Information needed urgently

A

Thread Starter

ali_qazi

Hi,

I recently got interested in process control. I have a question, can we measure a pressure drop in pipe by using two separate transducers rather then single differential pressure transducer? e.g. a water flown pipe has two pr transducers one at botom and one at top exit giving two pr readings. Pipe exits in tank open to atmosphere.

Is it right by taking readings of two pr transducers (psig) converting it into psia and then taking difference as pressure drop? If not how to caculate the pressure drop with two transducer readings?

Reply will be appreciated.
 
ali,
The basic problem is the difference between two large numbers generates a large error.

The use of a differential transmitter reduces this error.

Dennis
 
R

Robert Scott

As Dennis said, using two transducers is prone to larger errors because you have to subtract two fairly large numbers. However there are circumstances that can make this error acceptable. If both the transducers are gage pressure (with respect to ambient) rather than absolute pressure, then there is a chance that the two readings will be of the same order of magnitude as the difference between them. The problem comes if you are expecting a very small pressure drop but both pressures are relatively high. Can you provide some numbers? Also, what is your reason for wanting to avoid differential transducers? I can see if the measurement ports are far apart, you might not want to run air lines between them, and running wire might be simpler. Is that it?

Robert Scott
Real-Time Specialties
Embedded Systems Consulting
 
Thanks Dennis and Robert for your reply.

Dennis u have put words to what I had in mind ……… large errors, this might be acceptable on paper calculations with a word of caution however from practical point of view I know using two transducers is not the recommended way.

Robert, Yes u are right ……… Both the transducers are reading gauge pressures w.r.t ambient; one at the bottom of pipe and other at top position before the flow exits in tank open to atmosphere, distance between two transducers is 10m (quite far a apart).

The data is from vertical pipe upflow, with single phase (water) and two phase (air-water) flows. The issue is, knowing the contribution of hydrostatic and frictional pressure losses. Besides these pressure transducers readings there is differential pr transducer reading available from the middle position of pipe (to roughly calculate the gas fraction).

The numbers from two transducers (bar g) are:

Single phase Bottom: 1.09472 Top: 0.20213

Two phase Bottom: 0.46310 Top: 0.08694

Thanks again for the reply.

Ali
 
R

Robert Scott

Ah, I see the problem. You want to know pressure drop due to flow in the pipe. If you used a differential pressure transducer you would get that pressure drop directly. But if you subtract the readings from two gauge transducers that are 10 m. apart, you get an error due to the static head of 10 m. of water. For example, suppose you are using the two-gauge method and there is zero flow. The gauge at the bottom is going to read about 0.9 bar more than the gauge at the top just due to the static head. How does the differential pressure transducer avoid this error? Because the differential pressure transducer is fed with lines full of water, and the static head in those lines balances out the static head in the pipe. So the differential pressure transducer reads only the pressure drop above and beyond what is due to static head.

But this brings up another problem. You said that sometimes the pipe is full of a water/air mixture. In this case the static head in the pipe is different than the all-water case. If your lines to the differential pressure transducer are still full of water, then the static head in those lines will no longer accurately balance out the static head in the flow pipe, and the differential pressure transducer will produce a large error. This is a fundamental problem for either method (2-gauge or differential). There is no way to tell from pressures alone what the drop is due to flow because pressure difference due to static head appears just the same. The only reason the water-only method worked with the differential gauge is that you allow the water to enter the monitoring lines and create a static head offset to match that of the flow pipe. If you want that method to work when the fluid density is different, you have to ensure that the monitoring lines are filled with the same density fluid. Similarly, you could make the 2-gauge method work if you brought into the calculation some additional information that told what the density of the fluid was. Then you could subtract out the static head. There does not appear to be an easy solution to what you are trying to do.

Robert Scott
Real-Time Specialties
Embedded Systems Consulting
 
Top