Adafruit SGP30 Air Quality Sensor Breakout - VOC and eCO2 (3709)

The media could not be loaded.  Very easy to use (with Tasmota), fast response (seconds), and very sensitive. Spray a little disinfectant / deodorizer a couple of rooms away, or burn some toast, and in about 30 seconds this is in the "red zone". Open a window, and the reading drops back to a safe value just as fast. I used a Wemos D1 Mini Lite and this sensor; that's it, just two parts and the USB power adapter, cable, and enclosure. Added the readings to another project with a display, and also have it notify our phones if air quality goes over threshold.

While it says it measures eCO2, which I suppose it does, this does not seem to be related in any7 way to actual CO2 measurements. VOC's, of any type, cause the VOC and eCO2 measuremnts to spike. So if you wash your floor, expect the reading to jump.If you want to actually measure CO2, look elsewhere. You'll need a NDIR sensor like https://www.amazon.com/dp/B014PDFP9S/ref=emc_b_5_t to read CO2.

Schnelle Lieferung, funktioniert einwandfrei

Excelente

Quick summary: Expertly packaged by AdaFruit - super easy to integrate into whatever you want to do. BUT, if it's CO2 you are after, look elsewhere, this is misleading and oftentimes overreacts.Full review:So, I'm a bit of a CO2 nerd. I have multiple CO2 sensors (costing ~ £90 each) and I have been using them for years. I know how CO2 makes people feel, I know how it behaves, I'm all over it!I bought this SGP30 sensor because is was significantly cheaper than the sensors I have been using previously, so I thought I would give it a go and see what it's like.The SGP30 is actually a Volatile Organic Compound sensor, with estimated CO2 as a by-product. It's been beautifully packaged by AdaFruit, and with their libraries, it was a matter of a few minutes to set this up and get it reading values. It's a super neat board, and I like the 1.8V reference output too - nice touch. I2C interface, it just works right out of the box. I am using a Wemos D1 Mini (ESP8622 based board) and it works well.I can't fault the presentation of the device.As for the readings.... well... who knows? It claims 15% accuracy... and that may be true, but it's a noisy sensor, readings jump about all over the place - and not because I am breathing nearby. Take a look at the attached photo of a graph. Orange line is TVOC, thick red line is eCO2 from the SGP30. White line is CO2 measured by a different sensor sitting next to my test rig. Notice how noisy the orange and red traces are compared to the white. As you can see, this graph covers some night-time where there is no activity in the building. In terms of CO2, it either over-reads or under-reads compared to the 'real' amount of CO2. I *am* compensating for temperature and RH% in my test setup (SHT21 module you see in the photo).BUT, that said, it is a useful sensor for general trends, and it does pickup the VOCs well - on the graph, at about 05:00 there is a spike in values. Turns out this is because there was a car sitting outside the building for a while and a window open upstairs, so VOCs entered the building and were detected. The CO2 also went up as measured by the other device, but not by much. Likewise, at just after 08:00 there is a spike in real CO2 (white line), but barely anything on the eCO2 from the SGP30. I was in the room having breakfast!Whilst it works right out of the box, and the AdaFruit code works well, it can be fiddly to get the readings right - particularly the calibration of the baseline readings and the adjustment for temp and RH%. I have had this sensor for about a month, trying things before I am happy that I am getting the best I can out of it - compared to just 'an indication' of what's happening.So, I think for VOCs, it's a good enough sensor for the price, but for CO2, it really is just an estimate, and it's often wildly wrong.

Diese Teil hat mich Nerven gekostet. Nach dem der erste Sensor gar nicht ging, schickte mir der Händler kulanterweise einen neuen. Dieser funktionierte der zweite im Testbetrieb aber nur, wenn er alleine am I2C-Bus hing. In komplexer Schaltung störte er die anderen Sensoren. Ich las, dass es manchmal zu viel Pullup sein könnte und probierte den I2C Bus ohne programmierte Pullups einzustellen. Leider war es nicht erfolgreich.Parallel bestellte ich mir den SGP30 von Pimoroni - angeschlossen funktionierte dieser sofort!D.h. dieses Board ist Müll. Der Händler ist in Ordnung, das Produkt leider nicht.Da der Sensor nun mal da ist, bestellte ich mir einen Attiny85, vll. baue ich für einen separaten I2C-Bus nur für den Adafruit - Sensor auf.