# Instrumentation Amplifier

#### Danlar81

Joined Apr 19, 2019
87
Here Eric,
Here is the circuit, time domain and DC sweep simulations. The DC sweep is over R2. It does not start in the right place however when it is swept from 0 to 43 ohms. It looks ok if I start from 43 to 86 ohms though. I have about 3.464 mV, so I may be able to further fine tune, and this would explain why the DC sweep from 43 ohms to 86 ohms went from 5 Volts to approx 0.5V. I'm not entirely sure what to do about the DC sweep hitting the rail starting off high then going to 5 Volts. Any ideas?

#### ericgibbs

Joined Jan 29, 2010
18,661
Hi dan,
I would recommend that you do not submit that design for your assignment, it is far too complicated.
All you need is the differential pair followed by a single stage combining circuit.

I would like to post a full circuit for you, but as this is a marking assignment, I cannot.

Relook at that image I posted from the datasheet post #32.

E

#### Danlar81

Joined Apr 19, 2019
87
Hi Eric,
I'm not too sure about the DC sweep side of things although everything else seems ok. When R2 = 0 ohms there is approx 1.46 V at the R1/R2 junction which is right in the middle 1.5 mV either side. I wasn't expecting it to do that.

#### ericgibbs

Joined Jan 29, 2010
18,661
Hi dan,
At Mach zero the differential input will be 0V, at upper Mach 3mV.
At upper Mach you are requiring a 5V Vout, so that's an over all Gain of 5v/0.003 =1666
Consider setting the diff gain to say 500 the the final stage OPA to 3.2.

Post a circuit for this gain, using post #32 as a guide.

E

#### Danlar81

Joined Apr 19, 2019
87
Hi Eric,
Yes what your saying is correct.
I was just looking at circuit #32, and looks like another Instrument amplifier?
I had a look for a differential pair but only saw this. Not sure about what you mean with the combiner, however I can redo the gain and try again.

#### Danlar81

Joined Apr 19, 2019
87
Also, I didn't realize that the gain between each stage was multiplied. I though they were added. I set my gain at the IA for 1000 and 666 in the front end.

#### Danlar81

Joined Apr 19, 2019
87
Gee it is plain as day, I just wrote it in my book incorrectly.

#### Danlar81

Joined Apr 19, 2019
87
Ok Eric, I have 2x270k and 1x 1.1k for the front end and 150K and 150K for the IA. I get a value of 5.023V out but I have not adjusted anything yet. As for the DC sweep, I'm assuming you were expecting the same outcome? With the differentia pair and combiner would you mind explaining that a little further?

#### ericgibbs

Joined Jan 29, 2010
18,661
Hi dan.
Post #45 is a Difference Amplifier,
Your assignment is for an Instrumentation amplifier, as post #35.

Post a diagram of your #48 circuit.
E

#### Danlar81

Joined Apr 19, 2019
87
Hi Eric, I has not changed apart from the resistor values.

#### Danlar81

Joined Apr 19, 2019
87
Hi Eric, I have a bit of confusion with regards to what you said in relation to the differential pair and combiner. I Have looked a bit further and I believe this is what you are referring to about the differential amplifier.

I cannot seem to find what a combiner is though. When you mentioned this, I assume you did not mean to use this in place of the instrumentation amplifier. Were you inferring that these could be used in place of the summation amplifiers? If that is the case, these could connect directly to the IA front end I thought.

#### ericgibbs

Joined Jan 29, 2010
18,661
Hi dan,
You do not need 6 OPA's, 3 is all that is needed in order to design an IA amp.

E

#### Danlar81

Joined Apr 19, 2019
87
Hi Eric, apologies and appreciate your patience. So this could be connected across the resistor bridge, the offset would be fed into where the ground connection is, if I am following correctly? But there is one output here and the instrument amplifier has two inputs. I'm clearly missing something? Just not sure what it is.

#### Danlar81

Joined Apr 19, 2019
87
Hi Eric,
I think I will have to hand the assignment in as it is, as I have until midnight which is 20 mins away for me to submit. Perhaps after midnight when I have submitted, you may be kind enough to share the idea you had but could not mention. Anyway I'll keep trying for another 20 mins then I need to submit.

#### Ian0

Joined Aug 7, 2020
9,514
Hi Eric,
I think I will have to hand the assignment in as it is, as I have until midnight which is 20 mins away for me to submit. Perhaps after midnight when I have submitted, you may be kind enough to share the idea you had but could not mention. Anyway I'll keep trying for another 20 mins then I need to submit.
Just think how well you could have done if you had started sooner.

#### Audioguru again

Joined Oct 21, 2019
6,616
I agree that LM324/LM358 opamps have very poor quality specs to detect 3mV. One problem is that they are VERY noisy.

#### Ian0

Joined Aug 7, 2020
9,514
I agree that LM324/LM358 opamps have very poor quality specs to detect 3mV. One problem is that they are VERY noisy.
If he had to build the circuit, I wouldn't fancy his chances of getting it to work. As it's a DC circuit, a lot of the noise can be eliminated by filtering.

#### Audioguru again

Joined Oct 21, 2019
6,616
Opamp noise is higher at low frequencies including DC.
The noise from a lousy old LM324/LM358 is so high that it is not specified on their datasheets.

Here is the noise from the datasheet of the common TL084:

#### Attachments

• 58.4 KB Views: 1

#### Danlar81

Joined Apr 19, 2019
87
Hi Ian,
I did get it working eventually, but there is approx. 13mV on the output when I expect zero. The problems I had were around the DC offset. It took me quite a lot of trial and error to understand how each one was affecting the other. And it is extremely sensitive.

I don't really need the OA4 but kept it there so both +/- can be seen.

#### Ian0

Joined Aug 7, 2020
9,514
Then it will work properly if you use a better op-amp!
Your OA5 and OA6 stages contribute nothing except offset voltage and noise. It would work better if you remove them.