Follow-Up on AirLink/PurpleAir comparison

[jo...@johnkline.com]

Earlier, when I posted my AirLink to PurpleAir graphs, I said I would follow-up if the air worsened.

The air has worsened, and the graph shows the AirLink very closely following the A sensor of the PurpleAir (even more so than the PurpleAir B sensor).

Although one shouldn’t reach conclusions about the AirLink from a couple of comparison charts, I am personally skeptical of Davis’ claim:

“ACCURATE: Measures particulates as small as 0.3 micrometers with an accuracy of +/-10 micrograms per cubic meter in concentrations.“

I say this because there are ample studies that show, where air particles are dominated by those produced from wood smoke, PurpleAir concentrations read about double the true concentration.

Again, I’m not stating Davis got this wrong.  I’m just more skeptical of the claim than before these observations.

—John

[Tom Keffer]

Amazing how close they are tracking. I wonder what's inside them. Do they use the same laser sensor?

--
You received this message because you are subscribed to the Google Groups "weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/e633ece8-a3f0-4dfd-a58b-63df7abf7886n%40googlegroups.com.

[John Kline]

I believe the PurpleAir is a Plantower PMS5003 (PurpleAir states that they use PMS5003 and PMS1003 sensors).

The Davis looks like it could be a Plantower PMS7003.  I’m guessing from shining a flashlight to look at the Davis and comparing to images of the PMS7003).

If I guessed right as to the Davis sensor, there is a comparison here:

https://aqicn.org/sensor/pms5003-7003/

I don’t know reliable the above source is.  I note that two sensors are way off (both are PMS5003 if I got the colors right).

On Oct 2, 2020, at 3:06 PM, Tom Keffer <tke...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/CAPq0zEA8wq6RW-E%2BQ1Dd1KoBcxtdpArS%3DfDdLJT5tKa2EWS6sQ%40mail.gmail.com.

[vince]

I tend to see my PA 'a' and 'b' sensors very close to each other, but I average them when I feed the data to weewx.

My PA's JSON contains:

  "hardwarediscovered": "2.0+BME280+PMSX003-B+PMSX003-A",

  "hardwareversion": "2.0",

Run "curl http://purpleair-NNNN/json |  jq . | sort" for a more human readable version of all the output....