What is a Water Hammer?

You might know that most liquids are incompressible (or least barely-compressible), which means no matter how much pressure you apply, their volume doesn’t change. This can be useful, like in hydraulic cylinders, but that lack of “springiness” can also lead to catastrophic failure of pipe systems.

It’s easy to forget how heavy water is since we hardly ever carry more than a few ounces at a time. But if you add up the water in the pipelines of your city or even the pipes in your house, it makes up quite a bit of mass. And, when all that water is moving through a pipe, it has quite a bit of momentum. If you suddenly stop that movement—for example, by quickly closing a valve—all that momentum has nowhere to go. Since water isn’t compressible or springy, it can’t soften the blow. You might as well be slamming concrete into the back of the valve and the walls of your pipe. Instead of being absorbed, that sudden change in momentum creates a spike in pressure that travels as a shockwave through the pipe. Sometimes, you’ll even hear this shockwave as banging in your walls when you close a faucet or run the washing machine, hence the superhero-esque nickname, Water Hammer.

Banging pipes inside your walls can sound a bit spooky, but for large diameter pipelines that can be hundreds of kilometers long, that surge in pressure from a change in momentum can cause major damage. Let’s do a quick calculation: if you have pipeline carrying water that is 1 meter in diameter and runs for 100 kilometers (a fairly average-sized pipeline), the mass of water in the pipe is about 80 million kilograms. That’s a lot of kilograms. In fact, that's about 10 freight trains. Imagine you’re an operator at the end of this pipeline in charge of closing a valve. If you close it in a short amount of time, you’ve basically slammed those trains into a brick wall. And the pressure spike that results from such a sudden change in momentum can rupture the pipe or cause serious damage to other parts of the system. There’s actually another term for when a large spike in pressure ruptures a sealed container: a bomb. And water hammer can be equally dangerous. So, how do engineers design pipe systems to avoid this condition? Let’s build a model pipeline and find out. [Construction montage].

Here’s my setup. I’ve got about 100 feet (30 meters) of PVC pipe connected to the water on one end and a valve on the other. I also have an analog and digital gauge so we can see how the pressure changes and a clear section of pipe in case anything exciting happens in there. I mean civil-engineering-exciting, not like actual exciting. Watch what happens when I close this valve. It doesn’t look like much from the outside but let's look at the data from the pressure gauge. The pressure spikes to over 2,000 kilopascals or 300 psi. That’s about 5 times the static water pressure. It’s not enough to break the pipe, but way more than enough to break this pressure gage. You can see why designing a pipeline or pipe network can be a little more complicated than it seems. These spikes in pressure can travel through a system in complicated ways. But we can use this simple demonstration to show a few ways that engineers mitigate the potential damage from water hammer.

This is the equation for the pressure profile of a water hammer pulse. We’re not going to do any calculus here, but the terms of this equation show the parameters that can be adjusted to dial back these damaging forces. And, the first one is obvious: it’s the speed at which the fluid is moving through the pipe. Reducing this is one of the simplest ways to reduce the effect of water hammer. Velocity is a function of the flow rate and the size of the pipe. If you’re designing a pipeline, the flow rate might be fixed, so you can increase the size of your pipe to reduce the velocity. A smaller pipe may be less expensive, but the flow velocity will be higher which may cause issues with water hammer. In this case, my pipe size is fixed, but I can reduce the flow rate to limit the velocity. Each time I reduce the velocity and close the valve, the resulting spike in pressure decreases.

Next, you can increase the time over which the change in momentum occurs. One common example of this is adding flywheels to pumps so they spin down more slowly rather than stopping suddenly. Another example is to close valves more slowly. If I gently shut the valve rather than allowing it to snap shut, the pressure changes are more subtle. On large pipelines, engineers design the components and develop the requirements for operation of the equipment. This will almost always include rules for how quickly valves can be opened or closed to avoid issues with water hammer.

The final parameter we can adjust is the speed of sound through the fluid, also known as the wave celerity. This describes how quickly a pressure wave can propagate through the pipe. The wave celerity is an indirect measure of the elasticity of the system, and it can depend on the compressibility of the fluid, the material of the pipe and even if it’s buried in the ground. In a very rigid system, pressure waves can reflect easily without much attenuation. I’ve got flexible PVC pipe sitting on the ground free to move which is already helping reduce force of the water hammer. I can increase the flexibility even more by adding an anti-surge device. This has an air bladder that can absorb some of the shocks and reduce the pressure spike even further. Anti-surge devices are very common in pipe systems, and they can be as simple as a spring-loaded valve that opens up if the pressure gets too high. In water distribution systems for urban areas, water towers help with surge control by allowing the free surface to move up and down, absorbing sudden changes in pressure.

Plumbing is one of the under-acknowledged innovations that has made our modern society possible. When you harness the power of water by putting it in pipes, it’s easy to forget about that power. Water can be as hard as concrete when confined, and if you bang two hard things together, eventually something’s going to break. If you’re an engineer, your job is to make sure it’s not the expensive infrastructure you designed. Part of that means being able to predict surges in pressure due to water hammer and design systems that can mitigate any potential damage that might result. Thank you for watching, and let me know what you think!

How Soil Destroys Buildings

When most people think of property damage, they think about natural disasters. But what if I told you, there’s a slow-moving geologic phenomenon that causes more damage in the United States than earthquakes, floods, hurricanes, and tornadoes combined.

If you’ve ever been to a place where the ground looks like this, or if you’ve been in a building that looks like this or this, there’s a good chance you were in a place that had expansive soils. Just like these dinosaur toys, certain types of clay soils change their volume depending on moisture content. They swell when they get wet, and shrink as they dry. This is a microscopic mechanism where the shape and arrangement of the molecules actually change according to the amount of water mixed in. And, large portions of the U.S. Gulf coast and great plains have these kinds of soils. If you’re starting a foundation repair or road paving business, this is an important map for one very important reason: expansive soils break stuff.

Movement on its own and especially very slow movement is usually not a problem for structures. This is why we can lift buildings and even move them to new locations. What causes damage is differential movement. This is where certain parts of a structured move relative to each other. Differential movement leads to sticking doors and windows, cracked walls, and just general out-of-plumpness. And this is why expansive soils are so insidious because they don’t expand and contract evenly. For example, if your house sits on a concrete slab and you haven’t had any rain, the soils around the edges of the slab that are more exposed will dry out and shrink while the interior remains moist. Now you’ve got a foundation with no support around the edges. This breaks one of the fundamental laws of civil engineering, which says, and I quote, “You gotta have dirt underneath your concrete.”

Expansive clay isn’t just an issue for buildings. All kinds of infrastructure are at risk of damage from a shifting foundation. Leaking pipes can cause swelling of the soil, pulling apart joints and eventually leading to issues like sinkholes. Rainwater infiltrating through the cracks in roadways causes localized areas of swelling. This makes the road bumpy and uneven. Not even sidewalks, and by proxy rollerbladers, are spared. When designing to account for expansive clays, engineers not only have to know how much the soil can change in volume but also how hard it can push on anything sitting above, also known as swell pressure. So I’ve rigged up a little test so that we can see not only how soil swells, but also how much pressure it can exert. This apparatus called an odometer. It’s similar to a hydraulic cylinder, except I’m using dirt instead of oil, and I’ll use a dial indicator to measure how far the sample is able to move the piston. If you work in a soil laboratory, I’ll just apologize now for the rest of this video.

For my first test, I’ve got some soil straight from my own backyard. After all, there’s no place like a geologic unit containing abundant clay with high swelling potential. I put this in the oven to dry it out first, don’t tell my wife. Just kidding she knows whom she married. Now let’s put it in the apparatus and watch what happens. As it saturates, the soil expands over time, eventually reaching a 10% increase in volume over its dry state. Trust me, that’s enough to put a crack in the drywall. But, it’s really not that dramatic on video. So, to help illustrate these concepts a little better, I’ve got a bag of the instant viral video. That’s right I’m talking about Superabsorbent Polymer Beads, also known as Orbeez. These beads behave very similarly to expansive soils, except they’re way cooler than dirt in almost every way, even for a civil engineer.

First I tested these with no confining pressure and went a bit overboard. You can imagine if you built a house on this, you might get motion sickness every time it rains. It would wreak havoc on your structure. I tried it again with fewer orbeez, but it was still too much. This is an exaggerated view of what happens as water penetrates the subsurface and saturates an expansive soil. It’s hard to imagine anything that could avoid damage in this environment. So, let’s add some weight - and fewer orbeez this time so I don’t max out the range of my dial indicator. You can see that these fishing weights hardly make a difference. And that makes sense, right? A house probably puts more pressure on the ground below it than a few fishing weights. What about ten times that weight? It takes them a lot longer, but the orbeez are still able to swell to their full dimensions under this 20lb barbell, which is about the most my little acrylic oedometer can handle.

This is not just the case for orbeez by the way. Some clay soils have swell pressures on the order of megapascals (that’s hundreds of pounds per square inch). So you can see how big of a challenge these expansive soils can pose. There are lots of ways that engineers try to mitigate damage from these kinds of soils. You can simply remove all the expansive clay and bring in better soils for your project. You can grade the site so that water drains away from your structure, keeping moisture fluctuations down. You mix chemicals into the soil that limit its ability to absorb water. Finally, you can simply to build heavy enough to counteract the swell pressure and keep the soil from expanding. But as we saw in the demonstration, even a small amount of soil or in this case a colorful soil surrogate can lift a lot of weight.

I’m leaving out the simplest solution, which is simply to avoid expansive soils because it’s generally not feasible. It may be true in the parable that the wise man built his house on rock, but some civil engineer had to build a road to that guy’s house, and the engineer didn’t get to choose what kind of soil was on the way. Expansive soils are not a particularly newsworthy or exciting hazard (unless you’re the type of person who makes videos about dirt in your garage), but they still cause a tremendous amount of damage to buildings and the public infrastructure we rely on every day. They are one of the many factors taken into account when designing civil structures and the subject of ongoing research to find cost-effective and sustainable practices for mitigating the damage they cause. Thank you for watching, and let me know what you think!

How Quicksand Causes Dam Failures

When you think about quicksand, you probably don’t feel much fear or uncertainty. What was once a popular plot device in island and jungle movies of the 50s through the 80s has begun to fade from our collective curiosity. But, in civil engineering, quicksand is more than just a puddle of mud. It has a specific, and potentially more catastrophic, definition that can affect all kinds of infrastructure.

If you ever tried to build a dam in a creek or even in the storm gutter on your street, you know how hard it is. Water is not only powerful, but it’s elusive too. It seems to always find a way through. Like all fluids, water flows from areas of high pressure to low pressure, and a dam is a structure that separates those two conditions. This is a prime circumstance to induce flow, whether it’s through the dam itself, or underneath (through its foundation). Water flowing through soil is called seepage, and it follows some interesting and somewhat unexpected rules. So, to give you a better understanding, I’m building this demonstration out of clear acrylic sheets and some plumbing fixtures. This will essentially let us see a cross section through a dam so we can observe how the seepage behaves.

All dams have some seepage, so engineers need to be able to characterize it. Flow through soil follows Darcy’s Law, which is simple to understand, but complicated to calculate in two dimensions. Before computer models, engineers used an analytical tool called a flow net to estimate the behavior of seepage. Now software can do the work much more quickly and accurately. Darcy’s law says that the rate of seepage depends on the length of the flow path and the difference in pressure across it. The combination of these two factors is called the gradient. If the difference in pressure is small, like in this example, the seepage will be very slow. The bigger the difference in pressure, the greater the seepage flow rate will be. But, if you increase the difference by too much, eventually some strange things start to happen.

What’s happening here? I’ve talked about shear strength of soils in a previous video. Check that out if you want more detail, but here’s the gist: Soil is a granular material that has one main way of holding itself together: friction. Gravity pushes the soil particles together, creating friction which gives the soil strength. Seepage is the enemy of friction. The water gets between the soil particles and pushes them away from each other, reducing the friction and thus reducing the strength. This is important, because shear strength is really the only thing separating a solid from a fluid. In fact, when water pressure within the soil gets high enough to eliminate its shear strength all together, it’s called liquefaction, otherwise known as the “quick condition”. The soil is literally behaving as a liquid rather than a solid… it’s quicksand!

You can see why seepage is bad for dams and levees: engineers generally try to avoid building civil structures out of liquids. Soil liquefaction can be the starting point for a specific type of erosion called piping. As soil is carried away from the dam’s foundation, the seepage path gets shorter. Remember that the seepage depends on the gradient, which is a function of the length of the seepage path and the difference in pressure. If the pressures stay the same, but the path gets shorter, the gradient goes up, creating more seepage. This is a dangerous positive feedback loop. The erosion shortens the seepage path while the increasing seepage creates more erosion. Eventually the erosion pipe reaches the reservoir, leading to catastrophic failure of the dam’s foundation. This isn’t just speculation; piping is the number one cause of failure for earthen levees and dams.

How do engineers deal with seepage? One common ways is called a cutoff wall, a subsurface wall constructed in the foundation of dam. This is some footage from a previous demonstration I built that shows a cutoff wall in action. Notice the length of the seepage flow paths. They’re longer, which means the gradient is lower. That means less seepage and less chance for erosion. Thank you for watching, and let me know what you think!

How to Super-Size the Eclipse - Sun Funnel

If you live in North America, you know that this upcoming Monday is a big day for aficionados of astronomical phenomena: it’s a solar eclipse. And if you haven’t already, you need to get some of these glasses that let you look at the sun. But, the sun’s angular diameter in our sky is only half a degree. It’s smaller than your thumbnail at arm's length, and that’s pretty small. I won’t be in the path of totality this time, but I will be at an elementary school here in San Antonio enjoying the eclipse with students. And as usual when I talk to kids, I’m feeling compelled to try and make this experience a bit more visceral. Like mythbusters with explosions, I figure the best way to increase the excitement of a solar eclipse is to make it bigger.

The one job of a telescope is to gather a large amount of light and concentrate it so that you can see astronomical objects more clearly. That’s perfect for faint subjects like planets or nebulae, but not ideal for our nearest star, the sun. Even looking at the sun with no magnification can damage your vision, and magnifying those harmful rays is going to make the damage worse. So how can we get a bigger view of the sun safely? There’s two common ways: first you can use a solar filter that works just like eclipse glasses, blocking out the majority of the light so that you can safely see the sun through your telescope. But this way has a disadvantage: only one person can see at a time. Not ideal if you’re trying to share the excitement with a crowd. The second way is to use the telescope to project an image of the sun on a screen, and that’s the basis of the sun funnel.

Telescopes and binoculars use lenses or curved mirrors to concentrate light. The optics are usually aligned to focus the light onto the retina in your eye. But, if you move your eye out of the way, the light keeps going. Put a screen in front of it, and now you’ve got a projection. It really is as simple as holding up a screen in front of the eyepiece, but the sun funnel solves two practical problems: (1) it automatically keeps the projection screen in the correct orientation to the eyepiece even as you move your telescope around to follow the sun, and (2) it keeps anyone who doesn't know better from looking through the telescope at the sun and accidentally blinding themselves.

My design is based on a guide published on NASA’s website that I’ll link down below. They used a plastic funnel, but I decided to make my own using this thin-gauge aluminum sheet. I used pop rivets to fasten the edges. You can size the funnel based on your telescope, eyepiece, and how large you want the sun to be using a fairly simple formula. The eyepiece to my telescope slipped into the small end of the funnel and I connected it with a hose clamp. On the other end, I stretched a piece of vinyl shower curtain to act as a rear-projection screen. The last part of the project is to stop down the aperture on my telescope. Most reflector telescopes aren’t meant to focus the strong rays of the sun , and I didn’t want to overheat my secondary mirror. I’ll also be covering up the aperture altogether at regular intervals to make sure I don’t damage my telescope.

Here’s a test I did in my backyard. The sun is plenty bright enough to see, and big too! I think the kids are going to be impressed. I’ll post some pictures of the real eclipse in the sun funnel on my instagram next week but for now I can show a simulation of what it will probably look like. Even if you’re not looking at an eclipse the sun funnel is a fascinating way to observe an actual star up close and personal. I hope you liked this quick project, and for those of you in North America, I hope you get a chance to see the eclipse next week. Thank you for watching, and let me know what you think.

https://eclipse2017.nasa.gov/make-sun-funnel

How does land surveying work?

We use all kinds of smaller measuring tools in our everyday lives such as rulers, protractors, and tape measures. These tools work fine for home projects, but what if we need to layout something large like a road, bridge, dam, or pipeline?

Surveying is the science of taking big measurements. You’ve probably seen these guys on the side of the road looking through fancy equipment on a tripod. Almost any civil engineering project starts with a survey. This is to determine the legal boundaries between parcels of property. Surveying is also used to determine the location of existing infrastructure, and the topography and slopes of the land. Humans have always had a penchant for building big stuff. This means surveying is a career full of history and tradition. Behind every wonder of the ancient world was an ancient geometry nerd who laid out the angles and alignments during construction. Surveying is also how we created accurate maps of the continents like the Great Trigonometrical Survey of India. This took almost 70 years to complete. Everyone should aspire to accomplish something in your life that can be prefixed with the words “great trigonometrical.”

The ubiquitous tool for a survey is called a theodolite, and it’s one job is to measure the horizontal and vertical angles between points. Combine those angles with distances from a chain or tape measure, and you can triangulate the location of any point using trigonometry. Modern theodolites, called total stations, cannot only measure angles, but distance as well, and they have onboard computers to do the calculations and record the data for later use. When you see a surveyor peering through a funny telescope, it’s probably a total station, and he or she is probably sighting a reflector to record the location of a point. For long distances, these measurements have to be corrected for variations in earth’s gravity, refraction by the atmosphere, and yep, even the curvature of the earth. But don’t tell the flat-earthers. We’re sworn to secrecy along with NASA employees and airline pilots.

That’s just scratching the surface of sophistication with modern surveying equipment. With GPS and unmanned aircraft, surveying can get a lot more complicated. But I’ve got a few ways you can do your own topographic survey with fairly basic and inexpensive tools. Maybe you’ve got a drainage issue on your land or you’re planning a landscaping project. Or maybe you just want to exercise your God-given right to take measurements of stuff and write those measurements down on a clipboard. That’s my kind of recreational activity. My goal is to perform a leveling survey of my front and backyard, which is just a way to get the relative topography for an area. I laid out a grid of points on a map of my house and then transferred those points to real life using pin flags. Now I just need to pick my datum or base point and measure the relative difference in height between that point and all the others. I tried a few ways to do this and there are no sines, cosines, or tangents required.

First, a sight level which is essentially a combination of a telescope and a spirit level. To use it, first get a buddy or a willing spouse to hold a surveying rod on the point of interest. Now, look through the sight at a surveying rod and raise or lower the end until the bubble is centered on the line. Once it’s centered you know that you’re looking at a point that is exactly level to your eyes. Simply subtract the height of your eye-line with the height measured on the rod and that’s your elevation. It’s not a precision technique, but it is cheap and simple which the most you can usually hope for in any part of a home improvement project.

The next way I tried is a water level which is literally just a length of clear vinyl tubing filled with a liquid. As long as there are no bubbles or kinks in the line, the free surface at each end of the tube will self-level. I kept one end at my datum a fixed height and measure the height of the water at the other end as I carry it around to each of my points. It’s a little more unwieldy but it does have a distinct advantage, no line of sight required. You can use this method around corners or behind trees with no problem, and again, it’s a cheap and simple solution.

The third method to take a level survey worked best for me: my laser level. Here’s the thing: I really like lasers. I relish any chance I get to use them in a constructive way, and this is perfect. The laser level creates a perfect horizontal line that can be used to line up cabinets or tile, but it is also super easy to read on a surveying rod. You don’t need a helper, but you do probably need to wait until dusk unless your laser is really bright, or you have these sweet laser enhancement glasses. This isn’t the cheapest solution for a DIY land survey, but it is the fastest one I tried, and it’s a tool a lot of people already have.

Surveying is one of the oldest careers in the world, and also one of the most important. Why? Because land is important. If you own some, it’s probably your most valuable asset, and even if you don’t, you're pretty much stuck to it no matter where you go. As a career, surveying is a fascinating mix of legal knowledge, fieldwork, and technical challenges. And since most civil structures are too big for conventional measurement tools, the surveyor is one of the most important companions for the civil engineer. Thank you for watching, and let me know what you think!

How Do Sinkholes Form?

Some of the most complex civil engineering problems stem from the interaction of water and the ground. It sounds mundane but, there’s a good chance you’ve seen sinkhole on the news. How is it possible for the ground to simply open up and indiscriminately swallow anything or anyone that happens to be around?

We all know about erosion. This is the process that removes soil and rock from the earth’s crust and moves it somewhere else. And there’s a lot of ways this can happen: wind, landslides, abrasion, and scour. But here’s the thing, none of it compares to just the movement of water. Water is the great eroder. If you ever find yourself wondering how did this particular feature of the earth come to be here, or why is the ground shaped like so, or just why are things the way that they are, more often than not, the the answer is pretty much just water.

The ability of water to move soil or rock depends on several factors. The faster and more turbulent the flow, the more erosive it is. Larger particles like gravel and more resistant to erosion than small particles like silt or clay. Another important soil property is cohesion, or the ability of individual particles to stick to one another. Clay soils have more cohesion than sands, so they are more resistant to erosion. However, some clay soils are dispersive, which means they naturally wash away with water, making them particularly vulnerable to erosion. I love the standard test for dispersive soils, which is literally just to drop a clod of soil into a cup of water and see what happens. Finally, rather than physical erosion, some materials are soluble in water, just like sugar or salt, and can be eroded just by dissolving into the groundwater over time.

Most of us think about erosion on the surface of the earth, but erosion can occur in the subsurface as well. In fact, scientist and engineers have a very creative name for just such a process: internal erosion. If just the right factors come together in the subsurface, some very interesting things can occur, including sinkholes. But let’s look at a non-erosive example of groundwater movement first. This is a from a video I made before the channel was even called Practical Engineering. Water is flowing from the left side of the demo under an obstruction and over to the right. Notice two important things: first, the movement of water is slow. There’s not a lot of open space between all that sand, so it takes time for water to flow through it. Second, the sand is confined. Even if it wanted to move, there would be nowhere for it to go.

If those two conditions go away, that’s when sinkholes happen. Most natural sinkholes happen in areas with large deposits of carbonate rocks, like limestone. Over long periods of time, groundwater flowing through the subsurface can dissolve the rock, creating voids and open tunnels. In fact, this is how most caves are formed. These tunnels and voids create a significant change the character of groundwater flow. First, they allow water to flow quickly just like it would through a pipe, making it more erosive. Second, they create a space for soil to wash away. With those two conditions, any soil overlying a dissolution feature runs the risk of eroding away from the inside, eventually leading to a sinkhole.

But not every sink holes is formed through natural processes. In fact, many of the most famous sinkholes in recent times were human made. Just like a cave dissolved into the bedrock can act like a pipe and allow groundwater to carry away soil, an actual pipe can do the same thing. And actual pipes aren't limited to areas with a specific geology. If you could take a look into the subsurface of any urban area, you'd see miles and miles of water, sewer, and storm water drainage pipes. Unfortunately we can't see into the ground, so I built this demonstration so we can see for ourselves how this works.

All it takes is a little bit of settlement or shifting to create an opening in one of these pipes and allow internal erosion to start. Water moving through the pipe is able to dislodge the adjacent soil and carry it away. Notice that there's no signal on the surface that anything is awry. As more soil is washed away, the subsurface void grows. Depending on the type of soil and the speed of erosion, this process can take days to years before anyone notices. Many of our subsurface utilities are placed directly below roadways, and the paving often acts as a final bridge above the sinkhole, hiding the void below. It's only a matter of time before anything above is swallowed up.

Sinkholes aren’t the only problem caused by internal erosion. A specific type of internal erosion called piping is the most common cause of failure for earthen levees and dams, including Teton Dam in Idaho which killed 11 people and caused billions of dollars of damage when it failed in 1976. Maybe I’ll build a piping demonstration someday for a separate video. Internal erosion can be a natural process, but sometimes sinkholes can form to bad decisions, bad construction, or just bad luck with human made infrastructure as well. It’s just one of the complex failure modes that civil engineers must consider when designing a structure that might interact with water, the great eroder. Thank you for watching, and let me know what you think!

What's that Infrastructure? - Subsurface Utilities (Episode 4)

When you run a new utility line, whether it be electrical, gas, water, sewer, or communications, you basically have two choices for where to put it: overhead strung across poles, or below the ground. Today we’re talking about that second one, subsurface utilities. What’s that Infrastructure is a series where we divulge and discover the manmade world around us, and below us too.

Always call 811 before you dig (or the equivalent outside the US).

What's That Infrastructure: Transportation (Episode 1)

This is the first episode of a new series I started where I talk about miscellaneous pieces of infrastructure, including viewer-submitted photos. Click the WTI link at the top of the page if you'd like to submit your own photo!