Our cloud averages the data points into 5 minute averages. This is probably the difference from your script.
You are correct that we are piling through a lot of data right now bit it looks very good with our approach. I believe next week we will roll out some features that will make it easier to correct the Plantowers.
I just noticed that @Siriel_AirGradientās new algorithm showed up on my unit as part of that OTA update I did not realize I got. 
Iāll start monitoring for when I get updates now.
@Achim_AirGradient Would this be the most optimal set of settings since I know my sensor is from that batch? Or is using the EPA one stacked on top not recommended?
"corrections": {
"pm02": {
"correctionAlgorithm": "slr_PMS5003_20240104",
"slr": {
"intercept": 0,
"scalingFactor": 0.02896,
"useEpa2021": true
}
}
}
}
Yes, the EPA should be used on top. Your setting looks correct.
Thank you as always. I know I must drive you crazy 
When do you expect to start shipping indoor units with individually calibrated PMS5003?
And are you also planning to offer individual PM5003T calibration on the outdoor units?
I would like to buy one and would prefer to wait till you have the individual calibration set up.
But if you think itās going to take a long time, I might just purchase an indoor kit for now.
What is the authoritative source of what scaling factors and intercepts to use for the various batches?
I see a table for 3 different models here, but it says it is outdated:
Low Readings from Plantower PMS5003
Other articles that are linked from there donāt have a table.
In the Dashboard for Advanced Settings I see where I can pick different batches, but it only shows a Scaling Factor and the Offset is 0 for all of them
Update on PMS5003 Calibrations has some other values, but they donāt match up of the other 2 sources.
Iām looking to add this to the ESPHome config so those users can also have accurate data, but Iām confused as to what the calculation is, and what the scaling and offset/intercept values to be used are.
Ultimately, I think all of the relevant formulas and table of values should be updated in this page, which is supposed to have all off the Correction Algorithms used by AirGradient
Correction Algorithms
Sorry for the delayed response. Please take the ones from the dashboard with offset 0. These are the latest ones.
Is the formula the same from the first article I linked?
Is Intercept the same as Offset?
Does it still only apply if PM2.5 is below 31?
All of this should be included in the āCorrection Algorithmsā page
# Apply scaling factor and intercept
PM2.5_calibrated_low <- scaling factor * PM0.3_count + intercept
# Use the calibrated data only for PM2.5 < 31 ug/m3
IF (PM2.5_calibrated_low < 31)
THEN (PM2.5_calibrated = PM2.5_calibrated_low)
ELSE (PM2.5_calibrated = PM2.5_ugm3)
Just seeing this discussion and curious how we can know what batch our PMS5003 is from, to select the correct offset in the settings? I ordered a set of sensors to install on a board in early August 2024, and received them in late August. One of the offsets is labeled āPMS5003_20240826ā, should I use that one or the one prior? (Iām guessing the 20240104, but want to make sure)
You would have to open your unit up and look at the sticker, it will have the date code on it.
Hi All,
I just got my indoor unit up and running. PM 2.5 reads extremely low compared to other monitors. I did flash the update, and it is hooked up to the server. Iām not a coder, and short on time, so Iām hoping for an easy fix. Thanks.
Have you updated the PM2.5 PMS5003 calibration settings to suit your specific sensor (if necessary)?:
Dashboard
admin (for that monitor)
Advanced settings
PM2.5 Calibration formula - select the details of your PM2.5 sensor (you will need to open the monitor to read the details on the PMS5003 sensor label
Thanks Rog! Iāll give it a goā¦
What ti use? Specific bach calibration alone?
Or bach calibration + epa?
Latest August batchā¦
I would recommend + EPA as it seems to be giving more accurate readings