rjm21 (Customer) asked a question.

April 8, 2021 at 4:26 PM
FC-33 Signal Conditioner Calibration

I have an application where I need to read DC voltage to at least 10mV resolution. I'm using a VADT150-42-24 Voltage transducer which outputs 4-20mA to an FC-33 signal conditioner which then outputs 0-5VDC to the BRX BX-04AD2B Analog Input Card. I'm getting correct readings on the PLC end, but find that there is a lot of noise/jumping counts.

 

I do not believe it is a ground loop problem, these are the only components in the system. I also had an O-Scope on the Voltage source, it is not as noisy as the PLC is showing.

 

I think part of my problem is I don't understand how calibration works on the FC-33. If I hold the calibration button at the minimum of 0V (4mA from the VADT), the signal into the PLC clears up. However, then the FC-33 will only output that value until the DC voltage reads around 10V, as if the span is way off. I've attempted to set the maximum for the FC-33 following the instructions from the insert, but I've been unsuccessful.

 

Any thoughts?

 

I've attached a screen shot of a trend view showing WX0:U, you can see how much the count value jumps up and down. The trend view also shows the engineering units of VDC via RX0.35V

  • 4 answers
  • 457 views

  • Tinker (Customer)

    Yet another thought, You say DC voltage, how about just using a resistor voltage divider instead of the AC/DC transducer (being able to do RMS AC adds a lot of complexity)? Isolation might be a concern unless "grounds" are common with the PLC, but you do already own the FC-33, so you cold use it if you have to.

    Selected as Best

  • Tinker (Customer)

    I'm curious why you are using the FC-33, I guess you already had the voltage input module and the voltage sensor is only 4-20mA. However, 4-20mA across a 250 ohm resistor is 1-5V. Perhaps you need the isolation for some reason, but a resistor is much less expensive, and being a simple passive device removes some variables. I would at least try a resistor for testing to see what happens.

     

    you wrote: " If I hold the calibration button at the minimum of 0V (4mA from the VADT), the signal into the PLC clears up."

    My interpretation of the instructions are than one MUST do BOTH low and high calibration. It would probably best to have a process calibrator as a source. Personally, IF the FC-33 is actually needed I'd return it to factory calibration and do any tweaking needed in the scaling in the PLC. If results were unsatisfactory I'd next try some filtering on the PLC side, and if that didn't work, I'd reevaluate if that is really the right voltage transducer for the job.

    Expand Post

  • Tinker (Customer)

    Another thing, you wrote " have an application where I need to read DC voltage to at least 10mV resolution."

    The very minimal specs. for your 150 volt range transducer say the accuracy is better than 1% of full scale. 10mV is a LOT less than 1% of 150V, like an order of magnitude less. Considering what looks to be about 0.2V steps (plus minus smaller noise) in the data on your trend I'm guessing that is all the the transducer can do. Averaging a significant number of samples might get you a bit more resolution (like a crude sigma-delta ADC) but I make no guarantee of linearity and an order of magnitude is probably asking much too much.

    If your voltage is about 36 as shown on your trend, the 50V range transducer would probably get you a factor of 3 improvement, still only a third of what you want, but somewhat better, with averaging might get you close enough, maybe.

    Expand Post

  • Tinker (Customer)

    Yet another thought, You say DC voltage, how about just using a resistor voltage divider instead of the AC/DC transducer (being able to do RMS AC adds a lot of complexity)? Isolation might be a concern unless "grounds" are common with the PLC, but you do already own the FC-33, so you cold use it if you have to.

    Selected as Best

    • rjm21 (Customer)

      Thanks for this. I got an Elctrical engineer involved and after some trouble shooting, we discovered the "noise" (or maybe just inaccuracy) of the Voltage Transducer was the issue. We will be making a voltage divider to use instead. ​

 

Did this information resolve your issue?