I am modelling an exiting rural gravity feed water system on the remote islands on Vanuatu using EPANET. I initially I wanted to test myself using a simple system to help me remember how to use the model
I wanted to know how much water would be used a day in a system, so I created a largish tank, had some downstream take of with demands, and following a 24 hour period the top tank had dropped and this corresponded with the three node demands I entered.
Some of the physical existing systems I have seen ( which I will be modeling) have intermediate tanks 10,000 li.
In my simple model I had three nodes releasing a demand of 0.1 li.sec and when modelled it all looked reasonable and the flow from the tank to the initial link pipe was (considering the pattern) approx. three times the some of the individuals.
I then added a tank 2.4 dia full at 2.4 and the model successfully ran.
However the flow into the tank and the directly associated pipes increased flow by approximately a factor of ten
The ultimate drop in the top tank added to the quantity of the three end nodes, but I am concerned as I cannot explain why the addition of a full tank on one of the branches, with the maximum level below any of the upstream outfalls should make a massive difference.!!
I understand EPA NET will stop flow to a tank once it is at the Maximum level, so I thought in the case of a full tank, the flow in should be equal to the flow out to keep equilibrium.
The issue is it will make sizing pipes difficult
Just a further question, my simple model had end nodes with discharge, but also up stream tee junction junctions. I put the end demand into the end nodes only and I assumed the model would work out the required demand from upstream nodes (considering each branch) IS this correct, also do all nodes require to be designated a pattern or will the end node patterns dictate this?
I do appreciate any help on this Regards