Adafruit is open, safely ... and shipping all orders at this time! Read more!
0

irradiance measurement (w/m2) from the TSL2591
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

irradiance measurement (w/m2) from the TSL2591

by andreaguti on Thu May 02, 2019 10:58 am

Hi, I would like to know how to get the irradiance measurement from the TSL2591. I mean the sensor gives me LUX but I need w/m2 in the visible and the infrared.

thanks

andreaguti
 
Posts: 3
Joined: Thu May 02, 2019 10:54 am

Re: irradiance measurement (w/m2) from the TSL2591

by adafruit_support_mike on Fri May 03, 2019 2:08 am

That's toward the bottom of page 6 in the datasheet:

https://cdn-shop.adafruit.com/datasheet ... _EN_v1.pdf

Depending on which channel you use, there's a nominal counts-per-uW/cm^2 value.

adafruit_support_mike
 
Posts: 61107
Joined: Thu Feb 11, 2010 2:51 pm

Re: irradiance measurement (w/m2) from the TSL2591

by andreaguti on Mon May 06, 2019 12:44 pm

I didnt understand at all what you said.
I added an image which it is the sensor readings from the arduino when I put the sensor to sense in front of an iron which is 90*C, SO how Can I get the w/cm2 with the data which the arduino gives me
Attachments
sensor.PNG
sensor.PNG (32.02 KiB) Viewed 375 times

andreaguti
 
Posts: 3
Joined: Thu May 02, 2019 10:54 am

Re: irradiance measurement (w/m2) from the TSL2591

by adafruit_support_mike on Tue May 07, 2019 12:37 am

The datasheet lists the irradiance values at either 5338 counts per uW/cm^2 or 3474 uW/cm^2 depending on which input channel you use.

Basic math says each count then represents either 1/5338th or 1/3474th of a microwatt per square centimeter.

If the output from the sensor is 52, it means the sensor has detected either 52/5338 or 52/3474 of a microwatt per square centimeter.

adafruit_support_mike
 
Posts: 61107
Joined: Thu Feb 11, 2010 2:51 pm

Re: irradiance measurement (w/m2) from the TSL2591

by andreaguti on Tue May 07, 2019 10:09 am

we supposed it but we weren´t sure at all. THANK YOU SO MUCH!

andreaguti
 
Posts: 3
Joined: Thu May 02, 2019 10:54 am

Re: irradiance measurement (w/m2) from the TSL2591

by abferguson on Thu Jul 04, 2019 6:45 pm

Adafruit TSL2591 Gurus,

I'm using the TSL2591 sensor for the first time and am trying to understand how to apply the datasheet info to calculate W/m^2. But I'm a bit confused from reading this thread.

From adafruit_support_mike, "If the output from the sensor is 52, it means the sensor has detected either 52/5338 or 52/3474 of a microwatt per square centimeter." BUT the datasheet references channels CH0 and CH1 (column 3) for IR in Figure TSL2591 – 9. CH0 counts output with the 5338 reference and CH1 counts output with the 3474 reference.

Another way to read the datasheet to get IR uW/cm^2 is to choose either CH1 output (52) or CH0 output (137) and divide by their respective references. So, to get IR value, choose from 52/3474 uW/cm^2 OR 137 / 5338 mW/cm^2.

Using the same approach, white light uW/cm^2 would use the 6024 and 1003 count reference with the CH0 and CH1 counts, respectively.

I'm just trying to clarity on how to use Figure TSL2591 – 9. A clear example of how to use output from the two channels and the four count references would be very helpful.

Thanks

abferguson
 
Posts: 1
Joined: Thu Jul 04, 2019 6:18 pm

Re: irradiance measurement (w/m2) from the TSL2591

by Matadormac on Thu Aug 01, 2019 12:43 pm

abferguson wrote:Adafruit TSL2591 Gurus,

Another way to read the datasheet to get IR uW/cm^2 is to choose either CH1 output (52) or CH0 output (137) and divide by their respective references. So, to get IR value, choose from 52/3474 uW/cm^2 OR 137 / 5338 mW/cm^2.

Using the same approach, white light uW/cm^2 would use the 6024 and 1003 count reference with the CH0 and CH1 counts, respectively.



I am so confused now. Where in the data sheet does it give those figures for the Irradiance repsonsivity for IR (CH1)? Working with the June 2018 datasheet for the TSL2591, typical counts per uW/cm2 is shown as 154.1. Shouldn't that be the figure we use to divide the returned IR value from the sensor?

Am I looking at the wrong datasheet?

Regards

Mark MacKenzie

Matadormac
 
Posts: 30
Joined: Tue Aug 05, 2014 8:51 am

Re: irradiance measurement (w/m2) from the TSL2591

by Matadormac on Thu Aug 01, 2019 12:58 pm

Okay, answered my own question and this brings up another one.

Your forum answer was based on a TSL2591 data sheet from April 2013. This datasheet shows the irradiance responsivity to be:
TSL2591 2013.JPG
TSL2591 2013.JPG (27.32 KiB) Viewed 172 times
.

In the June 2018 TSL2591 datasheet the irradiance responsivity is markedly different:
TSL2591 2018.JPG
TSL2591 2018.JPG (24.72 KiB) Viewed 172 times
.

So now, how to tell which tsl2591 chip you are selling goes with which count value for irrandiance?? I am interested solely in the IR component in any case but have a number of tsl2591 breakouts from you from a year or so ago.

Since I don't have a way of back checking IR instrumentally I have to rely on the datasheets. I can understand a slight correction of values but this is at least an order of magnitude different and a real problem for me.

I hope you can shed some light (so to speak) on this issue.

Regards

Mark MacKenzie

Matadormac
 
Posts: 30
Joined: Tue Aug 05, 2014 8:51 am

Re: irradiance measurement (w/m2) from the TSL2591

by Matadormac on Thu Aug 01, 2019 2:56 pm

I just checked the tsl2591 datasheet that you have an URL to on the shop product listing and it is the correct latest datasheet of June 2018. So just the datasheet reference used earlier in this thread is stale dated. Boy, what a change in count values though!

Regards

Mark MacKenzie

Matadormac
 
Posts: 30
Joined: Tue Aug 05, 2014 8:51 am

Re: irradiance measurement (w/m2) from the TSL2591

by juandediosyanez on Tue Dec 31, 2019 11:54 am

Hello everyone, I write in the forum because I am with the same problem initially raised, and I think there is something wrong with the solution that has been raised. When the IR data is taken and divided by the number of counts / irradiance, a very low value is obtained which should actually be obtained.

In the terrestrial surface they should arrive on a sunny day about 1000 W/m2 and that is not the case applying the solution proposed in the forum.

You need to get the counts of channels 0 and 1, ADC or something like that I don't know how to get it with the library, and that value is what is divided, I don't think I understand, because doing it over IR does not give a real value.

Did you find any solution for this problem? How do they get irradiance in mW/cm2 or W/m2?

juandediosyanez
 
Posts: 2
Joined: Tue Dec 31, 2019 11:41 am

Re: irradiance measurement (w/m2) from the TSL2591

by zener on Tue Dec 31, 2019 3:38 pm

As far as the vast change in specs from one data sheet revision to another, I think you need to ask TI themselves. There may be more explanation on the data sheet, I don't know (I didn't look at it). It may be a difference in the IC itself or it may just be a correction. As far as converting lux to irradiance, there are issues regarding energy per wavelength, which is why it is not a short answer. If it was a simple factor then the datasheet would provide it. There is another issue which is mentioned here:

https://forum.arduino.cc/index.php?topic=180531.0 which is cosine error.

This statement was made above:
In the terrestrial surface they should arrive on a sunny day about 1000 W/m2
Well, yes and no. That is an ideal quantity, when the sun is directly overhead (perpendicular to the plane of the detector), on a clear day at a certain time of year. So that has to be considered since it has a huge effect on the measurement obtained.

It may be that there are better ways of measuring solar irradiance. I would think a small PV panel would be a more direct way. That would have to be characterized also of course.

zener
 
Posts: 4567
Joined: Sat Feb 21, 2009 2:38 am

Re: irradiance measurement (w/m2) from the TSL2591

by juandediosyanez on Wed Jan 01, 2020 9:07 am

True, it would have to be calibrated. I am making a modification on Adafruit's own library to obtain this data, according to the manufacturer's coefficients, it continues to give more than doubtful results. I wouldn't be surprised if the big change from datasheet 2013 to 2018 needs another big change in the future.

Thanks zener, and happy new year to all.

juandediosyanez
 
Posts: 2
Joined: Tue Dec 31, 2019 11:41 am

Please be positive and constructive with your questions and comments.