Airgradient vs Airthings View plus PM2.5 mesurements

[Martin_Moller_Larsen]

Hi

I have an Airgradient One in my living room and an Airthings View Plus in my bedroom. I was noticing that my view plus would go a lot up and down during the day when measuring PM2.5 while my Airgradient would always show 0/1

Today I tried placing them side by side in my living room and this is the result:

Furthermore I tried turning on the air purifier while doing this test in the living room and I then saw the View plus going down to 1/2 pm2.5 which makes me think it has the most correct measurements

Is this to expected by the Airgradient One? It may not seem like much to only measure wrong by 4/5 PM2.5 but when WHO says the average limit should not go above 5 PM2.5 this makes my Airgradiant too unprecise to actually use

Best regards Martin

[Achim_AirGradient]

Thanks for sharing.

In general I think it’s difficult to make a conclusion based on two low cost sensors which one would be more accurate without a reference device.

It would be interesting to test in medium and high PM concentrations as well. Can you create some particles and see how they behave then?

By the way, which firmware version are you on?

[hanuman]

Airgradient Plantower PMS5003 tech. specs say:
Accuracy: ±10%@100~500μ g/m³, ±10μg/m³@0~100μ g/m³

Airthings View plus tech. specs say:
Particle size detection range: 300 nm to 10 μm
Range: 0~200 μg/m³
Measurement error (PM2.5): 0 ~100μg/m³, ±10μg/m³,100 ~200 μg/m³, ±10%. Calibrated with a GRIMM using cigarette smoke source

Not sure what sensor model/make Airthings is using as their PM2.5 sensor but when comparing both Airgradient and Airthings the numbers seem in spec. Airthings is probably slightly more sensitive in the lower range.

Smoke a cigarette next to the sensors see what they say

[Ethan_AirGradient]

I’ve also found that the Airthings View Plus appears to have more ‘jumpy’ readings than both AirGradient monitors. I think it’s largely because it takes measurements at far less frequent intervals (I think on battery, it’s once every 10 minutes? I could be wrong, though). However, it doesn’t explain the differences at lower concentrations. I will run them side by side and see if my findings are similar.

By the way, the Airthings sensor is the PM2105L by Cubic Sensor.

[MallocArray]

PurpleAir has a correction algorithm that was developed with the EPA and AirGradient is planning on implementing it, although I’m not sure if it is in yet or not:
Correction Algorithms (airgradient.com)

In my cases, at lower concentrations it does increase the value in most cases. I’m not sure if Airthings is using it as well, but if AirGradient hasn’t fully pushed it yet, but Airthings has, that could also cause some difference in readings.

[Nastya_AirGradient]

We are currently working on EPA correction algorithm integration for both AirGradient Dashboard and Map and it will be released soon, but for now PM2.5 readings are uncorrected.

[Martin_Moller_Larsen]

I just retrieved the Airgradient Outdoor Air Quality Monitor and thought I would test that side by side with the others. This is the result

It seems like my indoor sensor just measures way too low compared to other sensors

[occam]

Since updating ESPHome 2024.8.0 (MallocArray) I’ve seen a shift up in all PM values!

PM 2.5

PM 0,3

PM 10.0

I use MIRV13 filters in my HVAC

[MallocArray]

Any increase to the PM2.5 values could be related to the addition of the EPA compensation algorithm that AirGradient is planning on implementing, but I may have beat them in adding it to the ESPHome firmware before the stock firmware.

But the increase in the other PM values wouldn’t be related to that.

Any chance there is an associated environmental change? Open windows, cooking, etc?
If not, may need to open an Issue with the ESPHome project itself.

[Achim_AirGradient]

Do you have outdoor data you can compare it with?

It might just come from a higher level outdoors that is reflected in the indoor values.

[occam]

I don’t have an on-site outdoor AQ monitor. I use the AirNow integration and the station is 30 miles away. I’m not seeing the same kind of upward shift in PM 2.5 and PM 10.0 from AirNow. We don’t smoke burn candles or have any other particle-creating devices. Cooking is detectable by peaks that subside quickly not shifts. I’ll watch both for a while longer and perform a correlation test

[matthosks]

My Airthings View Plus always shows higher values than AirGradient One Indoor. Airthings View Plus is sensitive to humidity for one thing. A hot shower will always increase the detected PM2.5 substantially which makes me think Airthings uses a corrective formula that throws everything off.

I was considering buying a new AirGradient One Indoor, but I will wait to see what happens after the corrective formula is added.

I have multiple Levoit 600s purifiers, and they consistently show higher PM2.5 than AirGradient.

[Nick_Windham]

I’m having similar issues with my new indoor AirGradient One since I set it up a few days ago. PM 0.3 works but PM 1, 2.5, and 10 don’t. Hopefully a firmware update will fix it. My non-glitchy and user friendly Qingping Gen 2 monitor is looking more attractive by the day. Considering returning this and just buying a couple more of those.

[Achim_AirGradient]

I looked at your monitor data and something seems to be off.
Please contact our support. You might need a replacement.

[Martin_Moller_Larsen]

I got the 3.1.9 now and it seems like the baseline is almost always 2 instead of 0 now. Does anyone else has the same experience?

[MallocArray]

I believe it now includes the EPA correction algorithm which does make it more likely to have a reading other than 0, since it takes humidity into account.

I believe it is acting as expected