Amplifing strain gauge with an LM 324 - not getting expected results

Discussion in 'The Projects Forum' started by MrJojo, Jun 6, 2014.

  1. MrJojo

    Thread Starter Member

    Jan 23, 2013
    45
    0
    Hello all,

    I'm using a strain gauge with a 5V power supply across the wheatstone bridge. at max load, I should be getting 2.5mV. Therefore, I decided to build an amplifier to range from 0-5V, with 5V being 2.5 mV. I have an LM 324 which I'm using as the op amp and I believe that those are great for DC and low frequency applications.

    Attached is an image of the current circuit I have bread boarded. First, I figured I would need a 2000 V/V gain with my amplifier. I also want to mock up the 2.5 mV output of my strain gauge, so I used a voltage divider on the 12V to get me 2.5mV.

    Math:
    Gain: 2000=(1+R3/R4); R3=1999*R4.
    Voltage divider: 2.5mV=(R2/(R2+R1)*12V; R1=4799*R2.

    With these, I found resistor values which are close to what I want.

    R1 = 1.05M
    R2 = 220
    R3 = 200k
    R4 = 100

    When I power on the circuit, my output is 10.5V. I checked the negative input of the op amp and I found that it is 5 mV while the positive terminal is 2.5 mV. I thought op amps want to have the negative and positive terminals at equal voltages, so why is the negative terminal twice as high as the positive?

    I went back and read the AAC's write up of divided feedback (http://www.allaboutcircuits.com/vol_3/chpt_8/5.html) and that basically reaffirmed what I thought about the positive and negative terminals wanting be equal voltage.

    What am I do wrong? Is my basic understanding of how the op amp works incorrect?

    Matt

    EDIT:
    I forgot to add the true gain and the voltage from the divider
    Voltage divider = 2.514 mV
    Gain = 2001 V/V
     
    Last edited: Jun 6, 2014
  2. crutschow

    Expert

    Mar 14, 2008
    13,003
    3,232
    The LM324 is not so great for high gain DC applications since it has a input offset of up to 7mV, and it would appear the op amp you have has an actual offset of 2.5mV. So you either need to add an offset correction circuit or use a lower offset op amp.

    Incidentally, your circuit will not amplify the differential output of a wheatstone bridge circuit. For that you need a difference (differential) amp circuit.
     
  3. enggricha

    Member

    May 17, 2014
    62
    1
    For almost exactly the same application I use a AD620 with the reference offset to about 2V to use it with a single 5V power supply. That way my useful range starts at 2V and i set the gain to take to max output to safely within range of upper power supply limits of the AD620.
     
  4. MrJojo

    Thread Starter Member

    Jan 23, 2013
    45
    0
    So something like this? I added the wheat stone bridge form the strain gauge, and then I am doing the DC offset via Millman's Theorem (Figured it was just a quick way to get it done without adding much more to the circuit).

    So I should be seeing 7.25V on U1.2 pin 5 and 7.25125 V on U1.1 pin 3. That means my difference would be .00125 V now instead of .0025 V. I will do the math later to double my gain - which is 2000 atm. I just wanted to post this to see if I'm on the right track at least.

    Math for gain
    Av = (1+2*(R)/R2)
    where R = 100k = R3, R4, R5, R6, R7, R8.

    Matt
     
  5. crutschow

    Expert

    Mar 14, 2008
    13,003
    3,232
    Note that such a differential amp is very sensitive to the ratio accuracy of the resistors. That's why a instrumentation IC, which has internal matched resistors, is normally used for that task, and I recommend that's what you use also.
     
  6. enggricha

    Member

    May 17, 2014
    62
    1
    not sure if you have figured this out...but you should start with a instrumentation amp like the AD620 for example
     
  7. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,532
    1,251
    There are circuit errors in your schematic. R7 and R5 connect to pins 1 and 7, not 2 and 6. Also, R2 is in the wrong place. This may not be an error, but the DC offset circuit is attenuating the output of the strain gauge by 50%. And if the opamp still is an LM324, they you still will have offset errors that could be greater than your signal.

    ak
     
    Last edited: Jun 20, 2014
Loading...