Public Lab Research note

Open-Lux REV-N: Full Circle

by donblair | April 21, 2014 07:02 21 Apr 07:02 | #10340 | #10340

(This is a quick note to post materials I've gathered after updating the Open-Lux design -- will flesh out this research note tomorrow ...)

Summary of current features (some of them new):

  • NeoPixel array is now arranged in a circle, surrounding RGB and NIR sensors; can control LEDs individually;
  • Added NIR LEDs, to facilitate "single pixel" NDVI measurements (e.g., of a single leaf);
  • Added a microSD card reader on back.
  • Internal pin breakout intended for a) adding audio jack in order to get data off the device via audio cable --> smart phone or to laptop, b) serial connection to Raspberry Pi, c) connecting to additional I2C devices, driving switches, measuring voltages, etc
  • Should also be possible to transmit data from device to laptop (and from Open-Lux to Open-Lux) via IR as well
  • Another pair of RGB and NIR sensors on back of board;
  • Added 3 buttons and a power switch that can easily poke out sides of an enclosure -- 3 buttons are intended to navigate through menu of pre-selected measurement / display modes -- 128 kb EEPROM chip can store several programs
  • Two headers for the Nokia 5100 LCD -- one accommodates the Adafruit Nokia 5110 breakout board; another accommodates a very common Nokia 5110 breakout board one can find on Ebay;
  • Screw terminal and resistor footprint can accommodate Adafruit 10K thermistor, or other common thermistor types -- temperature can be displayed on LCD, or conveyed via audio (buzzer) or light (using the ring of LEDs as a 'meter')

Quick links to additional info:


  • Mockup of very plain idea for a 3D case .... the front:


  • ... and the back (showing Nokia 5110 LCD):


  • Closeup of a 'sticker' that I'm imagining might be useful on the back of the device:


  • Info on the NIR LEDs:


  • Info on the RGB sensor:


  • Info on the LUX+NIR sensor:


  • OSHPark mockup of the front of the board ...


  • ... and of the back. Note the new microSD card reader, as well as the second pair of RGB and LUX-NIR sensors (for albedo measurements):


  • Just the silk:


  • Pointless inverse image of silk:


  • Schematic:



The circular layout will make concentrating the different LEDs towards a common point easier. good thinking.

Reply to this comment...

Yay! Yeah, I puzzled a bit over the best geometry of 'emitters and receivers' to use on the device. I have two photodiodes on the board, one of which does RGB, and the other does NIR. It's hard to have them too close together and still get the wiring layout to work out easily. But e.g. if trying to measure NDVI on a single leaf, I figure that we can illuminate the leaf with the IR LEDs (arranged in a Barnstar pattern, FTW!) and use Red from the NeoPixels. We can then measure Red using the 'R' channel on the RGB sensor (center-left) and NIR using the 'NIR' channel on the LUXNIR sensor (center-right); they won't both be receiving the same absolute amount of illumination, but since they're always a set distance apart, maybe we can just hope that that the difference in illumination will be relatively constant, so that we can assess (NIR-Red)/(NIR+Red) in a fairly consistent way.

But now the exciting threads here and here are making me think grandiose thoughts.

  • Check out Willie's awesome teardown of the Shinyei sensor, and the nice writeup of how it works ...

  • Could we make a 'shield' for the OpenLux that has a) some additional IR LEDs, oriented at an appropriate angle, with a collimating light baffle and an air intake (all 3D printed, I suppose), along with a place for a lens that would sit in front of the OpenLux NIR LEDs -- and then make our own, cheap particle detector?

  • Maybe with two lenses, or a reflector + a lens, we could simply fashion a 'lens cap' for the OpenLux that fit right on top, that works to detect particles in an air gap right in front of it? So that we can use the photodiodes and IR LEDs in their current configuration?

  • Is the 'heating element' in the Shinyei detector simply a way of providing a certain amount of air flow through the sensor? Or is it a way of accounting for the temperature of the air? A heating element is a battery suck (if we'd like to do this on a battery -- maybe not). If the element is to encourage air flow, maybe we could find a way to enhance air flow that doesn't require heating (i.e. locating the device near a window, or near a laptop fan, or etc). If it's to account for temperature, we do have a non-contact IR sensor on the OpenLux, which also has an internal thermistor ... and there's also a breakout for an external thermistor ... or I could add e.g. a TMP100 surface mounted sensor to the board ...

I'm also thinking that the ring of individually-addressable LEDs will make for a fun way of indicating measurement levels (of whatever). We could even have an old-school Knight Rider / Cylon effect ... and maybe make the buzzer make an associated 'swoosh swoosh' noise ...

Is this a question? Click here to post it to the Questions page.

Reply to this comment...

Follow-up -- there's some nice info on IR photodiodes (at least the sort that are in a typical IR remote), here -- including a link to a datasheet. Looks like those little guys are nicely designed so as not to be influenced by spurious pulses of light in the environment ... so maybe using the relatively simple (?) NIR photodiodes that are currently on the OpenLux for particle detection wouldn't work so well. That said, we could put the photodiodes and emitters on a breakout board. But it'd probably be hard to put that together, with lens and all, for less than the $2 that the Syhitech DSM501A is being sold for ...

Is this a question? Click here to post it to the Questions page.

Reply to this comment...

Login to comment.