AI CHARM Diagnostic - Primary Variable Inaccurate

AI CHARM with HART enabled, E&H Promag connected and auto-sensed correctly.

No diagnostics are indicated at the meter.  Instrument is functioning normally.

CHARM diagnostics indicate OInteg = BAD, and Primary Variable Inaccurate.

HART Device alarms are disabled.  Can someone tell me what is triggering this? Not sure how to diagnose. 

9 Replies

  • Sim,

    There are a couple of possibilities I can think of. Check the CHARM configuration and make sure the UNDERRANGE_PCT hasn't been set to zero (default is -3). That can trigger an uncertain status. The other is that the status is coming over the HART status from the device. There may be a "calibration required every xx" type of setting configured in the device.
  • In reply to Scott Thompson:

    Interesting. I'm wondering if the device Voltage drop is exceeding the fields side limit of 15 Volts, and is not able to drive the mA signal to match the actual PV value, hence a discrepancy between the analog signal and the HART PV value.

    This usually manifests at the higher end of the mA range as the voltage drop on the Input resistor (and some voltage drop in the Current Limiter circuit).

    I would compare the analog signal to the PV and if there is a discrepancy, figure out at what % value this happens at. Check the device specifications to see if there are some start up loads that might prevent proper initialization.

    One solution for higher startup or runtime power is to use the new DeltaV 3 wire Analog Terminal block. This block provides fused Injected power to the device, which bypasses the internal current limiting circuit of the CHARM and will provide full 24 volts at the field terminal. This can add up to 3 Volts of head room on the circuit. This terminal block was released with v13 and can be used on SIS or BPCS IO. It is strictly a hardware option so the DeltaV version does not matter. ( it works on any version).

    Not sure yet what the issue is, but if you have a second device in series in the current loop, the sum of voltage drops could affect the device's ability to drive the mA signal.

    Andre Dicaire

  • In reply to Scott Thompson:

    Hello Scott!  Underrange_pct is at -3, default.

    Device is posting no current diagnostics, but I'll keep digging -

    I was thinking a hidden HART status as well

      

  • In reply to Andre Dicaire:

    Thanks Andre, will check voltage.
  • In reply to SimHogan:

    Did you ever get to the bottom of this? I seem to have the same problem with a E&H Promass 200.
  • In reply to Cedric Dawnhawk:

    Hi Sim,
    Good Day & Hope you are doing Great...
    Same error has been observed in another Coral project & our findings to make "Low-Cut off" option disable in field device from AMS Device Manager.
    We have E&H Promass 300/500 Device rev 6 on which we have disabled "Low Cut Off" option which resolve / fix this issue.
    Kind request you to check in Device Configuration parameters to make this option off & refresh the DeltaV Diagnostics. I believe this will resolve issue.
    Kind request you to share your feedback once done.
  • In reply to Tushar Gohad:

    Thanks for your response. I had already started to think the low flow cutoff might be the problem. I think then when this is active it triggers the "Primary Variable Inaccurate" message in DeltaV Dagnostics, possibly via the HART response code status, which I suppose makes sense. I do not really want to disable the low flow cutoff. However I think there maybe an alternative solution by setting one of the ignore options in the HART_ERRORS parameter for that channel. I am not sure which - possibly "Ignore Loop Current Disparity". I will post again when I have some results.

    If anyone has any information about the correspondence between the HART response code status and the DeltaV ignore options I would be interested.
  • Hi Sim,
    Good Day & Hope you are doing Great...
    Same error has been observed in another Coral project & our findings to make "Low-Cut off" option disable in field device from AMS Device Manager.
    We have E&H Promass 300/500 Device rev 6 on which we have disabled "Low Cut Off" option which resolve / fix this issue.
    Kind request you to check in Device Configuration parameters to make this option off & refresh the DeltaV Diagnostics. I believe this will resolve issue.
    Kind request you to share your feedback once done.
  • In reply to Tushar Gohad:

    You asked for some feedback. I have been to site to investigate the four meters involved and my interpretation of the results are as follows:

    Examination of the diagnostics and event logs of the four meters showed that they were experiencing a number of errors:
    S862 Partly filled pipe
    S912 Medium inhomogeneous
    S842 Process limit

    These are all classed as ‘Out of specification’ errors (which is why they are prefixed S) and any one of them sets the “Out of specification” bit of the NAMUR NE107 condensed status which is returned in the HART communication. This is indicated in DeltaV diagnostics by the message Primary Variable Inaccurate. By default this sets the status of the signal from the CHARM to Bad, which then generates the IO error in the module. However there is also a parameter for each CHARM called HART_ERRORS which can be set to ignore various sorts of HART errors so that they do not produce a BAD status. In particular I found that the checking the option ‘Ignore Device Variable Status’ ignores all ‘Out of specification’ errors. I set this true for all four CHARM inputs. This seemed to clear all the IO input errors. It did not affect more serious errors on a faulty meter that still indicated bad status.

    In the Promass configuration it is possible to change the level of response to various error conditions to one of Off/Alarm/Warning/LogBook. However of the three above I could only find a setting for S912. Hence the blanket suppression of all ‘Out of specification’ by using the ‘Ignore Device Variable Status’ option.

    Low flow cut off and empty pipe detection were left enabled.

    Andrew Chadwick