I'm in my first year and I am trying to understand op amps, I did most of my readings with this book called Op amp for everyone, it was initially good with the basic stuff but once it gets to the op amps with single power supply, the book just doesn't make any sense to me anymore, it then for some reason jumps to simultaneous equations and derive some weird equations and then it just goes downhill for me.

So I know the very basic stuff like the inverting and non inverting configurations with ideal op amp characteristics but I'm looking for a book that becomes more advanced with baby steps, what book did you read to understand op amps when you were a beginner ? I'm finding the transition from beginner to slightly more advanced stuff very difficult.

 

Hello,

Perhaps the following book might interest you:
https://archive.org/details/fe_Op_Amp_Applications_Handbook_Newnes_Walt_Jung

Bertus

 

I'm in my first year and I am trying to understand op amps, I did most of my readings with this book called Op amp for everyone, it was initially good with the basic stuff but once it gets to the op amps with single power supply, the book just doesn't make any sense to me anymore, it then for some reason jumps to simultaneous equations and derive some weird equations and then it just goes downhill for me.

Simultaneous equations are something you're just going to have to learn to deal with if you intend to do very much with op amps; there's simply no way around that.

At a minimum, you'll need to get comfortable with solving two simultaneous linear equations in two unknowns for the purpose of calculating what resistor value you need in a particular circuit to achieve a desired result (gain, offset voltage, etc.). Sometimes, but less often, you'll need to solve a set of three simultaneous equations in three unknowns.

Just as important is learning how to derive the equations to be solved. Most often in an op amp circuit, we end up writing one equation for the voltage at the (+) input of the op amp, and another equation for the voltage at the (-) input. Since an op amp in a negative feedback circuit acts to keep its two inputs at the same voltage, we can solve the two equations simultaneously to find a resistor value needed to get a particular gain from the circuit, or a particular voltage offset on its output.

You might as well do whatever you have to do to get comfortable with all that, because you're going to be doing it unless you're content with being unable to deal with any but the most basic of circuits.

So I know the very basic stuff like the inverting and non inverting configurations with ideal op amp characteristics but I'm looking for a book that becomes more advanced with baby steps, what book did you read to understand op amps when you were a beginner ? I'm finding the transition from beginner to slightly more advanced stuff very difficult.

I was a beginner back in the late 1950's, with vacuum tubes and transistors; back then, op amps as commonly-used circuit components were at least a decade in the future, and I didn't have occasion to use them until the early 1970's. By then I was already fairly adept at circuit analysis (at least with simple circuits) so transitioning to op amps was no big thing-- it was just a slightly different application of what I already knew how to do.

I don't have a book recommendations for you; however, all of the major semiconductor manufacturers (TI, Analog Devices, Linear Technology, Maxim et al) have large collections of application notes and whitepapers containing tons and tons of good information that may be of help.

 

Thanks, something is missing in my understanding of analysing circuits that the book does not include, so the book I was reading started the chapter with this:

Single-supply systems do not have the convenient ground reference that dual-supply systems have, thus biasing must be employed to ensure that the output voltage swings between the correct voltages. Input sources connected to ground are actually connected to a supply rail in single-supply systems. This is analogous to connecting a dual-supply input to the minus power rail. This requirement for biasing the op amp inputs to achieve the desired output voltage swing complicates single-supply designs. When the signal source is not referenced to ground (see Figure 4–2), the voltage difference between ground and the reference voltage is amplified along with the signal. Unless the reference voltage was inserted as a bias voltage, and such is not the case when the input signal is connected to ground, the reference voltage must be stripped from the signal so that the op amp can provide maximum dynamic range.

At my level of knowledge this doesn't make any sense to me and I am honestly struggling to figure out what I should be working on.

 

Thanks, something is missing in my understanding of analysing circuits that the book does not include, so the book I was reading started the chapter with this:

At my level of knowledge this doesn't make any sense to me and I am honestly struggling to figure out what I should be working on.

Written that way, I'm not surprised it doesn't make any sense to a beginner.

A far more straightforward way of saying pretty much the same thing is to note that in a circuit powered only from a positive voltage V+ and ground, neither the op amp's inputs nor its output are free to "swing" both positive and negative relative to ground as they would have to do, for example, in a circuit designed to amplify an audio signal.

The output can't go negative because there is no negative supply voltage to allow it to do so; and, except in very unusual circumstances, neither can the inputs.

For an op amp to be able to operate off a single supply, the circuit must be designed so the inputs are always at some voltage between ground and V+ (actually, the upper limit is often not V+ itself but some level a volt or two below V+) as the input signal swings up and down. And the design must also cause the op amp's output to rest at some positive voltage between ground and V+, so it will have "room" to go both positive and negative relative to that resting state voltage.

This is what is meant by "biasing" the circuit, something we normally don't have to do with op amps powered both from + and - supplies. There, the inputs and the output can go both positive and negative with respect to ground, making the design job a lot easier.

 

Wow thanks for the explanation, that was very informative and easy to understand. I wish the so called beginner books out there didn't over complicated things so much.

 

Another reference on basic op amp usage that looks like it's pretty "beginner-friendly" is the Handbook of Operational Amplifier Applications by TI. It seems a bit more intuitive and somewhat lighter on the math than the one you are reading now.

 

Thanks, something is missing in my understanding of analysing circuits that the book does not include, so the book I was reading started the chapter with this:

At my level of knowledge this doesn't make any sense to me and I am honestly struggling to figure out what I should be working on.

My eyes started glazing over at that description! I can sure understand your search for another book.

Let's think about that dual-supply opamp that you started with. How is it ground referenced? There is no "ground" connection to the opamp so it has no idea what ground is! WE define some arbitrary node external to the opamp as our "ground" and then act like the opamp has telepathy and reads our minds to know what value 0 V is.

This raises the question of what the output is when it is open loop with Vdiff between the inputs? The theory tells us that it is Av·Vdiff relative to ground. But if it doesn't know what "ground" is, then what does this mean? If you tie the inputs together so that Vdiff is zero, then the output of the opamp will be at the voltage that it thinks is "ground". In general, that will not be the 0V reference that we have picked in the external circuit nor will it be halfway between the supply rails (which may or may not be close to the same thing that we chose as our 0 V reference).

But that's fine. We account for this with something called the opamp's input offset voltage. Image that you have an opamp powered by ±10 V rails and it has an open loop gain of 100,000. You connect the inputs together and tie them to your external 0 V reference. When you measure the output, you expect to find 0 V but you almost certainly won't. In fact, you will probably find that the output is railed in one direction or the other. Let's say the output is +9 V (and that this is the positive saturation voltage for this opamp). Now you connect a very small differential voltage between the inputs and discover that you need to apply -0.15 mV in order to get the output to be what YOU have defined as 0 V. You call this the input offset voltage of the opamp. Thought of another way, the opamp is acting like it had a fixed voltage differential of +0.15 mV on top of whatever the true voltage differential is that your circuit is applying. If you multiply this by the gain of 100,000, you discover that the opamp's preferred 0 V level is at what you would call +15 V.

In the vast majority of opamp circuit designs, we never even ask what the offset voltage is because it doesn't matter. We use negative feedback to force the circuit to virtually eliminate its impact.

So what changes when we go to a single-supply opamp?

The answer: NOTHING!

But, somewhat counter intuitively, this is not to say that our circuits don't have to change.

The problem isn't that the "ground" of the opamp is no longer the same as the "ground" of our external circuit -- we already know that it almost certainly never was. The problem is that we probably want to work with signals that are close to our external ground and perhaps even negative relative to it. This means that they are close to the lower power rail for our opamp and perhaps even below it.

This is completely analogous to wanting to work with signals in our prior dual supply example that are around -10 V and perhaps even lower. For instance, let's say that we wanted to work with input voltages between -11 V and -9 V. With our normal design techniques, we couldn't do so. But what if we introduced a new voltage source that we biased the input signal with to move it up into the range that the opamp can function with? Now the opamp works, but the output reflects not only the true input, but also the impact of the bias, so we need to further adjust the circuit to remove the part of the output that is due to the bias.

So, you might ask why some opamps are marketed as being "single supply" opamps? Normal opamps don't perform well close to the supply rails but that is where we want single-supply opamps to normally operate. Also, the span of the supply rail voltages is generally less for single supply operation than in dual supply circuits. So, for the most part, single-supply opamps are just opamps that have been designed to operate closer to rail-to-rail.

This might help you out: https://www.ti.com/lit/an/sloa030a/sloa030a.pdf

 

Thanks a lot WBahn,

The problem is that we probably want to work with signals that are close to our external ground and perhaps even negative relative to it.

Do you mean like that extra -0.15mV that we applied to the input so Vout would be 0 ?

This is completely analogous to wanting to work with signals in our prior dual supply example that are around -10 V and perhaps even lower. For instance, let's say that we wanted to work with input voltages between -11 V and -9 V. With our normal design techniques, we couldn't do so. But what if we introduced a new voltage source that we biased the input signal with to move it up into the range that the opamp can function with? Now the opamp works, but the output reflects not only the true input, but also the impact of the bias, so we need to further adjust the circuit to remove the part of the output that is due to the bias.

Sorry this part is confusing to me , as far as I know for example in a dual supply op amp, inverting configuration with + and - 10V rails, if we apply 2V signal then with the correct feedback resistor value, the output would be close to -10V or below it but saturated.
Here we have a signal with voltage between -9V and -11V and I assume we want to amplify it with a single supply op amp with +10V rail, but what's very confusing to me is how we can even amplify it with positive rail being only +10V, my assumption is that with a rail at like +30V and a biasing voltage that would allow the signal to swing about 15V, we would have an output about +/- 15V.
So the new voltage source for biasing, how do we go about adding that to the op amp ? Do we add this biasing voltage on top of our signal ? afterwards if we somehow remove the impact of the biasing voltage from our output, wouldn't it saturate again ?

 

I'm CMOS analog IC designer, hence I may not understand what are your real needs and what knowledge are you looking for (any architectures, configurations, parameters explanation?) in your discrete world, but a book I found helpful in general electronic, discrete as well, is The Art of Electronics by Horowitz an Hill (https://www.amazon.com/Art-Electronics-Paul-Horowitz/dp/0521370957). Chapter 4 is about opamps.

From my CMOS world, take a look here: https://payhip.com/b/5Srt. Maybe chapter 1 (specification) and chapter 5 (circuits) would be interesting for you (some parts of these chapters are in the free preview), but you may be already above that knowledge.

 

Most opamps, especially those intended for dual supply operation, are not rail-to-rail. They typically saturation about 1 V to 2 V before they get to the rail. Similarly, there are limits on what the input voltages can be relative to the rails. Depending on the design of the device, they may not be able to respond to voltages that are within a volt or two of one or both rails.

Now, let's forget about single-supply and dual-supply operations for a while. Let's consider the following problem.

You have a system that produces an output voltage relative to -5 V, meaning that if the signal is +1 V, the output that you get is -4 V and if the signal is -1 V the output you get is -6 V. The signal values can range from -2 V to +2 V, meaning that the output voltages you have to deal with range from -7 V to -3 V.

You want to build an opamp circuit in which the output voltage, relative to 0 V, is 4x the input voltage. So you want the output of your circuit to range from -8 V to + 8V.

Can you design a circuit that does this?

Now what if your opamp is powered by +/- 10 V rails and the voltages at the inputs must stay between -9 V and +9 V. Could you do that?

 

Thanks, something is missing in my understanding of analysing circuits that the book does not include, so the book I was reading started the chapter with this:

At my level of knowledge this doesn't make any sense to me and I am honestly struggling to figure out what I should be working on.

Your feelings are evident so not worth to discuss. It surprises me that you could not wade through the explanations in that book because I found it really enlightening.

Could I suggest you approach it again with no prejudices? Obscure as it could sound (not to me anyway) it could give way to an eureka moment on this subject.

The author, Ron Mancini, is a member of this forum but shows up not very frequently.

Maybe you are struggling with how to bias the inputs of a single supply opamp. If so, Google one of the many handbooks writne by Bruce Carter (he came after Ron Mancini), where he explains that precisely. That made for my difficulties to understand what I needed to start with the basics.

Buena suerte.

 

You want to build an opamp circuit in which the output voltage, relative to 0 V, is 4x the input voltage. So you want the output of your circuit to range from -8 V to + 8V.

Can you design a circuit that does this?

The input then must range from -2V to 2V, with op amp powered by +\- 10V considering output would saturate 2V before it gets to the supplied voltage.

Now what if your opamp is powered by +/- 10 V rails and the voltages at the inputs must stay between -9 V and +9 V. Could you do that?

If we are still talking about output being 4x the input voltage then the output would saturate at 8 or 9.

 

Your feelings are evident so not worth to discuss. It surprises me that you could not wade through the explanations in that book because I found it really enlightening.

Could I suggest you approach it again with no prejudices? Obscure as it could sound (not to me anyway) it could give way to an eureka moment on this subject.

The author, Ron Mancini, is a member of this forum but shows up not very frequently.

Maybe you are struggling with how to bias the inputs of a single supply opamp. If so, Google one of the many handbooks writne by Bruce Carter (he came after Ron Mancini), where he explains that precisely. That made for my difficulties to understand what I needed to start with the basics.

Buena suerte.

I would say 10 people would have 10 different experience with the same book they read, as good as the book is, it just doesn't do it for me, the words used and the way explained, come off as too technical and advanced for me, this doesn't mean I will never use this book, perhaps it's too early for me.
Thanks for suggesting a resource, I will look into it.

 

The input then must range from -2V to 2V, with op amp powered by +\- 10V considering output would saturate 2V before it gets to the supplied voltage.

The "signal" ranges from -2 V to + 2 V, but it is relative to -5 V. So the voltage that you have to work with ranges between -7 V and -3 V. You want your output to be -8 V when the input is -7 V and you want the output to be +8 V when the input is -3 V.

You should be able to do this design without worrying about saturation or the supply rails. We'll get to that.

 

The "signal" ranges from -2 V to + 2 V, but it is relative to -5 V. So the voltage that you have to work with ranges between -7 V and -3 V. You want your output to be -8 V when the input is -7 V and you want the output to be +8 V when the input is -3 V.

You should be able to do this design without worrying about saturation or the supply rails. We'll get to that.

So let me rephrase your question to make sure I'm understanding you, we have a system that has an input voltage range of +/- 2V relative to -5V(which I'm actually not really sure what that exactly mean) but the output of the system is in the range of -7V and -3V, we apply this voltage to an op amp that amplify its input by 4 and we want the output of the op amp to be between -8V and +8V.
So a -7V input would normally become -28V and -3V would be amplified to -12V so the output would be in the range -28V and -12V(that is if it's not the inverting amplifier), which is way below our -8V that we desire.

I really appreciate your patience in helping me out but I think I'm going way ahead of what I know about op amps here, all I have done so far is trying inverting, non-inverting and buffer configurations, taking signal from a function generator and getting an output about the voltage I expect using the formulas, but designing an op amp that turns -7V input to -8V and -3V to +8V... I don't even know what the first step is, I might be capable of answering this at my level but I'm really lost on this at the moment.

 

So let me rephrase your question to make sure I'm understanding you, we have a system that has an input voltage range of +/- 2V relative to -5V(which I'm actually not really sure what that exactly mean) but the output of the system is in the range of -7V and -3V, we apply this voltage to an op amp that amplify its input by 4 and we want the output of the op amp to be between -8V and +8V.
So a -7V input would normally become -28V and -3V would be amplified to -12V so the output would be in the range -28V and -12V(that is if it's not the inverting amplifier), which is way below our -8V that we desire.

I really appreciate your patience in helping me out but I think I'm going way ahead of what I know about op amps here, all I have done so far is trying inverting, non-inverting and buffer configurations, taking signal from a function generator and getting an output about the voltage I expect using the formulas, but designing an op amp that turns -7V input to -8V and -3V to +8V... I don't even know what the first step is, I might be capable of answering this at my level but I'm really lost on this at the moment.

If all you've done so far are the basic inverting, non-inverting, and voltage follower configurations, then you really aren't in a position to deal with handling the referencing and dereferencing that is needed for working with single-supply opamp circuits operated near ground. You are missing a couple of basic concepts still. In particularly, you need to get comfortable with the summing amplifier configuration. That shouldn't take you long. I think that's the only other thing you need for this, though the basic difference amplifier configuration might also be helpful. Once you are comfortable with those, how to tackle this problem will probably become a lot more obvious.