Why closed tanks are trickier than open ones
In a closed, pressurised tank the vapour space sits on top of the liquid, pushing down on it. A single pressure tapping would read liquid head plus vapour pressure — useless for level. The fix is a DP transmitter with its low-pressure side connected to the vapour space, so the vapour pressure cancels and only liquid head remains:
The complication is what fills that low-pressure impulse line. If the vapour is dry and non-condensing, the line stays empty — a dry leg, and the math is just like an open tank. If the vapour can condense (steam, most process gases), liquid collects in the line unpredictably, so it is deliberately filled and kept full — a wet leg.
The wet leg and zero elevation
A full wet leg of height Hwl and fill SG puts a constant head on the low side, even at zero tank level. That makes the differential negative at empty, so the range is shifted down — zero elevation:
Note the span (URV − LRV = SG·g·H) is identical to the open-tank case — only the zero moves. This is the opposite of suppression, and confusing the two is the single most common DP-level configuration error.
Worked example
Closed tank, process SG 0.9, 6 m level span, wet leg 7 m tall filled with condensate (SG 1.0):
- LRV = −(1.0 × 9.807 × 7) = −68.65 kPa
- URV = 0.9 × 9.807 × 6 − 68.65 = 52.96 − 68.65 = −15.69 kPa
- Span = 52.96 kPa — same as an open tank of the same liquid and height
- The transmitter must be configured with a fully negative, elevated range
Field notes
- Keep the wet leg full. A wet leg that boils off or drains reads high — one of the classic causes of a "creeping" level reading. Fill ports and periodic checks matter.
- Condensate SG isn't always 1.0. Glycol/water seal fluids and hot condensate differ; use the actual fill density.
- Remote seals eliminate the leg problem by using capillary-filled diaphragm seals on both sides — at the cost of temperature-dependent errors of their own.
- For an open (atmospheric) tank, use the open tank calculator instead.
Frequently asked questions
What is the difference between a dry leg and a wet leg?
On a closed (pressurised) tank the low-pressure side of the DP transmitter connects to the vapour space. A dry leg keeps that impulse line empty (gas only); a wet leg deliberately fills it with liquid (often a seal fluid) when the vapour would condense. A wet leg adds a constant negative bias that must be elevated out.
What is zero elevation?
When a wet leg puts more head on the low-pressure side than the tank puts on the high side at zero level, the transmitter sees a negative differential at empty. Zero elevation shifts the range down (a negative LRV) so 4 mA still corresponds to empty.
Why is the wet-leg span the same as an open tank?
The wet leg adds a constant offset to both LRV and URV, so the span (URV − LRV) is unchanged — it is still SG × g × H. Only the zero point moves.
Provided for reference and education. Verify independently before use in safety-critical work. See our disclaimer.