Pulse the leds at 500 Hz or 1 Khz after running them steady state for a few= seconds. Use a simple lock in amplifier or gated ADC on a cheap microcontr= oller to pick up the measurement, Now you have a rough measure of any backg= round, and the leds, without a shutter.
You could also use a few cheap dichroic filters to only look at the LED wa= velengths, TAOS and Agilent make cheap RGB sensor chips for this very purpo= se. The filters are on chip. The math to get color temperature, etc is in the a= pp-notes.
I strongly suggest a microscope slide beamsplitter, and you can get fancy w= ith plastic polarizers and plastic quarter wave plates as isolators. In mos= t cases, a few carefully placed pinholes will let the LED light dominate ov= er the outside light for the calibration sensor.
Yes, I have considered using my existing plastic window in this fashion. My chief worry then, though, is that now you would also get signal from the sample as well...
Hmmm... read with no sample, read with sample... nope, too complicated for the user!
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
That's a neat idea. OTOH, LEDs do emit sideways too, especially at the tip. Maybe Charlie could harvest that into a dedicated sensor, saving a path and a few optical parts.
If he synchronously modulates and detects the LEDs, he can also remove most of the effect of ambient light polluting his illumination source. We discussed that, ISTM, back the first time, when Charlie was first considering this project.
Yeah that's what I had in mind... But I was thinking of letting the sample light go straight through with the LED sensor at 45 degrees. And then having the sample sensor 'after' the beam splitter
\ sample | \ sensor |\------------> sample |\ | \ V LED sensor
Yup, lock-in detection seems to way to go. There shouldn't be any phase shifts to worry about, as long as he doesn't modulate too fast. You can most likely do the sychronous detection in software. (which depresses my analog heart.)
I wonder how repeatable the light coming from the side of the LED is?
Should be fairly repeatable -- I once wrote a ray-tracing program to analyze some ideas for LED package improvements. Total internal reflections produce a sideways "spray" off the tip that's hard to avoid.
It'll vary with die placement per part, but his calibration already factors that out automagically.
A further refinement is to cut grooves into the inside of the black tube. We ended up putting a screw thread into our graphite "black tube" insert when I was looking at the amount of beer foam that clung to the side of the beer glass.
Some brewers do measure it - but the last time I heard, not nearly enough of them to make Haffmans BV marketing entirely happy.
The screw-thread grooves kill the single pass low angle of incidence reflections which otherwise widen the acceptance angle of the almost non-reflective black tube ...
I've only been following Charlie's project tangentially, so this may already be part of it. But how about an intensity calibration step where the user puts a white reflective diffuser up against the output. You wouldn't have to do this all the time, but it could be used to set the current going to each LED. The nice thing is this needs no extra photodiodes... it's all in software.
Hi James, I don't think a mirror would be as good. Then there are all sorts of alignment issues. A diffuser (few pieces of white paper?) would give a more uniform reflection.
Many existing units do exactly that, but it is a step that is fraught with complications. Is the white target dirty? How do you signal to do it? Does this mean that you need additional switches, longer on times to give the user time to find the target, or other expensive options? Some units literally have a 'lens cap' to protect the front of the unit that is white to provide that calibration step. I want to make my unit a lot simplier to the user - just hold it against your target and press the button...
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.