How to compensate DP for elevation

Accurate measurement of differential pressure is never as simple as it seems.

Q: We need to measure differential pressure between two pipes at different heights. As shown in Figure 1, the tapping point of the low-pressure pipe is about ground level, and that of high-pressure pipe is 1.5 m above. What factors we should consider? The medium is water (liquid). I would like to know the best position/ place to install the differential transmitter. In my opinion, it should be installed on the low pipe tapping, as low pressure cannot rise against gravity to the transmitter.

Muneer Bhutta, WorleyParsons engineering


A: This is a very useful question because it gives me the opportunity to point out some of the general considerations that one should keep in mind when making differential pressure (DP) measurements. While the question deals with measuring the pressure difference between a high-pressure water line 1.5 meters above a low-pressure one, the answer will be applicable to any DP measurement between points at different elevations, no matter what the pipes carry.

To eliminate the effect of the quality of the DP transmitter used in the five configurations I will discuss, I will assume that in all cases (A to E in Figure 2), the same quality transmitter is used—0.1%  FS accuracy, smart, digital units, with zero and span settings accurately calibrated.

For each configuration, I will consider probable performance, cost, purge gas and maintenance requirements. Because the line pressures and the expected pressure difference were not given, I am assuming the high-pressure line is at 100 PSIG or less and the DP is 10 PSID or less.

Naturally, on other applications, these numbers can be anything else, but that doesn’t change the logic of the evaluation, only change the transmitter spans. However, the spans do affect the error (inaccuracy), because transmitter error is assumed to be 0.1% FS (% full span). In other words, if the pressures (P1 and P2 in option A) or the DP (in all options B to E) rise, the errors will also rise and if they drop, the errors will also drop in proportion.

We should also keep in mind that options D and E can only be considered if the supply pressure of the nitrogen (or other utility gas) exceeds the line pressures in the pipes.

The data in Table 1 is debatable and can vary with installation (i.e., material can settle in the wet legs), still, if we accept them, I would recommend the selection of the chemical sealed configuration (option C). Make the two capillaries the same lengths and locate them so they will always be at the same temperatures (no solar exposure).

If cost is the primary consideration, I would use option B and would make sure that the DP cell is located below the lower pressure tap so gas bubbles will rise, and insulate both legs or make sure that they are at the same temperature.

Béla Lipták


 A: I am assuming that in your situation: 

1) The pipes not connected (not the inlet and outlet of an exchanger or similar piece of equipment).

2)  The specific gravities (SG) of the water streams do not vary because of temperature changes.

The current installation of the transmitter (Figure 1) will give you very poor performance because air will be trapped in the vertical lines. If you manage to bleed out all the air (and it stays out) it might be possible to obtain consistent readings, but the moment any air gets in these legs, the reading will be off again.

2017 State of Technology Report: control systems

In this particular application, the best location for the transmitter would be slightly below the low-pressure tap, with the low side tubing running either horizontally (or ideally slightly sloped up) to the tie in point on the pipe, so that air cannot get trapped anywhere in the tubing. The taps should come out the side of the transmitter rather than from below as they do now. The high-side tubing should be run so it cannot trap air either – it should come out the side of the transmitter and run either slightly sloped or vertical up to the tie-in point on the high-pressure pipe.

Also, I don’t know what these pipes are connected to. If they are tied to the same piece of equipment and everything is filled with water, then the zero offset should be minimal. However, if they are not, then you could have a zero offset of up to 1.5 m of water.  During calibration, you’ll need to zero the meter under “no flow” and “zero DP” conditions for the DP cell to read properly.

In addition, your reading may vary with temperature. Since the pipes are not insulated, I doubt that the temperatures are very high. If the temperature doesn’t change much, the reading will be stable, however, if it rises, you’ll get an upward shift in readings due to the lower SG of the water in the system.

Hunter Vegas


A: You can use a remote diaphragm seal-type DP transmitter, which will eliminate errors due to liquid head in this installation with impulse lines. The pressure taps should be 3-in. on the bottom of the pipe and the diaphragm size should also be a minimum of 3-in.. Also, ensure that the capillaries are of identical lengths and that the transmitter and flange assemblies meet the design pressure and temperature requirements of the pipe lines.

Debasis Guha


A: To be able to measure the required pressure difference, you need to measure the two pressures and the specific gravity of the water in the impulse line at operating temperature.

1. The ideal location for the DP sensor is centered around the bottom connection.

2. The sensor should (if possible) be of the diaphragm type so that you do not have to constantly replace the sealing fluid.

Alejandro Varga


A: To learn more about elevation and suppression, see the tutorial dealing with calibration at

Jonas Berge


Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.


No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments