Cyril,

The simplest way to answer your question is by analyzing the simple ciruit, you have given to us. It is a so-called dual-input OR-gate in diode-resistor logic (DRL). How does it work?

Let us identify an interval of voltages, e.g. [0V, 0.4 V] as logical 0 or "L", and another interval of voltages, e.g. [2.4V, 5V] as logical 1 or "H" for input voltages, and intervals [0V, 0.8 V] as logical 0 or "L", and another interval of voltages, e.g. [2.0V, 5V] as logical 1 or "H" for output voltages. Let us furthermore assume that threshold voltages of diodes would be 0V .. 0.3V (Ge-Diodes, or zero-threshold diodes).

If now one of the inputs would be on "H"-voltage, then the voltage drop over the resistor (i.e. the output voltage) would be on (output)-H-level. Only if both inputs would be on L-level, the output would also be on L-level. You see: this is the performance of an OR-gate.

Would you replace the diodes by resistors, then you could no longer guarantee a similar performance, since then, you would have linear voltage-divider behavior. Think for instance that one input would be on H-level, the other on L-level. Then the output level would depend on the ratios of resistances, maybe just inbetween L- and H-level, but under no cirumstances, you could verify the OR-behavior.

The task of the diodes is thus to decouple the sources that feed the inputs of the circuit.

The deeper reason for such a behavior, not only in DRL, is that binary or m-ary AND or OR functions are no longer linear functions between two linear vector spaces (they may be bilinear or multilinear functions in some cases). I.e. you need suitable elements with non-linear behavior for implementation. The probably simplest and cheepest elements with such a behavior are diodes. This is the reason, why you find them in so many applications.

Hope that answers part of your questions. Regards, Michael

Hi Cyril

Don't know whether I understood your question correctly or not. Are you asking why in first place we have a complex multiple source and then have a complex circuit to make it work?

Cyril,

The simplest way to answer your question is by analyzing the simple ciruit, you have given to us. It is a so-called dual-input OR-gate in diode-resistor logic (DRL). How does it work?

Let us identify an interval of voltages, e.g. [0V, 0.4 V] as logical 0 or "L", and another interval of voltages, e.g. [2.4V, 5V] as logical 1 or "H" for input voltages, and intervals [0V, 0.8 V] as logical 0 or "L", and another interval of voltages, e.g. [2.0V, 5V] as logical 1 or "H" for output voltages. Let us furthermore assume that threshold voltages of diodes would be 0V .. 0.3V (Ge-Diodes, or zero-threshold diodes).

If now one of the inputs would be on "H"-voltage, then the voltage drop over the resistor (i.e. the output voltage) would be on (output)-H-level. Only if both inputs would be on L-level, the output would also be on L-level. You see: this is the performance of an OR-gate.

Would you replace the diodes by resistors, then you could no longer guarantee a similar performance, since then, you would have linear voltage-divider behavior. Think for instance that one input would be on H-level, the other on L-level. Then the output level would depend on the ratios of resistances, maybe just inbetween L- and H-level, but under no cirumstances, you could verify the OR-behavior.

The task of the diodes is thus to decouple the sources that feed the inputs of the circuit.

The deeper reason for such a behavior, not only in DRL, is that binary or m-ary AND or OR functions are no longer linear functions between two linear vector spaces (they may be bilinear or multilinear functions in some cases). I.e. you need suitable elements with non-linear behavior for implementation. The probably simplest and cheepest elements with such a behavior are diodes. This is the reason, why you find them in so many applications.

Hope that answers part of your questions. Regards, Michael

Thank you, Michael! Very comprehensive answer at a low circuit level:) BTW your insertion below

"Would you replace the diodes by resistors, then you could no longer guarantee a similar performance, since then, you would have linear voltage-divider behavior. Think for instance that one input would be on H-level, the other on L-level. Then the output level would depend on the ratios of resistances, maybe just inbetween L- and H-level, but under no cirumstances, you could verify the OR-behavior."

...is suitable more for RTL gate discussed here

https://www.researchgate.net/post/How_were_basic_Boolean_functions_implemented_in_electronics_What_were_the_general_idea_structure_and_operation_of_the_first_logic_gates

I invite you to join this question. Regards, Cyril

Hi, Vivek! It would be well if I illustrate the question by a series of pictures... and mybe I will do it to clarify the topic.

Saying "voltage source", I actually mean "circuit output producing voltage" (i.e., having low output resistance). Actually, we have only one voltage source supplying many digital devices (e.g., logic gates) but we can consider them as separate voltage sources. And the problem is if we can connect these outputs in parallel (join them), e.g. to realize OR function...

Hi Cyril,

the simplfication of the complex sources by using decoupling diodes is needed because you like to use the output of this sources in parallel. Like two batterys with unknown loading conditions. To avoid high currents between this loads with unknown state you have to decouple them. A simple way is to add series resistors to limit the current or two diodes as valves for the decoupling. Both are common way to solve the problem of balancing currents.

Hope that helps a bit. Regards Günter

Are you simply analyzing the logic, or are you trying to create a circuit?

If you are trying to create a circuit, why not just use a Logic AND/OR Gate?

This way you could have the best of both worlds... Unless I misunderstood.

Your question is a little confusing, as you are identifying circuit design, so if we knew your end result, we may be able to better help you :)

It prevents the reverse power which is reflected back from the front edges of the wave guide or any guided medium (electrically). Also as per the Kid Cowles. if you are implementing any logical operation circuit at the high frequency at that time also it isolates the input from the output.

Do write specific application as application wise the use of the same will change.

Dear Kid and Neetirajsinh, my mission in this world is to reveal the fundamental circuit ideas (I don't know why but it is true:) So, with this question I try to clarify why, in some cases, we should connect diodes between the input voltage sources and the circuit input... or why there are such diodes inside the "diode-based" IC logic gates (DL, DTL, TTL...)

I am looking for a more general answer that shows the fundamental idea, not its specific implementations (I am trying to see the forest for the trees:). I am sooner "circuit thinker" than "circuit doer":) although I have used this technique hundreds of times when mounting security systems in cars.

http://wiringinstructions.com/v2/training_guides/Understanding%20%20Diodes.pdf

http://ccs.exl.info/diodes_alarms.html

http://www.the12volt.com/diodes/diodes.asp

It also gives an extra protection to the source from worst case sink current which may flow to the source and damages the source...

So, a diode connected in series and in the same direction to a "general purpose" voltage source (i.e., allowing both sourcing and sinking current) makes it "only a sourcing source":)

It is interesting what happens if we reverse the diode... then it makes the source "only a sinking source":) But what does it mean? Is it a reliable connection? I think so...

generally as per the fundamental concept when you reverse bias the diode it will block the forward path electrically and it will provide the constant voltage across it, whatever the configuration will in which you use. You can't say sinking source. sink and source are two different terms. Sink side can work as a source and source can also work as a sink but at the same time a source can't work as source and sink.

Cyril Mechkov,

In your statement, both the cases are different.

- In first one in which place of the diode is there in between circuit and power supply, in this one that diode act as protection switch (or protection diode). It will allow the supply voltage or current to flow when the flow is from source to circuit or it will isolate the source from circuit to prevent damage.

- In the second case the same concept (forward bias and reverse bias) has been used to design the ICs n all. As in binary logic 5V is treated as logic 1 and 0V is treated as logic 0. When diode is in Forward bias it will allow full voltage to be given to following circuits so it will work as short circuit. When it is in reverse bias it will act as open circuit, so at the another side of the diode there will be either 5V or 0V.

Got it ?

@Cyril

As far as the current can be handled by diode yes...it can make a source to a sinking source... It seems logically correct..

Dear interlocutors, thank you for the so interesting thoughts. It seems the concept "source/sink" is relative and directly connected with the concepts of "ground" and "Vcc", i.e., when the load is grounded or powered by +V... only then it makes sense...

So, we can talk about a "source" and "sink" when we see only the one end of the load... Then, if we "push" a current into this point, we are a "source"; if we draw a current from this point, we are a "sink"...

This 2-terminal element produces a current that exits its one side and enter its other side; so it can be connected as we want - as a "source" or as a "sink"...

Take care with the use of the term sink and source. Mostly we talk about voltage sources. Also the sink is a low impedance termination, means voltage "source", working like a sink if connected to a voltage source. Try to define the diode circuit description with current sources or sink. This shall be a high imedance devices and diodes work as valves for the currents.

Thank you Cyril for this interesting topic, bringing us back to the roots of electronic circuit analysis.

Exactly, Guenter! I continue thinking about these arrangements. An example of a "voltage-less" sink is the lower NMOS transistor of a CMOS pair... but I thought that a voltage source acting as a current "sink" was a more exotic and not well-known example of a "sink"...

Mind well lower NMOS is the path which connects the output with the ground rail. We can't say that NMOS is kind of sink.

Maybe the NMOS transistor connected to ground is a kind of a "passive sink":)?

Neetirajsinh, you made me seriously think what is "source" and what "sink"...

Blocking the signal if its in reverse direction!!!!!!. Diode cant use as buffer the circuit like transistor.