PIPE SIZE FOR WATER AND AIR
Ralph P. Penney, PE, LSP
Our environmental profession/industry has relied upon existing techniques and equipment from established industries. Soil borings and monitoring wells came from the geotechnical industry, which used them for designing foundations. We added the monitoring well aspect with non-petroleum based grease and flush threaded PVC screens and risers. Recovery wells with submersible pumps came from the drinking water supply industry. The available pumps are still designed for high discharge heads to deliver 60 psi of pressure for domestic water supply systems. We would be content with 20 psi of pressure to pump the groundwater through carbon canisters or an air stripper. I seem to remember being told that regenerative blowers were developed by the U.S. Air Force.The academic training continues to focus on mineral and oil exploration, wastewater treatment and water supply. How many large wastewater plants have been designed and built since the EPA suspended the grant program in the mid-80s? How many environmental engineering college programs have problems involving regenerative blowers? We routinely use blowers for air stripping and soil vapor extraction (SVE), but they are primarily used for materials handling where small air leaks are not a problem. Most of the smaller blowers we use all leak contaminated air from around the stamped metal inlet and outlet manifold. Run your OVM or PID around the manifold of an operating SVE blower sometime.I have retrofitted numerous SVE systems that have not been designed correctly. Seventy-five percent of the time the piping is too small. There is no small, economically priced air duct available to handle the 100 to 300 cfm of air for our small SVE systems so we are forced to use PVC and steel pipe. The headloss caused by the piping causes the blowers to run hot which distorts PVC, a thermoplastic, and reduces the adsorption capacity of vapor phase carbon. It also causes the electric motor on the blower to use more electricity, which a client is always quick to complain about. “You never told me it would cost $350/month for electricity to run the system.”Again, has any engineer ever been taught how to size a water pipe in college? I think not! The water flow problems I remember focused on sizing the pump. The pipe sizes were always given. Water and air are very similar fluids except air is much more compressible and water is much more dense. For this brief article I shall treat them as the same. The major factors when considering the movements of fluids through pipe include lifting the fluid against gravity, expressed as a change in potential energy, a change of pressure to do work, and friction losses. The factors are best expressed in Bernoulli’s equation or the more precise energy equation as follows:
The equation basically states that the total energy of a fluid at one point in a piping system must equal the energy at another point. The P/ɤ and Z terms represent potential energy, which is the major factor when lifting water from a well to a treatment system located 20 to 50 feet above the groundwater table. The P/ɤ term is pressure usually expressed in feet of water. The hp term represents the energy supplied by a pump or blower in the form of pressure. The V /2g term represents the kinetic energy, which is based upon the velocity of the fluid. It is also used to calculate the hf friction loss created by the fluid dragging against the inner walls of the pipe and flowing through fittings such as elbows, tees, valves, etc. Headloss factors (K-factors) have been empirically determined for numerous diameter fittings in a laboratory. The fitting losses are calculated as K·V /2g. The K-factor for a 1” diameter, long radius, 90° elbow is 0.72. So the headloss for changing the direction of 10 gpm of water flowing at 4.1 fps through a 1”, long radius 90° elbow is 0.19 ft or 0.08 psi. If the pipe was only ½” in diameter the flow velocity would be 16.3 fps and the headloss would be 4.2 ft or 1.8 psi. Notice the large increase in headloss with the smaller pipe. The frictional losses are equal to the square of the flow velocity. The flow velocity is controlled by the volume of the fluid being moved and the diameter of the pipe it is moving through. For reference, a 5/8” or 3/4" garden hose with an open discharge being supplied with street pressure normally flows at six to eight gpm, or 6.3 to 5.3 fps, respectively.The maximum design flow velocities I normally use are 64 fps for air and 7.0 fps for water. How did I arrive at these values? Well, I simply plotted flow velocity versus headloss for a 2” pipe. Sure enough, the plot showed the headloss increasing exponentially to the second power. You would get a similar plot if you plotted miles per hour versus gasoline mileage for your car or SUV. I suppose that I should have integrated to find the maximum slope change or the maximum tangent deflection but I didn’t. I just looked at the plot and noted the point were the steep inflection point began. For air, it began at 90 cfm. So you don’t want to force more than 90 cfm of air through a 2” pipe. For HVAC designs, an air flow of 15 fps is used to limit noise at the diffusers.Keep these maximum design flow velocity values in mind the next time you design a groundwater treatment or an SVE system. A 200 cfm SVE system should be designed for 3” pipe. Forget about using ball check valves with their extremely high K values and those moisture separators rated for 200 cfm, but setup with only 2” pipe. If you have an operating system with undersized piping or too many fittings, look at retrofitting it with larger pipe.