Talk:Electronics/Op-Amps

From Wikibooks, the open-content textbooks collection

Contents 1 Reorganisation 2 Notation 3 Interested Contributor. 4 merge 5 Fully differential op amp 5.1 The challenge 5.2 Presenting FDA in a more attractive way

[edit] Reorganisation

Ok firstly this page is very confusing. So this is what I'm going to do. Firstly I'm going to remove ideas of coupling, oscillators, and offsets until the ideal op-amp is discussed.

Firstly oscillators because that clouds the use of Op Amps. They are first and foremost signal amplifiers. Secondly coupling because this is also a more advanced idea to real amplifier circuit and Op Amp have high CMRR, common mode rejection ratios which means coupling is generally unnecessary. Where as a real Common Emitter amplifier, for instance doesn't. Which means it is not neccesary. Lastly offsets, If this is your first sight of an Op Amp then how would know of or dream up an offset.

That which was removed is below. - IknowNothing

---

Oscillator When positive feedback occurs in an Op-Amp/Transistor it becomes an oscillator. Unless, of course, it becomes a Comparator.

Coupling ways of allowing an AC signal to go from on stage to another without changing the bias of the 2 stages.

Diode. DC voltage drop. When AC passes only allows half the waveform through. This is useful for constructing a power supply.

Capacitive coupling. DC does not pass through capacitors, but AC does. At low frequencies capacitors have high impedance, and at high frequencies capacitors have low impedance.

Transformer coupling. Inductor based. AC does not pass through inductors, but DC does. Low impedance at low frequencies and high impedance at high frequencies.

---

My next move is to move any comments of what an Op Amp is physically constructed from until a discussion of real op amps. Since it just clouds the issue. - IknowNothing

[edit] Notation

I have changed G_openloop to A_vo and use A_f for closed loop gain. - IknowNothing

[edit] Interested Contributor.

In am interested in contributing information on the "operational amplifier" used in a differential configuration. In particular, the case in which the input is a "floating voltage source" (i.e. magnetic readhead). This topology would be a variant of the difference amplifier whose input is typically two individual voltage sources each referenced to ground.

Do you think there might be any interest in this topic? - hgmjr

Yes. I don't quite understand what distinguishes the two cases you mention. I suspect I'm not the only one. What about the w:Instrumentation amplifier? Is that one of the two cases, or is that a third case? --DavidCary 21:52, 17 October 2005 (UTC)

[edit] merge

Since Circuit Theory:Op Amps Electronics/Op-Amps talk about exactly the same thing, should we merge them? --DavidCary 23:00, 17 October 2005 (UTC)

There is going to be alot of cross-talk between the Electronics and Circuit Theory books, although the later are being prepared specifically to tend to a different audience then the former. Also, I am working on creating an entire "curriculum" of sorts in Electrical Engineering, and I want to make sure that there is a firm foundation in place on these subjects before more advanced books on the topic get written.

I will admit that Op-Amps specifically do not play a particularly central role in the Circuit Theory book, or in the EE curriculum at large, and would conceed that perhaps the Op-Amp page in particular would not adversely affect the project if it were merged (or even removed entirely) from the Circuit Theory book. Other pages however, i would fight the motion whole-heartedly. --Whiteknight _TC^E 14:33, 19 October 2005 (UTC)

I don't know how you came to that conclusion. I'm an EE and I can tell you that the introductory circuit analysis courses that EEs take INCLUDES op-amps. It's a very basic part of the curriculum. What's the difference between the audiences between circuit theory and electronics? I don't see any. 70.18.253.241 22:09, 19 October 2005 (UTC)

Well, I'm an EE as well, and we learned about Op Amps one week, and the subject never once came up again. I was under the impression that it simply wasnt that big a deal considering the little treatment of it that I recieved in my education. I certainly don't think that there is enough material on Op Amps to span 2 books, so I wont bother trying to make that happen. Also, I came to the conclusion that the two books are focused towards different audiences for a number of reasons that I have enumerated (so far without opposition) in a number of places, including my user page, and the Circuit Theory talk page. --Whiteknight _TC^E 02:37, 20 October 2005 (UTC)

I read your criteria and it appears the electronics book version does treat it like a black box for the most part, which would fit your criteria. I really can't see you justify dropping it all together, I find it strange that your curriculum barely covered it, but I wouldn't exclude it just because yours didn't. They're an integral part of introductory circuit theory classes for a reason, and I think it'd be best not to question the judgement of such a unaninmously implemented program without some better justification than "well I personally didn't use it much after that." Your use page says you're a ECE (as opposed to just EE), so perhaps the exra digital slant is why you don't have as much experience with them? Njyoder 22:05, 21 October 2005 (UTC)

My only point is that if there is talk that the material is being covered adequately well in two books, and if the Circuit Theory book cannot cover the material any better then the Electronics book, then there probably isnt much reason to have it covered in 2 separate books. If the concensus is to merge then, we will merge the pages. --Whiteknight _TC^E 01:37, 24 October 2005 (UTC)

ooooo

It seems to me that this is a hardware/software problem. Keep both modules, with

• Electronics:Op-Amps describing the hardware aspect. Which Op-Amps are available, construction and price differences, which to use why (with mathematics for hardware by itself), max/min. statistics/ratings, their construction, even, maybe, different bases available, temperature range for use, for storage.
• Circuit Theory/Op Amps describing how to use them, the various applications (with applicable circuit-based mathematics).
• Links to other pages of WIKIBOOKS, and WIKIPEDIA, aiming for minimum duplication. Think of the many users who are not at all interested in the theory, only in what to do and how to do it quickly and at minimum cost.

QUITTNER 142.150.49.166 20:46, 22 November 2005 (UTC)

ooooo

[edit] Fully differential op amp

[edit] The challenge

(a copy of Talk:Circuit_Idea/Simple_Op-amp_Summer_Design#fully differential op amp)

David Cary wrote:

................................................

TI manufactures several 24-bit ADCs. All of their data sheets recommend using the OPA1632 "fully differential amplifier". (No surprise that it is also manufactured by TI).

A " fully differential amplifier" is very different from a " difference amplifier" or a "w:instrumentation amplifier".

Can Dieter's procedure be applied to a fully differential amplifier? At first glance, I'm guessing that there's one minor and one major change:

• Use equal feedback resistors Rf, one from the + output to the - input, and one from the - output to the + input. Connect the "common" input to the appropriate voltage source. (minor change to the " difference amplifier" )
• The sum of the gains = zero in a properly-designed fully differential amplifier circuit. (Right? major change)
• Calculate resistor values for each input, using Ri = Rf / |desired gain|.
• For inputs with positive gain, connect the calculated resistor value between that input and the + input on the amplifier.
• For inputs with negative gain, connect the calculated resistor value between that input and the - input on the amplifier.
• If all your inputs are differential pairs, then the sum of all the gains is now zero. Done. Otherwise add a ground resistor to bring the total gain to zero.

If that process looks good, please move it to Electronics/Op-Amps#fully_differential_amplifier. (Should we mention fully differential amplifiers on this Circuit Idea/Simple Op-amp Summer Design page? ) --DavidCary 23:15, 5 November 2007 (UTC)

................................................

David, I would like to congratulate you on your success; you have managed to revise and enrich the Daisy's (Dieter's) theorem! By the way, do you have a dog:)?

Two monts ago, a arrived to a similar conclusion that only unbalanced (asymmetric) differential amplifying circuits obey Daisy's theorem (see the old version of Simple_Op-amp_Summer_Design). If they were single-ended circuits, the sum of the input gains would constitute K; if they were balanced (symmetric) differential amplifying circuits (like the basic 4-resistor FDA circuit), the sum of the input gains would be 0. Only, Dieter didn't share my assertion and removed my insertion:(

Well, let's try to record these changes hoping that Dieter will miss out on them:). For a start, I have scattered the examples illustrating the quick design procedure. Circuit-fantasist 17:35, 16 November 2007 (UTC)

Shadow's design procedure only applies to Voltage feedback op-amps. It does not apply to Current mode op-amps, nor to video amps. Please keep the name. This is Shadow's procedure.

It's not a trick. The procedure only applies to the General Summing Amplifier. The gain equation is derived via VSA analysis and simplified for large op-amp gain.

Don't confuse this aproach with the comedy of error approach typically used in electronics. Djhk 18:13, 7 November 2007

Fully differential amplifiers were popular as video amps in the past; also popular inside ICs due to noise concerns.

You need a 4wire interface model.

Look up common mode and differential signals.

In telephony these are longitudinal and metallic signals.

I have rarely seen this discussed in Electronics.

The techniques are well known in telephony, but not publicly discussed.

The TI paper is a crude introduction.

If you're interested look for SLIC circuits.

Professional audio uses 3wire interfaces. The concepts are hidden by transformers.

In summary, this is a special topic. You need to learn some of basics.

Speculation is not good. Makes everyone look like fools. Drop the topic. Djhk 10:02, 22 November 2007

Djhk, I do not yet understand why the sum of the input gains of a fully differential amplifier is one instead zero. Please, elucidate me. Circuit-fantasist 18:59, 23 November 2007 (UTC)

The key to Daisy's theorem is the word ALL. Whenever you draw an incomplete circuit, the gain sum is indeterminate. Before submitting any circuit, test it. Build it on a simulator or better build the circuit. Keep the trash out of the wiki. Don't mention Daisy when speculating. Djhk

Djhk, I do not yet understand why the sum is 1. Please, log in before writing comments and finish them with ~~~~ (these four tildes will be replaced by your user name and the current date).

What do you think about the sketch below? Are there virtual grounds at the two FDA's inputs? Circuit-fantasist 08:56, 24 November 2007 (UTC)

"The sum of the gains = zero in a properly-designed fully differential amplifier circuit". Wrong! Daisy's theorem states that the sum of ALL gains is equal to one. This applies to the +out node, the -out node, or any other node. If we add two sets of gains, each of which has a sum of +1, how do we get a sum of zero. Is 1+1 = 0 ? Djhk 20:59, 24 November 2007

Djhk, look at equation 11 at page 10 of this TI article about FDA. For simplicity, assume R₁ = R₂ = R₃ = R₄ and Uocm = 0. Then, V_OUT+ = 0.5V_IN+ - 0.5V_IN-; the sum of gains is 0.5 - 0.5 = 0. Or look at equation 13 at the same page: V_OUT- = - 0.5V_IN+ + 0.5V_IN-; the sum of gains is again - 0.5 + 0.5 = 0. These calculations corroborate the speculation that the sum of the gains in a FDA is zero. Circuit-fantasist 17:37, 25 November 2007 (UTC)

Daisy's theorem applies to linear circuits. Any single voltage in a linear circuit can be expressed as a sum of gains times inputs. If ALL inputs are included in the equation, then the sum of the gains is equal to one. Taking two gains and adding them may result in any value. In your example the gains refer to different nodes. Write the linear equations for each output, include ALL inputs and you will find that the gains in each equation add to one.

Djhk 03:15, 3 December 2007 (UTC)

[edit] Presenting FDA in a more attractive way

Here is an idea for a more attractive circuit presentation in a case, if you want to discuss this nice circuit. Let's outline roughly the circuit operation.

To apply a differential input signal means to change the two input voltages to the same extent but to opposite directions. For example, let's choose a typical case when, in the beginning the two input voltages are zero. As a result, the FDA has set its output voltages zero as well; no currents flow through the circuit.

Then, we decrease simultaneously the upper voltage V_IN- under the ground and the lower voltage V_IN+ above the ground. At the first moment, the FDA is "surprised" and does not react to this "intervention" (its output voltages V_OUT- and V_OUT+ remain zero). As a result, the FDA's input voltages become V_- = - R₂/(R₁ + R₂) and V₊ = R₄/(R₃ + R₄).

After a while, the FDA "recovers" and, observing continously its input voltages, begins changing its output voltages: V_OUT- to the positive rail V_CC and V_OUT- to the negative rail V_EE. The FDA stops when it manages to zero (approximately) its input voltages (to make its inputs virtual grounds). By the way, there is another virtual ground point here - the middle point inside of the flying load R_L.

It is interesting that, in contrast to the ordinary op-amp differential amplifier, here both the power supplies produce currents flowing through the negative feedback resistors and the flying load. In this case, the (+) FDA's output "blows" a current while the (-) FDA's output "sucks" a current. If R₁ = R₃ and R₂ = R₄, the (-) FDA's output (or, maybe it is more correctly to say an "input") "sucks" all the current of the (+) FDA's output. Am I right?

It is interesting as well to draw more pictures when we vary other circuit quantities:

• add a constant common-mode input voltage to the varying differential input voltage,
• vary only V_IN1 or V_IN2 (converting a single-ended signal to a differential one),
• move simultaneously the sliders of the two potentiometers R₁-R₂ and R₃-R₄ (changing the circuit gain),
• vary the common-mode input voltage while the differential input voltage stays constant,
• vary the voltage V_OCM (not shown on the figure above) that controls the output common-mode voltage (keeping differential and input voltage constant), etc. Circuit-fantasist 16:15, 24 November 2007 (UTC)

Still a fantasy. Can we get a schematic? How about a simulation? Djhk 22:57, 24 November 2007 (UTC)