Water & Wastewater Treatment

50 | June 2014 | WWT | www.wwtonline.co.uk In the know Technically speaking: Reservoirs C ommencing in early 2013, and completed recently, a scheme to enlarge Black Esk reservoir, in southern Scotland, featured innovative use of precast piano-key (PK) weirs around the bellmouth sha rim. Scottish Water predicted a maximum deficit in the Black Esk water resource zone of 4.4 megalitres per day (Ml/day) for the planning period to 2031–32. Of the projects to address this, the enlargement of Black Esk reservoir was the largest. The increase in the storage volume by about 40% to 3,150 Ml increased the estimated yield to 19 Ml/d, which resolves the deficit and creates an estimated surplus of 1.1 Ml/d. The scheme increased the storage capacity of the reservoir by raising the full supply level, corresponding to the overflow weir crest level, by about 2.5 metres (m). An earthworks solution was recognised as providing a cost-effective means of heightening the embankment. The dam had originally been designed for raising symmetrically, raising the upstream and downstream shoulders by similar amounts about the centreline of the embankment. But the importance of maintaining water supply from Black Esk during this scheme precluded drawing down the reservoir fully to allow construction of a higher embankment on this principle. Through co-ordinated design, the amount of raising required to Focus on piano-key weirs A scheme to enlarge Black esk reservoir in southern Scotland featured innovative use of precast piano-key weirs around the bellmouth sha rim accommodate the probable maximum flood (PMF) and wave freeboard could be limited to 2.1 m, raising the dam could be restricted to the downstream shoulder and the need to include a wavewall – which had featured on the original dam – on the crest of the heightened dam could be avoided. Conventional labyrinth spillways have a large footprint, so are not well suited to implementation around the rim of a bellmouth. But recent research into PK weirs, in which the upstream and downstream ends of the zigzags are cantilevered beyond the foundation, led to the idea of adapting the concept to retrofitting PK weirs around the rim of a bellmouth. At Black Esk the new PK weirs pass the PMF of 183 m 3 /s with a flood surcharge of 0.97 m, saving 0.64 m from the amount of dam raising required had the bellmouth rim simply been raised. There are three flow modes for a bellmouth spillway, in order of increasing flows: • Control at the overflow weir crest, with the sha and tunnel running part-full • Orifice-type control at the throat of the bellmouth, with the overflow weir submerged and the bottom of the sha and tunnel running part-full • Tunnel-full control, in which the flow conditions are akin to a pipeline with a submerged bellmouth entrance running full. For low heads, weir control always occurs, with the rim of the bellmouth forming the overflow weir from the reservoir, and the sha and tunnel flowing part-full. For higher heads, control may first move to the throat of the sha and then tunnel-full control, or there may be direct transition from weir control to tunnel-full control. To check the higher part of the original rating, where the weir was drowned out by downstream flow conditions in the bellmouth sha and tunnel, analyses were undertaken of the two potential flow modes that would John Ackers TechnicAl DirecTor (hyDrAulicS) BlAck & VeATch cFD model of adopted full 24-cycle Pk weir www.wwtonline.co.uk | WWT | june 2014 | 51 bellmouth rim, on and around which temporary working platforms were deployed. ● A temporary crane platform adjacent to the bellmouth facilitated the extensive li ing required for removing the original debris boom and walkway, installation of the PK weir units, and the raising of the original valve tower. ● In order to maximise the weir length, the PK weir units were pushed to the outside edge of the bellmouth. This set the minimum inlet key width, which was chosen to allow a small section of in situ wall and base to be cast to stitch individual precast units together. The upstream overhang was limited to each be small enough to allow precasting, as well as offering a smaller external overhang which would be helpful during construc- tion. It was then realised that further simplification could be made by avoiding the handed pairs of units, making all 24 units identical, accommodating a 15° angle within the inward-projecting embayment and adjusting the concrete shoe arrange- ments around the original bellmouth rim to suit the 24-segment PK weir. ● The CFD model of the PK weir comprised only a repre- sentative 15° sector, extend- ing between the centres of successive inlet keys. CFD modelling of the PK weir did not include the sha and tunnel, largely because it would not be possible to achieve a sufficiently fine mesh in a model that included even half of the entire PK weir and sha , together with half of at least the upstream end of the tunnel. Instead, it was considered preferable to obtain an undrowned weir rating, then to apply an empirical treatment of the tunnel hydraulics (sup- ported by an original 1950s physical model test rating) and of the drowning of the PK weir. approximately half of the whole upstream/ downstream length to ensure the stability of each precast unit while being placed, transported, and stored on-site. ● Manufacturing the units in precast concrete allowed smaller bay widths than is feasible with in situ con- struction, as no scaffold- ing is required within the outlet key. Casting the PK weir units inverted allowed the casting of the rounded crest to be achieved with greater control than for insitu methods and also allowed thinner sidewall sections. The size of the units allowed two units to be carried on a standard articulated lorry without oversize load restrictions. cause this drowning: bellmouth throat control and tunnel flowing full. The analyses demonstrated that hydraulic control in the original spillway sha migrated, at a head of about 1.7m, from the ogee weir around the bell- mouth rim to tunnel full-flow control, without an intermediate region of throat (orifice-type) control. PK weirs have been developed over little more than a decade. At an early stage of design, a study tour was arranged to enable Scottish Water and the Black & Veatch design and construction team to see two PK weirs retrofitted at dams owned by Electricité de France near Limoges. The original ogee weir around the rim of the bellmouth sha comprised 12 straight segments, with angles of 30° between them. In order to suit this, the initial concept for the PK weir was based on accommodating a single weir cycle into each of the 12 segments, with each cycle of the PK weir positioned so that its outward-projecting bay (inlet key) was centred within the segment. The inward-projecting bays (outlet keys) straddled the 30° angles between the segments and were consequently truncated. In conclusion The graph (le) shows the routing of the PMF through the raised reservoir for two cases: a) the adopted design, with the 24-cycle PK weirs; b) an alternative design, based on raising the ogee weir crest, but simplifying the original debris boom arrangements. Hydrographs are given for the inflow, outflows for the alternative weir designs, and corre- sponding flood surcharges. The flood routing results in the graph illustrate the reduction in flood surcharge that oc- curs throughout the flood, lowering the peak level by 0.64m. During the rising limb of the flood, the outflow of course follows the inflow more closely with the PK weirs. For two to three hours around the peak of the flood, the outflow from the reservoir is governed by tunnel control, so the peak outflow with the PK weirs is marginally less than with the ogee weir. A sensitivity test was undertaken in which the PMF flow ordinates for the PK weir case were all increased by 10 percent. This results in a peak flood level being raised by about 0.3m, which would still be well within the residual freeboard provided by the dam raising scheme. This article is based on a paper presented by john Ackers to the 2nd International Workshop on Labyrinth and Piano Key Weirs, nov 2013, Chatou, France. Why 24-cycle design ● It was decided that the external overhang of the 12-cycle design would be excessive, so attention turned to a range of 24-cy- cle weirs, arranged in pairs with 30° angles between the pairs. These designs offered about the same ag- gregate weir length as the 12-cycle design, but with a smaller external overhang. Despite a slightly poorer hydraulic performance than the 12-cycle design, this design was favoured because the units would Why precast concrete ● The design of the PK weirs developed with ease of construction a significant consideration. The number of bays was determined in order to facilitate construction of the PK weirs as precast concrete units, despite the slightly poorer hydraulic performance. The adoption of identical 15° units allowed the use of a single mould and made sequencing of work simpler to manage. Working over water was minimised by ensuring that most of the construction activities occurred on the original Routing of the PMF through the raised reservoir

• Cover