We have no known records of how many extra people lived in the village while the work was being carried out. Temporary living quarters would have been built, almost certainly in a camp next to the reservoir, probably on the western side in the area now covered by trees. But we know that in September 1905, when work on the pipeline in Geltsdale and the water intake apparatus was at its peak, over 700 men were being employed. That then declined to 200 by April 1906. At that time, a modest village of huts was in position on the western edge of the site, believed to be several offices or stores, and four huts for accommodation. There were 40 men to a hut but unlike in previous months when “double banking” had taken place (one man occupied the bed by night, another by day), these contained single beds. Suggestions have been put forward that upwards of 12 public houses were opened to quench the thirst.

The first sod was cut on the site of the filter beds, in the village, in September 1904. Further up Geltsdale, temporary 3 foot gauge railways were introduced to help move materials – and men – around the area. One engine was called Swansea, another Kuroki. The S-shaped cutting, which is now a track leading down to Gelt bridge near the dutch barn in Geltsdale, was carved out for a railway line first and foremost. The bridge was also brand new – built to help construct the inlet workings further upstream. Another short line transported rock and stone from the quarry above the eastern side of the Reservoir, using a pulley system with a small engine. It is still possible to see the cut in the quarry where the pulley was housed, and an old piece of rail still sticks out of the ground there today. On the reservoir site itself, a 1,100 foot long ropeway, driven by an engine, was used to bring down stone from the nearby quarry, across 5 spans. One worker was paid 6d an hour just to keep it running. The descending stone was tipped out at the foot onto the crusher stage, and fed into the crusher and concrete mixer. Cement was brought in from How Mill station. A huge cement shed was also built.

Vast quantities of puddle clay – which is clay soaked with water and kneaded, usually by the heels of boots, to render it water-resistant - was transported from just to the south of Tottergill, near Nixon Head, to the western edge of the reservoir to help build the dam. Men loaded clay into 1 cubic yard wagons which then travelled down to the site. At its peak, 24 men worked here at any one time, with a small hut for protection against the weather. There were also numerous short railway lines on the main building site itself, all shifting materials.

There were major problems building the main dam. Frost and snow in the winter of 1906/07 held back work in digging the trench, with major slips occurring and plans having to be made for unforeseen continuous pumping out of leaking water from January 1906 until at least September 1907. 10,000 gallons per hour was sucked out – in order to allow men to keep digging down. Concrete was finally put in during the summer of 1907. But then more flooding took place and pumping had to resume. 

A diver was employed in the autumn, for four hours a day, to try to stem the leaks – particularly after a 7 minute failure in the pumps had led to the water in the trench rising by more than 20 feet. Pumping only finished once the trench had filled with puddle clay and the infill had reached ground level.

The maximum depth of the excavation was 130 feet. The total quantity excavated was just under 20,000 cubic yards. A gang of six men could take out 36, two cubic yard skips per 8½ hr day. Gelignite was 90 shillings per 100lbs, fuses were 3 pence each, a pick was sharpened for 6 pence, and 1000 detonators cost 30 shillings.

What was called a “wing trench” also had to be dug and a barrier built, along half of the western side of the Reservoir. This was because there were fears that some of the water might leak out through a bed of gravel which was found in the rock there. It proved to be a frustrating extra piece of construction, taking up valuable time and money. It was – and still is – a fraction over 2,000 feet long, which is about two thirds of the length of the reservoir, in fact longer than the main dam itself. It was finished by January 1909.

The Valve Tower was finally finished in April 1908. The formal completion certificate for the construction of the Reservoir was presented by the engineers, Manserghs, on July 28th 1909 (six days after the official opening).