• Not Answered

Intermittent Loss of Signal on Glycol Recirculation Pump Discharge Pressure Transmitter (Rosemount CD 3051)

Dear Sir,

 

I recently came across one of your publication on ‘’Pump Monitoring using Electronic Differential Pressure Transmitter’’ during a research to resolving a re-occurring fault on the discharge pressure transmitters of the glycol recirculation pump.

 

Below is the background story;

 

Design

 

We have two glycol recirculation pumps P83601 (No.1) and P83602 (No.2) both fed into a common header. Only one operates at a time. We have one suction pressure transmitter common to both pump suctions and two discharge pressure transmitters, PT-064 and PT-065, monitoring pump No. 1 and No. 2 discharge pressures respectively. Although both a exposed to the same pressure regime since they share a common header. Both have common operating scales (0 -100 barg), high alarm (set at 82.0 barg) and high high alarm (set at 85.01 barg). Operationally, when one pump is on duty, the other is on standby. Whichever one was running, the discharge pressure transmitter assigned to it monitored and tripped the pump in the case of abnormal condition. This is the operating philosophy.

 

Problem

 

We have had several re-occurring occasions when the PT-064 loss signal intermittently, momentarily and abruptly. This of course tripped the pump No.1. There is no appreciable increase in discharge pressure when this transmitter is trended on DCS. What is only seen is the dotted lines which signifies loss of signal. When the pump is reset and ran, it might be in operation for some days or weeks and then the issue re-occurs again. At times this might happened to both transmitters (PT-064 & PT-065) despite the fact that one is on duty and meant to trip the pump it was monitoring. This is really a serious issue as it has re-occurred in several occasions. I decided to take it up as a project to find a lasting solution to it.

 

Below are some troubleshooting carried out so far;

 

Troubleshooting Steps

 

1. Function tested PT-064 loop from field to DCS and confirmed okay.

2. Changed the signal cable, power cable and glands from the field transmitter to its field junction box.

3. Replaced PT-064 transmitter, twice, with new transmitters. However, problem persist after some time.

4. Swap PT-064 loop (faulty Tx) with PT-065 loop (good Tx) and the problem transferred to PT-065. Loops were returned to initial connection arrangement.

5. Changed barrier on PT-064 loop (DCS marchalling cabinet). Problem persist.

6. Increased the damping of PT-064 transmitter from default 0.40secs to 3.2secs. Problem persist.

 

I will be grateful if you can provide a more better solution to this issue. Thanks in advance.

 

NB: Both pressure transmitters are Rosemount CD 3051

 

Kind Regards,

 

Chris Adaro | Instrument/Metering Specialist | Maintenance Team | Sea Eagle FPSO Operations

Shell Petroleum Development Company | Offshore Division | Lagos - Nigeria  

T +234 807 025 3486|  Ext. 53486

Christopher.Adaro@shell.com | www.shell.com

5 Replies

  • Hi Adaro
    We face same problem, in discharge of reciprocating pumps, we found transmitter fail due to high discharge fluctuation and vibration. (pump pulsation damper failed )
    Regards
    Youri
  • In reply to MOH:

    Hello Youri,

    Thanks for your comments. The pump pulsation damper, in this case, is fully charged. Beside, if there is a high discharge fluctuation, it will reflect on the DCS and very visible when the discharge pressure transmitter is trended on DCS. I am still surprised with what could be responsible for this issue.
  • Your troubleshooting steps are proper. So the reason for that measurement failures seems to be in piping. Please reconsider the outlet pressure transmitters location - the required pulsation damping shall be provided.
  • In reply to adarochris:

    Hi Adaro
    the Best sampling rate in DCS is 1 seconed, and if fluctuation rate is faster than 1 second you can not see exact and correct value in trend.please check piping vibration and gauge for fluctuation in pressure.
  • Hi
    I also faced a similar kind of issue ,in that noticed the sample tapping of PT was just to discharge of control valve . The logic system (DCS or PLC )will work much faster than the trending ,so there could be a chance of missing trend,i would suggest to putting a filter timer and test run, tapping location and discharge flow pipe line study(flow ,vibration,air trap etc), or 2 out of 2 logic check with a new transmitter temporarily .