Water hammer, air entrapment and valve selection in pressure pipelines
Pressure pipelines (rising mains / force mains) and pumping stations have their own family of problems: capacity that sags, pressure surges after a pump trip or the closing of a valve, sub-atmospheric pressure that threatens older pipelines, pipelines that drain, valves that slam. The causes are linked — and the solution almost always lies in the combination of check-valve selection and air release and air admission.
Capacity that is never reached
Air collects at the high points in the pipeline and narrows the effective bore. The result: higher resistance, pumps that demand more energy and a capacity that stays below design — with no visible fault. Air release valves that keep venting air even under pressure restore the bore and with it the capacity. The right type in the right position is decisive here — a wrongly chosen air valve can quietly hold a pipeline below its design capacity for years. A good air release valve continuously vents air under operating pressure without admitting air locally at sub-atmospheric pressure — and is designed so that it does not slam shut after pump start.
Pressure surges and sub-atmospheric pressure
After a pump trip, pressure waves arise from the inertia of the water column: first sub-atmospheric pressure on the discharge side, then returning surges. Since variable-frequency ramp-up and ramp-down of pumps became standard, much has been gained — the serious risk remains the spontaneous pump failure, for instance during a power outage. Older pressure pipelines tolerate the sub-atmospheric pressure in particular poorly. Vacuum breakers limit it: a correctly chosen vacuum breaker opens only at a pre-calculated sub-atmospheric pressure critical to the system — and in many installations is a more reliable, lower-maintenance alternative to a surge vessel. The air-admission setpoint — including the adjustable range of the closing spring — is always tuned with the end user to the application and fixed before the materials are produced. Where the surge vessel is dropped, space is also freed up in the station, for instance for an extra pump and thus transport capacity. Surge-protection provisions and the right valve selection absorb the peaks.
Valves closing within the line
Not every pressure wave comes from the pumps. Gate valves are not control valves: their closing profile is far from linear, so the rate of velocity change (dv/dt) rises steeply just before full closure. The pressure rise is moreover highest at the location where the velocity change arises — limitation therefore belongs at the source, not at a distance in the station. Solution directions: close the final part of the travel very slowly, a valve with a linear closing characteristic, an over-pressure surge limiter at the valve concerned — and in practice above all: a check valve that is already closed at Q⁺ = 0, that is, before the local flow reversal at the valve.
Siphoning after pump trip
With a downward-sloping pipeline profile the flow can continue after a pump trip: the vacuum that forms in the pressure pipeline draws medium through the stopped pumps, and check valves open again while the installation is at rest. Air admission at the right point breaks the siphon; the check valve then holds the column.
Draining after pump trip
A pressure pipeline with a downward-sloping profile can drain after a pump trip. The next start then takes place in an empty line — pumps that cavitate, make noise and wear unnecessarily, and a filling phase that itself causes new surges. A back-pressure valve at a lower point keeps the pipeline filled after a pump trip, so that at start the pump works directly against a full column.
The check valve: closing before flow reversal
With a check valve a different dynamic is at play than the classic water hammer of a fast-closing valve. The valve must close while the flow is still just forward or around zero flow — before any appreciable return flow develops.
If the disc waits until there is return flow, it builds up speed and slams onto the seat. The surge then arises not from stopping the main flow, but from abruptly stopping that return flow at the moment the disc strikes the seat.
Two quantities determine the behaviour: the rate at which the flow decelerates, dv/dt, relative to the closing characteristic of the valve, and the return time of the pressure wave, Tr = 2L/a. The valve must be closed before the first pressure reversal within that return time.
For influent and effluent water EURAD selects check valves that travel freely for the greater part of their stroke and only cover the last 5–10° before the seat in a damped manner, via a dashpot. In this way two normally contradictory requirements go together: fast response to falling flow (little return flow) and a low impact velocity on the seat (no valve slam). Damping over the whole closing movement would instead make the valve too slow and give more return flow.
In practice that damping is seldom active. The valve design does the work: full bore already at a small opening angle and a favourable flow characteristic — among other things through a body that widens immediately after the seat. A small opening angle is only favourable if it does not come with permanently high resistance; otherwise the pumps are continuously pumping against it. Thus the valve closes calmly and in good time as flow falls, closed at Q⁺ = 0 — before the local flow reversal. The dashpot is there for the exception, not the rule.
The lever with counterweight keeps operation calm and supports the valve in closing. EURAD works with empirically tested resistance (head loss, in mwc) and opening angle (in degrees), plotted against the volumetric flow — including the dead weight of valve disc and lever. Where no fast return flow threatens, the disc can be allowed to open a few degrees further by choosing a lower counterweight, which leads to a lower energy uptake at the pump shaft.
The design also depends on how fouled the medium is. For influent water (sewage) a check valve is chosen with the main shaft positioned out of the flow, so that stubborn fouling — such as fibres from wet wipes — does not hinder operation. For effluent water (cleaner) the shaft can run in the flow, provided it is a slanted-seat design, so that the dynamic advantages described above are retained.
For large pumping installations the measure is therefore not the total closing time, but the response time, the closing angle at flow reversal and the impact velocity on the seat. A valve can close in half a second and still be virtually water-hammer-free — what is decisive is not how fast it closes, but that it is already almost on the seat before flow reversal and covers the last millimetres in a damped manner.
Selection per situation
Pipeline profile, pump behaviour, medium and certification requirements together determine the choice — from closing characteristic and damping to the type and placement of air release and air admission valves. EURAD advises and supplies on the basis of the application. A seemingly simple design can in fact contain a great deal of intelligence: easier maintenance, more reliable operation, simpler adjustability. The common thread: autonomously operating valves that — designed from fluid dynamics — respond to what happens in the pipeline, and keep doing so even during a power outage, exactly when the protection is needed most.
