# gain of inverting op amp?

#### Audioguru

Joined Dec 20, 2007
11,248
I think the 1.7 is a deliberate error to keep students from jumping at it.
Then the treacher is cheating more than the students.

Joined Dec 26, 2010
2,148
Could the list have been "doctored" away from a normal list of preferred values, removing the obvious 1.8 value to make a more difficult problem for the student?

I know this may seem artificial, but it's the sort of thing that educators seem to come up with in some parts of the world.

Edit: Snap. And yes, I rather agree with Audioguru, but the student may just have to dance to his teacher's tune if he wants to get credit.

#### Georacer

Joined Nov 25, 2009
5,182
Hey, don't talk about the teachers like they 're loving every minute of it. Boredom strikes in the most unexpected places.

But it's kinda funny this time:
Instead of taking two random standard resistors and finding their ratio, this guy took the trouble to alter the list to match his preferred ratio.

Joined Dec 26, 2010
2,148
I think were allowed a 1% error. My only question would be, with only being able to use these standard resistor values...if I use the 2ohm, the next size up would be a milliohm right? How could I use 2 X 10^1. Wouldn't that exceed my resistor values from the table?
If you are allowed up to two resistors for Rin and Rf, then this definitely can be solved, without going outside the given decade.

Perhaps you need to think about how the resistors might be combined.

#### Georacer

Joined Nov 25, 2009
5,182
Another constraint in full grayscale: "You may use no more than two resistors in combination"

What is that supposed to mean?

I 'd vote for using two resistors only in any power of ten.

Joined Dec 26, 2010
2,148
Maybe this is an examination for clairvoyants? Seriously though, using two resistors per element, four in total, there are exact solutions just using this list. Isn't that likely to be what the examiner was after?

#### Georacer

Joined Nov 25, 2009
5,182
Yeah, he was after rigging an OpAmp to source a few amps through a few ohms resistor.

Sigh, whatever. Let's go both ways here.

#### Ron H

Joined Apr 14, 2005
7,014
Maybe this is an examination for clairvoyants? Seriously though, using two resistors per element, four in total, there are exact solutions just using this list. Isn't that likely to be what the examiner was after?
I think you are correct. Nowhere in the statement of the problem does it say that he has other decades available.

Last edited:

#### JoeJester

Joined Apr 26, 2005
4,390
The problem statement began as ...
Using only standard values of resistors and capacitors, design ...
Then the two tables were titled Standard Resistor Values and Standard Capacitor Values.

Even if 1.8 was on the professors standard list, 1.0 and 1.8 would not equal 18, you had to do the decade multiplier. One of the problems the OP had was they thought there was no decade multiplier.

I agree 1.8 was replaced with 1.7 for the students to look elsewhere for the solution.

It would have been better if the professor stipulated that for this test, use standard values from the tables provided.

Joined Dec 26, 2010
2,148
With four resistors used in total, no decade multiplier is needed. Series and parallel...

#### steveb

Joined Jul 3, 2008
2,436
With four resistors used in total, no decade multiplier is needed. Series and parallel...
Yes, and that solution does not require the 1.7 ohm resistor, and does not require excluding the 1.8 ohm resistor. Hence, that is a typo, and one that will just confuse a student new to the subject. Hopefully, it was an honest mistake and not a pointless change.

I have to say, this problem makes a nice little puzzle, but it isn't very practical. Hopefully, that is the teacher's intent, and he is not really suggesting that parallel and series comninations are necessary in general.

The times where series and parallel combinations are nice is when you can make all resistors the same value. This allows one part specification to be used and it can improve tolerance issues also. This is particularly useful with surface mount parts because then you can buy one reel of resistors and save some money and the resistors on a reel tend to be well matched to each other, even better than their tolerance would indicate.

For example, to make a gain of 4, two 10K resistors in series for the feedback resistor and two 10K resistors in parallel for the input resistor would make a gain of 4 to very high tolerance in practice. It's likely that 1% resistors could give 0.1 % effective tolerance on the gain, with this approach.

#### JoeJester

Joined Apr 26, 2005
4,390
Unfortunately the problem statement also includes ... using two resistors.

#### steveb

Joined Jul 3, 2008
2,436
Unfortunately the problem statement also includes ... using two resistors.
No, it said "when trying to achieve a specific value of resistance, you may use no more than 2 resistors in combination".

So, the implication is clearly that you are allowed to create an effective Rf and and effective Ri using 2 resistors each, either in series or in parallel.

#### Georacer

Joined Nov 25, 2009
5,182
I wouldn't say clearly. But as you might already noticed, we 're alone here by now, the OP has left. So it's a free for all. Have at it.

#### steveb

Joined Jul 3, 2008
2,436
I wouldn't say clearly.

Yes, I should reword it. It is clear in hindsight. You have to go back and reread it a few times (at least I had to) before the meaning is clear. This is why I mentioned above that this problem is more like a nice puzzle than a practical problem. Good puzzles are confusing in foresight and head-slappingly clear in hindsight.