My readings with the TCS34725 have been all over the map. With the gain at 1x and integration_time at 700ms over a period of 12 hours my PAR out put ranges from 130 up to 430. Same location, same light, same dimming level. Sometimes the color readings are strange as well, like 95% blue 85% green 12% red. Then the next reading is different.
I feel like I'm missing a peice to the puzzle.
Arduino PAR and LUX Meters
-
- LED-Curious
- Reactions:
- Posts: 23
- Joined: Thu Dec 31, 2020 10:39 pm
@stevetheganjafarmer
Not familiar with your program parameters, just basic Arduino IDE.00schteven wrote: ↑Wed Jan 20, 2021 5:33 amMy readings with the TCS34725 have been all over the map. With the gain at 1x and integration_time at 700ms over a period of 12 hours my PAR out put ranges from 130 up to 430. Same location, same light, same dimming level. Sometimes the color readings are strange as well, like 95% blue 85% green 12% red. Then the next reading is different.
I feel like I'm missing a peice to the puzzle.
The TCS34725 Raw “Data” has to be “Read”, “Converted” then used to “Calculate” the required measurements (No separate Library file)... as shown in the Arduino “void loop” example below and... https://github.com/ControlEverythingCom ... S34725.ino
Code: Select all
void loop()
{
unsigned int data[8];
// Start I2C Transmission on the device
Wire.beginTransmission(Addr);
// Select data register
Wire.write(0x94);
// Stop I2C Transmission on the device
Wire.endTransmission();
// Request 8 byte of data from the device
Wire.requestFrom(Addr, 8);
// Read 8 bytes of data
// cData lsb, cData msb, red lsb, red msb, green lsb, green msb, blue lsb, blue msb
if (Wire.available() == 8)
{
data[0] = Wire.read();
data[1] = Wire.read();
data[2] = Wire.read();
data[3] = Wire.read();
data[4] = Wire.read();
data[5] = Wire.read();
data[6] = Wire.read();
data[7] = Wire.read();
}
// Convert the data
float cData = (data[1] * 256.0) + data[0];
float red = (data[3] * 256.0) + data[2];
float green = (data[5] * 256.0) + data[4];
float blue = (data[7] * 256.0) + data[6];
// Calculate the final lux
float luminance = (-0.32466 * red) + (1.57837 * green) + (-0.73191 * blue);
// Calculate RGB Percentage and CCT
// float blueP = blue * 100 / (red + green + blue)
// float greenP = green * 100 / (red + green + blue)
// float redP = red * 100 / (red + green + blue)
// float cct = 3810 * (blue / red) + 1319;
// Output data to serial monitor
Serial.print("Red color Luminance : ");
Serial.print(red);
Serial.println(" lux");
Serial.print("Green color Luminance : ");
Serial.print(green);
Serial.println(" lux");
Serial.print("Blue color Luminance : ");
Serial.print(blue);
Serial.println(" lux");
Serial.print("IR Luminance : ");
Serial.print(cData);
Serial.println(" lux");
Serial.print("Ambient Light Luminance : ");
Serial.print(luminance);
Serial.println(" lux");
delay(500);
}
The "Converted RGB Data" should "Calculate" Relative Percentage (RGB ratio) to be usable, e.g., R% = R * 100 / (R + G + B), G% = G * 100 / (R + G + B), etc.
-
- LED-Curious
- Reactions:
- Posts: 23
- Joined: Thu Dec 31, 2020 10:39 pm
I manage my esp32 and TCS34725 with esphome. It reports R% G% B% C% lux and color temp. Even with just that my readings are not consistent at all. All my set up variables do not change, same location, same light source. At the moment I do not have a dome over it.
@stevetheganjafarmer
- Salmonetin
- LED Enthusiast
- Reactions:
- Posts: 51
- Joined: Wed Jul 15, 2020 5:13 pm
...new nfo and no new likes then...back to home and back to mute Wilson...
...unos se quedaron con la vieja info del quote y otros con la nueva info que añadi a posteriori...o no...
...sigo actualizandola con nueva info que vi hoy... mass maderaa...
Thanks...
...unos se quedaron con la vieja info del quote y otros con la nueva info que añadi a posteriori...o no...
...sigo actualizandola con nueva info que vi hoy... mass maderaa...
Thanks...
Last edited by Salmonetin on Wed Feb 03, 2021 7:04 pm, edited 33 times in total.
-
- LED-Curious
- Reactions:
- Posts: 23
- Joined: Thu Dec 31, 2020 10:39 pm
Lowering the integration time to the lowest setting (2.4ms) and lowering the update interval (1s) I have been able to get more uniform readings. The individual colour readings still vary +/- 10% from second to second, it is better then the 80% variance I was experiencing with a longer update interval and a longer integration time.
@stevetheganjafarmer
Thanks for re-sharing this info, it compares to the following for "Hobbyists"...Salmonetin wrote: ↑Thu Jan 21, 2021 1:43 amreposted some erased posts... for your likes ...Thanks Bros...
sorry Bros maybe its offtopic here...but i looked at...
https://www.ies.org/fires/specifying-le ... -lighting/
...have a curious table... its old info...now there are better updated....
...Figure 9 shows the spectral responsivities of an AS7262 6-channel visible light spectral sensor with a spectral range of approximately 430 nm to 570 nm, as manufactured by ams AG of Premstaetten, Austria. While the spectral responsivities are not ideal Gaussian functions, it is clear that an instrument to directly measure radial basis function weights can be fabricated using existing technology.
https://www.adafruit.com/product/3779
...i dont see tutorial for that product...sorry...
https://ams.com/as7262
https://ams.com/as726xdemokit
https://ams.com/as726xdemokit#tab/documents
PD...este año veo novedades en sensores Sensirion los SHT40 que nuevos vendran el SHT41 y SHT45...y los SCD40 NDIR CO2....mas pequeños que los SCD30...
a ver si los de Sensirion sacan sacan este año el Smart-Gadget Kit con los SHT41 o SHT45 y hojala sacaran uno que tambien tuviera el NDIR CO2 SCD40...que para sensar los cultivos interiores nos vendria genial...
https://developer.sensirion.com/platfor ... pment-kit/
...y los modelos de nano lambda tambien me gustan mucho...diria que me gustan mas que los PBAR QUANTUM METERS....pero es mi solo pov...ustedes mismos...
https://nanolambda.myshopify.com/
https://nanolambda.myshopify.com/produc ... 2m_v1_temp
https://nanolambda.myshopify.com/produc ... 2m_w1_temp
PD1...PAR measures on 2021?... better go on PBAR measures...more extended than OLD and OBSOLETE PAR measuress...
PAR QUANTUM METERS are obsolete on actual times o led grow lights... PBAR QUANTUM METERS....better actually...just my pov...ustedes mismos...
...a los Bros muchos animos y cuidaros del covid...
Saludos desde Tenerife
reposteado ... para rezagados... pero daros mucha prisa que no durara mucho...
...postee varios sensores asi en mis posts borrados (por no gustarles a nadie)...y puede que algunos modelos sean mas caros que un quantum meter de rango extendido...pero creo juegan en ligas diferentes...y los caros lo son por algo...descubre el porque tu mismo...
for my pov better than TSL2561...the TSL2591...
https://www.adafruit.com/product/1980
https://learn.adafruit.com/adafruit-tsl2591?view=all
y claro que me gusta el TCS34725...pero creo puse hardware usado en un doc de la IES.org...que me dice mucho mas que el DLC...y claro que en DLC hay buena info...pero en IES.org...o en ASABE.org tambien y creo que mejor...yo mire el doc 240 y el 245 que aun no he podido leer...pues es de pago...y paso de pagar por ese...ya lo leere por otro lado...
el hardware del doc de la IES.org es el AS7262 y tambien os puse info...de ese y otros...en otros posts mios...
...y de ASM tambien hay Dev Boards muy interesantes aparte del AS7262... pero solo hablaba de ese modelo...mas caros los hay...tambien por algo lo seran... si relacionas calidad precio... es lo normal con cosas de cierta calidad... tu decides si te vale la pena o no...
no pretendo que compres o uses nada...solo os expuse mi pov e info...que gusto.. pues bien...que no..pues mal
y si encuentras algo de esa calidad a mejor precio...soy todo oido...puede que sean caros pero siendo lo que son ...tu ves alguno de esa calidad mas barato?...donde?...hazmelo saber...
este año mis post no creo duren mucho tiempo en foros o youtube...visto el exito del año anterior... y tampoco creo que postee mucho... pues tendre lio con otras cosas...
e function actived...fast mode en un ratito borrare...
Saludos desde Tenerife
...you will need at least a LUX meter but preferably a Spectrometer (ya puse info...si ves algo mejor o mas barato ..dame toque...) that can measure photons (ppfd)
There is a lot of youtube video and resources showing these products at work.
There is one problem however with this approach, these products require you to move the meters over your area to read the numbers in a given location, in short they measure on a single plane and dont take into consideration the angular distribution of the light source.
The polar distribution curve of a grow light can provide an indication of the type of distribution expected from the light.
What you need is an IES file from the manufacturer,
Data file created in the standardized IES (Illuminating Engineering Society) photometric format; contains measurements of light and light quantities.
These files can be used in a simulation program that allows you to move fixtures around your room and calculate the absolute ppfd for a given area before you ever need to turn a screw.
https://www.cutter.com.au/grow-room-des ... ht-layout/
https://dxboard.dialux.com/viewtopic.php?t=82606
seems to me there are Catalogs for Brands with IE files ...there are curious paneles leds de oficina...con curiosos SPD...y creo que tambien tienen IES FIles...
ya me gustaria que los las marcas de Horticultural leds sacaran IES files para jugar con el Dialux...pero no he visto sus IES files compartidos.... y estamos en 2021... a que esperan?...
...no son IES Files pero hay bastante info de lamparas de cultivo...gracias por el link...
https://www.designlights.org/horticultu ... ng/search/
https://www.sparkfun.com/products/14347
https://learn.sparkfun.com/tutorials/as ... 1611173272
https://ams.com/as7262
https://www.sparkfun.com/products/15050
https://learn.sparkfun.com/tutorials/sp ... 1611173272
https://ams.com/as7265x
PD...si tengo mas likes..quizas vuelva a repostear mi hilo borrado este año... si hay interes por el tema y un poco de amor para mi... ahora vuelvo al modo mute...nos vemos...
Paz y Amor...en 2021 mucho mas...
Saludos desde Tenerife
Inexpensive addition (IC Chip < $4.00 US) for hobbyists with DIY automation and I2C bus are; TS2561 equivalent Data to PBAR Quantum Meters and TCS34725 equivalent Data to Spectrometers. They have to be "Calibrated" for installed conditions (Acrylic Diffuser, Cut Filter, etc.) and in Academic Research (TCS34715) have been shown accurate to within one (1) percent when properly calibrated...
DIY PAR Sensor (TCS34715)... https://royalsocietypublishing.org/doi/ ... sos.160592
A UV/ IR Cut Filter or "Hot Mirror" provides the "required" PBAR or PAR range for the Sensors
Inexpensive addition (Photodiode < $1.00 US) for hobbyists with DIY milli-Volt Meters are; EAALSDSY6444A0 equivalent Data to PAR Quantum Meters and EAPDSZ4439A4 possibly equivalent Data to PBAR Quantum Meters. They have to be "Calibrated" for installed conditions (Acrylic Diffuser, Cut Filter, etc.) and in Academic Research (EAALSDSY6444A0) have been shown accurate to within one (1) percent when properly calibrated...
DIY PAR BAR (Ceptometer) with EAALSDSY6444A0 Silicon Photodiode... https://www.jove.com/t/59447/parbars-ch ... easurement
Good afternoon, my name is Miguel Angel, I have read this post from the beginning and it has been a long time since it began, I am currently a big fan of Arduino and all that it entails for parameter control, which I currently control an automated aquarium system The interest in the PAR meter is to find out the quality of my led screens. According to I have read, the BH1750 sensor is suitable with a simple conversion.
currently my BH1750, placed about 10 cm from the screen (emitter) gives me an output through the Arduino serial port of 37,450 Lux
Should I interpret it like this? 0.015 * 37450 = 561.75 PAR?
Kind regards to all
currently my BH1750, placed about 10 cm from the screen (emitter) gives me an output through the Arduino serial port of 37,450 Lux
Should I interpret it like this? 0.015 * 37450 = 561.75 PAR?
Kind regards to all
Hello,medlight wrote: ↑Wed Feb 03, 2021 1:15 pmGood afternoon, my name is Miguel Angel, I have read this post from the beginning and it has been a long time since it began, I am currently a big fan of Arduino and all that it entails for parameter control, which I currently control an automated aquarium system The interest in the PAR meter is to find out the quality of my led screens. According to I have read, the BH1750 sensor is suitable with a simple conversion.
currently my BH1750, placed about 10 cm from the screen (emitter) gives me an output through the Arduino serial port of 37,450 Lux
Should I interpret it like this? 0.015 * 37450 = 561.75 PAR?
Kind regards to all
Yes, that's the calculation and result if the BH1750 Sensor is properly "Calibrated" and the light is from an approximately 4000K (~ 3000K - 5000K) white LED.
The PPF/Lm factor varies for different Light sources and different White LED CCTs.
You can use the “Average” 0.015 (+/- 7% accuracy) for 3000K to 5000K White LEDs or you can use a custom, more accurate PPF/Lm Conversion Factor if you have a fixed Spectrum Light with a known PPF and Lumens, from manufacturer’s or posted test results (PPF divided by Lumens or PPFD divided by Lux).
For Example:
HLG Scorpion Diablo = 0.0156
Samsung Horticulture L2 = 0.0178
HLG QB 288V2 Rspec = 0.0154
HLG QB 648 Diablo = 0.0163
HLG QB288 V2 3000K = 0.0145
HLG QB288 V2 4000K = 0.0143
HLG QB 96 Elite V2 = 0.015
The "Calibration" is either
1. Against a Lux Meter or
2. A properly wired and programmed factory calibrated bare sensor (no Lens) or
3. A sensor with adequate Lens Correction Factor for any installed Diffuser Lens.
Lens Correction Factor is easily acquired with a properly operating BH1750 sensor.
1. Setup a fixed light source and fixed distance between the bare sensor, measure and record the result.
2. Place the Lens over the sensor without changing the distances, measure and record the results.
3. Calculate the % Transparency by dividing result #2 by result #1 to get a decimal result (Percentage when multiplied by 100).
4. Calculate the Lens Correction Factor by dividing 1 (one) by the decimal result (or 100% / xx% Transparency).
For Example; [2000 (#2) / 4000 (#1) = 0.5 ... 1 / 0.5 = 2.0 Cf]
I build my own PAR meter based on TCS34725 sensor: https://github.com/AVarfolomeev/tcs34725_par_meter2
I used clear acrylic resin, PLASTIK 71 PCB coating, to protect sensor PCB from water.
I mean, the PAR calculation from original MultispeQ article
float par = (0.65847 * cData) + (-1.60537 * red) + (-2.30216 * green) + (-0.50019 * blue);
is invalid, it like Lux not PAR.
The MultispeQ firmware don't use this equation, it use just clear channel and calibration multiplier.
I used clear acrylic resin, PLASTIK 71 PCB coating, to protect sensor PCB from water.
I mean, the PAR calculation from original MultispeQ article
float par = (0.65847 * cData) + (-1.60537 * red) + (-2.30216 * green) + (-0.50019 * blue);
is invalid, it like Lux not PAR.
The MultispeQ firmware don't use this equation, it use just clear channel and calibration multiplier.
Thanks for sharing your build info.avfv wrote: ↑Sat Feb 06, 2021 1:49 pm... https://github.com/AVarfolomeev/tcs34725_par_meter2
... I mean, the PAR calculation from original MultispeQ article float par = (0.65847 * cData) + (-1.60537 * red) + (-2.30216 * green) + (-0.50019 * blue);
is invalid, it like Lux not PAR. The MultispeQ firmware don't use this equation, it use just clear channel and calibration multiplier.
The Multispeq instrument uses TCS34715 Sensors with Hot Mirror UV/IR Cut filters of ~ 400 - 700nm, "AMS-TAOS USA, TCS34715FN + 700nm low pass filter", the TCS34725 are ~ 400 - 650nm, as such would have different calibration factors, less Red and Clear Wavelengths in their calibration equations, but almost the same Blue and Green. The Multispeq Beta uses the R-G-B-Clear data for the Lux and PAR conversions along with using the R-G-B % Values... https://github.com/Photosynq/Protocols-Arduino
The ClearData is used for increased accuracy, the AMS (TCS) color sensors' R-G-B-Clear measurements and resulting Lux values can be calibrated to an Accuracy of < 1.0 %... https://ams.com/documents/20143/36005/A ... 7c834cc60a
My initial TCS34725 prototype PAR Sensors are using the "Standard" RedData, GreenData and BlueData to calculate LUX and Percentage B-G-R. It also uses all four Data for better measurement resolution of PAR as it varies with the ratios of B-G-R wavelengths. It does currently require an additional Offset (Correction Factor) and would benefit from calibration with an actual PAR Meter, but so far tracks with Varying Light types, different spectrum with varying Red , Green or Blue wavelengths, Lux Measurements and a Lux Meter. Also the CCT (Kelvin) though not really required for an "Horticultural" light source is accurate to approximately + 200 Kelvin, tested against LEDs, Incandescent bulbs and Sunlight.
Simply add a translucent Acrylic Lens (Acrylite # 7328), at 32% Transparency its a "Glass factor" of 3.125 which increases your maximum range up to 1437.5 - 1562.5 umol/m2/shttps://www.reef2reef.com/threads/diy-p ... or.800942/
... bare sensor can't measure PAR above 460-500 umol/m2/s,...
DIY Epoxy Sealed PAR Sensors with Acrylic Lens, Peer Reviewed Documentation and Video... https://www.jove.com/t/59447/parbars-ch ... easurement
Last edited by PeteR_1 on Mon Feb 08, 2021 4:39 pm, edited 1 time in total.