Emerson Exchange 365

• Not Answered

DCS Analog Input Charms Always giving an offset value of 2% with no Diagnostics Error

I just wanted to check if anybody have experienced for the DCS Ai 4-20mA Charms reading a value with an 0.492104 % at 4mA and 102.058 % at 20mA.  The charm have no error in diagnostics nor led bad indication. This have been observed with the device connected  or direct injection of 4-20mA to the channel. 

EricT

• 28 Oct 2020 3:47 PM
• Cancel

1 Reply

• Andre Dicaire

  

  • 28 Oct 2020 5:43 PM

  The first question is what are you measuring the signals with? There's an old saying that a man with one watch knows what time it is, a man with two watches is never sure...
  
  Either all the CHARMS are off or the Meter is out of calibration.
  
  The error appears to be one of Span. The 2.058% error at 20 mA would show a 0.41% error at 4 mA. (I was curious)
  
  How are you measuring this: 4-20 mA calibration signal? Fluke digital meter in line reading mA? Fluke Digital Meter across CHARM measurement points reading 1 - 5 Volts?
  
  As I recall, the AI CHARM was designed with a high precision load resister that was not 250 Ohms. We added a second lower precision resistor in series to make up the difference so that a user can take a quick 1-5 Volt reading. So the 1-5 volt reading can introduce a small error that is not present in the actual signal read by the CHARM. If you adjusted a current to read exactly 5.00volts, that may not be exactly 20 mA. (if there was added resistance in the contact points of a voltage reading, adjusting the signal to 5.000v would require additional current. )
  
  If you measured in line with Fluke or from a calibration meter generating 20 mA, then either the Meter is out of calibration or the CHARMS are out of tolerance.
  
  The CHARMs are calibrated at the factory. Since all components are fixed there are no parts that can go out of calibration. Each circuit is checked and the calibration is done on the digitized signal. There can be no drift and hence there are no calibration pots. Since all CHARMS are calibrated to the same reference, an issue in this manufacturing step would explain a common error in all CHARMS produced during the period the error existed.
  
  My recommendation is that you check the calibration date of the meter used or send it out for calibration. Make sure you are reading the mA current and not an inferred voltage. Use one or more additional meters to get a "consensus" on the actual mA signal. If these readings all agree the CHARMS are out by 2% at 20 mA, I would log a call with the GSC.

  Andre Dicaire

  • Cancel