I thought users might be interested in a recent exchange, prompted by a review I left for a Classic. This is the review I left:
Blowin’ hot and cold
Unit seems to perform well, despite the fact that I had to mount it on S side of the house to get wifi signal. So it could probably do with some more screening from direct sunlight, even at this time of the year at 56°N, because internal temp has been running around 100°F in this UK September “heatwave”. But the readings have clearly shown the influence of Saharan dust on air quality!
Not sure yet why the readings do not correspond more closely with other stations nearby and an official station 3 miles NNE. . . maybe because I am only 125 m from a dual carriageway carrying 42,000 vehicles/day - but that should be far enough.
For the price, it should have replaceable sensors like the Flex (only $25 the pair!): but I chose Classic because I am not the slightest bit interested in your experimental VOC readings from the BME68x chip. The completely arbitrary Bosch iaq is for indoor use - for watching paint dry.
Re price: for 45% of the Classic price I am about to install another AQ monitor (with 3 redundant sensors) that is automatically replaced every 2 years - for an annual charge of another 45%, sure. But at least I will have confidence in the readings, long-term.
The biggest problem is that you now charge for API access: that is unnecessary and offensive. (Full disclosure: I was awarded a few more free points when I first queried it.)
Hope this helps.
This is the reply by PurpleAir, Inc.:
Thank you very much for your comment. We appreciate the feedback you have given us. However, we do want to touch on a few things below.
We do generally recommend that devices be placed in locations that are shielded from direct sunlight. However, the device also generated heat itself, which can affect its internal temperature readings.
We do sell replacement laser counters for the Classic for $45. Although you are correct, the replacement process for laser counters in the Classic is a bit more involved than in the Flex or the Zen.
Lastly, we want to make it clear that we do not want the transition to a paid API to affect sensor owners. Up until now, sensor owners have supported our company with their purchases. Now, we want to balance that support with API users who do not own sensors.
If you want to query the data from your own sensor, we want you to be able to do that. We are working on a system that will automatically allocate API points to sensor owners, but it has not been released yet. For the time being, please email [email protected], and we will manually grant you API points.
Reply to PurpleAir:
That was a surprise - I thought it had disappeared into thin air - but many thanks!
May I share our exchange - or at least the bit about the API - with other interested parties on a well-known weather forum?
I have been comparing readings from the Classic with my Gaia-A12, and I find they track pretty well.
My comment about “official” station readings can probably be explained by the fact that they frequently use the dominant pollutant - often ozone - to produce the AQI.
The nearby (125 m) road actually carries about 80,000 vehicles/day, but I have been unable to correlate AQI peaks with peak traffic hours so it would appear I am far enough away from it. On the other hand, I can usually tell within minutes when one of two neighbours light their open fires. . . and I am also very surprised to see that these devices sometimes register zero. That’s impossible - full stop.
So, the big question: is your device too sensitive to be of any practical use? After all, the EPA health warnings are based on a 24-hour average AQI (which I know you provide, but you do not emphasise).
Reply from PurpleAir:
Thank you for your email. I’ll try to provide some insights below:
-
Yes, you are welcome to share anything we said in the response. Regarding what we said about the API, if you intend to tell other people, I would recommend sending them this community article. It essentially outlines what was said in the response below.
-
There are a number of reasons that PurpleAir sensors and federal sensors can output different readings. However, we agree with what you have said here that federal machines are capable of measuring more than just particulate matter and will output an AQI value based on the dominant pollutant. PurpleAir devices, on the other hand, only report PM readings.
-
Lastly, we do not believe that our devices are too sensitive for practical use. In fact, we would argue the other direction, that the increased resolution of data is critical for real-world application. We feel that there are numerous areas around the world that are lacking air quality data. Our mission is to empower individuals with hyper-local air quality data, especially those in areas with a lack of data so that they can institute real change. That change can be found everywhere, including measuring the PM levels from your neighbor’s fire pit.
However, as you have mentioned, we do provide longer averages of data, including a 24-hour average, if you are not interested in higher-resolution data.
Please let us know if you need anything else or want to continue discussing air quality.
(With apologies to @ktrue for a direct quote re API.)