Timely recording for me. I was trying to get UV readings with a VEML6075. Reads nicely but discovered it is very directional. I'd need a solar tracker for it to measure solar UV consistently.
I want to thank you for your work. You've been so helpful, every time I want to buy a new sensor for a project there's a video of you trying the most common available and making a thoroughly made review about it. So, thank you you've made much lighter the work otherwise I've had to make in order to choose the right sensor for my project
Thanks again for a great video. I specially enjoy these compilation videos which compare similar devices and helps us followers save a lot of time searching for the info you kindly bring us in the usual well structured and easy to follow videos
Well that video brightened a dull day (sorry). Learnt a lot - never realised the sensors where so different. I’ll be following up the gesture sensor, thank you for publishing your code for that. One thing to note - a significant amount of ‘white’ paper is actually blue to give a clean cold look rather than a warm tint.
You are right for teh paper (AFAIK not in all countries equally). But here I measured the spectrum of the diodes without paper. So both effects would add up :-(
Thanks for dedicating your time to another great video! :) Right now we are developing a project that needs a light sensor and after a close examination we've went fo the same IC TCS34725.
Thank you for your effort and time spent on this Topic. Regarding your awning: What also could be helpful for you is the cloudage. I am measuring far IR Temp with a Melexis MLX90614 and environmental Temp with an BM280 or SHT35 (you would prefere the Swiss one). Then I calculate (or rather more guess) the cloudage.
@@AndreasSpiess him gokd sir...i, too was thinking about suggesting an IR sensor, as the luxmeter will not 'see' the IR that is heating your home...still, if it sees overall brightness, it would give a 'useful' correlation to insolation...
This video arrived in the perfect time for me, today I was researching the detection of gamma and beta radiation using photodiodes as a replacement for Geiger-Müller tubes, so this video was very handy! Congrats for the 200k! :D
In my current project I am designing a sensor for determining distance with a red 5mW laser diode. I would like to be able to use these readings to control the response of a small dc stepper motor for controlling an X/Y axis small milling table. I have a lot to learn and I am sure this information will set me in the right direction. I started thinking about this project only earlier today, so this video came out with perfect timing for my needs. Thank You for all of your help over the past few years that I have been developing my understanding of these concepts.
hi Andreas i like those videos when you talk about different sensors and compare them keep up the good work , i think your channel will become the reference in sensors boards for arduino and MCUs
Thanks for the video. I've been using the GY-49 / MAX44009 module for measuring lux in my greenhouse (to calculate the total moles of light the plants receive). It claims to have a range up to 188,000 lux.
@@AndreasSpiess your method (I use a similar one) allows us to demonstrate to our peers and students that's to "err" is not just natural but part of a healthy learning process. By brushing our mistakes under the carpet or worse, denying them, we create an atmosphere of mistrust and fear that only hampers learning. Honestly Andreas, I wish there were more UA-cam instructors (and teachers in general) like you out there. Humanity would be far less self-destructive if there were.
Hi Andreas, video with lot of tehnical details and extensive comparison of different sensor available on the market. I feel bad that I did not find your video for this long time.. Once again Thank you so much.. I see a German accent in your explanation. Danke schön.
I worked in lighting for 10 years. The best light sensor in lux, in my humble opinion, is the Texas Instrument OPT3001. I scanned across the entire visible spectrum, from 380nm to 780nm in 20nm steps, and the sensor actually fits almost perfectly the CIE-defined eye sensitivity curve. It is the best light sensor I could measure. In addition to adjusting to the sensitivity of the human eye, it can measure from 0.1 lux to 100,000 lux. This dynamic range of measurement is amazing.
superb explanation of content and inclusion of initial targets and end results for complete understanding! God bless you this video came as ray of hope for my works!!
An LED will also work as an ambient light detection sensor. First used IIRC by HP in calculators, measuring the voltage of 7-segment displays where the LED segment was off, to determine the current to drive the display segments, according to ambient light. So you wouldn't suffer retinal damage in the lab from a display readable in outdoor daylight.
Hello Andreas, I have worked for TAOs since long before we were purchased by AMS. You are correct the TCS3200 is outdated. Also the TCS3414 as well. Neither of these have IR filters because when they were designed the technology to deposit the filter directly on the chip was new and very expensive. You are right the TCS34725 is the only AMS part you had that is not obsolete. BTW the Avago part is really an AMS part. TCS3223 I believe. AMS now has a TIme of Flight device you should try. This is very interesting part. Oh, and that TSL2561 was used as the Ambient Light Sensor by Apple in the very first iPhone. Those have a very wide dynamic range because you have an analog gain setting and you can also set the integration time from 2.7ms all the way up to 65535 x 2.7ms. Great video! DE N5BOC
Thankl oyu for the info. I still struggle with both sensors out in the sunlight. As far as I see TCS34725 is saturated and the TSL 2561 sometimes provides strange values. I have to investigate further and also ordered some ND filters to reduce the light energy.
Very interesting video, many of these devices I have never used. Things have certainly come a long way since the OCP71 (or OC71 with the paint scratched off!). I have a WEMOS which monitors the temperature and humidity in my loft, I think I will add a light sensor to warn if the loft lights are left on when the hatch is closed. Bob
Very glad to have found this channel and this video. Would like to have seen more info on UV sensors, particularly for a weather station to detect the UV level outdoors.
At 10:19 you're talking about the TSL2591 as not recommended, but the red box on the left doesn't include that sensor. Looking at the AMS website, it actually says "Industry leading dynamic range which drives both ultra low light sensitivity and bright sunlight operation" about that sensor.
Hi Andreas, thanks for another great video. I just want to mention that LEDs can also be used as light sensor for some applications. I am using an array of 3 LEDs for an ultra low cost 3-channel spectrometer.
I use the same sensor on a GY-49.breakout board. However every Arduino library for the MAX44009 I downloaded had the same mistake in reading lux levels correctly. Looks like everyone copied the same error from the original library 😁 It's an excellent lux level sensor with a huge 22 bit dynamic range, ultra low current and IR/UV blocking filters.
here is the code I use, is it okay ? #define DATA_LUX_HIGH 0x03 #define DATA_LUX_LOW 0x04 float MAX44009::getLux(void) { int luxHigh = readI2CAddr(DATA_LUX_HIGH); int luxLow = readI2CAddr(DATA_LUX_LOW); int exponent = ( luxHigh & 0xf0 ) >>4; int mant = ( luxHigh & 0x0f )
// Request 2 bytes of data Wire.requestFrom(0x4A, 1); // Read 2 bytes of data luminance msb, luminance lsb if (Wire.available() == 1) { data[1] = Wire.read(); } // Convert the data to lux int exponent = (data[0] & 0xF0) >> 4; int mantissa = ((data[0] & 0x0F)
@andreas If you use a lux sensor for your awning you'll need to keep adjusting it every winter for when you want to sit out in the sun. It might be better to use a temperature sensor instead. Or maybe mix both (lux and temperature) and make some Node-RED flow for it! Sounds like a project! 😉
Thank you for another interesting video in a series that will hopefully be continued 👍 At last no sensor or web based service can replace the wifes skills and predictions 😁
@@AndreasSpiess yes you are right, I checked my setup (weather station) again and it's an SI1145 which I use outdoor. I use the TEMT6000 as indoor sensor
You can get filters for most wavelengths which is handy in some projects to filter out other types of light you are not interested in measuring and which can interfere with your results.
Hi, thank you for such an amazing video and information. I am working on a project to detect the color intensity of liquid in a cuvette, I have tested TCS3472, AS7341 and TSL2591, the TSL2591 appeared to be a better choice but still I am struggling to get stable and consistent results, the solution in the cuvette is a saltwater sample from my aquarium and an Alkalinity test reagent added to it to measure the water alkalinity in dkh, the color goes from blue to green and then yellow in alkalinity to acidic condition, so if I measure the blue color intensity in water I should be able to find out the alkalinity level, the cuvette is covered in a 3d printed holder with a whole on one side for sensor and other side for LED the LED color I've used is orange which works with blue color, there is no stray or ambient light reaching to the sensor besides the orange LED, if I read the value multiple time I get different results with slight variations, I tried to average out as well but each average of 10 sample is slightly different from previous one, would you be able to suggest some good sensor or tips that may help me stabilize the results so I can get the consistent values. I would really appreciate your help. Thank you
I do not know the answer. However, I would first check without the liquid to see if you get stable values and on which level (if not yet done). Then, I would introduce the liquid and check if the amount of light still is sufficient. Otherwise, you only measure random noise. Third, I would use an incandescent lamp because it has a very broad spectrum to see if the LED has a proper spectrum.
You know the wife is paying attention when she starts mocking your engineering choices. :D Thanks for the comparison -- I hadn't even heard of half of these. Time for another shopping round...
Grüezi from Zurich! I have a project currently based heavily on home sensors for monitoring house plants. Light is obviously an important variable to monitor so have been prototyping with lots of different breakout boards. Currently developing the design on a PCB and I've found that a lot of the lux sensors are hard to source when building yourself. Currently, I'm working with the BH1715 (poor results), VEML6030 (excellent I2C), and the TEMT6000 (analog, quite primitive). Would love to experiment with the UV as this is also an important aspect of plant health so will try to integrate a few extra sensors into the next version!
@@AndreasSpiess So much, always. Just finished up the design this weekend and have the VEML6075 UV sensor onboard :D. Once i'm at a point where it's worth seeing, I'll send you one for your mailbag!
Frohe Ostern, Andreas und vielen Dank für die tollen Videos. Ich freue mich, dass du die 200.000 Abonnenten erreicht hast - ein großartiger Meilenstein: Gratulation! Bleib gesund in dieser schwierigen Zeit! Vielleicht brauchen wir jetzt eine Definition von dBs(ubscriber), dann wünsche ich alsbald +3dBs ;)
I recently use VEML6030/VEML7700 ALS sensor with perfect human eye response. Also have a look about VEML6040 RGB sensor. All from Vishay. Nice packaging, but a little bad interface
Have you considered putting sensors in solar garden lights? You get a solar panel, a battery and an enclosure for a relatively cheap price. For instance temperature and moisture measurements in the greenhouse. You probably need a buck-converter to get the 3.3v needed from a 1.2NiCd or 3.2LiOn, but still. If the ESP went into deep sleep, and only sends occasionally, I guess most lamps could supply the power.
Good idea. I started to build a combination of a Lux, an UV, and an RGB sensor. In addition it communicates via 2.4GHz LoRa. The MPU is an Arduino Mini which Igels it’s Takt from an RTC Module. Still a lot of LEDs, so I do not know the current consumption. But it needs some space.
For proximity and lux sensors there also is the SI114X series. I'm waiting for an SI1145 which can even approximate the UV Index by combining the IR and Visible Light sensors. Frohe Ostern
Greetings Andreas, I am working on a personal research project but because of quarantine I can not buy any precision light measuring sensors and had to use the LDR that I already had at hand, but due to LDR's extreme variance with the change in temperature, it is far from precise, stuck at home and cannot buy anything, so I had to come up with an alternative, so I am using LEDs as a light sensor, I have tested many LEDs and came to a conclusion that the LEDs which light up at least voltage give the best results. I kindly propose the idea for one of your future videos to analyze and show how feasible it is to use an LED as a light sensor. Regards, Tanishq
Tanishq - maybe you are interested in this project - a solar tracker that uses LEDs as the sensor - he has lots of info on his webpage www.redrok.com/led3xassm.htm#led3xc3
Clear 10mm LED, two digital pins = very good detection of sunlight levels. In a nutshell, the LED is capacitively charged by feeding a reverse voltage into it (pins set to OUTPUT), then the decay of that voltage is measured as it slowly leaks into the (now) INPUT pins of the MCU. The more light there is, the faster it discharges. I've used it successfully for quite a few projects, no ADC needed.
@@marcelachternaam9249 Search for "LED as light sensor". The sparkfun article explains how it works, the instructable article shows a sketch, the stackoverflow describes the principle, pasting the text here: Set Port A output high set Port B output low // this makes sure the led is discharged wait 1mS set Port A output low set Port B output high // reverse bias and charge wait 1mS set Port B as input // Port B is high impedance input time how long for Port B to read low
Danke für das wieder sehr informative Video, Andreas es sind die Sätze wie, meine Frau machte Witze über meine Fähigkeiten die mich jedes mal schmunzeln lassen und ich sozusagen eine tiefe Verbundenheit verspüre, mir geht es ja genauso. Oder so Sätze wie in deinem LoRaWAN Video als es darum ging die Pins vom Draginoboard einzutragen "Otherwise you scratch your head an nothing goes" (Video 117) soooooo genial :-) Hey schöne Ostern und bleibt gesund!
Ich bin leider etwas spät für Osterwünsche. Danke! Ja, auf diesem Kanal sind viele ähnlich Interessierte Leute anzutreffen. Ich finde das sehr schön. UA-cam sei dank können wir uns so finden und etwas vernünftiges unternehmen...
The uni-t lux sensor has a white plastic lens, I wonder if it does affect readings in any significant way? Since if we were "to deploy" a device with a light sensor it should be covered with something to protect it imho. Btw. I'm running like 2 years now a LDR to do stuff on sunset/sunrise and after fine tuning of sensitivity levels in software it works fine. The program also checks the time so the trigger runs only these 2 times a day.
I assume they took the color of the plastic into account when they built the device. So far I do not yet know how to protect the sensors. I tend to use glas, but currently all shops are loched down :-(
Could you 'clear' up what is meant by 'clear' on the light sensor? I would think they mean 'white', but that's just a software combination of R, B, and G, so I'm unsure if they might mean something else.
Andreas thank you for your vídeo. Excelent. I used APDS 9960 for a project wich needs to get light colour temperature (Kelvin) it is an excelent sensor.
2:19 - It doesn't have to be specifically R, G and B. You could also measure at Cyan, Magenta and Yellow. Or any other mutually complementary hues. Basically, any three hues that are at 120º from each other on a colour wheel can be used to determine the dominant wavelength. But, as you know, reading just the wavelength won't allow you to distinguish (for example) red from salmon pink, or yellow from brown, since those differ in brightness and saturation as well. You can calculate those in software if the sensor outputs the three channels (not just its own internal calculation of the wavelength) but, as you mention later in the video, many of those sensors have very irregular overlap between the channels.
You are right. When I was young I worked for a company called Gretag where we developed color meters for the printing industry. But these facts are too complicated for such a short video ;-)
Isn't Macbeth from the 17th century? ;-) Actually, I vaguely remember seeing something made by Gretag before the -Macbeth. It wasn't related to printing, though. Maybe light meters or something related with photography? My father used to be a cinematographer (retired now).
ive got an "APDS-9960" but i cannot get it to work... from what ive read its actually only a APDS-9930.. which doesnt have the RGB functionality (what i purchaced it for!)..seems there a few fakes getting about...Thanks Andreas, i know what to look for what purchasing light devices now!
@@AndreasSpiess yes the APDS-9930 is a real component.. i meant the fakes are that the boards are marked and being sold as APDS-9960 even tho its only got the APDS-9930 onboard
Can I ask you a question about UV-C sensors? I built a germicidal UV-C sterilizer from an old microwave oven and a UV-C mercury vapor water sterilization bulb, but I am wondering about the actual 254nm irradiance level at the disinfection target distance. The manufacturer has ignored my requests for irradiance specs on their bulb, and reviewers like BigClive have found that many UV-C bulbs from China are fake and give off little or no UV-C light. I'd like to get a reasonably accurate UV-C irradiance reading without investing in a UV-C laboratory photometer. I did a little research and I found the 1918 ANALOG UV SENSOR BREAKOUT Board and that looked promising except that it has very poor sensitivity in the germicidal UV-C wavelength range. But then I saw that the GUVC-S10GD photodiode has a spectral range of 220nm - 280nm, which is ideal! Do you think the GUVA-S12SD photodiode on the 1918 ANALOG UV SENSOR BREAKOUT Board could be de-soldered and replaced with the GUVC-S10GD photodiode (or the GUVC-S40GD, or something similar) to get a better UV-C sensor breakout for germicidal UV-C measurement projects? NOTE: I don’t have enough electronics experience yet to understand what sorts of modifications to breakout boards are possible, so I’m reaching out to engineers & makers who have more expertise. It would be great to open source a diy project for building a UV-C photometer around an Arduino and some inexpensive off-the-shelf components.
There has to be this intuitive bond. I am currently working on a project in which I measure the UV-A and UV-B radiation, and if the brightness is exceeded, the shutters on the windows are to be automatically shut down until the UV-A and UV-B levels are again a comfortable level in the room reached. 🤗😘 In my Project I desire to use GY-6075 Light Sensor Modul and BMP280 for Weather Control with a Node MCU ESP32 Dev Board and a 433 MHz RF Transmitter to control the Somfy Window automation.
Hello Andreas, how are you? Your video is excellent, but I need to make a question and it is, how I do to calibrate the sensor of light TSL2561? Thank you for making this video, Greeting
What sensor would you use to detect color or brightness change on a little area (i.e. 1 cm2) on an LCD screen? Sensing speed is important. The faster the better. Color is unimportant, I need to detect if anything changed on the little area being watched or not. Sensor size is the smaller the better.
I really like and appreciate your videos. I am currently using TSL2561 sensor witch includes visibible + near infrared diode and another wide range infrared diode, which is very interresting because it lets me calculate the Lux by subtracting IR reading from Visible+IR. This help me measuring light illuminance in Lux with high precision by eliminating IR interference. This sensor also gives me the possibility to measure light power with good precision by using light+near infrared diode and wide range infrared diode and using a factor calculated for most common sun light spectre profil. The problem I have with this sensor is that the range is quite low and didn't exceed 40000 Lux which is not enough for day sun light especially here in Morocco. I didn't find this information in your video when you talked about TSL2561. Could you please give me your opinion about this issue and if there is a good alternative to this sensor for my application: Weather and fertilisation station for Agricuture. For more details here bellow a link to my collected data. Thanks a lot for your high level and extraordinary video content: www.abrobotics.be/siliconceptsarl/METEO/meteoframe.php
I just started using the sensor in full sun (and our sun is much less than yours). I also ordered ND filters for protection, but also to adapt the maximum Lux level to the capabilities of the sensor.
@@AndreasSpiess Ok, I didn't understand ND but I'll wait for you tests. I tried the same method by using small grey plastic part protection and it gives good results for illuminance in day light and also IR is not strongly influenced. The problem I have to solve now is to determine the factor to use for recalibrating Lux+near IR and IR sensor. If you see the charts you'll notice that IR reading is not always proportional to Lux especially when it is cloudy which indicates that IR is still measuring with good precision even with my DIY filter. Here is the link: www.abrobotics.be/siliconceptsarl/METEO/meteoframe.php
I found your video interesting. Could you advise me on this... At my job, we cut steel shavings that we stack on top of each other. I would like to build a device that, using a CCD sensor or camera, can vertically scan the stack of pieces and count how many are in the stack. What sensor would you recommend? Thank you.
The camera is not important and has to be selected by the program and HW you use. Most people would recommend to use a kind of "AI" for such a purpose. Maybe you find a similar project using google.
The TCS3200 may have use in non digital applications. Life is about change, and, that everything repeats itself. I know, that's odd, but if the digital age ended, (EMP?) you could essentially hook one up to an antiquated panel meter via transistor and smoothing filters.
Hello sir, thank you very much for this video, I need advice from you, I need to build a sonometer with decibel and i need a sensor that can give me analog values of sound, can you recommend me a sensors ?? thanks
No. Most only measure intensity and you have to calculate the watt produced by your solar cell. I would measure the watts produced by your solar cell for that purpose
Hi Andreas, I want to use the RGB sensor to detect the color of the water in a transparent tube, will it work? My idea is to read the PH/Alkalinity level of water of water test kit. If it works, which is the best RGB sensor choice?
I am no specialist in this matter. I think you have to try your particular scenario and also if you want to use reflected light or shine through the liquid.
Hi, the text at 4:00 is too small,
and thanks for the great content
Thanks for the feedback. I started to use OBS and did not get everything right. Next time I will change it.
@@AndreasSpiess you can just crop the view in your video editing software.
@@TheHellisYea but in OBS he can just drag it to be bigger part of the screen which is easier than editing software.
Timely recording for me. I was trying to get UV readings with a VEML6075. Reads nicely but discovered it is very directional. I'd need a solar tracker for it to measure solar UV consistently.
I like your videos, Andreas. They have a good "signal/noise ratio"
Cool, thanks
i.e. concise
These sensor / board comparison videos are my favorite. They're always informative and help me out in my projects.
Thank you!
Congrats: You hit the 200.000!! And again a great Video. Thanks for that and stay healthy. Have great easter holidays.
Thank you. As an engineer I think in dB. So the next important step is 1 mio ;-) Happy Easter, too!
Yes he deserve for it
I want to thank you for your work. You've been so helpful, every time I want to buy a new sensor for a project there's a video of you trying the most common available and making a thoroughly made review about it. So, thank you you've made much lighter the work otherwise I've had to make in order to choose the right sensor for my project
You are welcome. Glad my channel is helpful for you!
Thanks again for a great video. I specially enjoy these compilation videos which compare similar devices and helps us followers save a lot of time searching for the info you kindly bring us in the usual well structured and easy to follow videos
Glad you enjoyed it! These compilations help me too ;-)
Congrats Andreas!! Your videos should be within a Public Electronic Wikipedia as a reference! Greetings from Spain!
Maybe one day ;-) Thank you
Well that video brightened a dull day (sorry).
Learnt a lot - never realised the sensors where so different. I’ll be following up the gesture sensor, thank you for publishing your code for that.
One thing to note - a significant amount of ‘white’ paper is actually blue to give a clean cold look rather than a warm tint.
You are right for teh paper (AFAIK not in all countries equally). But here I measured the spectrum of the diodes without paper. So both effects would add up :-(
Thanks for dedicating your time to another great video! :)
Right now we are developing a project that needs a light sensor and after a close examination we've went fo the same IC TCS34725.
Very cool!
Thank you for your effort and time spent on this Topic. Regarding your awning: What also could be helpful for you is the cloudage. I am measuring far IR Temp with a Melexis MLX90614 and environmental Temp with an BM280 or SHT35 (you would prefere the Swiss one). Then I calculate (or rather more guess) the cloudage.
This is a very interesting concept! Maybe I add a few additional sensors to my Lux meter and compare their readings...
@@AndreasSpiess him gokd sir...i, too was thinking about suggesting an IR sensor, as the luxmeter will not 'see' the IR that is heating your home...still, if it sees overall brightness, it would give a 'useful' correlation to insolation...
I would love to see a Video on different sensor types and comparing the usefulness for this scenario.
Thanks for the informative video. I made a rgb sensing setup yesterday using the AS7262 chip. Though expensive, it is easy to work with and accurate.
I looked at the sensor in the past, but was driven away by the high price. Now I ordered one. Thanks for the tip!
Thanks, Andreas! That will help me for an advanced line following modul. See you next UKW conference hopefully!
We will see if hte borders are open till then ;-)
This video arrived in the perfect time for me, today I was researching the detection of gamma and beta radiation using photodiodes as a replacement for Geiger-Müller tubes, so this video was very handy!
Congrats for the 200k! :D
Thank you! And good luck with your project.
Worked so hard trying to fix the APDS9960 with no luck, thank you!!!
You are welcome!
In my current project I am designing a sensor for determining distance with a red 5mW laser diode. I would like to be able to use these readings to control the response of a small dc stepper motor for controlling an X/Y axis small milling table. I have a lot to learn and I am sure this information will set me in the right direction. I started thinking about this project only earlier today, so this video came out with perfect timing for my needs. Thank You for all of your help over the past few years that I have been developing my understanding of these concepts.
Maybe you look also at the TOF and Lidar sensors.
You have covered all sensors in my treasure. Nice explanation. Thanks !!! ✨✨✌
My pleasure 😊
hi Andreas i like those videos when you talk about different sensors and compare them keep up the good work , i think your channel will become the reference in sensors boards for arduino and MCUs
Glad you like them!
Thanks for the video. I've been using the GY-49 / MAX44009 module for measuring lux in my greenhouse (to calculate the total moles of light the plants receive). It claims to have a range up to 188,000 lux.
Thanks for the tip!
I love the way Andreas shares his own mistakes and that self-deprecating humour is wonderful!
Thank you. We all make mistakes. The difference is how we deal with it...
@@AndreasSpiess your method (I use a similar one) allows us to demonstrate to our peers and students that's to "err" is not just natural but part of a healthy learning process. By brushing our mistakes under the carpet or worse, denying them, we create an atmosphere of mistrust and fear that only hampers learning.
Honestly Andreas, I wish there were more UA-cam instructors (and teachers in general) like you out there. Humanity would be far less self-destructive if there were.
Hi Andreas, video with lot of tehnical details and extensive comparison of different sensor available on the market. I feel bad that I did not find your video for this long time.. Once again Thank you so much.. I see a German accent in your explanation. Danke schön.
Thank you for your nice words!
I worked in lighting for 10 years. The best light sensor in lux, in my humble opinion, is the Texas Instrument OPT3001. I scanned across the entire visible spectrum, from 380nm to 780nm in 20nm steps, and the sensor actually fits almost perfectly the CIE-defined eye sensitivity curve. It is the best light sensor I could measure. In addition to adjusting to the sensitivity of the human eye, it can measure from 0.1 lux to 100,000 lux. This dynamic range of measurement is amazing.
Thank you for the tip. I ordered one.
Happy Easter, Andreas!
Thanks for another great video!
You are welcome!
Thanks for a great video! I use VEML7700 or VEML6030 mostly for it's low sleep current. Not at Aliexpress but at Adafruit as a module.
Thanks for sharing!
superb explanation of content and inclusion of initial targets and end results for complete understanding! God bless you this video came as ray of hope for my works!!
Glad it was helpful!
An LED will also work as an ambient light detection sensor. First used IIRC by HP in calculators, measuring the voltage of 7-segment displays where the LED segment was off, to determine the current to drive the display segments, according to ambient light. So you wouldn't suffer retinal damage in the lab from a display readable in outdoor daylight.
I assume they use the same effect as the phototransistors. But I di not know that they are very sensitive.
Hello Andreas,
I have worked for TAOs since long before we were purchased by AMS. You are correct the TCS3200 is outdated. Also the TCS3414 as well. Neither of these have IR filters because when they were designed the technology to deposit the filter directly on the chip was new and very expensive. You are right the TCS34725 is the only AMS part you had that is not obsolete. BTW the Avago part is really an AMS part. TCS3223 I believe.
AMS now has a TIme of Flight device you should try. This is very interesting part.
Oh, and that TSL2561 was used as the Ambient Light Sensor by Apple in the very first iPhone. Those have a very wide dynamic range because you have an analog gain setting and you can also set the integration time from 2.7ms all the way up to 65535 x 2.7ms.
Great video!
DE N5BOC
Thankl oyu for the info. I still struggle with both sensors out in the sunlight. As far as I see TCS34725 is saturated and the TSL 2561 sometimes provides strange values. I have to investigate further and also ordered some ND filters to reduce the light energy.
Very interesting video, many of these devices I have never used. Things have certainly come a long way since the OCP71 (or OC71 with the paint scratched off!). I have a WEMOS which monitors the temperature and humidity in my loft, I think I will add a light sensor to warn if the loft lights are left on when the hatch is closed. Bob
Yea, the world became simpler and better in this respect.
Very glad to have found this channel and this video. Would like to have seen more info on UV sensors, particularly for a weather station to detect the UV level outdoors.
You can find other UV sensors. Just make sure that the protection does not block UV (e.g. Glas blocks UV)
Thanks for a lovely video on light sensors! Really appreciate your work and dedication.
My pleasure!
Very comprehensive as usual and a superb reference for the future.
Glad you think so!
Sure, Maaaaan. I like your form to explaint the sensors that I use, thank you very much
You are welcome!
At 10:19 you're talking about the TSL2591 as not recommended, but the red box on the left doesn't include that sensor. Looking at the AMS website, it actually says "Industry leading dynamic range which drives both ultra low light sensitivity and bright sunlight operation" about that sensor.
I just read the content of this website. It is more for professional use in products
a useful and educational video. liked it very much. The information you provided useful very well for me.
Glad to hear that! Have a nice Sunday!
Hi Andreas, thanks for another great video. I just want to mention that LEDs can also be used as light sensor for some applications. I am using an array of 3 LEDs for an ultra low cost 3-channel spectrometer.
Here is a link wiki.analog.com/university/courses/electronics/electronics-lab-led-sensor
Thank you for the link!
Always interesting ideas to get inspiration in the Arduino community thanks for sharing.
Glad you enjoyed it!
As always very informative and useful. Thank you Andreas :)
My pleasure!
Bloody hell, what a great video :-) Thanks a 1 000 000 for the content, very interesting and illuminating informations.
Glad you enjoyed it
Hello, I use the MAX44009 lux sensor from Maxim integrated, widely available from Aliexpress. I use it to adapt the brightness of my displays
Thanks for the info!
I use the same sensor on a GY-49.breakout board. However every Arduino library for the MAX44009 I downloaded had the same mistake in reading lux levels correctly. Looks like everyone copied the same error from the original library 😁
It's an excellent lux level sensor with a huge 22 bit dynamic range, ultra low current and IR/UV blocking filters.
here is the code I use, is it okay ?
#define DATA_LUX_HIGH 0x03
#define DATA_LUX_LOW 0x04
float MAX44009::getLux(void)
{
int luxHigh = readI2CAddr(DATA_LUX_HIGH);
int luxLow = readI2CAddr(DATA_LUX_LOW);
int exponent = ( luxHigh & 0xf0 ) >>4;
int mant = ( luxHigh & 0x0f )
// Request 2 bytes of data
Wire.requestFrom(0x4A, 1);
// Read 2 bytes of data luminance msb, luminance lsb
if (Wire.available() == 1)
{
data[1] = Wire.read();
}
// Convert the data to lux
int exponent = (data[0] & 0xF0) >> 4;
int mantissa = ((data[0] & 0x0F)
Great Work Andreas!
Thanks a lot!
This video is right up my alley
Good to know! Thanks.
@andreas If you use a lux sensor for your awning you'll need to keep adjusting it every winter for when you want to sit out in the sun. It might be better to use a temperature sensor instead. Or maybe mix both (lux and temperature) and make some Node-RED flow for it!
Sounds like a project! 😉
node red should be able to do some simple if-then-else to avoid wrong extractions...
Thank you for another interesting video in a series that will hopefully be continued 👍 At last no sensor or web based service can replace the wifes skills and predictions 😁
You bet!
Great test as always
Thanks for sharing 👍😁
My pleasure!
Yes there is another one which I prefer: The TEMT6000. I also use it to measure sunlight and control my sun roof
Another viewer in the comments below reported that this one maxes out quite early. Do you have a trick?
@@AndreasSpiess yes you are right, I checked my setup (weather station) again and it's an SI1145 which I use outdoor. I use the TEMT6000 as indoor sensor
Very comprehensive review.. .quite helpful to have for future reference
Glad you liked it
Thanks for such a nice sensors comparison video.
Hope this amount will buy you are beer during your vacation! Enjoy your summer :)
Yes, for sure. Thank you! I too hope you will have a pleasant summer
You can get filters for most wavelengths which is handy in some projects to filter out other types of light you are not interested in measuring and which can interfere with your results.
You are right. We live in a very good time.
Hi, thank you for such an amazing video and information.
I am working on a project to detect the color intensity of liquid in a cuvette, I have tested TCS3472, AS7341 and TSL2591, the TSL2591 appeared to be a better choice but still I am struggling to get stable and consistent results, the solution in the cuvette is a saltwater sample from my aquarium and an Alkalinity test reagent added to it to measure the water alkalinity in dkh, the color goes from blue to green and then yellow in alkalinity to acidic condition, so if I measure the blue color intensity in water I should be able to find out the alkalinity level, the cuvette is covered in a 3d printed holder with a whole on one side for sensor and other side for LED the LED color I've used is orange which works with blue color, there is no stray or ambient light reaching to the sensor besides the orange LED, if I read the value multiple time I get different results with slight variations, I tried to average out as well but each average of 10 sample is slightly different from previous one, would you be able to suggest some good sensor or tips that may help me stabilize the results so I can get the consistent values. I would really appreciate your help. Thank you
I do not know the answer. However, I would first check without the liquid to see if you get stable values and on which level (if not yet done). Then, I would introduce the liquid and check if the amount of light still is sufficient. Otherwise, you only measure random noise.
Third, I would use an incandescent lamp because it has a very broad spectrum to see if the LED has a proper spectrum.
You know the wife is paying attention when she starts mocking your engineering choices. :D Thanks for the comparison -- I hadn't even heard of half of these. Time for another shopping round...
No hurry. Logistics anyway does not work properly for the moment. I am waiting and waiting...
Grüezi from Zurich! I have a project currently based heavily on home sensors for monitoring house plants. Light is obviously an important variable to monitor so have been prototyping with lots of different breakout boards. Currently developing the design on a PCB and I've found that a lot of the lux sensors are hard to source when building yourself. Currently, I'm working with the BH1715 (poor results), VEML6030 (excellent I2C), and the TEMT6000 (analog, quite primitive). Would love to experiment with the UV as this is also an important aspect of plant health so will try to integrate a few extra sensors into the next version!
Good project. A lot to learn, I think.
@@AndreasSpiess So much, always. Just finished up the design this weekend and have the VEML6075 UV sensor onboard :D. Once i'm at a point where it's worth seeing, I'll send you one for your mailbag!
I really like the last sensor
Me too ;-)
Frohe Ostern, Andreas und vielen Dank für die tollen Videos. Ich freue mich, dass du die 200.000 Abonnenten erreicht hast - ein großartiger Meilenstein: Gratulation! Bleib gesund in dieser schwierigen Zeit! Vielleicht brauchen wir jetzt eine Definition von dBs(ubscriber), dann wünsche ich alsbald +3dBs ;)
Ich bin etwas spät für Osterwünsche. Dankeschön! Betreffend 200'000: Da denke ich in 10 dB und konzentriere mich auf die Million ;-)
I recently use VEML6030/VEML7700 ALS sensor with perfect human eye response. Also have a look about VEML6040 RGB sensor. All from Vishay. Nice packaging, but a little bad interface
Thank you for sharing your experience!
Nice video. Btw, don't worry, my wife also laughs with my experiments quite often.... so you're not alone ;-)
Good to hear! Not really dangerous for usi, I think ;-)
Thanks in advance for your comments
Have you considered putting sensors in solar garden lights? You get a solar panel, a battery and an enclosure for a relatively cheap price. For instance temperature and moisture measurements in the greenhouse.
You probably need a buck-converter to get the 3.3v needed from a 1.2NiCd or 3.2LiOn, but still.
If the ESP went into deep sleep, and only sends occasionally, I guess most lamps could supply the power.
Good idea. I started to build a combination of a Lux, an UV, and an RGB sensor. In addition it communicates via 2.4GHz LoRa. The MPU is an Arduino Mini which Igels it’s Takt from an RTC Module. Still a lot of LEDs, so I do not know the current consumption. But it needs some space.
For proximity and lux sensors there also is the SI114X series. I'm waiting for an SI1145 which can even approximate the UV Index by combining the IR and Visible Light sensors.
Frohe Ostern
Interesting sensor! Ordered ;-) Thank you
Greetings Andreas, I am working on a personal research project but because of quarantine I can not buy any precision light measuring sensors and had to use the LDR that I already had at hand, but due to LDR's extreme variance with the change in temperature, it is far from precise, stuck at home and cannot buy anything, so I had to come up with an alternative, so I am using LEDs as a light sensor, I have tested many LEDs and came to a conclusion that the LEDs which light up at least voltage give the best results.
I kindly propose the idea for one of your future videos to analyze and show how feasible it is to use an LED as a light sensor.
Regards,
Tanishq
Tanishq - maybe you are interested in this project - a solar tracker that uses LEDs as the sensor - he has lots of info on his webpage www.redrok.com/led3xassm.htm#led3xc3
@@LarryKapp1 Thank you for sharing this project.
Clear 10mm LED, two digital pins = very good detection of sunlight levels. In a nutshell, the LED is capacitively charged by feeding a reverse voltage into it (pins set to OUTPUT), then the decay of that voltage is measured as it slowly leaks into the (now) INPUT pins of the MCU. The more light there is, the faster it discharges. I've used it successfully for quite a few projects, no ADC needed.
Good idea. Thanks!
Do you have an example of this?
@@marcelachternaam9249 Search for "LED as light sensor". The sparkfun article explains how it works, the instructable article shows a sketch, the stackoverflow describes the principle, pasting the text here:
Set Port A output high
set Port B output low // this makes sure the led is discharged
wait 1mS
set Port A output low
set Port B output high // reverse bias and charge
wait 1mS
set Port B as input // Port B is high impedance input
time how long for Port B to read low
Danke für das wieder sehr informative Video, Andreas es sind die Sätze wie, meine Frau machte Witze über meine Fähigkeiten die mich jedes mal schmunzeln lassen und ich sozusagen eine tiefe Verbundenheit verspüre, mir geht es ja genauso. Oder so Sätze wie in deinem LoRaWAN Video als es darum ging die Pins vom Draginoboard einzutragen "Otherwise you scratch your head an nothing goes" (Video 117) soooooo genial :-) Hey schöne Ostern und bleibt gesund!
Ich bin leider etwas spät für Osterwünsche. Danke! Ja, auf diesem Kanal sind viele ähnlich Interessierte Leute anzutreffen. Ich finde das sehr schön. UA-cam sei dank können wir uns so finden und etwas vernünftiges unternehmen...
The uni-t lux sensor has a white plastic lens, I wonder if it does affect readings in any significant way? Since if we were "to deploy" a device with a light sensor it should be covered with something to protect it imho.
Btw. I'm running like 2 years now a LDR to do stuff on sunset/sunrise and after fine tuning of sensitivity levels in software it works fine. The program also checks the time so the trigger runs only these 2 times a day.
I assume they took the color of the plastic into account when they built the device. So far I do not yet know how to protect the sensors. I tend to use glas, but currently all shops are loched down :-(
Could you 'clear' up what is meant by 'clear' on the light sensor? I would think they mean 'white', but that's just a software combination of R, B, and G, so I'm unsure if they might mean something else.
You can see teh curve in the data sheet. It seems to be a widebanc sensor.
Good video Andreas. Greetings from Ensenada Mexico
Hello there! Thank you
Andreas thank you for your vídeo. Excelent. I used APDS 9960 for a project wich needs to get light colour temperature (Kelvin) it is an excelent sensor.
Thank you for sharing your experience!
2:19 - It doesn't have to be specifically R, G and B. You could also measure at Cyan, Magenta and Yellow. Or any other mutually complementary hues. Basically, any three hues that are at 120º from each other on a colour wheel can be used to determine the dominant wavelength.
But, as you know, reading just the wavelength won't allow you to distinguish (for example) red from salmon pink, or yellow from brown, since those differ in brightness and saturation as well. You can calculate those in software if the sensor outputs the three channels (not just its own internal calculation of the wavelength) but, as you mention later in the video, many of those sensors have very irregular overlap between the channels.
You are right. When I was young I worked for a company called Gretag where we developed color meters for the printing industry. But these facts are too complicated for such a short video ;-)
Gretag-Macbeth, later merged into X-Rite? I think I used a couple of their calibrators some 15 or 20 years ago.
Yes. Even before Macbeth...
Isn't Macbeth from the 17th century? ;-)
Actually, I vaguely remember seeing something made by Gretag before the -Macbeth. It wasn't related to printing, though. Maybe light meters or something related with photography? My father used to be a cinematographer (retired now).
Great vid Andreas!
I would really love instructional on how to make a colorimeter with ESP32.
This is a very special area, so the chance I will cover it is small.
@@AndreasSpiess Sad to hear but, thanks for the reply and thanks for your great videos.
great work Andres
Thank you!
wow , I thought I had alot of light sensor modules to test, wow Some of these I have never seen.
And during cleanup I found two additional ones ;-)
Very usefull these overviews of sensors.
Thank you!
ive got an "APDS-9960" but i cannot get it to work... from what ive read its actually only a APDS-9930.. which doesnt have the RGB functionality (what i purchaced it for!)..seems there a few fakes getting about...Thanks Andreas, i know what to look for what purchasing light devices now!
I also found aan APDS9930 during my sensor "reconciliation" . But I did not try it.It is only proximity and ambient light. So probably no fake.
@@AndreasSpiess yes the APDS-9930 is a real component.. i meant the fakes are that the boards are marked and being sold as APDS-9960 even tho its only got the APDS-9930 onboard
Hi Andreas could you please do a tutorial of a NBE-IOT hardware combined with a simple sensor????
Unfortunately we still have no NB-IOT in Switzerland.
really great and extraordinary content ... but have just a view sensors you already mentioned.
Thank you!
Can I ask you a question about UV-C sensors?
I built a germicidal UV-C sterilizer from an old microwave oven and a UV-C mercury vapor water sterilization bulb, but I am wondering about the actual 254nm irradiance level at the disinfection target distance. The manufacturer has ignored my requests for irradiance specs on their bulb, and reviewers like BigClive have found that many UV-C bulbs from China are fake and give off little or no UV-C light. I'd like to get a reasonably accurate UV-C irradiance reading without investing in a UV-C laboratory photometer.
I did a little research and I found the 1918 ANALOG UV SENSOR BREAKOUT Board and that looked promising except that it has very poor sensitivity in the germicidal UV-C wavelength range. But then I saw that the GUVC-S10GD photodiode has a spectral range of 220nm - 280nm, which is ideal!
Do you think the GUVA-S12SD photodiode on the 1918 ANALOG UV SENSOR BREAKOUT Board could be de-soldered and replaced with the GUVC-S10GD photodiode (or the GUVC-S40GD, or something similar) to get a better UV-C sensor breakout for germicidal UV-C measurement projects?
NOTE: I don’t have enough electronics experience yet to understand what sorts of modifications to breakout boards are possible, so I’m reaching out to engineers & makers who have more expertise. It would be great to open source a diy project for building a UV-C photometer around an Arduino and some inexpensive off-the-shelf components.
I have no experience with such sensors, so I am no help
There has to be this intuitive bond. I am currently working on a project in which I measure the UV-A and UV-B radiation, and if the brightness is exceeded, the shutters on the windows are to be automatically shut down until the UV-A and UV-B levels are again a comfortable level in the room reached. 🤗😘
In my Project I desire to use GY-6075 Light Sensor Modul and BMP280 for Weather Control with a Node MCU ESP32 Dev Board and a 433 MHz RF Transmitter to control the Somfy Window automation.
Thanks for sharing your experience. I try to add aslo a UV sensor to my Lux sensor to compare the readings...
Hello Andreas, how are you? Your video is excellent, but I need to make a question and it is, how I do to calibrate the sensor of light TSL2561? Thank you for making this video, Greeting
As always with calibrating you need a precise instrument and compare the results. I do not know such instruments, but I am sure they exist.
What sensor would you use to detect color or brightness change on a little area (i.e. 1 cm2) on an LCD screen? Sensing speed is important. The faster the better. Color is unimportant, I need to detect if anything changed on the little area being watched or not.
Sensor size is the smaller the better.
I do not know. Maybe you need a lens in addition
@@AndreasSpiess Thank you!
For analysis of food & fruit contents like fat carbohydrates etc which sensor is good for getting the wavelength of 700- 2300nm
I do not know these NIR sensors. I only used IR transistors which work on 900nM. The AS7263 covers 600-900nm for example.
It is helpful for my quantum photonics research.
Glad to help!
@@AndreasSpiess. God bless Sir.
excellent video, what would be the best sensor to measure the color of a liquid inside a test tube?
I do not know :-(
Frohe Ostern Andreas!
Danke. Leider bin ich für ein "Ebenfalls" zu spät :-(
what wavelength laser and what detector would you recommend for an outdoor laser wire during daylight hours?
They usually use IR lasers with filters.
I really like and appreciate your videos. I am currently using TSL2561 sensor witch includes visibible + near infrared diode and another wide range infrared diode, which is very interresting because it lets me calculate the Lux by subtracting IR reading from Visible+IR. This help me measuring light illuminance in Lux with high precision by eliminating IR interference. This sensor also gives me the possibility to measure light power with good precision by using light+near infrared diode and wide range infrared diode and using a factor calculated for most common sun light spectre profil. The problem I have with this sensor is that the range is quite low and didn't exceed 40000 Lux which is not enough for day sun light especially here in Morocco. I didn't find this information in your video when you talked about TSL2561. Could you please give me your opinion about this issue and if there is a good alternative to this sensor for my application: Weather and fertilisation station for Agricuture. For more details here bellow a link to my collected data. Thanks a lot for your high level and extraordinary video content:
www.abrobotics.be/siliconceptsarl/METEO/meteoframe.php
I just started using the sensor in full sun (and our sun is much less than yours). I also ordered ND filters for protection, but also to adapt the maximum Lux level to the capabilities of the sensor.
@@AndreasSpiess Ok, I didn't understand ND but I'll wait for you tests. I tried the same method by using small grey plastic part protection and it gives good results for illuminance in day light and also IR is not strongly influenced. The problem I have to solve now is to determine the factor to use for recalibrating Lux+near IR and IR sensor. If you see the charts you'll notice that IR reading is not always proportional to Lux especially when it is cloudy which indicates that IR is still measuring with good precision even with my DIY filter. Here is the link: www.abrobotics.be/siliconceptsarl/METEO/meteoframe.php
I found your video interesting.
Could you advise me on this...
At my job, we cut steel shavings that we stack on top of each other.
I would like to build a device that, using a CCD sensor or camera, can vertically scan the stack of pieces and count how many are in the stack.
What sensor would you recommend?
Thank you.
The camera is not important and has to be selected by the program and HW you use. Most people would recommend to use a kind of "AI" for such a purpose. Maybe you find a similar project using google.
which would be best sensitivity to detect light in bedroom when sleeping from curtain leakage?
ive gone for the Adafruit TSL2591 High Dynamic Range Digital Light Sensor as the Lux range is
from 0.000118 to 88,000
The TCS3200 may have use in non digital applications. Life is about change, and, that everything repeats itself. I know, that's odd, but if the digital age ended, (EMP?) you could essentially hook one up to an antiquated panel meter via transistor and smoothing filters.
I am not so sure if it is less prone to EMP. Still some silicone on it ;-)
Thanks! This video useful for me.I wonder how to create a strobe lux meter
I assume you need a sample and hold circuit used in ADCs to hold the maximum voltage of a sensor.
@@AndreasSpiess Thx
Hi, thanks for the video! Do you know where I can get an IR filter for the TCS3200?
No, I never had to buy one.
Great video! Thank you
You are welcome!
Hello sir, thank you very much for this video, I need advice from you, I need to build a sonometer with decibel and i need a sensor that can give me analog values of sound, can you recommend me a sensors ?? thanks
I do not know, but a quick goolge search showed that microphones or piezo elements are used for this purpose.
Can the TCS34725 detect the colour of liquids in transparent containers accurately?
I never tried. But if you shine light through it I assume it would work. But it only has 4 filters. Read the datasheet first.
awesome as always
Thank you!
Do you know any sensor for direct reading sunlight in watts?
No. Most only measure intensity and you have to calculate the watt produced by your solar cell. I would measure the watts produced by your solar cell for that purpose
Very good work, sir. I have a BH 1750 light sensor.
Is it possible to change the lux value of BH1750 sensor in W/m2 ? Thank you
I do not know. Maybe you consult its datasheet.
Hi Andreas,
I want to use the RGB sensor to detect the color of the water in a transparent tube, will it work? My idea is to read the PH/Alkalinity level of water of water test kit. If it works, which is the best RGB sensor choice?
I cannot answer you question. It is too specific, and I do not have a lot of experience with such sensors. Maybe google knows?
Do you think these RGB sensors would work well to read liquids? if this is the case, do you think we should add extra ilumination?
I am no specialist in this matter. I think you have to try your particular scenario and also if you want to use reflected light or shine through the liquid.
Well done. (again) Thanks.
You are welcome!
Thank you very much for this video! I found it very useful :)
Glad to hear that!
congrats on 200k
Thank you!
Frohe Ostern, Andreas!
Danke. Ebenfalls!