Connect with us

Op amp gain trouble.

Discussion in 'Electronic Design' started by bunny, Dec 1, 2005.

Scroll to continue with content
  1. bunny

    bunny Guest

    I am using the below schematic for a OpAmp TLC272CP. According to the
    formula of G = (R1/R2) the gain on this circuit should only be a matter
    of 11. however as you can see below it's closer to 15/16. I'm
    wondering what could cause this anomoly. I have put the circuit
    below(sorry for the Ascii) and what i put into the + of the opamp and
    what i got as output.

    Thanks



    base=2.0mv
    output=33.4mV


    +9V +9V
    | (+). |
    100k pot<------|\|
    | (-)| \_____.__R3____Meter__.
    | .-| / | |
    | | |/| | |
    | | | | |
    | .------R2-' |
    | | | |
    | R1 | |
    | | | |
    Gnd Gnd Gnd Gnd

    R1 = 1k
    R2 = 10k
    R3 = 10k
     
  2. No, the gain is (R1+R2)/R1. Which indeed is 11 in your case.
    If you check the datasheet, you'll see that the typical offset voltage of
    this opamp is 1.1mV. This offset gets added to the input voltage. So in your
    case, (2.0mV + 1.1mV) * 11 = 34.1mV. Almost exactly what you measured....

    Meindert
     
  3. bunny

    bunny Guest

    R3 is because i'm using one of those radio-hack board that has the stuff
    like switches and multimeter and led's and pots and crap all in it
    already with a breadboard int he middle and the multimeter reads mV's so
    the r3 10k resistor is meant to drop the voltage reading to the mV area
    so it doesn't destroy the multimeter on the board. I have a seperate
    digital multimeter which is what i used ot test for the values.
     
  4. bunny

    bunny Guest

    I have a radio-shack electronics lab with a breadboard in the middle and
    pots and stuff all around. It includes a multimeter that can only read
    mV's so i had to include 10k resistor in series with the multimeter to
    drop the Voltage down to a level the multimeter can safely read. I have
    a Digital Multimeter taht i used to take the values. The one on the
    board is analog.
     
  5. Pooh Bear

    Pooh Bear Guest

    You're forgetting about DC offset errors. A TLC272 is not suited to
    precision DC amplification.

    What's the point of R3 btw ?

    Graham
     
  6. Allen Bong

    Allen Bong Guest

    Please forgive my ignorance. I've downloaded the datasheet of TLC272
    from Texas Instrumenents.

    http://www.futurlec.com/TI/TLC272CP.shtml

    On P5 & P6, the VIO for TLC272C are TYP = 1.1mV and MAX = 10mV for both
    VDD = 10 & 5V.

    What is the parameter Rs stood for and how do I make VO=1.4V?
    P28 does memtion RS on a graph but I couldn't get any clue out of it.

    Will blowing hot air onto the chip while operating make the offset
    voltage raise up to 11mV?

    Allen
     
  7. Yuk. It's like doing some kind of arthroscopic surgery. Why not go to
    the horse's mouth.. TI's website?

    http://focus.ti.com/lit/ds/symlink/tlc272.pdf

    Rs would be source resistance. It's not particularly important for Vio
    except to notice that it's very low, so as not to include much in the
    way of input bias current effects.

    You make Vo = 1.4V by adjusting the input voltage(s). It's unclear how
    exactly they do the test. Another version of the data sheet for the
    same part has this footnote:

    * All characteristics are measured under open-loop conditions with
    zero common-mode voltage unless otherwise specified. Unless otherwise
    noted, an output load resistor is connected from the output to the
    ground pin.
    It could cause thermocouple voltages that happen to add to existing
    Vio. You should avoid thermal gradients and fast changes if you want
    to measure the actual Vio. But the effect will generally be << 1mV
    unless you're doing something silly like pointing a heat gun at the
    poor thing.


    Best regards,
    Spehro Pefhany
     
  8. Spehro Pefhany wrote...
    Perhaps he wanted the old 1994 version b of the datasheet,
    instead of the newer 2002 version e on TI's website? :)
    It'll also make a thermal gradient across the die, which will shift
    the Vgs voltage-matching of the input transistors, and the current
    mirrors. Vgs is a volt or two, so a 10mV difference is only 0.5 to
    1% change in Vgs. A quick glance at a few MOSFET datasheets shows
    the absolute value of Vgs can change 200mV for a 75C temperature
    change (in the sub-threshold region), so a 10mV mismatch shouldn't
    be too hard to achieve, with sufficient mistreatment of the part.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-