Need suggested schematic for a level shifing ampllifer

Discussion in 'The Projects Forum' started by AverageGuy, Aug 18, 2011.

  1. AverageGuy

    Thread Starter New Member

    Dec 27, 2009
    9
    0
    I have a battery monitor that is no longer working correctly. There doesn't seem to be a way to fix it and the monitor is no longer made nor in stock anywhere. The newer version from the company does not support a dual battery installation and I'm not going to buy two of them so I'm planning to build one myself. I have a current shunt that's working fine, so I'll use it to sense current flow (charging or discharging). Actually I have two, one for each bank. From my measurements I find that the shunt produces .5 mv for a two amp current. I'll be using the a dc-dc converter to give me a negative voltage for VSS and a regulator from the 12 V battery to provide vDD. I figure a gain of 50 will let me produce an output of +-2.5 V at 200 amps. I will then feed that into a ADC read by a microprocessor, most likely an Arduino. I need however to translate that to a range of 0-5 V What's the best way to do that?

    Thanks,
    Jim.
     
  2. #12

    Expert

    Nov 30, 2010
    16,257
    6,757
    Use an opamp circuit to introduce an offset.
     
  3. AverageGuy

    Thread Starter New Member

    Dec 27, 2009
    9
    0
    Would something like this work?

    [​IMG]
     
  4. #12

    Expert

    Nov 30, 2010
    16,257
    6,757
    Suffering from brain lock...working on it....anybody else is welcome to beat me to the answer.

    Edit: Got it. Taking time to draw it.
     
    Last edited: Aug 18, 2011
  5. #12

    Expert

    Nov 30, 2010
    16,257
    6,757
    First amp changes the signal to unipolar (one line). The second amp uses a current source to force the offset. Using Vss/.0025 might not be very steady. Any way to get -2.5 ma into the negative input node will work.
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You don't need a gain of 50; you need a gain of 2.5.

    See the attached. The LMC6864 is a quad RRIO opamp.
     
  7. #12

    Expert

    Nov 30, 2010
    16,257
    6,757
    Just spotted an error. The feedback resistor on the first amp needs to be larger to get the voltage gain you need. Perhaps 1k for the feeler resitors and 50k for the feedback resistor.
     
  8. #12

    Expert

    Nov 30, 2010
    16,257
    6,757
    Doing the math:

    .5mv/2amps= .00025 ohms for the shunt.
    200 amps times .00025 ohms = .05V per direction of flow
    2.5V/.05V = 50 (gain)

    I think .00025 ohms is an unlikely shunt value unless it's a piece of large copper wire, but I'd do it that way for a 200 amp feed!
     
  9. ifixit

    Distinguished Member

    Nov 20, 2008
    638
    108
    Therefore the shunt resistance must be 0.0005/2=0.00025Ω (0.25mΩ).

    I don't like the circuits suggestions so far, especially if the shunt is on the high side. In any case, a differential amp configuration should be used to ensure an accurate shunt voltage drop reading.

    I have attached a circuit idea for evaluation.

    Regards
     
    Last edited: Aug 18, 2011
    #12 likes this.
  10. #12

    Expert

    Nov 30, 2010
    16,257
    6,757
    Good way to use one less amplifier.
     
  11. AverageGuy

    Thread Starter New Member

    Dec 27, 2009
    9
    0
    The shunt is a copper strap. Maybe an 1/8 inch thick and 1/2" wide. The charger puts out over 100 amps and the inverter is a 2000 watt inverter. The total battery capacity is 440 amp hours. I used a good quality digital meter with a known load to measure the voltage at 2 amps. This isn't your hinky little 10 amp meter shunt. The wires leading to the shunt are 4/0 AWG. I think the math is correct.

    Jim.
     
  12. AverageGuy

    Thread Starter New Member

    Dec 27, 2009
    9
    0
    It's on the ground leg. Nice circuit. I probably won't move it to the high side.

    Jim.
     
  13. ifixit

    Distinguished Member

    Nov 20, 2008
    638
    108
    Do use the differential configuration so that you can have the circuits located in a handy spot away from the shunts. Use twisted pair wires to bring the voltage to where the amplifier is.

    Regards,
    Ifixit
     
    Last edited: Aug 18, 2011
    AverageGuy likes this.
  14. AverageGuy

    Thread Starter New Member

    Dec 27, 2009
    9
    0
    Thanks, I'm going to have to cogitate on how to redepoly the bias with a ground centric shunt. I've been away from op amp design for far too long. Actually when I was in school operational amplifiers were multi big bucks and consisted of discreet components. I introduced IC op amps in an ME class as a design for a whetstone bridge to implement a scale using strain gauges and they thought I was nuts. I showed them a working design. I think this was before the 741. I was using some Motorola parts but I have no idea now what they were.

    Thanks,
    Jim.
     
  15. ifixit

    Distinguished Member

    Nov 20, 2008
    638
    108
    Hi Jim,

    I remember to. Lets change our names to; AverageOldGuy and OldIfixit.:D Comparing the 741 to the modern versions is like comparing the first TVs to todays TV models.

    I have attached a more practical update. The small details are important when dealing with amplifying low voltages and ensure reliability and accuracy. Pick some parts for an opamp and a 2.5 volt reference that you can get from where you live and post your circuit. If you have questions, just ask.




    I have questions:
    1. What reading accuracy do you need?
    2. What temperature range will the circuit be operating in?
    3. Humidity?
    Regards,
    Ifixit
     
Loading...