offset error introduced in strain gauge amplifier

Discussion in 'General Electronics Chat' started by strantor, Apr 30, 2011.

  1. strantor

    Thread Starter AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    Please see the drawing:
    [​IMG]

    The portion drawn in back is existing and working. It is a load cell connected to a PLC via a strain gauge amplifier/signal conditioner. The idea is to have a display for this tension from the load cell. I plan to install the portion in red. I have been working over this plan with another technician on the other side of the world on this issue but he has become busy and unable to answer my email (or he's getting tired of my questions). He agrees with me that it would be better to install a 2nd amplifier rather than add to the output of the existing one and he agrees that it should work without connecting the exitation+ wire (to avoid the 2 amplifiers trying to supply excitation voltage) but he says that using a 3-wire amplifier will cause an offset error in the PLC or the new display or both (error proportional to cable length). He says I need to use another 4-wire amplifier like the existing one. Can anyone confirm this & explain why? I need this to work as I already have bought the display and it will only work with a 3-wire amplifier.

    Thanks
     
  2. GetDeviceInfo

    Senior Member

    Jun 7, 2009
    1,571
    230
    I would not do it that way. Your 4-20ma display should work by just dropping it in series on the signal line without the second amp. I modified a similar circuit recently where the complaint was that the PLC and the stand alone display were not agreeing. I ended up tossing the display (Redlion PAXD), creating a Labview based HMI that quered the PLC's AD converter over a Modbus connection (RS232).
     
  3. BillB3857

    Senior Member

    Feb 28, 2009
    2,400
    348
    Since the strain gauge amplifiers also source the drive current for the bridge, attempting to connect two amplifiers to the same strain gauge bridge is asking for problems. The prior suggestion of placing the display in series with the PLC 4-20ma loop makes perfectly good sense.
     
  4. strantor

    Thread Starter AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    The reasons for having 2 amplifiers are:
    The 4-20mA output of the existing amplifier goes to a 500Ω voltage divider (250Ω/250Ω) to produce a 0-5V output to the controller.

    My display has 250Ω series resistance which, if placed in series with the voltage divider, would skew it's output, effectively making it 750Ω (500Ω/250Ω). I can't correct for this error in software as it's outside the calibration range.

    I wish I could do it that way, but the "PLC" is actually a purpose-built proprietary programmed weird saur danfoss hyrdaulic system controller. Actual programming of thing is not possible; the "programming softawre" only allows you to adjust parameters like ramp speeds and such. there is also no way to connect an HMI to it.
     
  5. strantor

    Thread Starter AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    That's why I would leave off the +excitation wire from one of the amplifiers. It should work in theory (or only in my mind) but we shall see. I'm just going to hook it up and find out. I will post what I find
     
  6. strantor

    Thread Starter AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    For future reference, I installed the portion in red exactly as I drew it. It worked great. The effect on the calibration of the existing system was negligible; still well within established tolerances . The new display is dead-on accurate. Not sure where the other guy was coming from on this issue but it no offset error was noted.
     
  7. BillB3857

    Senior Member

    Feb 28, 2009
    2,400
    348
    Glad it worked out for you.
     
Loading...