A water tube level marks points at the same height by filling a clear plastic tube 10–20 m long with water, holding both ends up and noting where the water stands at each end, since water finds its own level. Used leapfrog along a route it can set a 0.2% grade of 10 cm per 50 m, but each setup adds a small reading error: at ±3 mm per setup, 50 setups can stack to 150 mm in the worst case and about 21 mm typically, so close the survey back on its start to check.
Part of Aqueducts and Channels
The water tube level is a long clear tube partly filled with water, used to find points of equal height over distances too long for a plumb-line level. It is cheap, needs no skill and is accurate enough for channel grades if used with care.
Water At Both Ends Of One Open Tube Stands At The Same Height
Open to the air at both ends, a connected body of water sits at one level. Hold the two ends of a tube at different places and the water surface in each stands at the same height above sea level, however the tube snakes between them. The tube does not have to be straight or level; only the two ends matter.
That makes it the tool for long distances, where the A-frame (A-Frame Level) must step along a pace at a time. The method on the aqueducts page is to mark level points along the route, then add the calculated drop: lower each stake by the drop per interval compared to the previous one.
What You Need
- Clear plastic tube, 10-20 m long, of even bore
- Water, with a few drops of dye or ink so the surface is easy to see
- Stakes at each station, and a ruler or marked stick
- A funnel for filling
- A partner for each end, and a notebook
Method: Fill, Purge Bubbles, Mark, Move
Step 1: Fill the tube slowly from one end with a funnel, holding the other end up, until water stands a hand’s width up in both ends.
Step 2: Remove every air bubble. Hold the tube up to the light and run your fingers along it, pushing bubbles out of the ends. A bubble breaks the connected column, and the two ends then no longer agree.
Step 3: Hold each end against a stake and let the water settle until the surface is still. Read or mark where it stands at each end, using the bottom of the curved surface at both ends.
Step 4: Mark both stakes at the water level. The two marks are the same height.
Step 5: Leapfrog. Keep the rear end on its stake, carry the front end forward to the next station, and repeat.
Step 6: Convert to grade. For a 0.2% grade, each stake sits 0.002 × distance below the previous one: 10 cm per 50 m. Stations should be no further apart than the tube reaches, so surveyed stations every 20 m or less are interpolated to the 50 m pegs.
Worked Example
1 km of channel at 0.2% needs a total drop of 1,000 × 0.002 = 2.0 m (Gradient Calculation covers grade arithmetic).
Suppose each setup is read to ±3 mm (an assumed figure for a careful team), and a 20 m tube means 1,000 ÷ 20 = 50 setups per km.
- Worst case, all errors the same way: 50 × 3 = 150 mm, which is 7.5% of the 2.0 m drop.
- Typical, random errors adding as the square root of the count: 3 × √50 = 3 × 7.07 = 21 mm, about 1% of the drop.
To check, survey back to the start by a different route. The difference between where you arrive and where you began shows the total error; if it is larger than you can accept, repeat the weak sections.
Common Mistakes
- A bubble left in the tube. The ends read differently by an unknown amount.
- A kink or a pinch. The water cannot level and gives a false reading.
- One end capped or squeezed. It holds water above its level.
- Sunlight on the tube. Warm and cold water have slightly different densities, and warm water can form bubbles.
- Reading from different sides of the surface at each end. Use the same part of the surface at both ends.
- Never closing the loop. Errors go unnoticed.
What’s Next
- Learn next: Optimal Gradient — choosing the grade to set
- Combine with: Channel Cross Section — what to dig once the line is staked