Beginner OpAmp questions

[gildasd]

Dumb questions... But these are too basic to be in spec sheets.
http://www.ti.com/ww/en/bobpease/assets/AN-31.pdf

Am I right in thinking that in the first case the Opamp will slam into it's Vcc+ and the second case it's Vcc-?

[tggzzz]

Dumb questions... But these are too basic to be in spec sheets.
http://www.ti.com/ww/en/bobpease/assets/AN-31.pdf

Am I right in thinking that in the first case the Opamp will slam into it's Vcc+ and the second case it's Vcc-?

No.

The best way to think about it is to assume Vo = A(V+ - V-), where A is a large value (the gain), and V+ and V- are the voltages at the non-inverting and inverting inputs.

For one case V- = Vo, so Vo =  A(V+ - Vo) and that can be simply rearranged to show Vo in terms of V+. The other case is left as an exercise for the student. See any introductory text on opamps, not data sheets not application notes.

[gildasd]

Cheers.
Sometimes one gets hung up on the most basic stuff...

[Horstelin]

The first one is a unity gain buffer:

http://en.wikipedia.org/wiki/Buffer_amplifier#Op-amp_implementation

The voltage you apply to the input is basicly tracked by the output (Gain A=1)


The second one's behaviour depends on the initial condition. It will either slam into Vdd or Vss, depending if (V+) - (V-) is positive or negative before supply voltages are applied. 

Hope it helps,
Urs

[w2aew]

Dumb questions... But these are too basic to be in spec sheets.
http://www.ti.com/ww/en/bobpease/assets/AN-31.pdf

Am I right in thinking that in the first case the Opamp will slam into it's Vcc+ and the second case it's Vcc-?

You might enjoy my video on the basics of op amp, and how to understand the operation of most op amp circuits:

[gildasd]

Thanks! I've already seen that video. Very good.

[w2aew]

Thanks! I've already seen that video. Very good.

OK, then it should be clear that the first configuration is a simple unity gain buffer, and the second case will slam the output to one of the rails, depending on the voltage at the inverting input.

[gildasd]

Last one for the "road":
The question for this one was; does the lamp switch on?

I think yes:
The Zeners are reversed:
- The top one lets 5,2/4=1.3V (let's say the forward current is 1/4 of the reverse)
- The bottom one lets  5.2V.
This means the current at V+ an V- are different. Op Amp tries to "compensate" error:
Current is released from +Vcc to the base of the transistor.
Transistor in "on". Light switches "on"

[dannyf]

A is at 5.2v and B is at 0.6v. Opamp output is low and the lamp is off.

- The top one lets 5,2/4=1.3V (let's say the forward current is 1/4 of the reverse)

Take an EE101 or even highschool level class. It would help you tremendously.

[gildasd]

A is at 5.2v and B is at 0.6v. Opamp output is low and the lamp is off.

- The top one lets 5,2/4=1.3V (let's say the forward current is 1/4 of the reverse)

Take an EE101 or even high school level class. It would help you tremendously.

Yeah, I know, but I had no time to get this ready... It sucks, I'd rather breadboard this than ask dumb questions.
We had insufficient time, spending 3 weeks doing intensice mechanics, stability, hydromechanics, diesel engines... It's hard to get back into Opamps with only 4 hours to get ready.
If I had full day, I would have redone all the basic circuits to memorise them... But no chance.

[gildasd]

Here are the test questions we had about Opamps, if anyone is interested.

They can be solved without a calculator.

[dave_birdi]

Here are the test questions we had about Opamps, if anyone is interested.

They can be solved without a calculator.

I'm actually a bit puzzled by that one, is the output from the op-amp meant to be 4.5 V?

[gildasd]

Yep, 4.5...
I'm doing a testing supply to try this one out, I disagree with my prof's result.

[dannyf]

They can be solved without a calculator.

Are today's students really that challenged?

[c4757p]

They can be solved without a calculator.

Are today's students really that challenged?

On one hand, I'm a bit saddened by people who can no longer do math without a calculator. I've been in an exam next to people who forgot to bring one, and they were acting like the world was going to end without it - I've forgot one before too, and I just did the arithmetic in the margin...

On the other hand, are they ever truly going to need that skill? It's a skill I'm proud to have, but it's not one that's really necessary these days...

[tggzzz]

They can be solved without a calculator.

Are today's students really that challenged?

On one hand, I'm a bit saddened by people who can no longer do math without a calculator. I've been in an exam next to people who forgot to bring one, and they were acting like the world was going to end without it - I've forgot one before too, and I just did the arithmetic in the margin...

On the other hand, are they ever truly going to need that skill? It's a skill I'm proud to have, but it's not one that's really necessary these days...

The ability to do "back of envelope" manual/mental calculations will always be highly beneficial. The good engineers are sufficiently numerate that they can do it, and in a meeting with them you will be left in the dust if you can't do it.

Of course, it is always necessary to realise when "back of envelope" approximations aren't sufficient and it is necessary to use a calculator, e.g. difference of two similar numbers such as an opamp's inverting and non-inverting inputs.

As for the concept that rote learning of the times tables is sufficient - fortunately better alternatives are used in the UK nowadays!

[dannyf]

I'm doing a testing supply to try this one out, I disagree with my prof's result.

No reason to try it out. The non-inverting end sits at 3v.

So you need to find a voltage (V2) that, together with 4.5v, yields 3v. That's pretty simple: think of the solution in terms of siemens (=1/ohm).

[jlmoon]

They can be solved without a calculator.

Are today's students really that challenged?

On one hand, I'm a bit saddened by people who can no longer do math without a calculator. I've been in an exam next to people who forgot to bring one, and they were acting like the world was going to end without it - I've forgot one before too, and I just did the arithmetic in the margin...

On the other hand, are they ever truly going to need that skill? It's a skill I'm proud to have, but it's not one that's really necessary these days...

I have done a few of those on restaurant napkins.. they work great!   

[macboy]

They can be solved without a calculator.

Are today's students really that challenged?

On one hand, I'm a bit saddened by people who can no longer do math without a calculator. I've been in an exam next to people who forgot to bring one, and they were acting like the world was going to end without it - I've forgot one before too, and I just did the arithmetic in the margin...

On the other hand, are they ever truly going to need that skill? It's a skill I'm proud to have, but it's not one that's really necessary these days...

Going off topic here. I once went to a Economics exam without my calculator. So I wrote all the equations and filled everything in... everything except the final answer. Got 100%. They didn't care that I had not given a single final answer; I knew what I was doing. This would not have happened on an Engineering exam!

Back on topic, for #2, the solution is:
* the positive input is at +3 V (a voltage divider is formed with two 10k resistors).
* the negative input is therefore at +3 V also (the op-amp is operating normally, not railed out).
* the two 10k parallel resistors make a 5k. There is 1.5 V across this (4.5 - 3), so current through it is 1.5/5000= 0.3 mA.
* current into negative input is (as always) "zero", so
* current through the V2 input 10k resistor is therefore 0.3 mA. Then 0.3*10000 = 3 V across this resistor (positive side on the right).
* The V2 input is therefore, 0 V

[tggzzz]

Going off topic here. I once went to a Economics exam without my calculator. So I wrote all the equations and filled everything in... everything except the final answer. Got 100%. They didn't care that I had not given a single final answer; I knew what I was doing. This would not have happened on an Engineering exam!

In every maths/physics/engineering exam I've had (with one exception, see below) exactly that would have happened. Everyone knew that g=10 (not 9.81) and pi²=10. Get your arithmetic wrong and, provided you had clearly shown your working, the examiner would deduct one mark for the error - and then follow through your answer to see if you had made any further mistakes. If no further mistakes then you got almost full marks even though 99% of the intermediate answers might be incorrect.

The exception was in the mid 70s when calculators were new and not allowed in exams. In the first exam they were allowed, there was one question which required the use of a calculator to get the right result (opamp input and output voltages). Most students didn't spot the necessity, and were correctly penalised because their engineering thinking was erroneous.

I also had a maths exam where the rubric stated "full marks can be obtained for answers to about six questions".

Eric Laithwaite said he always inserted a question that could not be answered well within the confines of a exam - and he expected his good students to recognise it and avoid it Any engineer should recognise where dragons are in residence, and go elsewhere.

[macboy]

Going off topic here. I once went to a Economics exam without my calculator. So I wrote all the equations and filled everything in... everything except the final answer. Got 100%. They didn't care that I had not given a single final answer; I knew what I was doing. This would not have happened on an Engineering exam!

In every maths/physics/engineering exam I've had (with one exception, see below) exactly that would have happened. Everyone knew that g=10 (not 9.81) and pi²=10. Get your arithmetic wrong and, provided you had clearly shown your working, the examiner would deduct one mark for the error - and then follow through your answer to see if you had made any further mistakes. If no further mistakes then you got almost full marks even though 99% of the intermediate answers might be incorrect.
...

At my uni, the thought was that incorrect answers kill people, so if the answer is wrong, the best you could get was about 1/10, even if all intermediate steps were correct. People rarely walked out of exams early. After finishing, you would go back and run your calculations again. When you finished that, you would do it again... until time ran out.

[gildasd]

Going off topic here. I once went to a Economics exam without my calculator. So I wrote all the equations and filled everything in... everything except the final answer. Got 100%. They didn't care that I had not given a single final answer; I knew what I was doing. This would not have happened on an Engineering exam!

In every maths/physics/engineering exam I've had (with one exception, see below) exactly that would have happened. Everyone knew that g=10 (not 9.81) and pi²=10. Get your arithmetic wrong and, provided you had clearly shown your working, the examiner would deduct one mark for the error - and then follow through your answer to see if you had made any further mistakes. If no further mistakes then you got almost full marks even though 99% of the intermediate answers might be incorrect.
...

We have the same issue with hydromechanics and calculating losses (Bernoulli and all that), during the exam, the student must show the method, the numbers might be your aunties land line and the Super-bowl averages...
In the end, any system becomes so complex the maths breaks down or gives you loony answers. The real world methods use empirical tables that have been refined since the dawn of the age of steam 

[gildasd]

They can be solved without a calculator.

Are today's students really that challenged?

On one hand, I'm a bit saddened by people who can no longer do math without a calculator. I've been in an exam next to people who forgot to bring one, and they were acting like the world was going to end without it - I've forgot one before too, and I just did the arithmetic in the margin...

On the other hand, are they ever truly going to need that skill? It's a skill I'm proud to have, but it's not one that's really necessary these days...

Going off topic here. I once went to a Economics exam without my calculator. So I wrote all the equations and filled everything in... everything except the final answer. Got 100%. They didn't care that I had not given a single final answer; I knew what I was doing. This would not have happened on an Engineering exam!

Back on topic, for #2, the solution is:
* the positive input is at +3 V (a voltage divider is formed with two 10k resistors).
* the negative input is therefore at +3 V also (the op-amp is operating normally, not railed out).
* the two 10k parallel resistors make a 5k. There is 1.5 V across this (4.5 - 3), so current through it is 1.5/5000= 0.3 mA.
* current into negative input is (as always) "zero", so
* current through the V2 input 10k resistor is therefore 0.3 mA. Then 0.3*10000 = 3 V across this resistor (positive side on the right).
* The V2 input is therefore, 0 V

You are right, this is the "class" method.
The Bob Pease examples are nearly useless in the exercises we get...
(http://www.ti.com/ww/en/bobpease/assets/AN-31.pdf):
Page 1, Difference Amplifier.

 

[tggzzz]

Going off topic here. I once went to a Economics exam without my calculator. So I wrote all the equations and filled everything in... everything except the final answer. Got 100%. They didn't care that I had not given a single final answer; I knew what I was doing. This would not have happened on an Engineering exam!

In every maths/physics/engineering exam I've had (with one exception, see below) exactly that would have happened. Everyone knew that g=10 (not 9.81) and pi²=10. Get your arithmetic wrong and, provided you had clearly shown your working, the examiner would deduct one mark for the error - and then follow through your answer to see if you had made any further mistakes. If no further mistakes then you got almost full marks even though 99% of the intermediate answers might be incorrect.
...

At my uni, the thought was that incorrect answers kill people, so if the answer is wrong, the best you could get was about 1/10, even if all intermediate steps were correct.

Which leads directly to the "... but it is the right answer because the computer/calculator says so" mentality. The old motto is "to err is human, but to really foul it up you need a computer".

In the real world, critical engineering calculations would be verified firstly by you and subsequently others! Economic calculations are, unfortunately, a different matter.

It sounds like your uni was more interested in seeing how effective you are as a calculator than as a well-rounded engineer.

People rarely walked out of exams early. After finishing, you would go back and run your calculations again. When you finished that, you would do it again... until time ran out.

Wouldn't everybody? I've never seen anyone walk of of an exam early - but not for the reason you mentioned!