Inconsistent measurements between PMS5003

Breakout boards, sensors, other Adafruit kits, etc.

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
User avatar
jbbrown9
 
Posts: 3
Joined: Sun Sep 05, 2021 1:29 pm

Inconsistent measurements between PMS5003

Post by jbbrown9 »

I've been working with 2x PMS5003 devices for about a week in hopes of comparing inside vs outside air quality and the effectiveness of various methods of improving air quality indoors. I can't reconcile the 2 sensors and am wondering if there could be a problem with the sensor or if it's possible there's a setup issue.

Each one is connected to a raspberry pi 0 w and is the only device connected to it. The sensors have been right next to each other for this round of testing.

I'm attaching charts of the last ~12 hrs of testing. The green and blue lines are my 2 sensors. The 4 purple lines are the 4 closest Purple Air stations, which are spread out over an area of a few miles, so it's not surprising that they don't match exactly.

When I had the sensors indoors overnight, the green/blue were fairly close together at the bottom. Readings were mostly <=10, with a few odd spikes on only one sensor (which I can work around, if I have to).

The bigger problem came when I moved them outdoors at the vertical blue line. They both jumped up into the neighborhood of the Purple Air sensors, but the PMS5003 sensors are very far apart from each other. When I set very high smoothing, they look like flat lines at ~65 and ~110.

Curiously, the pm10 readings are close together. I am only starting to log pm10 from Purple Air now, so don't have that for comparison, yet.

Bottom line is these sensors match for pm10, but do NOT match for pm2.5. In other rounds of troubleshooting, I've tried blowing them out, swapping around sensors between pi devices, swapping power supplies, etc, and couldn't find a pattern in other hardware changes affecting the readings. The same sensor would always be higher before and after those changes.

Any thoughts about what to do next?
Attachments
Screen Shot 2021-09-06 at 7.40.28 AM.png
Screen Shot 2021-09-06 at 7.40.28 AM.png (28.92 KiB) Viewed 112 times
Screen Shot 2021-09-06 at 7.48.33 AM.png
Screen Shot 2021-09-06 at 7.48.33 AM.png (23.68 KiB) Viewed 112 times

User avatar
jbbrown9
 
Posts: 3
Joined: Sun Sep 05, 2021 1:29 pm

Re: Inconsistent measurements between PMS5003

Post by jbbrown9 »

Now that I'm charting Purple Air pm10, I realize these sensors are putting out very high values for pm10. They're around 1000 when the nearest Purple Air station says around 40. I'm not sure either is right, but 1000 is almost certainly much higher than reality.
Attachments
Screen Shot 2021-09-06 at 8.44.34 AM.png
Screen Shot 2021-09-06 at 8.44.34 AM.png (61.99 KiB) Viewed 111 times

User avatar
jbbrown9
 
Posts: 3
Joined: Sun Sep 05, 2021 1:29 pm

Re: Inconsistent measurements between PMS5003

Post by jbbrown9 »

One last set of data, here. I ran them for a couple days outside in some pretty bad air. The purple air stations showed the air quality varying through the day. The values shifted quite a bit depending on which way the wind was blowing. Visually, the shape of the purple air charts matched what I could see out my window.

While the purple charts air seemed believable, the PMS5003 sensors gave noisy data, didn't match each other, and drifted apart over time. When I moved them inside (at the vertical blue line), they did drop very low. (In this instance, the outside air improved shortly after I moved them, but the readings from the sensors dropped immediately when I moved them inside.) I repeated that patten a few times. Outside: high. Inside: low. But, the actual readings are more or less useless. I had hoped to use these to measure how effective some air purifying systems are during times of very bad air, but the two sensors can have a 100% difference from each other when they're right next to each other.
Attachments
Screen Shot 2021-09-16 at 1.41.10 PM.png
Screen Shot 2021-09-16 at 1.41.10 PM.png (39.07 KiB) Viewed 90 times

Locked
Please be positive and constructive with your questions and comments.

Return to “Other Products from Adafruit”