Talk:Circuit Idea/Virtual Ground

From Wikibooks, the open-content textbooks collection

Jump to: navigation, search

Contents

[edit] What virtual ground is (after Wikipedia page of the same name)

(moved and adapted by Circuit-fantasist from Wikipedia Virtual ground discussion page)

[edit] Virtual versus analog ground

Virtual ground implies ground potential at the node. Adding a component to a ground will have no effect if the other component connection is ground. There is no potential difference, hence no current flows. This may not be true for a virtual ground. In an inverting amplifier adding a capacitor between ground and the virtual ground at the (-) op-amp input can make the amplifier unstable. Stray capacitance from a probe is often sufficient. A warning should be added. Djhk 18:26, 23 September 2007 (UTC)

Djhk, you hate virtual ground but this is a phenomenon, fact that exists whether or not we like it. What do we do when we haven't an access to a real ground? Then we use a virtual ground (a typical example is creating of a split supply by one voltage source). Sometimes, we use a "shifted" (e.g. "lifted") ground instead an exact ground.
In a broadest sense, any circuit point having a voltage that is kept artificially (e.g., by means of a negative feedback) can be named virtual ground. In a literal sense, a virtual ground is a result of subtraction of two voltages having opposite polarities. We can say that a virtual ground is actually an active ground. Every active thing in this world is more stable to influence than passive one. Circuit-fantasist (talk) 17:34, 24 November 2007 (UTC)
You are confusing Virtual ground with analog ground. You need to split the topic into virtual and analog ground; combining the two creates confusion. Djhk 02:49, 3 December 2007 (UTC)
What do you mean when you say "analog ground"? What is the difference between a virtual and an analog ground? Circuit-fantasist 16:27, 3 December 2007 (UTC)
Analog ground is just "normal" ground, i.e. 0V. Virtual ground behaves like a nullator connected to ground. -Roger 16:34, 3 December 2007 (UTC)
Roger, I am not sure if the more abstract concept of nullator can help understanding the worldly virtual ground phenomenon unless you explain what a nullator actually is. Well, let's try to "decode" what "The inputs of an ideal opamp (with negative feedback) behave like a nullator, as they draw no current and have no voltage across them" means. Maybe, the Baron Munchhausen's idea of bootstrapping can help us. Circuit-fantasist 13:05, 5 December 2007 (UTC)

[edit] Analog ground

Roger, you are right. There is nothing special in the term of "analog ground"; it is just a piece of wire. Analog ground is nothing else than a separate wire connected to the real (ordinary, normal, common) ground and earmarked only for analog devices; another (just the same) wire is intended only for digital devices. In this way, analog and digital grounds are two branches of the common ground. No matter whether the ground is analog, digital or just a common ground; it is a static, passive "component" that can't resist influences, if they are too significant or, if the ground internal resistance is too high. Circuit-fantasist) 13:05, 5 December 2007 (UTC)

[edit] Virtual ground

Virtual ground is something else (excepting that it has the same zero voltage as an "ordinary" ground). Virtual ground is a unique worldly, technical, electricity and finally an electronics phenomenon; there is a great philosophy behind this idea, which deserves a special attention. In electronics, contrary to the "static" ordinary ground, virtual ground is a "dynamic" point whose potential is kept zero by artificial means (e.g., parallel negative feedback). For example, in the popular circuit of an inverting amplifier, an op-amp "observes" the virtual ground point and makes all that is possible to keep it at zero voltage (more precisely speaking, "...to keep it equal to the potential of the non-inverting op-amp input...") If we "blow" a current into this point, the op-amp will decrease its output voltage to "suck" a current from the virtual ground point; if we "suck" a current from this point, the op-amp will increase its output voltage to "blow" a current to the virtual ground point.

Generally speaking, great (circuit) ideas do not depend on the specific (component) implementation (tube, transistor, op-amp, etc.). They are non-electrical; they are eternal. In order to reveal the philosophy behind circuit ideas, we have first to show to visitors (that are human beings) examples from their human routine. These analogies will serve as educational mainstays that help understanding.

That is why, I have written the lengthy introductory part of the article about virtual ground (see the old version [1]) where I have shown various situations illustrating virtual ground phenomenon. Only, someone has moved them to the end of the article [2] and finally, somebody else has removed them at all [3]. As I can see, in this way, the article will gradually become the initial poor and "sterile" page [4].

This is the great paradox of Wikipedia: every narrow-minded and even anonymous visitor can edit and even remove frivolously and irresponsibly a text that is created by an editor having a philosophy about the phenomenon discussed. That is why, I have abandoned Wikipedia and started Circuit Idea. Circuit-fantasist 13:05, 5 December 2007 (UTC)

[edit] The philosophy behind the phenomenon

In the Virtual ground wikipedia page I have just tried to reveal the secret of the phenomenon, to show that it is not a "black magic" but a well-known everyday occurrence. We mustn't forget we write this encyclopedia for ordinary human beings, not for computers. Human beings can easily understand abstract electronics concepts by conveying their knowledge of life to concrete circuit phenomena. For this purpose, we have first to show some typical exemplary situations from human routine that are analogous to the phenomenon discussed, then to generalize them into a philosophy behind the phenomenon (everything in this world has own philosophy) and finally, to show the concrete circuit solutions.

Special electronics sources present concrete circuit solutions; an encyclopedia has, first of all, to reveal the philosophy behind them. In the case of the phenomenon discussed here, we have first to show in the most general (non-electrical) way that, in this case:

  • if there are two opposite (positive and negative) power sources and
  • they are connected each other by a resistive medium,
  • a zero point (virtual ground) appears somewhere along the medium and
  • the resistive medium wastes energy continuously as the carrier flows continuously.

In electronics, all the circuits with parallel negative feedback (all the op-amp inverting circuits) operate in this way.

Contrary, all the circuits with series negative feedback (all the op-amp non-inverting circuits) operate according to the opposite great phenomenon - bootstrapping:

  • if there are two identical (the both positive or the both negative) power sources and
  • the one of the sources follows the other (no matter if they are connected each other),
  • a zero difference ("differential virtual ground") appears between them and
  • no energy is wasted as no carrier flows.

Combine the two phenomenon and you can explain all the circuits with negative feedback.

A month ago, I tried to show, in a human-friendly manner, this truth about the great virtual ground phenomenon (it deserves that) by means of various analogies. Circuit-fantasist 09:10, 15 December 2007 (UTC)

[edit] A voltage divider viewpoint

For me, a virtual ground can be created either by buffering (to the degree necessary) a voltage divider to provide dual rail supplies or using negative feedback to make the potential between two points zero. For the first case the explanation is as simple as the voltage divider equation, very little elaboration should be necessary. Roger 23:32, 15 December 2007 (UTC)

You are right: the bare electrical device - the voltage divider, after a "little elaboration", is the simplest circuit for creating a virtual ground.
Only, if we want to obtain a true zero-voltage virtual ground, we have to "move" down its ground by another negative power source (to "pull" down the lower resistor R2). Actually, this "shifted" voltage divider acts as a parallel voltage summer that sums two opposite (weighted) voltages according to the superposition principle:
VOUT = V1.R2/(R1 + R2) + V2.R1/(R1 + R2) = 0, therefore V1/R1 = -V2/R2
If we respect the proportion above, the output of this nice passive circuit will serve as a virtual ground; it is imperfect, "static", "soft", influenceable, etc. but yet it is an artificial, virtual ground because it has zero voltage potential without a direct connection to the "true" ground. It can serve as an analog ground but yet it is a virtual ground. Analog ground is a true, natural ground while a virtual ground is an artificial ground. Can I conclude we have come to an agreement?
This text is very important as it dethrones a widespread misconception - that the virtual ground is an electronics phenomenon inherent only of op-amp inverting circuits. Virtual ground is an electrical phenomenon; it had existed plenty of time before the ubiquitous op-amp to appear. Circuit-fantasist 08:07, 19 December 2007 (UTC)

[edit] An op-amp viewpoint

Again, a virtual ground can be created either by buffering (to the degree necessary) a voltage divider to provide dual rail supplies or using negative feedback to make the potential between two points zero. The second case is simple too. Vo = A*(V+-V-), let A->infinity then (V+-V-) must -> 0 if Vo remains finite. Or just use two of the basic negative feedback opamp assumptions: Rin is infinite and V+ = V-. Roger 23:32, 15 December 2007 (UTC)

You have suggested using the classical viewpoint at op-amp circuits with negative feedback that has been reigning from dozens of years in electronics resources: Vo = A*(V+-V-), let A->infinity then (V+-V-) must -> 0 if Vo remains finite.
The trouble of this explanation is that it does not explain anything. It works well, if you have no time for lengthy explanations (or, if you just do not like to explain circuits) but yet you would like your students (readers) to get to know something about op-amp circuit operation; then you just toss this one-line "explanation" to them and they will not ever bother you asking "naive" questions. Another example is Ohm's law. You can present it as a fraction a = b/c and then to speculate how a depends on b and c without showing that a is a flow-like, b is a pressure-like and c is an impediment-like quantity, without speculating what causes what, how a voltage and current sources behave, what the resistor does, etc.
But this approach fails completely, if you want readers to understand really the circuit operation. For the first time, I saw this text in 1975 when I entered the university and was trying without success to understand what an op-amp really does in this strange circuit with two resistors. What is wrong with this presentation?
The problem is that there is no TIME in this Vo=A*(V+-V-) explanation. The op-amp is considered as a non-inertial, proportional device that acts momentarily (the input and the output voltage change simultaneously). Only, if you try to understand the circuit in this way (how the op-amp has achieved the equilibrium), you will get into a vicious circle. The paradox is that, in order to understand op-amp amplifying circuits, we have to think of the op-amp not as an amplifier but as an integrator, as something slow, inertial and "thinking" that observes the virtual ground and reacts if it tries to change. The best way of understanding the op-amp behavior is to imagine that we perform its functions (i.e., applying the synectics's empathy). The truth about op-amp circuit operation is hidden rather in the transition time interval where the op-amp reacts to a sudden input change than in the next calm settled state when it "sleeps" (...to be continued...) Circuit-fantasist 08:07, 19 December 2007 (UTC)
I see no problem with using the differential amplifier equation to prove/explain virtual ground. If you're explaining virtual ground in the context of an opamp circuit then you should assume your audience understands how opamps operate. If they have a problem understanding it at this point, then their understanding of opamps is probably not sufficient. In which case they should go back and understand the ideal differential amplifier equation.Roger 16:17, 17 December 2007 (UTC)

[edit] Voltage divider versus op-amp viewpoint

I usually think of the two examples of virtual ground as being completely different, but if you prefer to think of them as two special cases then thats just as valid. The negative feedback case to me is the "truer" virtual ground since it behaves exactly like ground (within limits) and is actually at 0V. Roger 16:17, 17 December 2007 (UTC)

[edit] Is virtual ground a relative concept?

The divider case is a bit different since the voltage is usually not at 0V, but we simply decide to make every voltage relative to the virtual ground. So if we have a 10V supply and divided it down to 5V we could define 5V to be our new (virtual) ground and now we have a +5V, -5V supply. There isn't really any circuitry involved, it just goes back to the fact that voltages are always relative to some reference. Roger 16:17, 17 December 2007 (UTC)

Wonderful, this is another clever trick that complements the example above. There, we have "pulled" down the lower resistor by another additional negative power source, in order to "move" down the divider's output voltage until it becomes equal to zero (a virtual ground). Here, you suggest just to abandon the true old ground and to make the existing 5V output voltage serve as a new (virtual) ground. I think this text (especially, "...voltages are always relative to some reference...) can describe very well Fig. 4b. Circuit-fantasist 18:04, 17 December 2007 (UTC)

[edit] A virtual ground is another ground

Only, I am a little confused on this arrangement... Is this really a virtual ground? I'm asking this question because, when we create a virtual ground, there is a true ground as well and the two grounds have zero voltages (regarding to what?); no, they have the same voltages. Virtual ground is another ground; it is a "copy" of an existing true ground. In this arrangement we have only one ground. What kind of ground is it? Virtual or true? It seems that it is a true ground... Circuit-fantasist (talk) 18:34, 17 December 2007 (UTC)

It seems, in order to say that we have a virtual ground, another (real) ground has to exist; so, virtual ground is a counterpart of a real ground. In all the circuit diagrams of the article (see the old version), there is a real ground (the common point between the two power sources) and there is a virtual ground (even if it is a voltage divider's output). Circuit-fantasist 18:47, 18 December 2007 (UTC)

[edit] Can a virtual ground be relative to itself...

I think this is getting back to the problem that the term "virtual ground" isn't clearly defined, perhaps anywhere. I'll get back to this issue when I reply to Wtshymanski's post below. Anyway, yes, it is a virtual ground, relative to itself. Relative to real ground it is +5V (i.e. there is a p.d. between real and virtual ground). Of course in your example real and virtual ground just happen to have the same voltages relative to real ground, hence no confusion. See [5] for examples. Roger 18:57, 17 December 2007 (UTC)

[edit] ...or a virtual ground can be relative only to an existing ground?

The "relative to itself" concept sounds very intriguing and philosophical but I am not sure if I have made sense of its meaning. For now, I continue holding "relative to existing ground" concept. I have the feeling that I have finally realized what a virtual ground is and how to define the term. Here is my story about the scope of the virtual ground term; this is a tale about the evolution of a natural ground into a virtual one (we might move later this text to Ground although this page is maintained by very difficult wikipedians that are not inclined to such natural discussions). Circuit-fantasist 18:47, 18 December 2007 (UTC)

[edit] A tale about virtual ground evolution

...Voltage is a differential quantity; so, to process "true" voltage signals we need differential-input (-output) devices. If we want to deal with single-ended voltages (potentials) that is usually more conveniently, we need a voltage "mainstay" (a ground) where to connect one of the device terminals and regarding to which to measure the voltages (potentials). For this purpose, we usually use one of the power source terminals as a natural, true, genuine ground; thus, we can deal with only positive or only negative voltages. If we want to have both positive and negative voltages, we use an internal power source point as a ground (for this purpose, we usually connect two or more voltage sources in series and in the same direction, see Fig. 4a). If there is no such a compound power source, we create an artificial but still real ground, for example, by using the bare voltage divider (see Fig. 4b). This ground is a real one simply because there isn't any other ground (more precisely speaking, there is such a point somewhere inside the source but we have no access to it). Finally, we "clone" the ground (no matter it is true or artificial) by making another circuit point (virtual ground) follow the ground; for this purpose, we usually make an op-amp with negative feedback do this donkeywork:)... Circuit-fantasist 18:47, 18 December 2007 (UTC)

Retrieved from "http://en.wikibooks.org/wiki/Talk:Circuit_Idea/Virtual_Ground"