In re: I2C light sensors--the BH1750 seems to be an easily obtained part conveniently mounted on eval boards (see Adafruit and others). I think it's amazing that TI didn't include a single application example, other than suggesting this part would conveniently mate up with a microcontroller. Apparently, TI marketing took the day off when the datasheet was being prepared. (grin) Finally, @IMSAI Guy, someday you should do a "clean up" video or videos where you address some of the questions raised in the comments. A lot of the questions don't justify a full episode, but I think there are enough cliff hangers left that you could string several short topics together and make a full video. You'll notice how quickly I volunteered you for additional work. I'm always trying to be helpful that way.
Looks like 0.001 to 1000kHz range output... so 1Hz to 1MHz. What's the easiest way to translate the scale from 1Hz-1Mhz to 1Hz- ~20kHz for the full range of intensity/radiance?
Since his 'scope is set to 1 mS/div, it would appear that the highest frequency he shows is about 3 kHz (kind of hard to see), I'd say the highest frequency is already in human hearing. The low frequency range also seems to be well below human hearing, though.
I did some work with these in the 1990's, and found that, using the 100 diodes, it would give a measurable signal, useful above the dark current, in my garage, so dark that you could not make out your hand in front of your face...also, there is a standard filter from Hoya, a blue temperature conversion filter, that gives a rather flat spectral response from say 430 mm to 800 mm...
I miss all of these oddball goofy ICs that the silicon valley guys were churning out in the 80's & 90's. BBDs and sensors and tone decoders... ... oh my. :D
They're still chugging them out, they just all speak I2C, SPI, TTL, or a one off odboll protocol. Looking at you 1-Wire, and RGB LEDs! Then there's the whole part where NFC chips don't actually transmit anything. They modulate their power draw, and the transmitter uses that to determine the bitstream.
Can these handle direct sunlight? According to wikipedia, that's 120.000 lux. I'm looking for something to measure sunlight in my outdoor hydroponic garden.
3:24 onward: He fingers Fig. 2 Spectral Responsivity mouthing figures that we can see, without ever connecting wavelengths to spectral (rainbow) colours. The curve reveals that the chip sensitivity peaks in the near infrared, suggesting it could be used in an overheating sensor.
Great vid tuts, but when you are displaying your hands and waving them or your fingers all over the place it is extremely distracting and off-putting. Especially when we are wanting to see just the diagrams and or components as well as tech spec sheets. Can you use pointers instead? I know you have to pick up boards and components and move them and even write calcs but with your hands and fingers taking over the screen it's almost impossible to concentrate on what you are explaining and teaching with so many fingers and hands shaking and moving all over the place. Especially when your hands and fingers are taking up nearly all of the screen. It would be great if you could maybe lessen the hand and finger content.
Yes indeed that is a cool chip but I can't quite seem to think of how I would use it. How do they measure light intensity to make a spectral graph like that? Since light sensors have different response at different wavelengths and since light emitters have different emission at different wavelengths the only way I can see to correlate the two is by use of some basic physical phenomena like black body radiation or absorption.
It could be used in a number of musical applications. It could be used to make a theremin-type instrument, as an oscillator for a synthesizer, a variable clock signal, light-controlled LFO, perhaps to control a pulse width modulated compressor/VCA?
Microcontrollers have hardware counters that can count pin transitions and they have hardware timers that divide the crystal frequency. Together, you get a fairly accurate frequency counter. To get more than 8b resolution and/or a reasonable dynamic range, you might need to supplement the hardware with software and a few bytes of RAM. If ultimate calibration is needed, the color dependence presents a problem that a single microcontroller might be unable to solve.
I have a couple of the light sensor types in clear packages. from different suppliers, the clear package are really nice. I use mine for a photographic light meter.
for space applications as well. it would be easier to use optical gates in some cases since electrons are harder to get to operate at colder temperatures than photons
In re: I2C light sensors--the BH1750 seems to be an easily obtained part conveniently mounted on eval boards (see Adafruit and others).
I think it's amazing that TI didn't include a single application example, other than suggesting this part would conveniently mate up with a microcontroller. Apparently, TI marketing took the day off when the datasheet was being prepared. (grin)
Finally, @IMSAI Guy, someday you should do a "clean up" video or videos where you address some of the questions raised in the comments. A lot of the questions don't justify a full episode, but I think there are enough cliff hangers left that you could string several short topics together and make a full video.
You'll notice how quickly I volunteered you for additional work. I'm always trying to be helpful that way.
Can we connect the output to an audio amp and define the output to be within the human hearing range?
Looks like 0.001 to 1000kHz range output... so 1Hz to 1MHz. What's the easiest way to translate the scale from 1Hz-1Mhz to 1Hz- ~20kHz for the full range of intensity/radiance?
Since his 'scope is set to 1 mS/div, it would appear that the highest frequency he shows is about 3 kHz (kind of hard to see), I'd say the highest frequency is already in human hearing. The low frequency range also seems to be well below human hearing, though.
I suppose you could hook it up to an audio oscilator. (?)
@@jafinch78 the output divider can make it f0/100, so the range would be from 10Hz to 10kHz... mostly audio, though 10Hz is a bit low :-).
I did some work with these in the 1990's, and found that, using the 100 diodes, it would give a measurable signal, useful above the dark current, in my garage, so dark that you could not make out your hand in front of your face...also, there is a standard filter from Hoya, a blue temperature conversion filter, that gives a rather flat spectral response from say 430 mm to 800 mm...
I miss all of these oddball goofy ICs that the silicon valley guys were churning out in the 80's & 90's.
BBDs and sensors and tone decoders...
... oh my.
:D
They're still chugging them out, they just all speak I2C, SPI, TTL, or a one off odboll protocol. Looking at you 1-Wire, and RGB LEDs! Then there's the whole part where NFC chips don't actually transmit anything. They modulate their power draw, and the transmitter uses that to determine the bitstream.
You could make a cool optical theremin type instrument with this.
OOOOH Could make a light meter for blind photographers, of which there seems to be a great number.
You Brits! I love this! LOL! Best comment of the day!
@@TonyBarr99 Thanks for the compliment.
These were used in blood glucose meters in the 80’s and 90’s
Can these handle direct sunlight? According to wikipedia, that's 120.000 lux.
I'm looking for something to measure sunlight in my outdoor hydroponic garden.
I have some very old LED's packaged in TO-92 packages with gold plated leads. The packaging material is transparent with a red tint. They're very odd.
Reminded me of the "Lumichord" from the 1964 film, "The Time Travellers". Here: ruclips.net/video/S39OH7MrzRo/видео.html
3:24 onward: He fingers Fig. 2 Spectral Responsivity mouthing figures that we can see, without ever connecting wavelengths to spectral (rainbow) colours. The curve reveals that the chip sensitivity peaks in the near infrared, suggesting it could be used in an overheating sensor.
Great vid tuts, but when you are displaying your hands and waving them or your fingers all over the place it is extremely distracting and off-putting. Especially when we are wanting to see just the diagrams and or components as well as tech spec sheets.
Can you use pointers instead? I know you have to pick up boards and components and move them and even write calcs but with your hands and fingers taking over the screen it's almost impossible to concentrate on what you are explaining and teaching with so many fingers and hands shaking and moving all over the place.
Especially when your hands and fingers are taking up nearly all of the screen. It would be great if you could maybe lessen the hand and finger content.
This is a neat RGB sensor IC often sold on a PCB allowing I2C, AS7262 6-channel Spectral RGB sensor...cheers.
they're not black chips they are white chips
Very interesting, I'll never use them or know what to do with them but still interesting. :)
Yes indeed that is a cool chip but I can't quite seem to think of how I would use it.
How do they measure light intensity to make a spectral graph like that? Since light sensors have different response at different wavelengths and since light emitters have different emission at different wavelengths the only way I can see to correlate the two is by use of some basic physical phenomena like black body radiation or absorption.
they use a monochromator ruclips.net/video/jHsQxFlDuHQ/видео.html to input various frequencies of light
It could be used in a number of musical applications. It could be used to make a theremin-type instrument, as an oscillator for a synthesizer, a variable clock signal, light-controlled LFO, perhaps to control a pulse width modulated compressor/VCA?
Huh, never seen anything like that. Interesting. Will have to read into these more. ~775nm as well.
That was really neat. Thanks.
Microcontrollers have hardware counters that can count pin transitions and they have hardware timers that divide the crystal frequency. Together, you get a fairly accurate frequency counter. To get more than 8b resolution and/or a reasonable dynamic range, you might need to supplement the hardware with software and a few bytes of RAM. If ultimate calibration is needed, the color dependence presents a problem that a single microcontroller might be unable to solve.
I used a 14 pin in a light activated alarm in high school. WAY back. Really cool to show the internals of a chip.
Are these any different than the photodiode made by Cherry Semiconductor for use in Kodak Instamatic cameras?
Just as an FYI, I found the squelch or noise cancellation kicking in when you were speaking to be really annoying
none used, it sounds OK to me
@@IMSAIGuy uh, oh... maybe RUclips is A/B testing me with some audio "improvements"
I didn't know about these, but sounds like a v-f converter with an opto on the front end. cool.
I have a couple of the light sensor types in clear packages. from different suppliers, the clear package are really nice. I use mine for a photographic light meter.
Good old times, when parts were simple and easy to use...
Yep, very kewl.
Bring back the LM3909!
here ya go: ruclips.net/video/2WWUhsus9ic/видео.html
@@IMSAIGuy
Thanks!
фантастика... чего только не придумают
That's super cool. Could be useful to control synthesizer parameters, BBD, or perhaps for PWM compressors?
for space applications as well. it would be easier to use optical gates in some cases since electrons are harder to get to operate at colder temperatures than photons
Is there any difference between the no-latter, A & B variants?
The no-suffix has an absolute frequency tolerance of ±20%, the -A has ±10% and the -B has ±5%.
Unfortunately they are no longer made :(
Available on Aliexpress. One in an SOP-8 package would cost me the equivalent of US$55, or NZ$89. An 8 pin is at least double that.
Cool.
👌👍
That’s fun! Totally dig the different IC’s out there! I appreciate the video!