Many thanks,
I really appreciate your help.
Edit: I’m getting this error message,
Can I solve it by myself?
Did you use the new link above to get the new version code?
If yes, then comment of line 77 (put // in the beginning of line 77)
No, I just updated it with its own link.
Just used the provided link and now it works as expected.
Thanks!
1 Like
I am still experimenting and deciding which is the best way to distribute the new series of drivers that use Library files. Using HPM will solve the issue for installing releases, but the problem is with installing the Beta versions of the drivers. In this case I have forgotten to update the "importUrl" link inside the driver, so it points to the code version where the libraries used are included at the end of the driver code.
1 Like
Thanks for the drivers. Just bought the VINDSTYRKA yesterday, and been recording metrics since.
Do we know what the units are for airQualityIndex ?
And what the values can be for airQualityLevel ?
Is it based on this chart from Understanding TVOC: What You Need To Know About Volatile Organic Compounds
2 Likes
If you look at the driver itself, @kkossev put in a link to the chart being used:
With that being said, I still don't really know what it's measuring. I seem to get really high index values that don't really correspond to smells or pm25.
3 Likes
After a week of using the Vyndstyrka, I'm not sure I trust the Air Quality Index at all... it routinely goes into "hazardous" (very high) levels of TVOC... just from making toasts or cooking spaghetti sauce. If that's enough to bring the air in "hazardous" zone, I don't know what hazardous means...
Above are the plots from the VOC (in ppb) as measured by my AirThings device, and the PM2.5 (in ug/m^3) as read by the Vindstyrka, below is the Air Quality Index, which is supposed to be total VOC...
1 Like
Read the PDF linked here:
It explains how the sensor inside works.
Ah, so it is not the AQI as linked before [AQI Basics | AirNow.gov], it is a "VOC Index". AQI typically has nothing about VOC.
I would say that the driver is misleading then. It should not show this data as airQualityIndex, but rather as VOCIndex or something like that.
2 Likes
@mboisson you are right.
VINDSTYRKA does not report "Air Quality Index", but VOC or tVOC index values. The same applies to Aqara Air Quality sensor ...
I will do some more research during the weekend and will change the driver, the numeric value attribute will be named "tVOCindex" probably... as it is definitely not a TVOC measured in ppb ( parts of gas per billion parts of air),
Do you think that the 'Index Category' value from the table that you linked corresponds to the "Index Value" as measured by the Ikea sensor?
4 Likes
I do not think they are the same scales. That one was from another vendor, Kaiterra. It seems to be an absolute scale (50 or less is good, 51-100 is moderate, 100-150 is high, …)
By comparison, the sensirion document states
On the VOC Index scale, this offset is always mapped to the value of 100, making the readout as easy as possible: a VOC Index above 100 means that there are more VOCs compared to the average (e.g., induced by a VOC event from cooking, cleaning, breathing, etc.) while a VOC Index below 100 means that there are fewer VOCs compared to the average (e.g., induced by fresh air from an open window, using an air purifier, etc.).
So that sensor produces a measurement which is relative to your own baseline… assuming nothing changed in the air around the sensor, I would expect the sensirion sensor to stabilize at a value of 100, regardless of the actual concentration of VOC in the air. A pretty useless measurement, I think.
But I may be wrong, there are a lot more details on their algorithm in this document https://sensirion.com/media/documents/4B4D0E67/6436C169/GAS_AN_SGP4x_BuildingStandards_D1.pdf#page3
But I haven’t had time to read it all
1 Like
For what I understand from reading this is that the sensor does not discriminate between specific VOC nor between harmfull and harmless ones. Furthermore, its sensitivity is not the same across all VOC. So with all those limitations, they decided not trying to report an absolute value which would be misleading and inacurate. Instead they choose to mimic the human nose.
This is more suitable as a trigger for automation related to odors, not air quality per say. (bathroom fan, cooking, cat litter)
2 Likes
I have updated the driver to version 2.1.5 (2023/09/02 9:24 PM") :
- removed airQualityLevel ( the "Good", "Moderate"," Unhealthy...." ... text )
- airQualityIndex replaced by sensirionVOCindex
- the sensirionVOCindex is deleted from the events if the polling is disabled
- added thresholds for the PM 2.6 and the sensirionVOCindex (can be adjusted for less frequent reporting)
The preferences should look like this:
If it is working OK, let me know and I will put this version in HPM for easier installation and updates.
3 Likes
Working great for both sensors. 
"VINDSTYRKA Air Quality Monitor" driver can now be installed from HPM.
If the driver was already installed manually, do a 'Match Up' in HPM.
3 Likes
Was getting really high IAQ readings in one room. Logs showed it jumped every 15 minutes. Found out my wife put an automatic air freshener right next to the VINDSTYRKA.
1 Like
That means that the sensor works! 
1 Like
Works fine. Thank you!
I moved my VINDSTYRKA over to Hubitat from the Ikea Dirigera hub.
Pretty cool that your driver shows info that even the Ikea app/hub doesn't display.
I did the match up and update via HPM and now for some reason the new 'sensirionVOCindex' attribute isn't triggering events for Maker API where almost everything else is? I'm seeing this on 2 different hubs (2.3.5.152):
Edit: Seems removing and then re-adding the device to the MakerAPI instance did the trick. That's interesting. I wonder how many other of my devices that have gotten new states aren't pushing the data...
