Back to

SCK 2.1 - TVOC and ECO2 (CCS811)

Hey everyone!

This is a quick update for anyone seeing weird values on their Indoor Air Quality (TVOC and ECO2) sensor. The sensor needs a 48h burn-in to stabilize. You can check the datasheet for more information while we work on updating our docs page with all this information. Also after each reset/reboot, it can take up to 20min for the TVOC and ECO2 to stabilize.


We are working to improve our FAQs section with all this information

1 Like

A month ago a received my first smartcitizen kit 2.1. After some experimenting I installed it in our kitchen at 2 meters high. The sensor values seemed alright, although the EC02 and TVOC where a little bit high. PM, ECO2 an TVOC reacted as you could expect when we were preparing a meal, toasting a bread or brewing a beer.

Wednesday a received a second sensor which a planned to use outsite. But first I installed it beside the first sensor, just to look at the bandwith of the measures. Most of them are inside the 5-10% difference. That’s acceptable. But the values of the EC02 en TVOC (same sensor) differs very much. The oldest sensor measure at the same moment 1795 ppm and 444 ppb. The newer sensor measures 488 resp. 28 ppb. The kit has run now for 3 day’s, so it couldn’t be the burn-int time. You can find the data at: and

I use a lenovo tablet powersupply whith a quality usb cable on the new kit.

Hi @peterveeling

Thanks for the detailed explanation!

@oscgonfer just released a few weeks ago a section around the ECO2 and TVOC sensor that might help you

RE sensor id:9897
CO2 readings are not real - over 1000 ppm in the woody rural area. The sensor runs for months now.
So the VOC must be suspect too.
Sound dB recordings are not frequent enough - the result is unrealistic averaging.

Anybody similar problems? My sensor is showing way over 2000ppm next to another sensor showing around 600 in a well ventilated space. Cleaning or some calibration necessary?

How long has the sensor been set in that location? Did you have them anywhere else before?

As you might know, the sensor had an initial burn-in time, and after it needs a stabilisation process after each collocation

We tested more than 200 sensors in our lab in the past months, and all of them converged after of maximum of one week