleair.com/json - OR - you will need to specify the sensor's ThingSpeak READ key using the key parameter in the querystring. To update the Owner's Email on a sensor that was previously registered contact us.

NOTE: Please limit multiple threaded applications to a few threads at any time to reduce load on purpleair.com or thingspeak.com servers. If you do not limit the amount of threads, you may be blocked on the firewall.

The following is a list of the fields and a description of their values contained in the JSON data:

"ID":1234, // PurpleAir sensor ID
"ParentID":null, // The PurpleAir sensor ID of the "parent" entry in the case of Channel B
"THINGSPEAK_PRIMARY_ID":"1234", // The Thingspeak channel ID for primary data of this sensor
"THINGSPEAK_PRIMARY_ID_READ_KEY":"XXXX", // The Thingspeak read key for primary data of this sensor
"Label":"name", // The "name" that appears on the map for this sensor
"Lat":null, // Latitude position info
"Lon":null, // Longitude position info
"PM2_5Value":"1.07", // Current PM2.5 value (based on the
"State":null,  // Unused variable
"Type":"TYPE",  // Sensor type (PMS5003, PMS1003, BME280 etc)
"Hidden":"true", // Hide from public view on map: true/false
"Flag":null, // Data flagged for unusually high readings
"DEVICE_BRIGHTNESS":"1", // LED brightness (if hardware is present)
"isOwner":1, // Currently logged in user is the sensor owner
"A_H":null, // true if the sensor output has been downgraded or marked for attention due to suspected hardware issues
"temp_f":"xx",  // Current temperature in F
"humidity":"xx", // Current humidity in %
"pressure":"xx", // Current pressure in Millibars
"AGE":29831, // Sensor data age (when data was last received) in minutes
"THINGSPEAK_SECONDARY_ID":"1234", // The Thingspeak channel ID for secondary data of this sensor
"THINGSPEAK_SECONDARY_ID_READ_KEY":"XXXX", // The Thingspeak read key for secondary data of this sensor
"LastSeen":1490309930, // Last seen data time stamp in UTC
"Version":"2.47c", // Current version of sensor firmware
"LastUpdateCheck":1490308331, // Last update checked at time stamp in UTC
"Uptime":"5210", // Sensor uptime in seconds
"RSSI":"-68", // Sensor's WiFi signal strength in dBm

"Stats": // Statistics for PM2.5

"{
\"v\":1.07, // Real time or current PM2.5 Value
\"v1\":1.3988595758168765, // Short term (10 minute average)
\"v2\":10.938131480857114, // 30 minute average
\"v3\":15.028685608345926, // 1 hour average
\"v4\":6.290537580116773, // 6 hour average
\"v5\":1.8393146177050788, // 24 hour average
\"v6\":0.27522764912064507, // One week average
\"pm\":1.07, // Real time or current PM2.5 Value
\"lastModified\":1490309930933, // Last modified time stamp for calculated average statistics
\"timeSinceModified\":69290 // Time between last two readings in milliseconds
}"
}

Another way to get the data is to use ThingSpeak.com and to do this, you will need the API Key and channelID. These two pieces of data are available from PurpleAir’s JSON.

More info on using ThingSpeak’s API is here: https://www.mathworks.com/help/thingspeak/rest-api.html

NOTE: Please limit multiple threaded applications to a few threads at any time to reduce load on purpleair.com or thingspeak.com servers. If you do not limit the amount of threads, you may be blocked on the firewall.

We use tweezers to access the SD cards.  They are not meant to be accessed frequently so they are a bit harder to get to.  Also when you replace the SD card or powering on when not connected to WiFi you will want to make sure there is a blue LED light that flickers next to the SD card from time to time (the logging period). If it flickers three times, the SD card itself was not loaded properly and not logging data. It is helpful to have them connected to WiFi to make sure they are logging correctly, but realize this may not be available.  In order to get any future firmwares update you will need to be able to connect the sensor to a network as well.

The SD Card version of the PA-II (PA-II-SD) has a built in real time clock and OPENLOG serial logger. The SD card contains data in CSV format with the following headers:

UTCDateTime, mac_address, firmware_ver, hardware, current_temp_f, current_humidity, current_dewpoint_f, pressure, adc, mem, rssi, uptime, pm1_0_atm, pm2_5_atm, pm10_0_atm, pm1_0_cf_1, pm2_5_cf_1, pm10_0_cf_1, p_0_3_um, p_0_5_um, p_1_0_um, p_2_5_um, p_5_0_um, p_10_0_um, pm1_0_atm_b, pm2_5_atm_b, pm10_0_atm_b, pm1_0_cf_1_b, pm2_5_cf_1_b, pm10_0_cf_1_b, p_0_3_um_b, p_0_5_um_b, p_1_0_um_b, p_2_5_um_b, p_5_0_um_b, p_10_0_um_b

Firmware version 4.02 - Header Error

PA-II-SD sensors running firmware version 4.02 contain a header error on the SD card. To fix, open the csv file and find the header "hardwareversion". You will need to delete this header and move all following header columns to the left one. The "gas" header in the last column should now line up with the last column of data.

Header Descriptions

UTCDateTime: The Date and time derived from the Real Time Clock and synced with NTP where possible (in UTC).

Mac_address: The MAC address of the WiFi module on the sensor (used as an ID for the unit).

Firmware_ver: Firmware version of the control board.

Hardware: Hardware the control board has detected.

current_temp_f: Current temperature in F.

Current_humidity: Current Humidity in %.

Current_dewpoint_f: Calculated dew point in F.

Pressure: Current pressure in millibars.

Adc: The voltage reading on the analog input of the control board.

Mem: Free HEAP memory on the control board.

Rssi: WiFi signal strength in dBm

Uptime: Firmware uptime in seconds.

Pm1_0_atm: Channel A ATM PM1.0 particulate mass in ug/m3

Pm2_5_atm: Channel A ATM PM2.5 particulate mass in ug/m3

Pm10_0_atm: Channel A ATM PM10.0 particulate mass in ug/m3

Pm1_0_cf_1: Channel A CF=1 PM1.0 particulate mass in ug/m3

Pm2_5_cf_1: Channel A CF=1 PM2.5 particulate mass in ug/m3

Pm10_0_cf_1: Channel A CF=1 PM10.0 particulate mass in ug/m3

P_0_3_um: Channel A 0.3 micrometer particle counts per deciliter of air

P_0_5_um: Channel A 0.5 micrometer particle counts per deciliter of air

P_1_0_um: Channel A 1.0 micrometer particle counts per deciliter of air

P_2_5_um: Channel A 2.5 micrometer particle counts per deciliter of air

P_5_0_um: Channel A 5.0 micrometer particle counts per deciliter of air

P_10_0_um: Channel A 10.0 micrometer particle counts per deciliter of air

Pm1_0_atm_b: Channel B ATM PM1.0 particulate mass in ug/m3.

Pm2_5_atm_b: Channel B ATM PM2.5 particulate mass in ug/m3

Pm10_0_atm_b: Channel B ATM PM10.0 particulate mass in ug/m3

Pm1_0_cf_1_b: Channel B CF=1 PM1.0 particulate mass in ug/m3

Pm2_5_cf_1_b: Channel B CF=1 PM2.5 particulate mass in ug/m3

Pm10_0_cf_1_b: Channel B CF=1 PM10.0 particulate mass in ug/m3

P_0_3_um_b: Channel B 0.3 micrometer particle counts per deciliter of air

P_0_5_um_b: Channel B 0.5 micrometer particle counts per deciliter of air

P_1_0_um_b: Channel B 1.0 micrometer particle counts per deciliter of air

P_2_5_um_b: Channel B 2.5 micrometer particle counts per deciliter of air

P_5_0_um_b: Channel B 5.0 micrometer particle counts per deciliter of air

P_10_0_um_b: Channel B 10.0 micrometer particle counts per deciliter of air

PA-II NOTES:

Each sensor contains two identical laser counters, hence channel A and B. If these two channels do not agree to some extent then there is something wrong with one or both channels.

Plantower PMS sensor notes:

ATM is "atmospheric", meant to be used for outdoor applications

CF=1 is meant to be used for indoor or controlled environment applications

However, PurpleAir uses CF=1 values on the map. This value is lower than the ATM value in higher measured concentrations.

At this time, when the sensor is out of network range and recording to the SD, the data is solely stored on the SD card.  When you come back into network range, the data that was logged when offline is not synced up to the cloud. The SD card would have to be pulled to extract that data. 

It is possible to swap out the SD card that comes with the sensor.  The card must be formatted for FAT32 - 

Mac Tutorial: https://www.youtube.com/watch?v=DA48hcfZv0g   
Windows 10 Tutorial: https://www.youtube.com/watch?v=Vd3OOwOzDp8