It has been brought to my attention that several users do not believe
that SWMM 5 is correctly simulating the behavior of a long, pumped force
main during times when the pump rapidly cycles on and off. When the pump
shuts off, SWMM shows that flow continues to exit from the end of the
force main for several minutes at a decreasing rate. This causes it to
de-pressurize at some point along its length since flow is leaving but
not entering the main during this period of time. When the pump starts
up again, there is a lag and then a gradual increase in flow out of the
main for several minutes until such time as the main completely re-fills
and re-pressurizes once again. The net result is that the initial flow
rate seen at the pump when it turns on again is significantly attenuated
at the outlet of the main.
The input data for an example showing this behavior is listed below. It
has a 50,000 foot long flat force main divided into 10 conduit sections
of 5,000 feet each. A fixed downstream boundary condition with water
level 0.1 ft above the top of the force main is used. A typical diurnal
DWF pattern is fed into the pump's wet well and the Type 3 pump turns on
at 5 feet and turns off at 1 feet. If you run this example, graph the
time series of flows for the pump and the last force main segment with
the X-axis interval adjusted to focus on the period from 16 to 18 hours
so you can more easily observe the behavior just described. Note that
during periods of prolonged pumping, when the main remains fully
pressurized, the outflow from the main is perfectly matching the varying
inflow from the pump.
An objection has been raised about this behavior, based on the belief
that once the main is pressurized, the outflow should equal the inflow
at every instance of time, including those times when the pump is turned
off (implying that the main will not de-pressurize during such times).
SWMM begs to differ. When the pump shuts off, the water in the main is
moving at about 2 ft/sec. The momentum contained in this column of
water will continue to carry it out of the downstream end, until such
time that frictional forces dissipate its forward motion. This is simply
Newton's Second Law at work (a body in motion tends to stay in motion
unless subjected to an opposing force). During the time that the pump is
off but the main keeps discharging water, mass continuity says that
somewhere along the main there must be empty volume that becomes
available to make up for the volume being discharged. Thus the main no
longer remains full throughout its length, so when the pump cycles on
again, the flow at the pump will not be immediately seen at the outlet
since the main is no longer pressurized throughout. Instead, some period
of time will elapse before the main refills and can then start obeying
"outflow equals inflow".
The question being put to the group is: Is SWMM 5 modeling this scenario
correctly? Has anyone has observed this type of behavior in actual
practice (i.e., flow continuing to issue from the end of a long (e.g.
several miles) force main after the pump feeding it has been shut off
and then gradually ramping back up over several minutes to match the
pump flow once the pump turns on again)?
==========================================================================
[TITLE]
[OPTIONS]
FLOW_UNITS CFS
INFILTRATION GREEN_AMPT
FLOW_ROUTING DYNWAVE
START_DATE 07/01/1999
START_TIME 00:00:00
REPORT_START_DATE 07/02/1999
REPORT_START_TIME 00:00:00
END_DATE 07/03/1999
END_TIME 00:00:00
SWEEP_START 01/01
SWEEP_END 12/31
DRY_DAYS 0
REPORT_STEP 00:00:30
WET_STEP 00:05:00
DRY_STEP 01:00:00
ROUTING_STEP 0:00:05
ALLOW_PONDING NO
INERTIAL_DAMPING NONE
VARIABLE_STEP 0.00
LENGTHENING_STEP 0
MIN_SURFAREA 0
NORMAL_FLOW_LIMITED SLOPE
SKIP_STEADY_STATE NO
IGNORE_RAINFALL NO
FORCE_MAIN_EQUATION H-W
LINK_OFFSETS DEPTH
[JUNCTIONS]
;; Invert Max. Init. Surcharge Ponded
;;Name Elev. Depth Depth Depth Area
;;-------------- ---------- ---------- ---------- ----------
----------
2 0 0 0 1000 0
3 0 0 0 1000 0
4 0 0 0 1000 0
5 0 0 0 1000 0
6 0 0 0 1000 0
7 0 0 0 1000 0
8 0 0 0 1000 0
9 0 0 0 1000 0
10 0 0 0 1000 0
11 0 0 0 1000 0
[OUTFALLS]
;; Invert Outfall Stage/Table Tide
;;Name Elev. Type Time Series Gate
;;-------------- ---------- ---------- ---------------- ----
Outfall 0 FIXED 3.1 NO
[STORAGE]
;; Invert Max. Init. Shape Shape
Ponded Evap.
;;Name Elev. Depth Depth Curve Params
Area Frac.
;;-------------- -------- -------- -------- ----------
-------- -------- -------- -------- --------
1 -10 22 0 FUNCTIONAL 800
0 0 0 0
[CONDUITS]
;; Inlet Outlet
Manning Inlet Outlet Init. Max.
;;Name Node Node Length N
Offset Offset Flow Flow
;;-------------- ---------------- ---------------- ----------
---------- ---------- ---------- ---------- ----------
2 2 3 5000 0.013
0 0 0 0
3 3 4 5000 0.013
0 0 0 0
4 4 5 5000 0.013
0 0 0 0
5 5 6 5000 0.013
0 0 0 0
6 6 7 5000 0.013
0 0 0 0
7 7 8 5000 0.013
0 0 0 0
8 8 9 5000 0.013
0 0 0 0
9 9 10 5000 0.013
0 0 0 0
10 10 11 5000 0.013
0 0 0 0
11 11 Outfall 5000 0.013
0 0 0 0
[PUMPS]
;; Inlet Outlet Pump
Init. Startup Shutoff
;;Name Node Node Curve
Status Depth Depth
;;-------------- ---------------- ---------------- ----------------
------ -------- --------
1 1 2 InfluentPump
ON 5 1
[XSECTIONS]
;;Link Shape Geom1 Geom2 Geom3
Geom4 Barrels
;;-------------- ------------ ---------------- ----------
---------- ---------- ----------
2 FORCE_MAIN 3 130 0
0 1
3 FORCE_MAIN 3 130 0
0 1
4 FORCE_MAIN 3 130 0
0 1
5 FORCE_MAIN 3 130 0
0 1
6 FORCE_MAIN 3 130 0
0 1
7 FORCE_MAIN 3 130 0
0 1
8 FORCE_MAIN 3 130 0
0 1
9 FORCE_MAIN 3 130 0
0 1
10 FORCE_MAIN 3 130 0
0 1
11 FORCE_MAIN 3 130 0
0 1
[LOSSES]
;;Link Inlet Outlet Average Flap Gate
;;-------------- ---------- ---------- ---------- ----------
[DWF]
;; Average Time
;;Node Parameter Value Patterns
;;-------------- ---------------- ---------- ----------
1 FLOW 15 "diurnalcurve"
[CURVES]
;;Name Type X-Value Y-Value
;;-------------- ---------- ---------- ----------
InfluentPump Pump3 10 30
InfluentPump 20 25
InfluentPump 30 20
InfluentPump 40 10
InfluentPump 50 0
[PATTERNS]
;;Name Type Multipliers
;;-------------- ---------- -----------
diurnalcurve HOURLY .5 .4 .3 .2 .1 .1
diurnalcurve .5 1 1.2 1.1 1.0 1.4
diurnalcurve 1.7 1.5 1.2 1.1 .8 1.1
diurnalcurve 1.6 1.7 1.3 1.1 .9 .7
[REPORT]
INPUT NO
CONTROLS NO
[TAGS]
[MAP]
DIMENSIONS -139.899 39986.136 2937.878 40003.401
Units Feet
[COORDINATES]
;;Node X-Coord Y-Coord
;;-------------- ------------------ ------------------
2 500.691 39998.450
3 663.377 39997.384
4 809.869 39997.384
5 992.985 39997.384
6 1155.172 39997.384
7 1338.288 39997.384
8 1504.527 39998.450
9 1720.214 39997.384
10 1978.220 39998.450
11 2250.655 39998.450
Outfall 2589.358 39998.450
1 7.363 40000.904
[VERTICES]
;;Link X-Coord Y-Coord
;;-------------- ------------------ ------------------
[LABELS]
;;X-Coord Y-Coord Label
-2029.817 43740.130 "Influent pump 1" 1
"Arial" 10 0 0
--------------------------------------------------------------------------------------