I am considering whether or not I should use fire hydrant test for calibration or if I can simply collect field data during peak hours. From what I understand this would be dependent on the amount of velocity of the water traveling within the pipes. So my question is at what velocity's can the modeler be assured that they would be capturing the impact of ,pipe roughness, closed valves and demand errors to a significant degree. Because if the velocity in my system is reaching that velocity during peak hours I could simply set out a bunch of pressure and flow loggers during peak hour and calibrate based on that instead of having to conduct a bunch of fire flow tests.
For HGL measurements to be useful for model calibration (adjusting roughness, finding closed valves and connectivity errors), the head loss between the source and the measurement point needs to be significantly larger than the error in measurement. For distribution pipes (6 to 16 in.) it is difficult to do this without inducing a high flow as in a hydrant flow test. On the other hand, if your system isn't designed for fire flow (2 to 6 in.), you get significant head loss during normal conditions. For large transmission mains, (> 16 in.), then hydrant flow tests don't impact HGL very much and you need to rely on peak demand times or conduct C-factor tests over long distances.
To add to Tom's answer - the following article has a list of many resources for calibration tips including some journal papers that contain advice like that given by Tom here: Water Model Calibration Tips
Regards,
Jesse DringoliTechnical Support Manager, OpenFlowsBentley Communities Site AdministratorBentley Systems, Inc.