DP Type Level Measurement

Best for what purpose? Surely it depends upon application?

Otherwise, there would only be one way of doing it.

DP Level transmitter, dual head DP or flange mount? The DP transmitters are physically the same, they're mounted differently, but the differences should be taken into account:

1) Is there a flange to mount to?
For retrofits on existing tanks, the question becomes what kind of pressure tap is available? The issues involved in emptying a tank, cleaning or purging it for welding on flange, clean-up and recommissioning can be significant and costly, so most people are inclined to use what they've got. If there's a threaded connection down low, it's likely to be used because it's there and it's available.

2) Isolation valving
A DP transmitter can be readily valved off (2 way instrument valve in the tubing/piping) for removal/cleaning/whatever

A hot tap ball valve for removal of a flange mounted DP level transmitter is a much more expensive proposition.

3) Difference in component cost

A DP from one major manufacturer lists at $1,879 USD.
- Add a valve, tubing and fittings.

A flange mount DP level transmitter from the same manufacturer lists at $2,342 diaphragm flush to flange), a 4" extension is a $272 adder.

- isolation valve?

4) freeze protection

Depending on the medium, its process temperature in the tank and the climate, freeze protection might be necessary. Exposing the contents of a dead-headed impulse tube to outside-the-tank elements is far more likely to expose the contents to freezing temperatures (depending . . ), whereas a flange mounted DP level transmitter that locates its diaphragm at or near the tank wall (frequently through the lag of tank insulation) does not directly expose the medium to the outside-the-tank weather or climate temperatures.

5) zero error

A welded flange won't move up or down in elevation so the datum, the transmitter's zero reference point, will not change.However, DP level transmitters connected with tubing/piping can change, even if inadvertently.

For example, the tubing gets damaged and during the repair the transmitter get remounted at a different elevation than it was originally, shifting the zero.

Determining an exact elevation with respect to the tank is many times, guess-work. I have seen a tech get a rotating laser, set it up and get an exact point on the tank for the port elevation, but not always.

6) Temperature error

A flange mount DP level transmitter has fill fluid the diaphragm at the end of the extension tube and the sensor element. That's a closed system and is subject to temperature error. Much of that error on typical 4" extension tubes is compensated for in modern transmitters that measure body temperature if pressure body is the same as or to the fill fluid temperature.

Impulse tubing to a dual head DP should have no temperature error because the tubing is open to the tank. Any medium expansion/contraction due to temperature changes will be 'seen' as specific gravity changes, which is an inherent issue in any DP level measurement.

A diaphragm seal connected with a capillary will have a definitive temperature error. I mention this because I've seen it used numerous times.

7) Potential for damage resulting in leaks

The opinion of those I've dealt with is that flange mounted DP level transmitters are less subject to damage that could result in a leak than are tube connected DP.

Historically, tank level has been done with flanged mounted transmitters. Tube-connected DP is the exception, not the rule, in my experience.

When I had a discussion with one engineer on this topic, his view was that the $400 saved by installing a DP transmitter, rather than a flange mount, was outweighed by the increased risk of tank leakage due to damage to components outside the tank.