scroll to: first paragraph
wheel_drwg.jpg
the Centrefold image linked to this page

the
engineering

of the
living swimming

pool at El Valero by
Trev Miller
designer and constructor

This document is at www.wavesculptor.com /pools/El_Valero_tech.htm.
headings and scroll to items scroll to the top of the window when clicked.
Images with italicised captions link to larger, pop-up images.
within paragraphs pop-up an explanatory window.
Contact the author easily from any on-line computer.


Related pages on www.wavesculptor.com:
The El Valero pool story describes in images and everyday language the effort to create this living sculpture;
its biological systems are explained on the living swimming page.
Some words about why I want to avoid chlorine appear on the bathing in bleach page.
preamblescroll to: page contents | this | next
Neither writer Chris Stewart, my patron for this project, nor I, the constructor, are fans of trendy, arcane technology with built-in obsolescence. So my intention was to keep El Valero pool workings as simple as compatible with longevity, visual interest, operating efficiency and environmental benignness. Quickly conceived, mass-produced, short-life, energy hungry components whose main selling points are availability and cheapness were not part of our plan. A Water Wheel is not available off a hypermarket shelf but it will outlast many off-the-shelf swimming pool pumps, and, in combination with the primal object next to it that is the filter, it is much more efficient.

Aficionados of CAD will appreciate that nearly all aspects of the pool were designed in this environment, many from the point of conception, using the screen for brainstorming.

So, starting at where much of the energy that runs the system is collected...

the photovoltaic (PV) system: the solar trackerscroll to: page contents | previous | this | next
This project was going to consume much more electricity than the house, which already had a just about adequate solar electrical system; the house was also far enough from the pool that it was easy to agree that the best way to power the project was an additional, dedicated custom solar power system. A tracking system was considered to be more exciting, cost effective and reliable in the long term, than double the number of static panels and all the deep-cycle batteries that would be needed to distribute the same power throughout the same period as the tracker operates. These advantages were considered to offset the considerable cost of the tracking mechanism.

A tracking system outputs power with much less variation than does a fixed array, between the extremes of sunrise, noon and sunset; its more constant output is closely matched to the constant motor loads of the pool system, whilst requiring little investment in or maintenance of batteries. Just enough storage capacity is provided to buffer the variation in output of the panels with small clouds and opacity of the atmosphere and provide the heavy pulse required to start the induction motor loads.

Shortly after the sun rises over the mountains, the pool circulation begins; when the sun sets behind the mountains to the west, or is shaded by sufficient cloud, the circulation stops. The pond and filter life does not appear to suffer from the lack of water circulation at night. This use pattern contrasts strongly to household use, where a large proportion of electricity demand will be when no collection is occuring, requiring a proportionally large investment in storage.
TrevTrackerFrame.jpg
building El Valero tracker 		The El Valero tracker can follow the sun accurately at any time of year, even though powered in only one axis. It will follow the sun on its summer or winter arcs or equinox plane. It carries 8 laser-etched, 85W monocrystaline BP Solar PV modules. An 800W sine-wave inverter converts their output to household voltage ac at 95% efficiency; this voltage can be cabled better than low voltage over the 100 or so metres to the Water Wheel or pump motors which it runs. The electronic control circuitry for the system is custom built. The tracker itself experienced hardly a moment's outage due to malfunction during its first three years, despite the control system still living on its prototyping board!

designing and building the trackerscroll to: page contents | previous | this | next
Just before I began this project I had been developing my own solar tracker in France. I applied what I had learnt to this one instead. The tracking geometry was determined with an atlas and calculations and verified with a model consisting of a piece of water pipe, to sight the sun through, hinged to a precisely aligned post. Perhaps I experienced the feelings of the prehistoric astronomers as I found the tiny sun remained visible all day through the equally tiny aperture of my tube with its restricted movement, despite the vast expanse of sky to lose it in! The mechanism was drawn up on my Apple Mac and then I set about building the thing beneath the orange trees on the terrace where it now stands. In practical terms, the remote terrace, to which deliveries had to be by tractor, was far from ideal as a workshop, but offered the spiritual delight of working among orange trees near rivers... that is, when I was not being grilled, bitten, blown away, flooded, drenched or left in the dark....

Meanwhile, the solar panels back at my camp remained lashed to rickety, roughly south-facing wooden trestles weighted with boulders... the occasional diabolical winds here having already scarred them by playing with them like kites. I doubt many small commercial models of solar tracker would withstand the ferocious gusts that hit El Valero several times a year. A case for a wind turbine rather than solar electrics, perhaps? No: much of the rest of the year, there is hardly a breeze. At this junction of two precipitous river valleys, high in the mountains, you'd think the rivers themselves might actually provide power, if not water. Think again: the rivers dry up in the summer and their use would risk the wrath of the Hydrografica, the bureaucrats managing them. The sun, however, is reliable enough and still unregulated here....

My thanks to Tramas Techno Ambiental (TTA), Barcelona - an independant co-operative specialising in pay-as-you-use solar-electrical installations for villages, for help with sourcing materials for the solar collector.

the Water Wheelscroll to: page contents | previous | this | next
In the context of this pool, we are concerned with a wheel that lifts water from a low canal and tips it into a high one, using energy supplied by an electric motor, in turn supplied by the solar collector described above. It is not a wheel of the kind that extracts energy from a flow of water to power a mill or generator or such like; it does, however, have that certain familiarity of form about it. Despite being near the confluence of two rivers, there is no year-round, reliable, permissable source of water power available at the site of the pool. The Water Wheel circulates the water around the pool system in a closed loop by lifting it 3 metres from behind a weir forming the lowest part of the water conditioning pond and tipping it into a sand filter bed. From there, the water continues its way to the swimming pool, to the pond and back to the wheel by gravity. wheel_318.jpg
completed water wheel (still enlargment)

!1.5MB! video of this as animated gif

The water lifting wheel principle is many thousands of years old but has been effectively obliterated by the centrifugal pump. The pump is more compact and therfore cheaper to make and instal, can have any desired output, irrespective of the dynamics of the body of water, and is more easily matched in mechanical characteristics to an electric motor or internal combustion engine.

why reinvent the wheel?scroll to: page contents | previous | this | next
I chose to reinvestigate and reinvent the water lifting wheel because of its romance factor but this enabled me to sell its other important advantages over equivalent, stark looking and sounding machines whose unprepossessing kind tend to be banished from human presence. As a streamlined, low speed machine, my wheel causes very little turbulence in the water and so wastes relatively little of the power driving it. An equivalent centrifugal pump spinning at several thousand rpm passing all its flow at great speed through a very small area of impeller wastes a considerable proportion of its input power by creating turbulence which heats the water it passes; anyone who has felt the temperature of a small body of water repeatedly recirculated through such a pump with no opportunity for cooling will appreciate this. As the energy source for this project was to be expensive solar photovoltaic panels, sacrifice of capacity to mechanical wastage was considered to be worth overcoming.

My design of The Waterwheel is such as to give a flow of water that would, industrially, be supplied by a pump consuming several kilowatts. The Waterwheel motor consumes about half of one kilowatt. How such efficiency is made possible by the gravity filter is explained under the filter heading, below.

Making The Waterwheel was neither exclusively work nor leisure. During its development and construction, I was determined to ignore commercial pressures upon me as much as possible. The Waterwheel became a device that was robust yet economical in materials, visually intriguing, yet basic enough to be built in a minimally-industrial area, and ultimately, about twice as efficient as the best centrifugal pumps designed for use with swimming pool filters, in terms of volume of water circulated per amount of energy supplied:   a performing sculpture taking full advantage of computer-aided design but neither art nor engineering alone.

wire not woodscroll to: page contents | previous | this | next
Naturally enough, I began by wanting to make The Waterwheel from environmentally sound wood. But the limited range of timber I could find in stock local to the El Valero project tended to be imported, or not farmed, or not of the quality to which I was accustomed or not competitively priced. I rant about the forests that could cover more of Andalusia but which are continuously consumed by goats and fire (such as this example), leaving few trees to achieve the proportions necessary for cropping for constructional use. The problems of wood-rot and insect damage also inclined me to look to less vulnerable materials, especially as timber preservatives and aquatic ecosystems somehow don't seem to go together.

In Spain, steel, in contrast to wood, is cheaper than anywhere else I've worked; it is of reasonable quality and available in a great variety of forms. Steel is also easily protected from corrosion by galvanising. So I looked for a way of using steel for The Waterwheel, but in a way that would not detract from the ambience of the pool environment. I ended up taking my inspiration for the core of the assembly from the tensioned, spoked bicycle wheel. This accurately supports a fabricated, galvanised lattice-work rim to which the buckets and the drive are coupled. The drive to the wheel was coupled to the rim to eliminate the substancial, high-torque hub-drive gearing typically encountered in wheels. The fabricated stainless rim-gear is turned by galvanised general purpose chain. The speed and torque characteristics of the wheel are matched to the motor by means of two reduction stages involving a morse chain and a poly-V rubber belt. The motor is a single-phase induction motor of no-load power consumption of 250W.

The wheel bearings are custom machined from nylon bar stock, which has the considerable advantages of working well when wet and needing minimal attention. One may easily rotate the empty 200 kg-plus wheel with gentle fingertip pressure, when the reduction drive mechanism is disconnected from the motor - which one would otherwise also be trying to spin at one-and-a-half-thousand rpm!

The wheel buckets feature hydrodynamic form, volumetric transfer efficiency and small-scale fabricability. They are of 0.5 mm stainless sheet, hand-formed on specially made dies, and assembled using polyurethane mastic and stainless rivets. The buckets are fitted to the end of slim but rigid, solid steel dipping arms which also carry the vent tubes to reduce air entrapment in the buckets. These features mean that very little of the energy supplied to the wheel is wasted as turbulence. The buckets transfer 95 percent of their capacity to the collection canals, when sufficient water is in the lower canal. Surplus energy supplied to the water tends to provide aeration, both as the buckets empty forming a film of water and in swirling down the stack feeding the filter.

the biologically active sand filterscroll to: page contents | previous | this | next
Apart from the waterwheel, the sand filter is the component of the El Valero pool system that arouses most curiousity. It resembles a 3 metre high, long-necked conical flask or decanter of the kind used for serving wine. It was fabricated in situ from and finished in a terracotta-like mineral colour. Though the resultant form has a function, it is exploited creatively, as in its accompanying waterwheel. (The filter image shown on the Pool Story page)

The compact, pressurised filter of a conventional swimming pool is actually a very energy hungry element of that system; it requires several times the pressure and hence power of the El Valero system to force a similar flow of water through a small but thick bed of sand. The amount of extra investment that would be required to provide a manyfold larger solar PV system to furnish such a difference in pool operating power may begin to justify the investment in unique technology in this application. The biologically active gravity filter functions perfectly with the large flow at small head from the waterwheel. The additional head capacity of a pump feeding the filter would be wasted.

The use of a filter passing water through a larger bed at a slower rate has useful implications for the biology of the system, as well as the energy consumption. The romance of the waterwheel is thus inextricably bound with the supply of water to a large area organic filtration system operating at low pressure.

why the filter shape?scroll to: page contents | previous | this | next
In order to pass a large volume of water slowly through a filter bed, a relatively large bed area is required. The El Valero pool filter bed has an area of 3 square metres, many times that of a fast filter that might be used with a similar sized pool. The intruiguing shape of the filter arises as a result of assimilating this area of filter into a system with finite water capacity. As the system starts when solar power is available, the level of the water above the filter matrix will rise, particularly when the filter needs cleaning, up to its limit of 2 metres. If the sides of the filter were a cylinder, the amount of water stolen from other parts of the pool and pond system would result in an unacceptable level drop outside. The filter vessel is thus constricted above the bed itself, to sufficient to accommodate a person for inspection purposes. The refinement of the form was compliant with transforming it subjectively from a potential industrial monster into a companionable visual entity.

As with most filter beds, a system of valves is provided for reversing the direction of flow of water through the filter. This has the effect of seperating the particles of the matrix of the filter and dislodging build-up of debris that would eventually block it as effectively as clay. A specially designed integral rotary rake, operated from outside the vessel during washing, further agitates the surface, where most entrapment occurs, and relevels the matrix surface after washing to ensure uniform flow. It is intended that the filter be washed when the maximum backpressure has been reached, and water begins bypassing it. The wash water emerges initially from the filter as opaque brown sludge - startling comparison to the crystal clear water that normally emerges. The wash water (with an acceptable odour of seaweed) may be either used for fertilisation in the garden or the farm or, to conserve water, returned to the front-end of the pool system treatment pond where it settles out overnight.

the vacuum cleanerscroll to: page contents | previous | this | next
In the configuration in which construction was begun, the El Valero pool was to have an easily retractable cover. This cover was to prevent the ingress of much light leading to algal growth and wind and animal borne debris, particularly during periods of hostile weather when the pool was out of the minds of its users. But, during the first season during which the pool was filled but the cover not yet installed, Chris, my patron, fell in love with the autumn leaves and spring blossom skating in formation over the clear water on still mornings. I had to abort the cover part of the project. The result of this, and the large amounts of overhanging vegetation, was that an extremely robust means of continually cleaning the pool had to be developed. Originally, it was intended that a hand-operated siphon suction - "vacuum" - tool be used to clean the bottom and feed the debris to the treatment pond, about a metre below, but this tool turned out to be far too time consuming to keep pace with the constant build-up of sediment. settler212.jpg
vacuum cleaner bag?
To boost the water flow and head between the pool and the treatment pond, such that an automatic suction-propelled vacuum cleaner could be used, a conventional swimming pool pump had to be installed - also serving as a back-up for the Water Wheel. But the sheer amount of debris in the pool would have resulted in clogging a normal vacuum cleaning system at very inconveniently short intervals. To overcome this, a vacuum vessel of aboout 0.25 cubic metre capacity was built onto the suction port of the pump, with flow designed to sediment larger particles and a screen to entrap more. Though complicating the system, this feature permited the pool floor to be vacuumed alternately with Water Wheel operation, whilst still circulating the water, and the collected debris to be easily emptied at intervals of over a month.

scroll to: page contents | previous | this
this site ©2004 Trev Miller -

W3C-validated HTML 4.01   W3C-validated CSS

Page history: First set in html: 2003, October 8th; Minor mods: 2004, August 15th

End of Page - scroll to: top and page contents