I’m a 1st year Undergraduate Aeronautical Engineering student in the UK, and one of the courses that makes up my degree is Analogue Electronics.
I’m consolidating my notes for this course at the moment, and I’m OK with most of it. Op amps (operational amplifiers, but you knew that) are a bit confusing at the moment. Our lecturer tells us it’s not important to even care about how they work, just to know how to answer the questions and what sort of answers we’ll be getting.
Going over some old exam questions, I realise I can’t work out what voltage the output of inverting & non-inverting amplifiers will saturate at.
I know that V[sub]out[/sub] is saturated fully + if V+ is more +ive than V- and vica versa. But I’m not sure how to work out when this is.
My google-fu isn’t up to finding this information, I keep finding just general information about operation amplifiers and that doesn’t help.
*Going over some old exam questions, I realise I can’t work out what voltage the output of inverting & non-inverting amplifiers will saturate at…
I know that V[sub]out[/sub] is saturated fully + if V+ is more +ive than V- and vica versa. But I’m not sure how to work out when this is.*
When I was in school, one of the tips I got in analyzing Op-amp circuits was to assume a “virtual” ground between the inverting and non-inverting inputs. This virtual ground shorted the inputs to the same voltage but allowed no current to flow between them. The instructor would draw it in a circuit as a vertical, arrowheaded line between the two inputs, wiping out the op-amp entirely (disconnecting the output as well); we were to assume this line fixed the voltage to be the same at each end but that no current would flow thru it.
The output pin was not connected at all to this virtual ground, but as you likely know an op-amp operates properly only when there is a feedback path from the output pin to one of the inputs, which provides a link to complete the circuit.
Full saturation of an op-amp usually occurs at a very tiny voltage (70 microvolts is a typical figure), so for all practical purposes (e.g. a voltage comparator) if the positive input is forced to be higher than the negative, the op-amp is saturated. An analysis of a simple comparator can be found here
Am I missing something here? My experience with OP amps is somewhat limited and from a long time ago but won’t the output saturate at the supply voltage? Suppose you have an amplifier with a gain of 10 and with ± 10 volt supply. Then the output will saturate when it is ± 10 V. which occurs when the input is ± 1 V or more.
Yes, that does make sense now! I was the one missing something, I think. The topic was only skimmed over in lectures, it’s not that large a part of the course.
I think the problem was that I was attempting to do the questions by memorising the methods and not really understanding what Op amps are and how they work (as suggested by my lecturer). Reading into them now, it seems to make much more sense. Guess I just learn that way.
Thank you for your help, David Simmons and CJJ*, the link you provided is very helpful. I think I found a very similar page also on www.allaboutcircuits.com.
Not quite - usually the output voltage range is slightly narrower than the supply voltage range. But some op-amps can do it (“rail-to-rail output”). You’ll have to look at the data sheet to find out.
I think we can ignore that refinement for the purposes of the exam under discussion. But yes, all such devices begin to become highly non-linear as they approach their supply voltage.
When we’re doing these questions, we assume that the op amp is an ideal op amp, which I think means it saturates at pretty much the supply voltage. Is this correct?
Anyway, the question in…er…question says “At approximately will the output of each amplifier saturate”. I assume this means that I can simply state the supply voltage.
IIRC (and it’s been a long time) what matters in op-amp circuits is the feedback. You can treat the op-amp itself as a black-box and as ideal. That is you can consider it as effectively having infinite gain, infinite input impedance and perfect common mode rejection.
On following CJJ’s link (looks excellent) I see feedback gets mentioned straight after the comparitor. Op-amps are fun, it’s discreet component circuits that are a pain.
Op-amps will saturate at different voltages depending upon the particular device, the supply rail voltages, the temperature, and the output load. Some op-amps are designed to swing the output all the way to -V, some will swing all the way to -V or +V, but most will have a diode drop or two between the maximum output voltage and the rails. Typical figures: A TL071 op-amp with +/-15V rails will swing the output between about +/-13.5V with a 10kohm load; downrate that to +/-12V (typ.) with a 2kohm load.
CJJ’s mental model of the op-amp is sound, though it should be added that the op-amp inputs will drift apart as soon as the output is saturated. Watch out for this if you’re driving an op-amp output into clipping, as some op-amps can only handle a few volts between their inputs before they die.
For most applications you don’t need to know what (differential) voltage the op-amp inputs will saturate at, as it will be the maximum output voltage divided by the open-loop gain, and this last factor is a very large number that can’t be relied on to be a particular value, it will just lie within a broad range.
Comparators usually have open collector outputs, and need a pullup resistor. They’re a bit like op-amps, but sacrifice linearity and low noise for speed of switching. Op-amps make poor comparators (though good enough for most purposes), and comparators make appalling op-amps.
Then you’re lecturer is a fool, and he’ll just be turning out engineers with half-decent degrees but no intrinsic understanding of what they’re doing. If you don’t understand the basics of op-amps intuitively, then study and play with them until you do. If you’re doing any sort of analogue engineering, then this is the stuff you need to know. You don’t need to know it all in your first year though…
You will have reached a reasonable level of understanding of op-amps if you can read an op-amp datasheet and understand and know how to use all the given parameters. If you’re using an op-amp type that’s new to you, always read the datasheet first, as some of them have some little features and quirks that need taking into consideration.
Yup. Although I suppose an ideal op-amp wouldn’t saturate at all.
Fridge I don’t think Witty is designing real circuits here, exam questions probably don’t require you to have memorised data sheets. Though an understanding that real devices are in fact limited (and in what ways) could be a valid question.