PurpleAir
purpleair

The [PurpleAir](https://www.purpleair.com/) is one of many low cost devices that use laser optical particle counters to estimate particulate matter mass concentrations in air for PM2.5 and PM10. Other similar devices include the [Dustduino](https://publiclab.org/wiki/dustduino), [Speck](https://publiclab.org/notes/chrisbartley/04-15-2014/speck-particle-monitor), [Dylos](http://www.dylosproducts.com/?gclid=Cj0KCQjw5NnbBRDaARIsAJP-YR_IQiQC0876PaQmo6lWAuRPJ5FG3oh65dc7rsf95xSqDd0Wdqrs6OYaAsxtEALw_wcB) and more. The PurpleAir costs around $200 and provides real time air quality data that can be used to understand indoor or outdoor environments and conduct environmental investigations on air quality. **The intention of this page is to create a central location to discuss successful (and unsuccessful) use cases of the PurpleAir and problem solve together. Please contribute any experiences you've had with the PurpleAir. Perhaps consider visiting https://publiclab.org/simple-air-sensor to see the start of a barebones kit that is kicking off the development of a well-supported, open source air sensor similar to the PurpleAir.** ## Why use one? Every state in the US is mandated by the Environmental Protection Agency to create air quality sensor networks to monitor the six pollutants that the EPA defines as hazardous - Ground-level ozone, Carbon monoxide, Sulfur oxides, Nitrogen oxides, Lead and Particulate Matter. Read more about [particulate matter](https://publiclab.org/wiki/pm#Background+Information) here. However, due to the huge cost upwards of $25,000 associated with Federal Reference Method (FRM) and Federal equivalent method (FEM) sensors, most cities only have a handful of sensors, sometimes just one, measuring each category of pollutant. In addition, FRM sensors often have a lag of several months for quality control before the data is publicly accessible. Optical sensors provide a way to get near real-time data on the particulate matter in your home or community and to have autonomy over the locations being monitored. ##How it Works [This page](https://publiclab.org/wiki/optical-pm) does a great job of explaining how laser optical particle counters work in detail. But, the most important things to know when deciding to use an optical particle counter like the PurpleAir is that: 1. The data the PurpleAir (and other optical counters) produce is an _estimation_ of particulate mass concentration that relies on several assumptions for shape, diameter and density. The quality of your data will depend on those assumptions as well as environmental considerations such as humidity, light and temperature. 2. Because of the fact that optical counters rely on these assumptions, the data produced by them are _not_ FRM or FEM certified. That being said, there have been a number of academic studies that have quantified the performance and limitations of some of these optical sensors to help you choose which one to use depending on what you are trying to sense and what level of performance you are looking for. The EPA has an excellent summary of these studies in this table [here](https://www.epa.gov/air-sensor-toolbox/evaluation-emerging-air-pollution-sensor-performance). A higher R^2 value means better performance. [This page](https://publiclab.org/wiki/optical-pm) has some more links of studies to check out if you scroll a bit. [Here](http://www.aqmd.gov/docs/default-source/aq-spec/field-evaluations/purpleair---field-evaluation.pdf) is a study specific to PurpleAir. ## Use Cases - PurpleAir maintains a [map](https://www.purpleair.com/map?#0.27/0/-30) of every user's data from around the world. Check it out. - A use case in [Utah](https://www.good4utah.com/news/local-news/dirty-air-monitors-are-they-telling-us-the-whole-truth/382744964) - A use case in [California](https://cleantechnica.com/2018/07/09/my-new-favorite-gadget-a-purple-air-air-quality-monitor/) - [This facebook group](https://www.facebook.com/groups/purpleair/) includes people who own a PurpleAir and discuss their experiences - [This blog](https://www.wxforum.net/index.php?topic=33031.0) has some dialogue about various use issues Please add your experiences with PurpleAir too! ## Suggestions for conducting particulate matter investigations [This page](https://publiclab.org/wiki/general-environmental-monitoring-study-strategies) summarizes important considerations before starting an investigation. To add onto those, specifically in regards to air quality, here are some suggestions. Before ordering a sensor, check out the data that is freely available to you online. [Visit your state's department of environmental management website](https://www.epa.gov/home/health-and-environmental-agencies-us-states-and-territories) and their air quality section. With some poking around, you can find their ‘Air monitoring network plan’ which will show you the _exact locations_ of all of their different FEM/FRM sensors and what pollutant they are measuring. This can help to identify gaps in their monitoring network that you may be interested in. [AirNow](https://www.airnow.gov/index.cfm?action=airnow.main) has national daily data from FEM monitors. You can search for your zip code and see the live air quality forecast. This can be used to help inform your day to day movements especially if you are an asthmatic. You can also download data* from FRM monitors [here](https://www.epa.gov/outdoor-air-quality-data/download-daily-data). This is really cool because you can specify what data you want and it will generate an Excel file for you. You can then start making plots and figures to compare different time periods and locations. [Check out this post](https://publiclab.org/notes/jiteovien/08-01-2018/air-quality-data-visualization-no-coding-necessary) which used this online EPA data to create some visualization of the monitoring in Rhode Island. *NOTE: As mentioned earlier, the FRM data has a lag for quality assurance reasons. For example, the most current Rhode Island data is for May 2018. (It is August at the moment this is being written). Everyone, whether or not you own a Purple Air has free access to the Purple Air's database of global sensors. Read more about this in the activity [Download & Analyze your Purple Air data](https://publiclab.org/notes/jiteovien/08-28-2018/download-analyze-your-purple-air-data?_=1535473842). ## Questions Questions can be either frequently asked questions, or "next step" challenges we're looking to solve. [questions:purpleair] ## Activities Activities show how to use this project, step by step. [activities:purpleair] ## Other things to know about using your Purple Air The PurpleAir sensor has two 'channels' that measure data. Channel A and Channel B. Each channel has a 'primary' and 'secondary' data set. The data that Channel A measures is described below: PrimaryData - field1: PM1.0 (CF=ATM) ug/m3 - field2: PM2.5 (CF=ATM) ug/m3 - field3: PM10.0 (CF=ATM) ug/m3 - field4: Uptime (Minutes) - field5: RSSI (WiFi Signal Strength) - field6: Temperature (F) - field7: Humidity (%) - field8: PM2.5 (CF=1) ug/m3 This is the field to use for PM2.5 SecondaryData - field1: 0.3um particles/deciliter - field2: 0.5um particles/deciliter - field3: 1.0um particles/deciliter - field4: 2.5um particles/deciliter - field5: 5.0um particles/deciliter - field6: 10.0um particles/deciliter - field7: PM1.0 (CF=1) ug/m3 This is the field to use for PM1.0 - field8: PM10 (CF=1) ug/m3 This is the field to use for PM10 particles/deciliter is a particle count per volume measurement. ug/m3 is the particle mass concentration - the popular method of measuring particulate matter. Channel B measures the exact same primary and secondary data. This is likely just done for robustness just in case one of the channel fails or has an error. These setup instructions are included with the Simple Air sensors and can be reprinted Purple_Air_Instructions.pdf ...


Author Comment Last activity Moderation
warren "Hi! I had to look up "adsorb" -- (it means accumulating on the surface rather than interior) but that makes sense -- basically @guolivar, you're sa..." | Read more » over 5 years ago
warren "Hi, any updates on the high humidity tests? Thank you! " | Read more » over 5 years ago
Cbarnes9 "Hello everyone, sorry for the late replies; I am posting the field study results from AQMD that I was referencing. Their tests indicate that the PM..." | Read more » over 5 years ago
jeffalk "@cfastie is certainly correct in his comments that each of the purple air devices reads, in general, higher than the Grimm. (Almost all slopes (coe..." | Read more » over 5 years ago
jeffalk "I just reviewed this conversation and would like to add to it and would like comments from @sagarpreet and @guolivar if possible. About the USEPA ..." | Read more » over 5 years ago
cfastie "The field evaluation report linked above includes 78 graphs comparing the results of a Purple Air sensor with a $25k GRIMM optical particle counter..." | Read more » over 5 years ago
stevie "Hi @Cbarnes9 thanks for sharing out the questions! I'm also curious where you're referencing the field test for PM10 to be biased high? Is it in t..." | Read more » over 5 years ago
Ag8n "The purple air monitors have a low r2 value for the pm10 testing. This may lead to the skewing that you are talking about. I corresponded with th..." | Read more » over 5 years ago
guolivar "Following on @jeffalk answer, please expand on your question. In my opinion (and I've worked in aerosol science for a number of years) the main poi..." | Read more » over 5 years ago
guolivar "As far as I know Plantower hasn't disclosed that (to be fair, I don't think any of the low-cost manufacturers has done either). From the field test..." | Read more » over 5 years ago
jeffalk "I'd appreciate learning what in the document you referenced indicates to you that the PM10 is "biased rather high". Thanks " | Read more » over 5 years ago
jeffalk "Hi. Given that you mention "silica" and you are at UW Eau Claire ( western wisconsin sand mining...) I take your question to really ask whether on..." | Read more » over 5 years ago
jeffalk "I could be wrong but I do not believe optical particle counters use spectroscopy as the method is usually described. At any rate a number of years..." | Read more » over 5 years ago
Cbarnes9 "I found the information about the 640 nm wavelength lasers quite interesting, thank you. I have been in communication from people from Lane Regiona..." | Read more » over 5 years ago
prashant0768 "Laser scattering based PM sensor has tendency to show inaccuracy in particle between 2.5-10 um in size. Since Purple air as per specification i se..." | Read more » over 5 years ago
nanocastro "Hi. Sorry for the late answer. We are using the PMS7003 but the 5003 is almost the same except for the size. We were using Arduino Mega with three..." | Read more » over 5 years ago
karlssonviking "Hi... This is Viking Karlsson and I am new here and I need support from you peoples. I was surfing the net and at random land over here and find th..." | Read more » over 5 years ago
guolivar "Here is a page about the PMS5003 and 7003 ... https://aqicn.org/sensor/pms5003-7003/ The difference seems to be in their internal layout. In that ..." | Read more » over 5 years ago
cfastie "There is another version of this sensor with the same specifications and the same price but it is about half the size. Does anyone know any reason ..." | Read more » over 5 years ago
warren "Thanks, everyone! " | Read more » over 5 years ago
kkoerner "So, great timing, I was just working on doing this with an Arduino mega today (here are some pictures). I was going to do a more extensive writ..." | Read more » over 5 years ago
jywarren "Wow, this got a reply on my posted question real fast. Thank you!!!!  On Friday, October 26, 2018 at 12:45:33 PM UTC-4, notifications wrote: --..." | Read more » over 5 years ago
yamanoorsai "This sensor comes with a serial output. Adafruit has an excellent tutorial on how to interface to an Arduino:https://learn.adafruit.com/pm25-air-qu..." | Read more » over 5 years ago
warren "There's been some strong interest in the Plantower PMS5003 which is the sensor in the #purpleair -- in case it's interesting -- https://publiclab.o..." | Read more » over 5 years ago