EEVblog Electronics Community Forum
General => General Technical Chat => Topic started by: Holister on January 20, 2020, 09:59:49 pm
-
Hello!
So i bought this chip capable of reading UV radiation on (GYML 1145https://cdn-shop.adafruit.com/datasheets/Si1145-46-47.pdf (https://cdn-shop.adafruit.com/datasheets/Si1145-46-47.pdf)) and I set to contrast the values it delivered with those of the monitoring stations near my city (http://www.meteochile.cl/PortalDMC-web/otros_pronosticos/grafico_radiacion.xhtml?estacion=360019 (http://www.meteochile.cl/PortalDMC-web/otros_pronosticos/grafico_radiacion.xhtml?estacion=360019))
I noticed that at 5pm there was a radiation of 3, but when I pointed the sensor directly at the sun, it marked 7.
My question is, do you think that professional uv monitoring stations mark maximum radiation at 12 o'clock because of the fixed position they are in?
Probably if the sensor was moved at an angle to the sun, there would be a more accurate measurement of the radiation.
Edit: added link to datasheet
-
I can't find any reference to "GYML1145". Does the chip directly read out the value in terms of a UV index?
It seems more of a calculated value than a direct measurement: https://www.epa.gov/sunsafety/calculating-uv-index-0 (https://www.epa.gov/sunsafety/calculating-uv-index-0)
If you look for UV index graphs, they seem to have quite a similar shape to what you've linked: https://en.wikipedia.org/wiki/File:UV_Diurnal_Erythemal_Dose_Rate_Per_Latitude_graph.png (https://en.wikipedia.org/wiki/File:UV_Diurnal_Erythemal_Dose_Rate_Per_Latitude_graph.png)
-
The environmental monitors I'm familiar with use a lens/diffuser with angular/spatial response correction (and require careful siting to avoid shading/reflections/etc that affect readings).
The resultant graphs will show a similar "normal" curve response naturally anyway, simply due to angular-related differences in atmospheric attenuation etc (e.g. higher UV levels at noon when the sun's overhead and radiation is travelling through less atmosphere; lower levels mid morning/afternoon as it's travelling through more atmosphere; and lowest levels after full dawn / before the beginning of sunset, as UV has to travel through the most atmosphere).
-
Unless your sensor has a response like this image, obtaining solar irradiance is not straightforward.
The recommendation is to use a simple LDR (it has a much wider spectrum width) with adequate external filtration and, for each system, perform a calibration procedure.
https://en.m.wikipedia.org/wiki/Solar_irradiance#/media/File%3ASolar_spectrum_en.svg
-
Unless your sensor has a response like this image, obtaining solar radiation is not straightforward.
The recommendation is to use a simple LDR (it has a much wider spectrum width) with adequate external filtration and, for each system, perform a calibration procedure.
https://en.m.wikipedia.org/wiki/Solar_irradiance#/media/File%3ASolar_spectrum_en.svg
I don't think LDRs are very sensitive to UV (100-400nm): https://electronics.stackexchange.com/questions/211894/what-colors-or-wavelengths-are-photoresistors-sensitive-to
You need specific materials to allow UV to pass (quartz, etc.).
-
My mistake. I have considered that the OP want a W/m2 result of solar irradiance.
Now I see that the main goal is get UV only. It is the problem of starting to read and imagine old projects |O
I have used the si1133, but the result are not so great. You always need a proper mechanical assembly with optical diffusers, etc.
The IC is more recent than si1145, and have mutch more information.
https://www.silabs.com/documents/public/application-notes/AN968-Si1133-UV-Index-Sensor-Electrical-and-Optical-Design-Guide.pdf (https://www.silabs.com/documents/public/application-notes/AN968-Si1133-UV-Index-Sensor-Electrical-and-Optical-Design-Guide.pdf)