For my project, I need to observe the lux inside a double-pane window. Last time, I used 'Adafruit TSL2561 Digital Luminosity/Lux/Light Sensor Breakout'* but its recommended maximum operating temperature is 70 deg C.

Since I will be placing this window under a solar simulation for about 30-45 minutes under Standard Test Conditions (1000 w/m2 at 25 deg C constantly) and, I need a sensor that can withstand and operate under 90 degrees happily (without degrading or causing weird results like negative lux vaules).

Are there such lux sensors?

Link to the previously used sensor: https://www.adafruit.com/product/439

  • \$\begingroup\$ Can you reflect the light to a sensor outside the enclose space and calibrate it for the reflection and transmission losses? Piping in cooling air may be possible without overly perturbing the in window space energy balance. \ \$\endgroup\$ – Russell McMahon Aug 17 '17 at 12:21
  • \$\begingroup\$ I don't think its a possibility since I have to do more experiments next week. Seems like an interesting idea, though. \$\endgroup\$ – Shreea Sharma Aug 17 '17 at 13:38
  • \$\begingroup\$ may take minimal effort. Small mirror inside - or two. Sensor on glass or near it wit a mask around to ensure light comes from intended source. Possibly an internal mask. ("Mask' is probably fancy term for pieces of cardboard coloured black and some sticky tape :-). Also see my answer re using photodiodes. Digikey are US based by supply Europe etc. (I take it you are the Delft researcher that Google search flags up.) \$\endgroup\$ – Russell McMahon Aug 17 '17 at 13:53

Photodiode + electronics solution:

An optical sensor with good spectral bandwidth and 125C max temperature rating can be obtained by using a photodiode with separate electronics. A very good performance unit is mentioned below by way of example, but lower cost solutions with performance that meets you requirements are available.

The ODD-#W-2 photodiode has a 125 degrees C maximum operating temperature and 400-100 nm spectral response. The response varies with wavelength but can be linearised in software (or with hardware) and spectral range could be limited with optical filters if desired. These cost $27.41 in unit quantity, in stock at Digikey - but far lower cost diodes are also available if unit cost matters.

BPW24R $4.50/1 125C max opn, UV to IR response.

BPW20RF $6.30/1 125 C max opn, UV to IR

A selection of in-stock photodiodes with 125C max operating temperature are listed here


Optical sensors including integral electronics:

This Digikey search offers optical sensors with 105C or 125C max temperature . A possibly major concern is that these are usually targeted at IR sensing and the spectral response may be too narrow, depending on your application.

Wider spectral range sensor: This -40C to 105C operating range sensor claims "near human eye spectral response" and is in stock at Digikey in unit quantities. Intersil ISL76671.

Sensitivity is high (100 lux max) but a suitable ND filter would allow this to be extended.

enter image description here

Other approaches:

In lieu of high-temperature-rated sensors you might consider:

  • Reflecting the light to a sensor outside the enclosed space, and calibrate it for the reflection and transmission losses. Mirror can be perhaps 5% loss , or under 1% if front silvered. Transmission through window glass depends on material and wavelength but is usually under 10%. Calibration should allow this method to work adequately for most applications.

  • A heatsink of small area extending either through the glass or embedded in the frame so minimal thermal flow occurs apart from via the sensor.

  • Fibre-optics for conveying light to an external sensor.

  • Piping in cooling air (or a recirculating feed) may be possible without overly perturbing the in window space energy balance. A temperature sensor monitoring the light sensor would allow minimal air (or other fluid) flow.



Excellent TI/Burr Brown app note

Another Ti/Burr Brown Photo diode app note - you do not need to use their nice but no doubt expensive IC.

Sharp Photodiode/Phototransistor Application Circuit

Microchip - AN951 Amplifying High-Impedance Sensors – Photodiode Example

Hamatsu some ideas

Hofoo. Who? App notes and data galore. May even be useful.

Ideas galore

AD - somewhat OTT but provides ideas
2 MHz Bandwidth PIN Photodiode Preamp with Dark Current Compensation

Some relevance:


  • \$\begingroup\$ @pjc50 And has received one upvote :-) (Perhaps from you. If so thanks). C'est la vie. I don't care about additional rep per se - but upvotes on an answer serve as a sign to others that an answer may be worth looking at if they have a requirement. \$\endgroup\$ – Russell McMahon Aug 18 '17 at 13:17
  • \$\begingroup\$ @ShreeaSharma If you are in Delft - list member Wouter van Ooijen is in Holland and is a very competent electronics engineer who may be able to advise re local sourcing of components and implementation. I'll refer this thread to him and see if he has any comment. \$\endgroup\$ – Russell McMahon Aug 19 '17 at 21:03
  • \$\begingroup\$ I am indeed in the Netherlands, but I have nothing to add to your excellent answer! \$\endgroup\$ – Wouter van Ooijen Aug 19 '17 at 22:00

It could maybe work with an ambient light sensor like this one:


It's rated -40-100C.


Here is the link of TI's ambient light sensors: "http://www.ti.com/lsds/ti/sensing-products/optical-sensors/light-sensors-overview.page". OPT3001 sensor (one of TI's sensors) can operate between -40 and 85 C degrees. It might be solve your problem.

Or maybe you can design a basic heat exchanger with aquarium pump, copper pipes and PC CPU cooler to keep your sensor cold if you have enough space.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.