DP Flowmeter for Steam Application

[rossuwahab]

Hi,

I have one venturi flowmeter for steam application, in which currently the reading shows abnormal trend whereby reading sometimes is high (not tally to the process), but it'll go back to normal after a while. I have verified the flowmeter, replace the condensate pot, replace the safety barrier but still the error is there. The one thing i noticed is that the impulse line has no insulation, will this be the root cause to my problem?

Thanks, appreciate your help.

 

[hacksaw]

What suggests that the process conditions are not changing?

 

[danw2]

Root cause impulse line not insulated? No.

For a high DP reading (which is not a real process change) there has to be
- something that makes the low side reading lower than it should be so that when the low side gets subtracted from the high side, the remainder is higher than it should be
- something that makes the high side higher than it should be

In steam service with condensate filled impulse legs, making the high side higher than it should be is not likely, particularly for an intermittent change (it's not like gas service when condensate can produce an offset pressure on one side or the other).

Things to consider:
- blockage in the low side impulse line (scale) or port to which it connects or in a root valve, or physical damage like a crimp in the tubing.
- leak on the low side, like a fitting on the condensate pot, plug/vent/bleed on the transmitter, plug/vent on the low side of a 3 or 5 valve manifold,
- loss of liquid head in the low side impulse tube, or if the condensate pot were to empty
- trapped air in the transmitter itself, in the space between the sensing disphragm and the process head, which can happen when the ports are horizontal (or vertical facing down) and the process heads have not been vented.

You mentioned uninsulated impulse tubing. Is either condensate pot insulated?

 

[roydm]

Shouldn't you have 2 condensate pots?
I have seen similar where the level in one impulse line dropped as the condensate flashed off, remember it only takes a fraction of an inch to make a big zero shift.

 

[Niwot]

A photo of the installation would be helpful. Can you let us know the % of error you're seeing and over what scale (i.e. 25% of flow, 50%, 100%?). The lower the flow, the larger the accuracy impact on imbalanced water legs. Is your pressure and temperature constant or are you compensating for changes in density (pressure and temperature measurement)?

All posters make good points, but I go back to danW2's comment about a low side blockage. It is tricky because a blockage can trap a static pressure so it appears that you are reading a "live" DP when the velocity changes are only being sensed by the high side. "Normal" flow could be indicated only when the trapped pressure is equivalent to what it should be sensing at a particular velocity/flow rate. The DP transmitter doesn't know any better.

If this is a start/stop service, water in the fill legs could be suctioned out unevenly when the steam is shut off - the steam collapses and creates a vacuum that can pull more water from the low side impulse line. I have seen cases with DP where steam flow is registering positive even when the line is shut down (no flow), indicative of imbalanced water legs. Are you still registering a positive flow when the line is shut down?