Low side current sensing

Discussion in 'General Electronics Chat' started by DumbDummy77, Mar 16, 2010.

  1. DumbDummy77

    Thread Starter Member

    Jan 21, 2010
    43
    0
    Hi

    I am current using inverting amplifier circuit to sense the current level. The Rsense is 10mOhm and I use the gain about 100 so that means I will get 1V when 10mV go into input of non-inverting of op-amp.

    At first time, I useD 10k for non-inverting and inverting input respectively, I use D1Mohm for the feedback. I useD 5V for now. To my surprise, I only got about 540mV which I did not expect. The output should be 1V.

    Then I reduced the resistor values down 1k for both inputs and 100k for the feedback. Again I still get the same reading. The output do climb up or down when the input voltage up or down respectilvey but the output should be 1V when 10mV is on the input. I suspected it must be do with my op-amp chip - LM324N but I am not entirely sure to be honest. What is your recommedation?

    DD77
     
  2. Ghar

    Active Member

    Mar 8, 2010
    655
    72
    Have you tried switching the inputs?
    You could be trying to output a negative voltage and if you're using a single voltage supply your LM324 won't go negative.

    Edit:
    What am I saying, you only have one input...
    Try a non-inverting configuration?
     
    Last edited: Mar 16, 2010
  3. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    Can you post a schematic of the circuit you are using? That will greatly facilitate this discussion.

    hgmjr
     
  4. Bychon

    Member

    Mar 12, 2010
    469
    41
    The input offset voltage of a 324 can be as high as 7 millivolts. If it was 5.6 millivolts, that would explain everything.Try an OP-07. Their input offset is guaranteed to be less than .15 millivolt and they have zeroing terminals to get really accurate.
     
  5. DumbDummy77

    Thread Starter Member

    Jan 21, 2010
    43
    0
    Here is my sketch of circuit

    I am going to find the reason behind this problem in the meantime...

    DD77
     
  6. DumbDummy77

    Thread Starter Member

    Jan 21, 2010
    43
    0
    Just looking at my textbook,

    To determine the voltage error at output by using three parameters such as input offset voltage, input bias current and input offset current.

    For LM324N, the worst case parameters

    Input offset voltage - 7mV
    Input bias current - 250nA
    Input offset current - 50nA

    RB2 = R1 // R2

    So R1=10K and R2=1M

    RB2 become 9.9kohm

    Applying in three formula

    V1error = (10k-9.9k)*250nA = 25nV

    V2error = (10k+9.9k)*(50nA/2) = 497.5uV

    V3error = 7mV

    Again = 100

    Voutputerror = 100(25nV+497.5uV+ 7mV) = 750mV

    Is the correct?

    I am not convinced because I am expecting to get summat like 400mV instead because I get 1mV at the input and get about 570mV which should be 1V.

    DumbDummy77
     
  7. DumbDummy77

    Thread Starter Member

    Jan 21, 2010
    43
    0
    To be precise, the input voltage is 10mV and the output is 537mV and I need to confirm my calcualtion match to what I have observed. Can anyone please double-check my calculation in case there is error?

    DD77
     
  8. Bychon

    Member

    Mar 12, 2010
    469
    41
    Your gain is 1+1M/10k
    The input bias current flows from each input terminal, one through a 10k resistor and one through a 9.9k resistor. It's the difference in bias currents that adds to error, and it's usually not much because you used similar value resistors on each input. 250na x .1k = 25 uv is more like the correct calculation for bias current.
    And the input offset current is the real bias current error contributor. 50na x 10k = 500uv
    7mv + 25uv + 500uv = .007525
    .007525 x 101 = 760mv max error

    so, do you want to do some more math or try an op-amp that doesn't have all these input offset voltages?
     
  9. DumbDummy77

    Thread Starter Member

    Jan 21, 2010
    43
    0
    Please continue some math and I am interested to hear what op-amp IC you would recommend me to use. Out of curiosity, can the voltage error can varies from 0 to 760mV? If yes, how can it varies?


    DD77
     
    Last edited: Mar 17, 2010
  10. Bychon

    Member

    Mar 12, 2010
    469
    41
    I already recommended an op-amp. As for how can the error voltage be less than the maximum guaranteed by the manufacturer, ask the manufacturer why all their op-amps don't come off the production line with the worst possible performance they can get away with.
     
  11. DumbDummy77

    Thread Starter Member

    Jan 21, 2010
    43
    0
    Hi

    I have just ordered a couple of 0P07 OP-AMP and Hopefully it comes this week. In the meantime, I have tried to use TL084 op-amp and I could not understand why the voltage swing to 5V regardless of signal on the non-inverting input. Why is that?

    Is it must be related to common mode voltage?

    DD77
     
  12. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    The TL07x and TL08x series of amplifiers cannot "see" inputs that are within 3v of -Vcc, or -V, or Vee. If you have the negative supply terminal grounded, then the lowest it can see is 3v.

    If you wish to use a TL084 to sense input near ground, you must use a dual supply. +5/-5 minimum, +15/-15 preferred.

    [eta]
    If you are not using any of the other amps in the IC, you should turn them into voltage followers so that they do not oscillate by themselves. For each unused amp, connect the output to the inverting input, and connect the noninverting input to a reference voltage that is halfway between +V and -V. A couple of 4.7k or 10k resistors in series across your supply will work just fine.

    Don't forget to use 0.1uF bypass caps between +V and ground, and -V and ground, right at the amp.


    You're amplfying the Vsense signal quite a bit. If you are planning on eventually using it for controlling PWM, you will have problems, as you will need a very, very fast opamp to keep up.
     
    Last edited: Mar 20, 2010
  13. DumbDummy77

    Thread Starter Member

    Jan 21, 2010
    43
    0
    Hi

    I have received my OP07 today and have replaced the existing op-amp (LM324N). As I increased the current up to 1A and the output voltage did not climb up because it was struck at 12V. I double-checked the circuit and it was ok. I could not understand what was the problem. :confused:

    dd77
     
  14. DumbDummy77

    Thread Starter Member

    Jan 21, 2010
    43
    0

    Hi

    I have been looking at a datasheet of TL084 and I could not find what you have said about op-amp cannot see signal below 3V. According to my understanding, what you have said is known as input common mode voltage (If I am mistaken, please correct me) deals the +/-15V signal.

    Furthermore I don't intend to use dual rail for my application as I intend to use single rail op-amp.

    DD77
     
  15. DumbDummy77

    Thread Starter Member

    Jan 21, 2010
    43
    0

    I am struggling to understand or find the information about what you said TL084 cannot deal voltage below 3V otherwise it will latch up to postive rail. How can I exactly determine the input common mode voltage for single and dual supply?

    DD77
     
  16. Bychon

    Member

    Mar 12, 2010
    469
    41
    Read the data sheet! Texas Instruments page 6 says, Electrical characteristics, Vcc = +/- 15V
    Common mode input voltage range: minimum +/- 11 V Typical: +/- 12 V Maximum +/-15 volts.

    15 minus 11 is 4

    The chip is only guaranteed to work if the inputs are at least 4 volts away from the power supply voltage.
     
Loading...