Simple Engineering Calculators

Tank Spillover — Gooseneck Cascade Flow

Steady-state flow through the gooseneck (inverted-U) spillover pipes that cascade produced water between oilfield disposal tanks — two parallel trains, T1→T3 and T2→T4. Each pipe is the smaller of a weir limit at the crest and a full Darcy–Weisbach + Colebrook–Whiteenergy balance; the floating oil pad adds (lighter) head that still helps push water over the crest. A bisection solver finds the source level that passes exactly the inlet flow, flags overflow when even a full tank can’t, and reports settling and storage-to-alarm times alongside a live schematic.

Cascade schematic

TRAIN 1TRAIN 2vent26T127.6 ft927 bblHLA 28HLSD 31T320.0 ft672 bblHLA 28HLSD 31vent26T229.6 ft994 bblHLA 28HLSD 31T420.0 ft672 bblHLA 28HLSD 313000 m³/d3000 m³/d3000 m³/d3000 m³/dpump outpump out08162432

Train 1 — T1 → T3

Source level
27.6ft

927 bbl (water + oil pad)

Cascade flow
3,000m³/d

Capacity 11,345 m³/d · weir (crest-limited)

Head above crest
0.58ft

Effective (oil-corrected) head over the crest

Driving head
6.58ft

Total differential driving the flow

Velocity
3.80ft/s

Re = 233,874

Settling time
1.0hrs

Water retention at this inlet

Storage to HLA
0.5hrs

Source + downstream tank, to high-level alarm

Storage to HLSD
0.7hrs

…to high-level shutdown

Train 2 — T2 → T4

Source level
29.6ft

994 bbl (water + oil pad)

Cascade flow
3,000m³/d

Capacity 10,246 m³/d · weir (crest-limited)

Head above crest
0.58ft

Effective (oil-corrected) head over the crest

Driving head
6.58ft

Total differential driving the flow

Velocity
3.80ft/s

Re = 233,865

Settling time
0.8hrs

Water retention at this inlet

Storage to HLA
0.5hrs

Source + downstream tank, to high-level alarm

Storage to HLSD
0.6hrs

…to high-level shutdown

Equivalent length

Straight pipe
50.0ft

Up-leg + down-leg + connecting run

Fitting eq. length
149.1ft

ΣK = 3.00, as ΣK·D/f

Friction factor f
0.0134

Colebrook–White at the duty point

Total equivalent length
199.1ft

Straight + fittings

How the numbers are made & assumptions

Flow model — min(weir, Darcy)

Two limits are computed and the smaller governs. Weir (crest-limited): when the effective head above the crest is below the pipe diameter the pipe runs partially full at the crest and behaves like a weir, Q = Cd·Aseg(h)·√(2gh) with Cd = 0.62. Darcy (energy-limited): a full Darcy–Weisbach energy balance from source to receiving tank using the driving head, pipe friction (Colebrook–White) and the minor (fitting) losses. Low downstream level ⇒ weir regime; high downstream level ⇒ back-pressure regime; head above crest past the diameter ⇒ full pipe.

Oil-pad head correction

Oil floats and is skimmed separately, so only water goes through the pipe — but the oil pad still presses down. Because it’s lighter (SGoil/SGwater< 1) it adds head at a reduced rate: effective head above the crest = (water − crest) + (SGoil/SGwater)·oil pad. That lets flow start even when the water surface sits below the crest.

Steady-state solver & overflow

A bisection search finds the source-tank water level whose cascade flow equals the inlet flow (mass balance). If even a full tank can’t pass the inlet, the train overflows and the reported cascade flow drops to the actual full-tank capacity.

Settling & storage times

Settling = water volume in the source tank ÷ inlet flow (retention for oil/water separation, water only). Storage to HLA / HLSD = the spare volume from the current levels up to the alarm / shutdown setpoint, across the source anddownstream tanks of the train, ÷ inlet flow.

The math runs internally in US units (ft, ft³/s); flow is shown in m³/d to match the field convention. Pipe IDs come from standard NPS/schedule tables.

Read before you rely on these numbers

  • These are steady-state estimates for planning — verify against a detailed hydraulic model before procurement.
  • The weir coefficient (Cd = 0.62) is a sharp-crested approximation; a real gooseneck crest may differ.
  • Fitting K-values are nominal; foul/scaled pipe and partial-vacuum siphon effects aren’t modelled.
  • Oil/water separation quality drives the real settling requirement — the settling time here is retention only.