Francis Turbine Cavitation: Quantifying Wear Cost and Optimizing the Operating Point

In a Francis turbine, cavitation begins when local pressure drops low enough for vapor bubbles to form inside the water passage. As those bubbles collapse again in higher-pressure regions near the runner outlet, trailing edge, crown, band, or draft tube, they generate intense micro-impacts on the metal surface. The visible result is pitting, roughness, coating loss, and later weld repair. The less visible result is a machine that is quietly losing economic value before the maintenance order is raised.

For operations teams, the important point is accumulation. A few hours in a mildly aggressive zone may be acceptable. A few hundred hours in the same zone shorten inspection intervals, accelerate surface deterioration, and move the next outage forward. That means cavitation is not only a hydraulic phenomenon. It changes the true cost of every dispatch decision made by the plant or trading desk.

The technical reference is the cavitation number sigma. On site, operators compare the available sigma margin with the sigma required at the actual operating point. Because sigma risk changes with net head and therefore with the specific hydraulic energy Hn, the same MW target is not equally safe at every hydraulic condition. A point that looked acceptable during one head regime can become costly in another.

A Francis hillchart should be read as more than an efficiency document. In normalized coordinates such as Q11 and n11, it shows how the operating point moves as head and discharge change. Once sigma bands, cavitation observations, and admissible envelopes are layered onto that chart, the hillchart becomes a practical map of wear exposure rather than a static test-sheet.

Near the BEP, internal flow is usually cleanest and the hydraulic loading is most balanced. Move far below BEP and part-load effects such as draft-tube swirl, vortex rope, and pressure pulsation become more likely. Move too far above BEP and overload cavitation may appear near the runner outlet where local pressure recovery worsens. In both directions, the plant may still generate, but it is now buying revenue with higher mechanical risk.

That risk is not only hydraulic. Off-BEP operation increases fluctuating hydraulic forces on the runner, guide apparatus, and adjacent structure. These oscillatory loads increase alternating von Mises stress in critical regions and consume fatigue cycles more quickly than a point close to the center of the admissible island. For that reason, a hillchart cavitation zone should be understood as a combined hydraulic and structural wear zone.

The most useful wear model starts from maintenance reality. Suppose a runner that would normally require cavitation repair every 30,000 operating hours is now expected to need repair every 18,000 hours because dispatch repeatedly enters aggressive cavitation bands. The economic consequence is not abstract. It includes earlier weld build-up, machining, coating, NDT, mobilization, crane time, and the value of lost generation during the outage.

From there, the plant can convert damage into a wear rate in EUR/hour. If the accelerated maintenance cycle is expected to add EUR 240,000 of maintenance and outage cost over a season and the unit spends 2,400 hours in the corresponding damage zone, the wear penalty is already EUR 100/h before counting the efficiency loss caused by a roughened runner surface. That is large enough to change a dispatch decision that looked attractive on gross market revenue.

The model can be made more granular with green, yellow, orange, and red operating zones weighted by sigma margin, BEP distance, vibration, or inspection history. This is the point where maintenance optimization becomes operational optimization. Once wear is priced in the same unit as market value, the plant can compare the extra MWh against a real damage penalty rather than leaving the maintenance budget to absorb it later.

A spot-price-only optimizer assumes the machine is indifferent to how it earns the next megawatt-hour. A Francis turbine is not indifferent. Two operating points can produce similar hourly revenue while imposing very different cavitation exposure, pressure pulsation, and fatigue loading. If wear is ignored, the optimizer will often chase the most aggressive output whenever the market spikes.

Consider a simple case. A one-hour dispatch move during a price peak adds EUR 180 of gross margin. If the same move increases expected cavitation and fatigue cost by EUR 120/h and adds another EUR 40/h of hydraulic deterioration cost, the net value is already close to zero. Repeated across a season, the plant has optimized price while consuming runner life.

This is why the best operating point for turbine cavitation cannot be defined by efficiency or spot price alone. It must include damage memory: sigma margin, time spent in aggressive hillchart zones, off-BEP structural loading, and the cumulative consumption of fatigue cycles. Without that, the algorithm is maximizing a partial objective.

DAMagedOpt is built around a combined objective. Instead of selecting the point that maximizes instantaneous revenue only, it evaluates the feasible operating points that maximize net value after expected wear cost is included. The platform combines hillchart position, BEP distance, sigma-related cavitation risk, Hn, SCADA signals, and plant constraints into one operating recommendation.

In practice, DAMagedOpt integrates a damage cost model directly into the dispatch optimizer. Candidate operating points are scored by both expected revenue uplift and expected wear penalty in EUR/hour. A point that looks attractive on price but sits in a poor cavitation zone can therefore be downgraded or rejected automatically. You can see this current-versus-recommended logic on the interactive demo.

That same logic also explains the business model. The pricing page shows the uplift-based structure: the commercial model follows measurable value creation rather than charging a fixed fee disconnected from plant results. For operators, O&M teams, and energy traders, that creates a shared language between revenue and mechanical protection.

If your Francis unit is still being dispatched with efficiency curves, operator habit, and spot-price logic handled in separate workflows, there is a good chance cavitation cost is being underpriced. A free uplift estimate is the fastest way to test where the plant is losing net value rather than just gross output.

Open the demo to compare the current operating point with a healthier recommendation, review the uplift-based pricing, and contact the team if you want a plant-specific discussion around sigma, BEP distance, hillchart cavitation zones, or hydraulic wear cost assumptions.