Problem

When simulating a Pump start-up scenario, the flow rate at the end of the transient simulation is significantly greater than the nominal flow rate at the start. Why does this happen?

Problem ID#: 36519

Solution

The main cause of this discrepancy is likely due to differences in pipe
friction coefficients. HAMMER must use Darcy-Weisbach friction
coefficients for the transient simulation, so a conversion takes place
using the initial conditions. In the case of a pump startup event, the
pipe flows will typically be zero for the initial conditions. In this
case, HAMMER converts the user-entered pipe friction coefficient
(usually Hazen Williams) to an equivalent Darcy-Weisbach f coefficient.
However, HAMMER uses a tolerance to check if pipe flows should be
considered zero. This tolerance is found in the "flow tolerance" field
in the transient calculation options. If the pipe flows are below this
tolerance value, they will be considered zero and a Darcy-Weisbach
coefficient will be computed for the transient simulation. However,
sometimes the computed initial conditions may not be accurate enough for
the flows to be properly considered as zero.

To see this, go to Tools > Options > Units and enter '10' for the flow
display precision. Then, look at the initial condition flow values - you
may notice that some are closer to zero than others. If many of these
are above the "flow tolerance", HAMMER will not be able to compute the
proper Darcy-Weisbach f coefficient. This will cause the pipe friction
to essentially be different than what was used when the pump nominal
head/flow values were observed (when turning the pump on and computing
initial conditions). To observe this in the model, compute the transient
simulation, open the "output log" under Report > Transient Analysis
Reports and then scroll down to the section titled "Pipe Information".
You will observe an invalid friction coefficient for pipes whose flow
was zero but not close enough to zero to be considered as such. To
resolve this problem, you could either try a smaller value for
"accuracy" in the steady state/EPS solver calculation options (which
will compute flows closer to zero if the simulation is able to balance)
or by using a larger value for "flow tolerance" in the transient
calculation options (such as 0.01L/s). After making such a change and c
omputing the transient simulation, you should re-check the friction coefficients in the Output log to ensure that they are correct.

For more information on modeling a pump startup event, check the
following Technotes (depending on the version of HAMMER you're using):

https://communities.bentley.com/products/hydraulics___hydrology/w/hydraulics_and_hydrology__wiki/3276.modeling-a-pump-start-up-transient-event-in-bentley-hammer-v8i-tn.aspx