The Energy Equation
Real flow loses energy to friction, and a pump is the only way to buy it back.
Before this Lesson
Last time, two gauges on a level main disagreed because of a reducer. Here are two more, a mile apart on a level 8 in. main with no reducer at all. Same pipe, same flow, same elevation. The upstream gauge reads 60 psi, and the downstream one reads 52 psi.
Bernoulli's equation says they should agree. The pipe never changes diameter, so the water can't change speed, and it never changes height. Yet 8 psi has gone missing, and every water main in the country does the same thing. Where did it go?
The toll friction takes
Bernoulli's equation assumed no friction, and real water has plenty. It drags along the pipe wall and churns against itself in eddies, and that work turns into heat. The Hook's missing 8 psi is ft of head, enough to warm the water by about two-hundredths of a degree Fahrenheit. The energy isn't destroyed, but it never comes back as pressure, speed, or height.
So the books need one more entry. The Handbook's Energy Equation, for steady incompressible flow, puts it on the downstream side:
The Handbook calls "the head loss, considered a friction effect": all the head the water gives up between the sections. It sits with section 2 because the water paid it on the way: what arrives plus what was spent equals what set out. Pipe Friction computes for a length of pipe. Here, a problem states it.
A loss is never negative, so unless a pump stands between them, total head always falls in the direction of flow. Run that backward and it tells you which way the water moves: from higher total head to lower, and the difference is . Pressure alone doesn't decide it. Water draining down a vertical pipe usually moves toward higher pressure, because it gives up elevation head faster than friction takes it.
Now the Hook. In a level pipe of one diameter, continuity fixes the velocity head and the elevation head never changes, so the whole loss comes out of the pressure head. The Handbook gives that case on its own: . A row of standpipes, which the Handbook calls piezometers, shows it: the water in each stands a little lower than in the one before.
The line through the water levels is the hydraulic grade line (HGL). It marks at every point. One velocity head above it runs the energy line, the total head, which the Handbook also calls the total head line and many problems call the energy grade line (EGL). Here both slope down at the same rate, and each drops by the head loss from one standpipe to the next.
Reducers and risers, with a loss
The reducer problem from Continuity and Bernoulli's Equation comes back with one more term. Pick two points, find each velocity head by continuity, turn the known pressure into a head, and solve for the other. Going downstream:
The loss always subtracts here: it's head the water arrives without. Often, though, the known pressure is the downstream one. A fixture at the top of a riser needs some minimum pressure, and the question is what the bottom has to supply. Solve the same equation for the other end:
Read it as a bill. The bottom pays for the pressure the top needs, the climb, any speed-up, and the loss. A sloping line works the same way, since only the rise enters the equation, not the slope.
A reservoir draining through a pipe
A favorite exam setup: a reservoir drains through a pipe, maybe ending in a nozzle, into the open air. Pick point 1 on the reservoir's surface and point 2 in the jet just outside the outlet. Both are open to the atmosphere, so both gauge pressures are zero. The surface drops so slowly that its velocity head is negligible. What's left is height turning into speed, minus the toll:
With no loss, this is the ideal of Tanks, Siphons, and Weirs: the water leaves as fast as if it had fallen from the surface. Every foot of loss is a foot it doesn't get to fall. The velocity is the jet's, so the flow is that velocity times the outlet's area, which is the nozzle's if there is one.
Pumps add head, turbines take it
Friction only ever takes head away. A pump is the one thing in a pipeline that puts it back: its impeller does work on the water, raising the total head by the pump head, . A turbine runs the other way. The water does work on its blades, and it takes out the turbine head, . Both go into the energy equation:
The left side is what the water has to spend, and the right side is where it goes. The Handbook prints its energy equation only for flow "with no energy input (e.g., no pump)," so these two terms are yours to add, and a machine the system lacks gets a zero.
Between two reservoir surfaces, the pressures are atmospheric and the velocities negligible, so only the elevations, the machine, and the loss remain. A pump lifting water to a higher reservoir has to supply the lift and pay every toll on the way:
A turbine below a dam gets what's left of the drop after the tolls: . The pump head is what Pump Head and Power turns into a power.
Two more questions move point 2. For the pressure at a point on the discharge line, put point 2 there and count only the loss between the reservoir and that point. For the highest elevation the pump can deliver the flow to, let the water arrive at an open surface, where the pressure and velocity heads are zero, and solve for .
Losses at fittings
Not all the loss happens along the pipe. Every elbow, valve, and tee makes the water turn or squeeze, and the eddies it sheds churn energy into heat. The faster the water, the harder it churns, so the loss grows with the velocity head. The Handbook's section on Minor Losses in Pipe Fittings, Contractions, and Expansions writes it as
where , which many problems write , is the number of velocity heads the fitting costs. The Handbook adds each to the energy equation beside , so a line's total loss is its friction plus every fitting's share. Problems usually state that total as one head loss, and this Lesson's means the total. The Handbook says a specific fitting's "will be provided in the problem statement," but it does print the values where a pipe meets a reservoir:
Adapted from the FE Reference Handbook, Minor Losses in Pipe Fittings, Contractions, and Expansions
Water swept around a sharp corner pulls away from the wall and has to spread out again, wasting half a velocity head. A rounded bell guides it in, and a pipe poking into the reservoir is worst, since water reaches its lip from every side, even from behind.
At an exit, the water carries its whole velocity head into the reservoir, and the still water churns all of it away, so . Count that energy once. With point 2 on the receiving reservoir's surface, where , the exit loss carries it. With point 2 in a free jet, there's no reservoir to churn it, and it stays in the equation as . Never both.
When fittings are the only losses, as in a short pipe of one diameter between two reservoirs, the surfaces again drop the pressures and velocities. The difference in level pays for every coefficient:
The same formula runs backward to measure a coefficient. Put gauges on either side of a valve in a level pipe of one diameter. The elevation and velocity heads match, so the whole pressure drop is the valve's loss:
A fitting's stays put as the flow changes, so its pressure drop follows the velocity head, which goes as the square of the velocity, and so of the flow.
The grade lines
The two lines from the standpipe picture can now map a whole system: the HGL at , and the energy line one velocity head above it. Along a pipeline they follow a few rules:
- Along a pipe, both slope down with friction, parallel where the diameter doesn't change.
- At a fitting, both step down by its loss. At a pump, both jump up by , and at a turbine they drop by .
- At an entrance, the energy line starts at the reservoir's surface and steps down by the entrance loss. At an exit, it drops by the whole velocity head onto the receiving surface, which the HGL meets.
- The gap between them is the velocity head, so it widens in a smaller pipe.
The picture's best trick is showing pressure. The vertical distance from the pipe up to the HGL is the pressure head there. Where the pipe climbs above the HGL, is less than , so the pressure is negative: the water is under suction. In the figure, the pump sits above the lower reservoir's surface, so the last stretch of its suction line runs above the HGL.
Choosing the form
Every problem in this Lesson used one equation, trimmed to fit. Pick two points where you know the most, such as free surfaces, jets, and gauges, write the full equation, and strike whatever is zero. With losses including any entrance and exit, the usual systems trim to these:
| System | Two points | What remains |
|---|---|---|
| Level pipe of one diameter | Two gauges | |
| Reducer, expansion, or riser | Two sections | Every term, with |
| Reservoir draining to the air | Surface, jet | |
| Gravity flow between reservoirs | Both surfaces | |
| Pump between reservoirs | Both surfaces | |
| Turbine between reservoirs | Both surfaces |