op amps / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

op amps

1 2 next

Move Thread

LAN_403

IamSmooth

Sat Feb 08 2014, 07:21PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

I have seen the basic circuit for a differential op amp and one for gain.
Can I simply take a non-inverting opamp setup and have Va and Vb for inputs, so the output is the gain times (Va-Vb)? Or, does Vb have to be set to ground for the non-inverting style?

testtest

Sat Feb 08 2014, 08:01PM

Registered Member #3271 Joined: Mon Oct 04 2010, 02:29AM
Location: Canada
Posts: 159

See below. original image did not show up.
Admin: OK to remove !

Sulaiman

Sat Feb 08 2014, 08:03PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

The common way to achieve what you want is this

Your circuit will give Av.(Va-Vb) PLUS Va ...

i.e. if Va = Vb, the output will be Va (=Vb) ... NOT Av(Va-Vb) = 0

IamSmooth

Sat Feb 08 2014, 08:08PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

Thanks. I did see this. I was curious if I could just change the ground to my Vb. I have just tried this and it does not seem to work. Thus, I will use the setup you have referenced.

testtest

Sat Feb 08 2014, 08:17PM

Registered Member #3271 Joined: Mon Oct 04 2010, 02:29AM
Location: Canada
Posts: 159

Ref: national Semicond applications

here is a generalised version. If you set R1=R3 and R2=R4 you will get what you want.
Gain is R2/R1

Mind offset and stability issues.

Also note I have the invert input on top unlike your diagram.

edit; looks like my attachment did not show up....

Dr. Slack

Sat Feb 08 2014, 08:28PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

Sulaiman wrote ...

The common way to achieve what you want is this

Your circuit will give Av.(Va-Vb) PLUS Va ...

i.e. if Va = Vb, the output will be Va (=Vb) ... NOT Av(Va-Vb) = 0

I've captured the drawing at that link and very slightly redrawn it

to show the symmetry and emphasise what it is doing. The op-amp will work to keep both its inputs at the same voltage. That means that as the top pair and bottom pair of resistors have the same ratio, the *difference* between the output terminals will be -g times the difference between the input terminals. Now you can put the output ref terminal to where you like. Often it goes to ground and you have a floating to single-ended (that is - referenced to ground) converter. You could also regard the circuit as adding the output reference to -g * the input difference to give the output voltage. Note that the ratio g can be bigger or smaller than 1.

Just my halfpen'orth anyway. I can never remember configurations, so whenever I come to want to use any sort of differential amplifier, I first draw this symmetrical form, then move it around until it does what I want.

Shrad

Sat Feb 08 2014, 09:59PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

for your general use about instrumentation amplifiers, here are some appnotes you might find useful...

there are equivalent op-amps in most brands so you are not stuck with linear except for some very specific ICs

IamSmooth

Sun Feb 09 2014, 06:57AM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

I've gotten most of my opamp to work but I am having one issue. What I have so far:

I made a constant current source of about 1ma using an NPN and opamp. I use this so the voltage drop on the RTD is linear with the change in resistance. I made a three wire hook-up for an op amp to eliminate the lead resistance. Both of these units work and I have a voltage drop on the RTD of about 0.1mv at room temperature.

The problem I am having is putting this through a non-inverting opamp for gain. My rails are 5v and 0. My R2 and R1 values are 5k and 100R. I would expect a gain of 50 (1+R2/R1) which my MC3403 should do. The gain on the datasheet is 20 minimum; 200 typical V/mv. Yet, the most voltage I get out is about 3.74v.
If I use R1 = 5k and R2 = 200R, my gain works. For Vin of 0.12v I get a little over 3v.

I would expect to get close to 5v with my initial R values. Am I asking the opamp to do something it can't do with these supply voltages? I get the same issue if I use +5/-5v on the rails.
How close to the upper rail voltage can I get my input signal if I keep upping the gain?

Dr. Slack

Sun Feb 09 2014, 08:53AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

There's a lot to be said for reading *all* of the datasheet. 3.74v under those conditions (driving a heavier load than the data sheet quotes maximum output voltage at - remember the feedback has to be driven as well as any explicit load) is a bit more than the datasheet says is typical, and is way more than is guaranteed. MC3403 does not have a R2R output. Use more supply volts, or a R2R amplifier.

You quote the gain as 20 min 200 typical. Those gain figures are for +/- 15v supplies, the +5v supply figures are 10 min, 200 typ. Not that the difference will be noticable in your application, but, make sure you read the right bits of the data sheet.

Shrad

Sun Feb 09 2014, 07:56PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

heat the NPN holding it in your fingertips for a couple minutes and measure the current again...

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob