Today I measured Vf of 16 unmarked through hole diodes. I have no data on them at all, and as they're unmarked I suspect some may be seconds / out of spec. The results surprised me...
In short... why? Tamb = about 22C Results were rounded to nearest 5mV. The meter is a uncalibrated cheapie, the datrons are due for repair. Packages include plastic, epoxy & glass. All look like
Didn't find your answer? Ask the community — no account required.
T
Tim Williams
Physics, of course!
For junctions of equal doping (concentration and profile), one would expect Vf to vary as log(area). High current diodes have lower voltage drops at a given current, since the current density is lower, and therefore lower on the V-I curve (weighted by junction size).
Doping isn't constant, though. High speed diodes will have more abrupt doping (leading to lower avalanche/breakdown voltage* and faster recovery), and higher concentration, which leads to a greater built-in potential, and still higher voltage drop than a rectifier diode of otherwise identical dimensions.
If you have a high voltage power supply, test them as zeners. Use a pretty small current ( of spec. The results surprised me...
A
amdx
Clear glass? Light intrusion? Mikek
This email has been checked for viruses by Avast antivirus software.
http://www.avast.com
T
tabbypurr
Ahh. Handy to know. You might have noticed solid state physics was not one of my strong areas :)
NT
T
tabbypurr
Some probably did get light on the junction. If I get another round tuit I should remeasure them and see how they respond to light, see what Vr they break down at etc.
NT
T
Tim Williams
I've measured 1N914 dark and under intense light (point blank from a 1W white LED, should be brighter than direct sunlight) and found only a few mV difference (measurable, but not much more than thermal due to handling). I don't think the junction extends very far to the edge of the die, so it makes a very poor PV cell. (Can anyone confirm if it's a traditional planar construction? I shouldn't think it's got the junction extending to the edge, not for anything Vf > 30V.)
Set your DMM for DC current and see how much photo-current you get.
George H.
P
piglet
The diode PV response will be much greater at 800-900nm wavelength light which white LEDs do not produce much of.
piglet
T
Tim Williams
Below the noise floor. When the meter is below 400mV, circuit resistance is dominated by diode leakage (it drops to the "high impedance" range). So the current is vanishingly small.
In contrast, a 6AL5 measures a healthy -0.5V and 2.5uA (exponential region), just like a semiconductor diode with light or a thermal gradient (as odd as that might sound :) ). Measurements of which, did involve high value shunt resistances (using 1M external, or by nulling methods).
Efficiency peaks in that range, but quantum efficiency is nearly flat from
400 to 800nm. Even then, efficiency only drops arithmetically for shorter wavelengths, not really a "much greater" difference. 3dB not 10dB.
Unless you mean that's an observed effect for these kinds of diodes, which I wouldn't discount: the light has to travel through more thickness (edgewise into the die) before it reaches the junction, and attenuation is quite steep in Si above 800nm or so.
That, of course, depends on construction; some diodes may be diced more or less cleanly, sawn versus cleaved, etc., so that sometimes the junction is closer to an edge and the device is rather photosensitive. Which might suggest other "spooky" behaviors, like trying to use these diodes in high impedance circuits, and having them fail under strange circumstances (like every time the sun shines through a window, or only when people use flash photography near it).
So that said, it might be interesting to examine a wide swath of diode types and see if there's a wide variation in photosensitivity. *Shrug*.