What kind of problems can an orifice plate have when using natural gas?

[daniel27ad]

Hello, I'm noob around here, hope I'm posting in the right part of the forum.

This client has a natural gas pipeline with funny operations. Flows constantly spike to high values (say from 0 to 3000 MSCFD), stay there for less than 1 minute, and then falls back.

I'm thinking that there might be a problem with the metering (client says everything's fine though), but I haven't found any specific problem related to the use of orifice plates on gas pipelines that may give me a clue.

* I'm way far from the actual location of the pipeline, so I cannot check problems on site.

Thanks for your help.

 

[danw2]

Short duration events like this are most likely not real process events.

To look into this, the exact data path needs to be established. Is it an analog DP transmitter to flow computer (with line pressure and temperature compensation) that talks a digital protocol to a DCS with an OPC server that dumps into an OPC client database?

Any device, a flowmeter, flow computer, other inputs to a flow computer (temp or pressure) wireless gateways or data acquisition can have a fail-safe mode, where signal goes full upscale or full downscale if the meter or a component detects a failure.

The chart is not necessarily scaled to the exact flow element/flow meter range, so a fault indication could conceiveably max out at the peak value shown on a data analysis graph.

It isn't clear whether multiple readings are acquired during these peak-faults, or whether the peak-faults are single point readings. Single point blips are difficult to track because of the instantaneous nature.

Another factor can be noise induced on a periodic basis on top of non-faulted real signals. The twice weekly start-up of a 1,000HP pump motor used to spike several signals in the area that were traced to single ended analog input card whose ground reference was earth ground. Apparently the earth ground potential shifted during the motor startup enough to drive low level DC signals off scale.

 

[hacksaw]

ditto the sbove assessment, you need to look at what is happening at the station equipment wise