John, why don't you just immerse them in a can of black painting, let-it dry a couple of hours and enjoy the benefits ?
You could add comments like "Our regulation stages are using specially treated reference diodes" in the audiophile marketing blabla, and keep the brand of the painting like a big secret 🙂
I suppose the subject has been drained, now, and that we can return to more serious discussions, like the psychology of aliens ?
Esperado, it is SY who uses unpainted LED's as an input tube bias resistor, not me! '-) I only use them to bias output stages. About 60dB difference in sensitivity. I think that they look pretty in the circuit.
Because of course it's based on classified military technology and they couldn't tell you even if they wanted to. 😛
se
I recall a repair job on a transistor radio that was humming.
Turned out some black paint was missing.
Modern transistors encapsulations are perfectly opaque.
Dan.
Yesterday night, in one episode of NCIS at the TV : "If I told-it to you, I will be obliged to kill-you.".
I feel like Sidney Zweibel (as played by Jeff Goldblum) in Buckaroo Banzai, as he figures out the origin of the Red Lectroids. No, don't you get it? The lights in the sky are LEDs! Orson Wells hypnotized us to think they are UFOs.
But seriously, the effect is small but not nonexistent. If you want to use a LED as a photodiode you will get best results with some reverse bias (photoconductive mode) or zero bias (photovoltaic mode), and progressively poorer response with forward bias. I'm not sure if the intrinsic quantum efficiency as a photodetector changes that much with bias, but whatever feeble photocurrents do flow are developing a voltage across a rather low impedance that can be read as the tangent to the V versus I curve. So not only is the photocurrent small with respect to the forward currents, but the impedance is small as well. So providing a bias source for a modest current generator, which is the prevalent application, it will be a very small effect. And as I've corroborated with my own recent measurements, the noise level is so low at least for the HLMP-6000 parts as to suggest no need, though it won't hurt, to low-pass filter local to the base of the bipolar in the generator --- the rbb' and half re noise of the transistor, and the emitter resistor thermal noise, are going to dominate.
Actual measurements of the photodetection effect with a forward-biased LED are fraught with gothchas as has been discussed, since the electric fields are going to couple significantly and their effects have to be carefully subtracted. The strategy of putting some optical filtering between emitter and detector and noting the difference with and without is a nice reality check.
Anyone for a lock-in amplifier?
But seriously, the effect is small but not nonexistent. If you want to use a LED as a photodiode you will get best results with some reverse bias (photoconductive mode) or zero bias (photovoltaic mode), and progressively poorer response with forward bias. I'm not sure if the intrinsic quantum efficiency as a photodetector changes that much with bias, but whatever feeble photocurrents do flow are developing a voltage across a rather low impedance that can be read as the tangent to the V versus I curve. So not only is the photocurrent small with respect to the forward currents, but the impedance is small as well. So providing a bias source for a modest current generator, which is the prevalent application, it will be a very small effect. And as I've corroborated with my own recent measurements, the noise level is so low at least for the HLMP-6000 parts as to suggest no need, though it won't hurt, to low-pass filter local to the base of the bipolar in the generator --- the rbb' and half re noise of the transistor, and the emitter resistor thermal noise, are going to dominate.
Actual measurements of the photodetection effect with a forward-biased LED are fraught with gothchas as has been discussed, since the electric fields are going to couple significantly and their effects have to be carefully subtracted. The strategy of putting some optical filtering between emitter and detector and noting the difference with and without is a nice reality check.
Anyone for a lock-in amplifier?
Because I was in class the day they taught about the difference between reverse bias and forward bias. Apparently, you missed that day. 😀
I just tried it both ways (with and without DC bias) and I got a reading on both with my scope and my HP339 meter.
Last edited:
If someone from Japan told you, he'd have to kill himself.
Classic is a matter of wind heading.
Yes I noted something funny. You stated that you got no signal. I get lots. Everyone else who has tried it has shown photo sensitivity. The conclusion is very simple YOU ARE WRONG.
With the cardboard blocking the flash there is no signal.
Way back in the days of Popular Electronics they even had a project to use an LED as both the transmitter and detector for a light beam communication.
Dmitri showed almost zero signal. Mooly got a minuscule signal. Your results look more than a bit funny, but perhaps you don't see why.
And note that the receiver in the PE project was not a forward biased diode.
No the only funny thing is you said you used a strobe. So did I. With a strobe the signal is not minuscule. Everyone else has a signal. Why did you not get one with a strobe?
You also stated you don't get a signal from a forward biased diode. Again not the results of the other folks. Why don't you repeat the tests and see where you have erred?
What part of the trace do you not know how to read or understand? Perhaps I can help you.
You are masterful at the "left out important detail." 😀 Hint: does one expect an electron to be liberated with the absorption of a photon?
TBH, I wasn't looking at -130dB with illumination right up against the LED, so it could well have been there. But since that's ridiculously low, who cares?
TBH, I wasn't looking at -130dB with illumination right up against the LED, so it could well have been there. But since that's ridiculously low, who cares?
- Status
- Not open for further replies.