VOC Reading level, not index

[guitartoys]

Hello,

I understand the rational of using the Voc Index, since you can’t tell good vocs from bad.

But my purpose is to have this monitor vocs in the room where I am 3D printing.

So, having the index be based on a prior period doesn’t seem to be the best thing for me.

While I am interested in changes, I think I would want to have the actual value, because if the values remain consistently high, and then the device basically uses the prior period to establish the new norm, won’t this then make the readings look low, even if the room is still high?

I certainly don’t print ever day. And often print PLA, but when I’m printing things like PETG, ABS, Nylon, etc. I’d really like to know how much of those gases are in the room.

Am I not understanding this correctly? Should I just change the VOC index learning curve to 720 hours?

Is there a way I can get the VOC value in place or, or maybe in addition to the VOC index?

Thanks.

M

[MallocArray]

I don’t personally think any of it will be very helpful. I print with PLA, PETG, ABS, and ASA and never seen any related movement of VOCs before during or after prints. So unless the Nevermore Mini is just flawless, I don’t think this sensor really picks up what 3d printing is emitting.

[guitartoys]

Interesting observation. As yes, I hadn’t noticed that much movement with one exception.

I use hairspray on my plate, and oh boy, the Airgradient certainly senses that.

But I do notice that the QingPing seems to notice when I’m printing stuff other than PLA

[Altair_AirGradient]

Hi @guitartoys! Welcome to the community, and this is a good question.

Your understanding is broadly correct. The VOC Index is designed to show changes relative to the air the sensor has recently experienced. A value around 100 represents the learned baseline, while higher values indicate that the current VOC signal is worse than that baseline.

With a sustained VOC source, the VOC index algorithm will gradually adapt and establish a new baseline. Extending the learning period to 720 hours should slow that adaptation, which helps with occasional printing events, but it would still not turn the reading into an absolute VOC concentration. However, I should mention that my colleague @Ethan_AirGradient recently tested the SGP41 and found that adjusting the learning time offset didn’t appear to have a significant effect. (Correct me if I’m wrong, Ethan!)

We analysed the raw SGP41 signal against a reference instrument here: How Accurate is the Sensirion SGP41 TVOC Sensor?

This conclusion part of this article is very useful in this context. The raw signal is therefore the closest available option to an “absolute” output from this sensor but this is still a log-scaled resistance value, not an absolute concentration. In the dashboard, the raw signal is called “TVOC (ticks)”. There’s also TVOC (ppb), this ppb value is converted from the Index to ppb using Sensirion’ own conversion formula, but this is still not an absolute concentration but Sensirion’ attempt at making the sensor output compliant with green building certifications like WELL and RESET:

It is also important that the SGP41 is “broadband sensitive” which means it cannot identify whether the detected gas is styrene, lactide, alcohol from hairspray, or another compound. Its sensitivity also differs substantially between compounds. PLA printing commonly emits lactide, for example, and the sensor does not respond strongly to it. Hairspray often contains alcohols and other compounds to which metal-oxide VOC sensors respond much more strongly, which would explain the clear increase you observed.

Different consumer monitors also process the sensor signal differently but use the same underlying sensor (as there’s nothing else low cost in market for TVOC detection, at least as far as I know!) I would point out that Qingping monitors use the same sensor (SGP41), but they do use their own algorithm on top of the Sensirion conversion index.