>>> Join Us in the Fight Against Air Pollution

AirGradient Forum

PM0.3 discrepancy between firmware version 3.0.1 and 3.0.8

I have two indoor units right next to each other. The 3.0.1 always reports positive, non-zero 0.3 readings. The 3.0.8 reports zero most of the time, with the occasionaly sub-1 reading.

Did this algorithm change in the firmware, or is this likely to indicate some other issue?

A few questions to give you a proper answer:

  1. can you post the charts for the 0.3 over at least 48 hours. Both devices.
  2. what is your outdoor pm2.5 / what city do you live (then we can look it up)
  3. does your home use any air filters

In general different batches of pm sensors deviate on the count. The manufacturer then calibrates the ug/m3 values. So there can be a difference but only to some extent and that’s why I would need to know above to make a better judgement.

Also please flash on both devices the 3.0.8 firmware.

One of the units initially did not return any PM values. I updated it to the latest firmware, and I thought at first it resolved the issue but really it just replaced “no value” with “incorrect zero value.”


These are in an open air outdoor courtyard right next to each other. I don’t have 48 hours of them on the same firmware, but there is 3 hours of them on the same firmware.

Nearby air monitoring stations:

That does not look right. Just to be completely sure it’s not the board or firmware could you swap the two PM sensors to confirm it’s the PM module?

I will then send you a new one.

I have the PM sensors swapped now, and it looks like it is the issue. The “Bad PM Readings” board with the swapped sensor is reporting values, and the “Working PM Readings” board with the swapped sensor is not.

I’ll let it run tonight (inside) and will post the chart tomorrow.

Yes this is definitely broken.

Please send me an email with your order number and I will send a replacement. Please also send a photo of the serial number on the PM module so that we can check which batch it’s from.

Swapped overnight:

You can see the switchover clearly. The readings are lower after the swap because I brought them indoors.

The number under the barcode on the bad sensor:

PMS5003-2023121801137

Thanks. Replacement is on the way.