Author Topic: opamp confusion  (Read 5290 times)

0 Members and 1 Guest are viewing this topic.

Offline cravenhavenTopic starter

  • Contributor
  • Posts: 40
  • Country: au
opamp confusion
« on: June 14, 2015, 02:39:30 am »
I am trying to create a cct to amplify and level shift an input. It is a DC coupled cct so I cant use any caps in the signal lines.
I have been using LTSpice to simulate the cct but for the life of me I just cant get it to operate as I want.
As can be seen, the input for simulation is a 0.1hz sine wave with a 1V swing about 10V (ie half supply rail), but the output is only swinging a couple of microvolts albeit at a lower central voltage.

What I would like to be able to do is to amplify the signal over a range of about 0.5 to 5 and level shift so that the output swings from 0 up to whatever its max is.
« Last Edit: June 14, 2015, 02:41:57 am by cravenhaven »
 

Online IanB

  • Super Contributor
  • ***
  • Posts: 11895
  • Country: us
Re: opamp confusion
« Reply #1 on: June 14, 2015, 02:50:18 am »
I have enormous trouble trying to figure out what your circuit is trying to do, but one question leaps out: why do you say that your 10 V signal is about half the supply rail, when your diagram clearly shows a 0 to 10 V supply on the op amp? [Edit: which I see you have just modified  :) ]
 

Offline matseng

  • Frequent Contributor
  • **
  • Posts: 563
  • Country: se
    • My Github
Re: opamp confusion
« Reply #2 on: June 14, 2015, 02:57:48 am »
Do you have the expected voltage swing at the junction of R3/R7 (the positive input of the opamp)?
 

Offline oPossum

  • Super Contributor
  • ***
  • Posts: 1417
  • Country: us
  • Very dangerous - may attack at any time
Re: opamp confusion
« Reply #3 on: June 14, 2015, 03:00:20 am »
R1 must go to a virtual ground at half the supply voltage. Or simply replace it with resistors of equal value to both supply rails.
 

Offline KerryW

  • Regular Contributor
  • *
  • Posts: 112
  • Country: us
Re: opamp confusion
« Reply #4 on: June 14, 2015, 03:55:43 am »
Consider just the offset. to start with.
When your input voltage is at 10.0 volts, you gave 10 volts on R7 and on R3, and therefor on the + input of the op amp.  You should then see 10V on the - input, and 35V on the output!

Drop the divider (R3, 4 and5), add a resistor from the + rail to the - input, say 5500 ohms.  Multiply all existing resistors by 10 (R2 becomes 5K etc)
« Last Edit: June 14, 2015, 03:58:03 am by KerryW »
One accurate measurement is worth a thousand expert opinions
- Adm. Grace Hopper
 

Offline cravenhavenTopic starter

  • Contributor
  • Posts: 40
  • Country: au
Re: opamp confusion
« Reply #5 on: June 14, 2015, 04:47:45 am »
Thanks guys for the quick reply.
To back up a little. I was designing a cct for a current monitor and I needed to offset and amplify the output signal for display but couldnt get it to work properly. So I decided to try a simple offset amplifier which I initially copied from a sample design on the web (http://electronicdesign.com/boards/non-inverting-level-shifter-requires-only-one-op-amp-one-supply-voltage) and modified to suit the voltages I wanted to use. But I cant get it to work in the simulator hence my question.

So the aim of the cct is to amplify the signal from v2 which is a 1V sine wave offset to +10V (ie sine wave between 9 and 11 volts).
Resistors R4 and R5 are meant to create a virtual ground at half supply voltage ie 10v. I mistakenly had set the supply voltage at 10v when I first posted the picture, so I corrected it as soon as I realised.
R1 and R2 are to set the gain as for a standard noninverting opamp. I also considered that R1 should be connected to the virtual ground and have tried it both ways with no meaningful result. (Actually it does improve things but the output is still only between 8.22 and 8.25, so about 1.5% of the input.
The voltage at R3/R7 is between 9.2 and 10.2, so its half the input and centred at approx the virtual ground voltage of 9.38.
If I disconnect the input signal V2 (remove R7) then the + input =9.3v, the - input = 9.03 and the output = 8.23.
When I increase the value of all the resistors by 10 with v2 disconnected then all the input and output voltages approx =10v.
 

Offline KerryW

  • Regular Contributor
  • *
  • Posts: 112
  • Country: us
Re: opamp confusion
« Reply #6 on: June 14, 2015, 05:11:59 am »
That circuit is a summing amplifier, it adds the signal and +15V to generate the offset.  What you want is to subtract an existing offset, which is what the resistor from the + rail to the - input is for.  Try it.
One accurate measurement is worth a thousand expert opinions
- Adm. Grace Hopper
 

Offline NoItAint

  • Contributor
  • Posts: 33
  • Country: us
Re: opamp confusion
« Reply #7 on: June 14, 2015, 07:15:29 am »
 

Offline KJDS

  • Super Contributor
  • ***
  • Posts: 2442
  • Country: gb
    • my website holding page
Re: opamp confusion
« Reply #8 on: June 14, 2015, 07:21:37 am »
The bottom of the 200R resistor should be tied to somewhere that's at the midpoint of the supply, not to the negative rail.

Offline cravenhavenTopic starter

  • Contributor
  • Posts: 40
  • Country: au
Re: opamp confusion
« Reply #9 on: June 14, 2015, 08:00:35 am »
Thankyou all,
Kerry and Possums solutions got me back on track. I increased all the resistor values by 10 or 100 and shifted the bottom of R1 to the midpoint rail instead of -ve rail and all started working.

Kerry, I tried the 5500 resistor to the + rail, but dont know where to tie the R1 resistor from the "-" input to. I tried a few options but the output is either flat or slight signal but at 14mv wrt ground.

Thanks for the link to the calculator noitaint.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19528
  • Country: gb
  • 0999
Re: opamp confusion
« Reply #10 on: June 14, 2015, 02:09:35 pm »
Here's an idiot's guide to scaling and level shifting with op-amps.
http://www.ti.com/lit/an/sloa097/sloa097.pdf
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf