In the fourth episode of the Neutrium podcast Trevor and Matt continue their discussions on process hydraulics, covering the calculation of the friction factor and its subsequent use in determining pressure loss in a pipe.

Transcript available

Welcome to the fourth episode of the Nutrium Podcast. I'm Trevor Walker and I'm Matthew Kidd. And this week we will continue our discussions on hydraulics, this time looking at calculating pressure drop in pipes. Okay before we get started let's quickly recap what we covered last episode. First off we used some pseudo physics to demonstrate that pressure is a key variable in hydraulics as it provides the force to move a fluid from point A to point B. Then we went on to talk about a guy named Daniel Bernoulli, who aside from making his father insanely jealous, also developed an equation that describes the energy balance involved in hydraulics. This led to a discussion of flow regimes, of which there are three, laminar, turbulent and transitional, with the flow pattern varying across the regimes. Laminar flow is where the fluid flows along in straight lines without much mixing, eddies or vortices. Turbulent flow is the exact opposite to laminar, and transitional flow has elements of both laminar and turbulent flow. We also noted that the calculation of pressure drop or more specifically the friction factor calculation will vary depending on the flow regime. We will discuss this in more detail later in this episode. And lastly we introduced a dimensionless number called Reynolds number which can be used to determine the flow regime a particular fluid flows in. As we covered last episode, the Reynolds number is king among dimensionless numbers and measures the ratio of momentum forces to viscous forces in a fluid. If you're a process engineer and ever commit one dimensionless number to memory, make it the Reynolds number, which happens to be the density times velocity times characteristic length, i.e. the momentum of the fluid, divided by the kinematic viscosity, i.e. the viscous forces. Now, I say characteristic length because the Reynolds number can be calculated for a range of different situations, like either flow in a pipe or natural convection up a wall. For flow in a pipe, which happens to be what we are discussing today, the characteristic length is the diameter of the pipe, which makes sense when you remember that the numerator of the Reynolds number is the momentum of a slice of fluid and that momentum is equal to mass times velocity. Just to give you a little background on the Reynolds number, it was actually introduced as a concept in 1851 by Sir George Gabriel Stokes of Stokes Law and Navier-Stokes equation fame. Stokes was an interesting character and I'm sure we'll cover him in a future episode, but briefly he was an Irish mathematician born in 1819 who roamed around the UK drinking Guinness and making major contributions to the field of optics and fluid dynamics until his death in 1903. However, as you may have noticed, it's called Reynolds number, not Stokes number, and this is because it was popularised by another Irishman called Osborne Reynolds around 1883 in his paper titled "An Experimental Investigation of the Circumstances which Determine whether the Motion of Water in Parallel Channels Shall be Direct or Sinuous and the Law of Resistance in Parallel Channels". Don't you just love paper titles? Reynolds who happened to be one of the first professors in the UK to hold the title of Professor of Engineering constructed an experimental rig that allowed him to establish water flow through a transparent tube in which he would inject dye and observe the flow patterns over a range of flow conditions. In watching the dye Reynolds determined that the flow pattern changes significantly as you go from a Reynolds number of around 2100, i.e. laminar flow, through to a Reynolds number of around 4000, i.e. turbulent flow. This observation was key to accurately calculating pressure drop as it indicated a need to characterize things like friction factor over two distinct flow regimes. So once we've calculated the Reynolds number of our pipe and determined whether the flow is laminar, turbulent, or deity forbid, the transitional regime, the next step is to determine the resistance of flow and this means calculating the friction factor. They say there's no such thing as a free lunch and so it is with moving fluids around. it's going to cost you some energy. The next step is figuring out how much energy, in the form of pressure, it's going to take you to push your fluid through the pipe to where you need it. To do this, you're going to need a tool called friction factor. The friction factor is a number which helps relate the flowing fluid in the pipe to how much pressure we're going to lose from friction. The friction factor is well understood, and there are many correlations we can use to calculate it, unless we're in the transitional flow regime, in which case we need to get to a sensible flow regime and figure out what the friction factor is for sure. The friction factor is proportional to the pressure loss in the pipe, so the higher the friction factor, the higher the pressure loss. Generally, the friction factor is highest for low Reynolds numbers, and lowest for high Reynolds numbers. This makes sense intuitively if you remember that for a low Reynolds number you might have a dense, slow-moving, very viscous fluid through a thin pipe. So imagine trying to suck honey through a straw. You'll have very low Reynolds number and you'll need a big pressure difference to move that honey. On the flip side, to get a high Reynolds number, you have a low density, fast moving, low viscosity fluid through a larger pipe. So that might be like sucking air through a straw, which is going to be a lot easier. Even a theoretical pipe with perfectly smooth walls will extract some energy as you push a fluid through it due to no-slip boundary effects. Now there are two types of friction factor you might come across in the engineering world, Darcy friction factor and Fanning friction factor. The Darcy friction factor might also be called the Darcy-Westbach friction factor or the Moody friction factor. The only difference between the two is a factor of four. The Darcy friction factor is four times the Fanning friction factor, so converting between them is easy. The important thing is to be clear which friction factor you're dealing with. There can be only one, so we like to deal with the Darcy friction factor. There's a few names attached to these friction factors, and each has been influential to the study of fluid flow. Henry Darcy and John Fanning, who our friction factors are named after, are two 19th century engineer architect types, and both have to their credit the building of a major water distribution system for a city. Darcy built a water distribution system for Dijon in France, while Fanning designed the water system for Manchester in the United States. With the importance of understanding pressure losses to the successful design of a water system, it's not hard to see why both found the time to work on unlocking the secrets of where all the pressure was going. Wiesbach was a teacher of mathematics and mechanics, and refined the work of Darcy into the Darcy-Weisbach equation, which is ultimately how we relate friction factor, the fluid, and flowing conditions to pressure loss. Lewis Moody was an engineer and professor. The American Society of Mechanical Engineers, ASME, gives out an award named after him for original papers which further the field of fluid engineering. His most famous work is a chart, which we'll get to in a minute. Before we can calculate the friction factor, we need to figure out how rough our pipe is. Are we dealing with a nice smooth stainless steel or some kind of ancient foul brickwork sewer pipe? Roughness is a measure of the lumpiness of the surface of the material. It is a kind of average bulk property, not something you could actually measure with a ruler. You can think of it as a measure of the average of the distance between the peaks and troughs of the pipe wall material. The values of pipe roughness for different materials have been determined experimentally and you can look up values of roughness for various pipe materials and use them in combination with your preferred friction factor calculation methodology. Roughness is usually quoted as absolute roughness for a certain pipe material. You often need to convert that to relative roughness by dividing it by the pipe diameter. The relative roughness gives you the ratio of the pipe wall roughness to the overall pipe diameter, and it's dimensionless. When you are looking for the roughness of pipe materials, it's important to avoid imitations. There are other relationships which calculate flow. Manning's roughness and the Chezy coefficient are used for flow in open channels, and the Hazen-Williams roughness is used as an alternative pressure loss calculations in pipes and ducts. They're similar in concept to our pipe roughness, and they aim to do the same job of characterizing the surface for the purpose of quantifying fluid flow. But these roughness numbers are not interchangeable, and you have to be careful with your sources to make sure you don't get them mixed up. So before we move on to calculating the friction factor, here's John to talk about our sponsor, ManyTricks. ManyTricks is a great software development company whose apps do, you guessed it, many tricks. Their apps include Butler, Chemo, Leech, Desktop Curtain, TimeSync, Usher, Moom, NameMangler, Resolutionator and Witch. There's so much to talk about for each app, but they make. We're just going to touch on some of the highlights of five of them. Name Mangler. You've got a whole bunch of files you need to rename quickly, efficiently, and in huge numbers. Name Mangler is great for creating staged renaming sequences with powerful pattern matching, showing you the result as you go. And if you mess it up, you can revert back to where you started and you can try again. Usher can access any video stored in iTunes, Aperture, iPhoto, and on any connected hard drives on your Mac, allowing you to easily group, sort, tag and organise them all in one place. Install some plugins and there's no need to convert anything in an iTunes format to watch it. So if you've got a video collection scattered across platforms, drives and formats, then Usher can help you neatly sort it out in one app. Which You should think about which as a supercharger for your Command + Tab app switcher. If you've got 3 or 4 documents open at once in any one app, then which's beautifully simple pop-up quickly lets you pick exactly the one you're looking for. Moom makes it easy to move any of your windows to whatever screen positions you like, halves, corners, edges, fractions of the screen and then you can even save and recall your favourite window arrangements with a special auto-arrange feature when you connect or disconnect an external display. It's awesome. Resolutionator is their newest app and it's gloriously simple. A drop down menu from the menu bar and you can change the resolution of the display that you have to anything you like. The best part though is you can even set the resolution to fit more pixels than are actually there and it's still very usable. It's handy when you're stuck on a laptop screen and you need a bit more screen real estate. And that's just five of their great apps. There's still another five to check out. All of these apps have free trials and you can download them from manytricks.com/engineered and try them out before you buy them. They're all available from their respective pages on the site or through the Mac App Store. However, if you visit that URL, you can take advantage of a special discount off their very helpful apps exclusively for Engineered Network listeners. Simply use ENGINEERED25, that's ENGINEERED the word and 25 the numbers, in the discount code box in the shopping cart to receive 25% off. This offer is only available to Engineered Network listeners for a limited time, so take advantage of it while you can. Many thanks to ManyTricks for sponsoring Nutrium and the Engineered Network. Okay, when we have our flow regime and our relative roughness, we are finally ready to calculate the friction factor for our pipe. To calculate the friction factor, we use a different method depending on the flow regime our fluid is in. For the laminar flow regime, we use Darcy's equation, where the friction factor is equal to 64 divided by the Reynolds number. A simple analytical equation that works every time. In laminar flow, the fluid at the pipe wall is essentially stopped, so the pipe roughness doesn't matter, and doesn't factor into the friction factor. You could think of the fluid as acting as a gap filler, essentially smoothing out the pipe. For the uncooperative transitional flow regime, we examine the entrails of a goat and consult the oracle to determine the friction factor. Other techniques include taking the turbulent friction factor, which will be on the high side of the true answer, or using a weighted combination of the laminar and turbulent friction factors. For the turbulent flow regime, we have options for determining the friction factor. We can use a chart or one of many correlations. These are all based on the painstaking measurements of many interested people who set up hundreds of rigs and gathered tens of thousands of data points. At the end of it all is the heroic work of Colebrook and White, who produced the Colebrook-White equation, which brought all these data points together into one convenient equation. Well, maybe not so convenient. It turns out the equation is implicit, meaning the friction factor we are trying to find is required in the expression used to calculate itself, and so we have to iterate to get an answer. I couldn't really find anything interesting to say about Colebrook or White, so we can assume that they were perfect engineers, boring, lifeless people with no interesting features except for their mindless pursuit of technical accuracy. We salute you. In 1944, Lewis Moody took the time to chart out the results of the Colebrook-White equation, forming the popular Moody chart, which shows the friction factor against Reynolds number for various pipe roughnesses. This is the chart we mentioned before when we were talking a little about Moody, And if you've ever had to study fluid dynamics, it's a certainty that you've seen this chart. You can have a look for yourself over at the Nutrium article, "Pressure Loss in Pipe". The Moody chart has been the backbone of most pipe flow calculations for engineers from its creation until computational power reached a point where friction factor could easily and usefully be calculated from a formula. A 50-year span as one of the most useful tools for hydraulic calculations is not a bad run. In the age of the modern computer, we don't need to look up charts anymore. If precision is the aim of the game and time is no object, then solving the Colebrook-White equation iteratively is your best bet. But if you're calculating the friction factor over and over again, like you would in a segmented hydraulic model, this might slow you down too much. Fortunately, hundreds of brave engineers and scientists have attempted to produce explicit equations that accurately approximate the Colebrook-White equation. These are equations that don't require us to do any iterations and allow us to speed up our calculations. From this daunting list of approximations we have Serg Heide's equation Chen's equation Zygrang Sylvester equation The Harland equation Churchill equation The Swami Jain equation The Wood equation The Eck equation The Jain equation A different Churchill equation The Round equation The Barr equation The Manadili equation The Monzon-Romeo-Royo equation the Guda-Sonat equation, the Vantanka-Kuca-Zadek equation, the Buseli equation, the Avki-Kargoz equation, and the Evangelides-Papavengelou-Tzimpopovs equation, just to name a few. Oh man, I think I need a lie down. Now you may say that which one you use depends on the level of accuracy you need weighed up against the computational power you have available, but you're wrong, because SirGuide's equation is the correct answer. It gives great accuracy over a very wide range of Reynolds number and roughness values, and is fine for almost all practical computational time constraints. There are many other correlations that promise a higher level of accuracy, but I question the need for higher accuracy given the uncertainty in the assumed roughness values of the pipe material and the data that the relationship was based on. So there you go, we have our friction factor. If you were in laminar flow, we did simple calculations with Darcy's equation. If you're in turbulent flow, you grabbed your trusty Moody chart or fired up the old Fortran for your Colebrook-White or Serg-Hydes equation. Now we can calculate the pressure drop in our pipe. Believe it or not, but once you have determined the friction factor, the rest is relatively easy. For pressure loss in terms of head, which strictly speaking should actually be referred to as head loss, we multiply the friction factor, the pipe length on diameter ratio and velocity head. Now velocity head is a term that will come up a fair bit when discussing pressure loss, particularly pressure loss from fittings, which we will cover next episode. Velocity head is a measure of the dynamic pressure and is equal to the velocity squared divided by two times gravity. If you're sitting there now and thinking this sounds pretty familiar, you would be right. Velocity head is one of the components of Bernoulli's equation, which we covered last episode. Dynamic pressure, which is what velocity head measures, is just kinetic energy. Framing kinetic energy in this way allows us to directly compare kinetic and potential energy. When we say one velocity head, what we mean is the pressure consumed to take a fluid from rest to the flowing velocity, or how much potential energy is converted to kinetic energy when a fluid starts flowing. If you're a bit confused, don't worry, many people get confused when discussing dynamic pressure. All you need to remember is that velocity head is the difference in pressure when you change the velocity of a fluid. Now you know how to calculate pressure drop the correct way, let's quickly look at another method you might come across in industry. The Darcy friction factor and the pressure drop equation are the accepted standard for calculating pipe pressure drop, but it isn't the only method available. When you work in the water or fire systems industries, you will find that they shy away from the Darcy friction factor and use the Hazen-Williams coefficient we mentioned earlier. The water industry is one of those industries that loves to develop its own simplified correlations rather than depend on more widely applicable and accurate relationships. So in around 1906, Alan Hazen, an American hydraulics expert, and Gardner Stuart Williams, an American civil engineer, developed the Hazen-Williams equation, an imperial correlation for pressure drop that does not depend on Reynolds number. The equation itself is quite simple requiring only multiplication and division and the only inputs you need are pipe length, flow rate, hydraulic diameter and the Hayes and Williams coefficient. You can see why it would have been popular back before the invention of the computer when the alternative was to perform an iterative hand calculation of the Colebrook-White equation. The downside is it's limited to water and does not include temperature and viscosity corrections. I still come across people using the Hayes and Williams equation from time to time. It is particularly common when discussing fire water ring mains where clients will specify various Hayes and Williams coefficients for different sections of pipe. I personally tend to avoid it though because it is a bit too approximate for my liking and we have many fine solutions for the Colbrick-White equation that are easily calculated in a spreadsheet or hydraulic package. So now you know how to calculate pressure loss in a straight pipe. You can use your Reynolds number with your pipe roughness information to get the friction factor, then throw it all into the Darcy Weisbach equation and figure out your pressure loss. But what happens when you need to turn a corner? That's something for next time when we look at pressure loss through pipe fittings and some other tricky situations like two phase flow. So this has been another episode of the Nutrium Podcast. As always, we've added links to the related reading in the show notes. If you have any topics you'd like us to cover or general suggestions for the show, please contact us via the contact form at Nutrium.net or engineered.network. Network. (upbeat music) (upbeat music) (upbeat music)