Stress Behaviour of Surface Penstocks in High-Head Hydropower Stations: A Practical Review
Overview of the Literature
This study by Cheng Zhenglong, Shi Changzheng, Hu Lei, and Xie Yinghan addresses a question that has cost real money and real schedule days on more than one project I have been involved in: how does a surface penstock actually behave under the combined action of gravity, internal water pressure, and temperature, and where does the design code leave room for judgement? The authors use three-dimensional finite element analysis on a high-head diversion-type station, and their conclusions are direct and useful. They confirm that for high-head stations, the stress in the penstock and its support rings under normal operating conditions is dominated by internal water pressure, and the magnitudes are significant. They also confirm that the provision of expansion joints and sliding supports allows axial free deformation, so that uniform temperature action produces almost no thermal stress. The critical warning, and the part that I would underline in any design review, is that under non-uniform temperature action the pipe will develop noticeable transverse bending deformation, so the supports must permit some transverse sliding and be fitted with transverse restraint devices to prevent transverse thrust and pipe slippage. The authors specifically flag strong solar radiation zones as a risk area.
Interpretation of Technical Points
Load Case Dominance
The finding that internal water pressure dominates the stress state is not a surprise to anyone who has sized a high-head penstock, but the study puts a number on it. For a high-head station, the hoop stress from internal pressure scales directly with the head, while gravity and temperature effects scale with pipe diameter, wall thickness, and span. When the head is high, the pressure term wins by a wide margin. What this means in practice is that the wall thickness and the support ring design are governed almost entirely by the pressure case, and the other loads are secondary in the strength check. However, secondary is not negligible. The support ring is a local discontinuity, and the stress concentration factor there must be evaluated carefully, because the pressure-induced stress is already high. A poorly detailed support ring on a high-head penstock is a crack waiting to happen.
The Role of Expansion Joints and Sliding Supports
The study's confirmation that uniform temperature produces almost no axial stress when expansion joints and sliding supports are properly provided is a classic example of a design principle that is easy to state and hard to execute. The expansion joint must be located and sized to absorb the full expected axial movement, and the sliding supports must have a friction coefficient low enough that the residual axial force does not overstress the anchor blocks or the expansion joint bellows. In my experience, the failures in this area are rarely in the calculation; they are in the details. The sliding surface gets corroded or contaminated, the friction coefficient doubles, and the anchor block or the expansion joint takes a load it was never designed for. I have seen a penstock anchor block crack because a sliding support was installed with the wrong bearing material and the friction force was three times the design value.
Non-Uniform Temperature and Transverse Bending
The most valuable contribution of this paper is the warning about non-uniform temperature action. When one side of the pipe is heated by the sun and the other side is in shade, a temperature gradient develops through the wall and around the circumference. This gradient produces a bending moment that tries to curve the pipe in the transverse plane. If the supports are rigid in the transverse direction, the pipe will either develop a large transverse reaction force on the supports or, if the support has insufficient hold-down, it will lift and slide off. The authors recommend allowing transverse sliding and providing transverse restraint devices. This is exactly right, and it is a detail that is often omitted in the design of penstocks in tropical or high-altitude sites where solar radiation is intense. I have seen a penstock in a high-altitude site in South America where the transverse restraint was a simple angle bracket, and the pipe had worked itself sideways by several millimetres over a year of operation, enough to cause a visible misalignment at the expansion joint. The fix was to add a proper transverse restraint with a sliding pad and a stop.
Integration with Engineering Practice
Design Review Checklist
Based on this study and my own experience, I would add the following items to any penstock design review:
| Check Item | Why It Matters | Typical Pitfall |
|---|---|---|
| Support ring stress concentration factor | Pressure stress is high; local discontinuity can push stress above yield | Using a simple beam formula for the ring instead of a shell analysis |
| Expansion joint travel and location | Must absorb full axial movement from temperature and anchor block displacement | Joint travel undersized for the actual temperature range at site |
| Sliding support friction coefficient | Determines residual axial force on anchors and expansion joint | Bearing material not specified for the site environment |
| Transverse restraint design | Prevents pipe slippage under non-uniform temperature | Restraint omitted or too weak for the expected transverse force |
| Solar radiation shading | Reduces non-uniform temperature gradient | No consideration of site orientation and sun path |
A Real Project Lesson
I recall a project in the Middle East where a high-head penstock was designed with sliding supports but no transverse restraint, on the assumption that the transverse force would be small. The site had intense solar radiation, and the pipe was a large-diameter, relatively thin-walled design. Within the first year of operation, the pipe had moved transversely by enough to cause a visible gap at the expansion joint flange. The owner was concerned about leakage, and we had to shut down the unit, re-align the pipe, and install transverse restraints. The cost was significant, and the schedule impact was worse. The lesson is simple: the transverse restraint is not an optional detail; it is a required component for any penstock in a high-radiation zone, and it must be designed for the actual transverse force that the non-uniform temperature gradient can produce.
Key Questions and Reflections
The study raises one question that I would like to see addressed in future work: what is the actual magnitude of the non-uniform temperature gradient in a real penstock under worst-case solar loading, and how does it vary with pipe diameter, wall thickness, and surface coating? The finite element model in this study uses a prescribed temperature field, but the real gradient depends on the solar absorptivity of the coating, the wind speed, and the thermal mass of the pipe. A dark, uncoated pipe in still air will have a much larger gradient than a light-coloured, insulated pipe in a breeze. I would also question the assumption of a linear gradient through the wall. In reality, the gradient is nonlinear, and the bending stress distribution is not linear either. A more refined model would use a coupled thermal-structural analysis with a realistic solar load.
Study Insights and Implications
The practical takeaway from this paper is that the design of a high-head surface penstock is governed by internal pressure, but the details that prevent failure are the ones that control temperature-induced deformation. The expansion joint, the sliding supports, and the transverse restraints are not afterthoughts; they are critical components that must be designed with the same rigour as the pressure-containing shell. For the engineer on the delivery side, this means that the fabrication and installation drawings must clearly show the support details, the expansion joint installation procedure, and the transverse restraint configuration. I have seen too many projects where the penstock shell is fabricated to a high standard, but the supports are installed incorrectly because the details were not clear enough on the drawings. The result is a pipe that is structurally sound but mechanically unstable.
Reference Value and Outlook
This study is a useful reference for the design of high-head surface penstocks, particularly for the treatment of temperature effects. The authors' recommendation to allow transverse sliding and provide transverse restraints is a sound engineering principle that should be incorporated into design guidelines. For future work, I would like to see a parametric study that covers a range of pipe diameters, wall thicknesses, and support configurations, so that the transverse force can be estimated from a simple chart rather than requiring a full three-dimensional finite element analysis for every project. That would be a practical contribution that design engineers could use directly. In the meantime, the lesson from this paper and from the field is clear: when you are designing a penstock in a sunny climate, think about the sun, and give the pipe room to move sideways.
Zhuojin Pipe Fitting Co., Ltd