Due to high demand expect some shipping delays at this time, orders may not ship for 1-2 business days.

GUVA-S12SD calibration
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

GUVA-S12SD calibration

by earthres on Thu Oct 21, 2021 2:35 pm

This past summer I had a question about getting UV index values from a GUVA-S12SD sensor. Adafruit's documentation suggests that based on the gain of the transimpedance amplifier that converts the very small photocurrent from these devices to a usable output voltage, the UVI value is V_out/0.1. A reply to my post identified this calculation as just an approximation. But in fact, it's a very poor approximation. This graph shows some calculated UVI data collected during September at a latitude of 40N, using V_out/0.1, A UVI of more than 8 around solar noon is WAY too high for this time of year and latitude, even under clear skies. A more reasonable UVI would be around 5 or so.
Commercial UVI meters cost several hundred dollars and it's unlikely that calibrations of a single-detector UV sensor with such a device will be the same under a variety of sky/sun conditions. Over time, results from this sensor module taken around solar noon, pointing straight up at a clear sky, could be compared with published UVI values for your location, but even these "calibrations" will show a LOT of variability.
This isn't a criticism of the GUVA-S12SD sensor module, which is very easy to use. But it's not really appropriate to claim that it can be used as a reliable UVI meter. For looking at diurnal variability of UV radiation measured as power observed by this particular sensor, the module performs quite well, with a "cosine response" that is surprisingly good for a detector that's not covered with a diffuser.
GUVA_UVI.jpg (14.11 KiB) Viewed 51 times

Posts: 19
Joined: Fri May 28, 2021 10:48 am

Re: GUVA-S12SD calibration

by sj_remington on Fri Oct 22, 2021 9:47 am

The UVI scale is an extremely crude and limited measure, intended primarily for warning purposes.

But, it would be useful to have a side by side comparison of the UV sensor response together with the output of a calibrated instrument, over the course of a year or so.

Posts: 191
Joined: Mon Jul 27, 2020 4:51 pm

Please be positive and constructive with your questions and comments.