2 diodes in series: why ?

[Nounours18200]

HI,

I have read in this topic: https://www.eevblog.com/forum/chat/vd1212-diodes-replacement/

that a good replacement for a VD1212 diode is to use 2x1N4148 in series.

I apologize for my lack of knowledge, but why a single 1N4148 is not enough ? I do not understand why to connect 2 in serie, because a diode pass or does not pass the current ?

Thank you for your explanation...

[tautech]

The Vf required is the clue to why two in series are required. 

[bd139]

Exactly that.

That's the only characteristic required most most cases for these. They are usually used to give two semiconductor junction drops in amp biasing circuits which is conveniently the same as the B-E drops on the PA transistors added together (or near enough).

You can also replace them with a BJT and a couple of resistors (or a trimmer) for a precise drop but it's hardly worth the effort in a reasonably well designed circuit. Look up "Vbe multiplier".

[ledtester]

I apologize for my lack of knowledge, but why a single 1N4148 is not enough ? I do not understand why to connect 2 in serie, because a diode pass or does not pass the current ?

Diodes can have different V-I curves. THe forward voltage (the point in the V-I curve where the diode starts conducting) for the 1N4148 is around 0.65 V whereas the forward voltage for the VD1212 is around 1.3 V. Putting two 1N4148 diodes in series will approximate the V-I curve of a VD1212.

The same thing is going on with LEDs -- red, yellow, green, blue LEDs all have different voltage drops ranging from around 2 V to 3.5 V.

Update: Here's are some V-I curves for LEDs. Note that the curves start bending at different voltages.

[001]

The same thing is going on with LEDs -- red, yellow, green, blue LEDs all have different voltage drops ranging from around 2 V to 3.5 V.

Is it common constant voltages for LEDs? Can I use LED as a zener diode with low capasitance?

[bd139]

Yeah you can do that. It's quite handy. You can also use LEDs as varactors.

[Someone]

The same thing is going on with LEDs -- red, yellow, green, blue LEDs all have different voltage drops ranging from around 2 V to 3.5 V.

Is it common constant voltages for LEDs? Can I use LED as a zener diode with low capasitance?

Yeah you can do that. It's quite handy. You can also use LEDs as varactors.

Noting they have a "bonus" photodiode effect which can make those uses a little complicated.

[AG6QR]

Is it common constant voltages for LEDs? Can I use LED as a zener diode with low capasitance?

To a reasonable approximation, the frequency of light emitted by a LED is related to its Vf in the same way that the frequency of a photon is related to its energy expressed in electron-Volts.

The Planck constant allows you to convert from one to the other.

Physics is pretty cool.

[ledtester]

To a reasonable approximation, the frequency of light emitted by a LED is related to its Vf in the same way that the frequency of a photon is related to its energy expressed in electron-Volts.

In researching my answer I came across this SE thread which suggests that Vf is determined more by the resistivity of the materials used in the manufacturing of the LED:

https://electronics.stackexchange.com/a/396384/95488

Lots of good comments and food for thought there.

[AG6QR]

In researching my answer I came across this SE thread which suggests that Vf is determined more by the resistivity of the materials used in the manufacturing of the LED:

https://electronics.stackexchange.com/a/396384/95488

Lots of good comments and food for thought there.

There are lots of good comments, and a lot of them talk about how Vf is related to the wavelength of the light produced.  It's a common undergrad (or even high school) physics experiment to use colored LEDs to measure the approximate value of Planck's constant.

https://www.google.com/search?q=led+planck+constant+experiment

But to get an accurate value, you need to determine the threshold voltage that just barely begins to produce light output, as explained in one of the links from the Google search.  Here's one of the experimental procedures: https://www.scienceinschool.org/2014/issue28/planck .  Others suggest a more crude approach: http://pages.physics.ua.edu/lab10x/ph102/PDF/Plank%27s_Constant_LAB.pdf

The Vf at a nominal current of 20mA or so may be significantly above the threshold voltage, and may be influenced by a number of other factors, including resistivity of materials.  Still, even at reasonable working currents, you'll find that the Vf is still related to the wavelength of light produced.  Everyone agrees that, at similar light outputs and currents, red LEDs have a lower Vf than green ones, which have a lower Vf than blue ones.  Everyone agrees that Vf is related to the bandgap voltage of the doped semiconductors, and that the color produced is also related to the bandgap voltage.

[Nounours18200]

usefull info !

I currently only have 1N4149 : can I use them (in series) instead of 1N4148 ?

The datasheet  shows that they have the same Vfmax = 1.0A, but i am not totally sure...

Thanks for your replies,

[rfeecs]

The frequency of light emitted depends on the bandgap.  Vf also depends on the bandgap, but also depends on current.  As you lower the current, Vf goes down, all the way to zero.  So the number for Vf you get is arbitrary, depending on the current you choose to measure it.

In these experiments, they choose a current that works out to give approximately the right answer.

Here is yet another proposed school experiment that uses a potentially more accurate way to measure bandgap by varying the temperature:

https://hal-amu.archives-ouvertes.fr/hal-01266699/document