PM2.5 and PM10 caping at 150 ug/m3

[PetitSphincter]

Using a Dyson Purifier Humidity+Cool with this integration. Works like a charm. But during the 2 years i've used this integration i noticed something.

The value of PM2.5 and PM10 are caped at 150 ug/m3 while the dyson cloud app or the display screen can show values up to 999 ug/m3. I just reproduced it right now by vaping right into the filter :

While the value displayed on the screen was caping at 999 ug/m3

All the value between 999 ug/m3 and 150 ug/m3 on the screen were at 150 ug/m3 on HA. Once it decreased under 150 ug/m3 values are the same, so it's not a problem of proportion just caping for values above 150.

[dotvezz]

Interesting! I'll look into this. For the time being would it also be possible to get a screenshot from your Dyson app showing the 999 value? I'm curious how it registers, and it could help us in understanding the signals.

[PetitSphincter]

Downloaded the app back (trying to get rid of cloud apps :) ) and destroyed my lungs but here's a screenshot :

It's quite hard to reach 999 ug/m3, hopes that 700 is enough (got to destroy my lungs even more) And here's how it shows in HA :

You can clearly see when it caps around 19:35

[dotvezz]

Wow, okay that helps contextualize things a bit. I also got a chuckle realizing that it's literally off the chart.

One last question, are you able to run this tool while sacrificing your lungs for science one more time?

If you'd rather not, I totally understand. I don't vape but I have some friends who do, and I KNOW they'll get a kick out of vaping at my fan lol.

[PetitSphincter]

I'm not home currently coming home next wednesday, i just put a reminder for next week to try this out

[codyc1515]

You could also just spray something in the air, like “air freshener”.

Thanks @PetitSphincter …

[PetitSphincter]

This tend to raise VOC more than PM. Plus it's really hard to reach high level of PM with it. At least with air freshener i have

[PetitSphincter]

Struggled a bit to find the IP but after an angry IP scanner here's the results :

It seems like the same thing is happening here. The third scan is the one where it was way above the 150 ug/m3 (700 ug/m3). But here this point is at 0149 which is, i suppose, 149 ug/m3. All the other line where below on app, and coherent with the tool

[dotvezz]

@PetitSphincter thanks a ton for that info. It's really interesting, and kinda frustrating. Some devices to have values they send over MQTT which are flagged as "high resolution" in the mobile app. I'll investigate those values and their availability. If I'm able to work out a solution, I'll get back to you here!

[RedEarth1]

I've noticed this too, so I monitored my fan using MQTT explorer while giving the fan a quick squirt of deodorant and here's one of the messages:

{"msg":"ENVIRONMENTAL-CURRENT-SENSOR-DATA","time":"2024-05-01T18:12:07.000Z","data":{"tact":"2963","hact":"0054","pm25":"0149","pm10":"0149","va10":"0013","noxl":"0007","p25r":"0999","p10r":"0999","hcho":"NONE","hchr":"NONE","sltm":"OFF"}}

pm25 and pm10 are the values which were showing in HA (both 149) but as you can see p25r and p10r as both showing 999 which is also what was showing in the Dyson app at the time, so maybe these are the 'uncapped' values that it uses??

[dotvezz]

@RedEarth1 Yep, p10r and p25r are the "High Resolution" values. Thanks for looking into it!

The only reason I don't just immediately switch this integration to using them is because I'm not sure about the availability of them on the different devices and generations.

[DennisOtto-dev]

Just bought the Purifier Humidify-Cool Formaldehyde two days ago and I am experiencing the same issue regarding the PM-readings. Yesterday there was a spike of more than 270 PM2.5 (looking at the Dyson App), but the highest PM2.5 value in my HomeAssistant during this time was only 132. The same with PM10: Dyson App shows more than 270, HomeAssistant shows 104. Looks like all values are off.. HCHO: Dyson App shows 0.011 and HomeAssistant shows around 0.018 VOC: Dyson App shows 3, HomeAssistant shows 4.74 NO2 looks correct, but for now I don't have any spikes to compare.