|
[Sponsors] |
Since Michael Faraday, scientists have watched the curious patterns that form in a vibrating liquid. By adding floating particles to such a system, researchers have discovered spiky, hedgehog-like shapes that form near the surface. At low amplitudes, the surface patterns resemble the typical smooth rounded lobes one would expect, but as the wave amplitude increases, spikes form in the tracers, driven by the motion of the waves. (Image and research credit: H. Alarcón et al.; via APS Physics)
Bubbles in a pure liquid don’t last long, but with added surfactants or multiple miscible liquids, bubbles can form long-lasting foams. In soapy foams, surfactants provide the surface tension gradients necessary to keep the thin liquid layers between bubbles from popping. But what stabilizes a surfactant-free foam?
New work finds that foams in mixtures of two miscible fluids only form when the surface tension depends nonlinearly on the concentration of the component liquids. When this is true, thinning the wall between bubbles creates changes in surface tension that stabilize the barrier and keep it from popping.
In mixtures without this nonlinearity, foams just won’t form. The new results are valuable for manufacturing, where companies can avoid unintentional foams simply by careful selection of their fluids. (Image credit: G. Trovato; research credit: H. Tran et al.; via APS Physics; see also Ars Technica, submitted by Kam-Yung Soh)
Starfish won’t win any sprints, but they’re actually quite good at moving around as they hunt for prey. Without brains, starfish are led by their feet, which pull in the direction of food they scent. Each foot is connected to what amounts to an internal hydraulic system within the starfish. With a combination of secreted adhesive and pumping, the starfish can trundle along. (Image and video credit: Deep Look)
In the decade since the Deepwater Horizons oil spill, scientists have been working hard to understand the intricacies of how liquid and gaseous hydrocarbons behave underwater. The high pressures, low temperatures, and varying density of the surrounding ocean water all complicate the situation.
Released hydrocarbons form a plume made up of oil drops and gas bubbles of many sizes. Large drops and bubbles rise relatively quickly due to their buoyancy, so they remain confined to a relatively small area around the leak. Smaller drops are slower to rise and can instead get picked up by ocean currents, allowing them to spread. The smallest micro-droplets of oil hardly rise at all; instead they remained trapped in the water column, where currents can move them tens to hundreds of kilometers from their point of release. (Image and research credit: M. Boufadel et al.; via AGU Eos; submitted by Kam-Yung Soh)
This gorgeous photograph of Kelvin-Helmholtz clouds was taken in late December in Slovenia by Gregor Riačevič. The wave-like shape of the Kelvin-Helmholtz instability comes from shear between two fluid layers moving at different relative speeds. Here on Earth, clouds like these are often short-lived, but we see similar structures in the atmospheres of gas giants like Jupiter and Saturn. (Image credit: G. Riačevič; submitted by Matevz D.)
In flight, birds must adjust quickly to wind gusts or risk crashing. Research shows that the structure of birds’ wings enables them to respond faster than their brains can. The wings essentially act like a suspension system, with the shoulder joint allowing them to lift rapidly in response to vertical gusts. This motion keeps the bird’s head and torso steady, so they can focus on more complex tasks like landing, obstacle avoidance, and prey capture. (Image and research credit: J. Cheney et al.; submitted by Kam-Yung Soh)
There is an interesting new trend in using Computational Fluid Dynamics (CFD). Until recently CFD simulation was focused on existing and future things, think flying cars. Now we see CFD being applied to simulate fluid flow in the distant past, think fossils.
CFD shows Ediacaran dinner party featured plenty to eat and adequate sanitation
Let's first address the elephant in the room - it's been a while since the last Caedium release. The multi-substance infrastructure for the Conjugate Heat Transfer (CHT) capability was a much larger effort than I anticipated and consumed a lot of resources. This lead to the relative quiet you may have noticed on our website. However, with the new foundation laid and solid we can look forward to a bright future.
Conjugate Heat Transfer Through a Water-Air Radiator
Simulation shows separate air and water streamline paths colored by temperature
It turns out that Computational Fluid Dynamics (CFD) has a key role to play in determining the behavior of long extinct creatures. In a previous, post we described a CFD study of parvancorina, and now Pernille Troelsen at Liverpool John Moore University is using CFD for insights into how long-necked plesiosaurs might have swum and hunted.
CFD Water Flow Simulation over an Idealized Plesiosaur: Streamline VectorsIllustration only, not part of the study
Fossilized imprints of Parvancorina from over 500 million years ago have puzzled paleontologists for decades. What makes it difficult to infer their behavior is that Parvancorina have none of the familiar features we might expect of animals, e.g., limbs, mouth. In an attempt to shed some light on how Parvancorina might have interacted with their environment researchers have enlisted the help of Computational Fluid Dynamics (CFD).
CFD Water Flow Simulation over a Parvancorina: Forward directionIllustration only, not part of the study
One of nature's smallest aerodynamic specialists - insects - have provided a clue to more efficient and robust wind turbine design.
Dragonfly: Yellow-winged DarterLicense: CC BY-SA 2.5, André Karwath
The recent attempt to break the 2 hour marathon came very close at 2:00:24, with various aids that would be deemed illegal under current IAAF rules. The bold and obvious aerodynamic aid appeared to be a Tesla fitted with an oversized digital clock leading the runners by a few meters.
2 Hour Marathon Attempt
• RANShttps://www.cfd-online.com/Forums/bl...1&d=1610557096
• MRF
• Compressible
• K-Omega SST
• Subsonic
• Inlet T = 300 K
• Inlet p = 1 atm
• Mass flow = 0.1 Kg/s
• Rotation Speed = 50 000 rpm
Hi Alexey,
I have a problem (again) when i am following the instructions as given in https://github.com/mrklein/openfoam-...ase-&-Homebrew In particular, I have followed the steps without any problem until when I had to apply the patch with git: git apply OpenFOAM-v1912.patch When I opened the patch file, I show the flag: 404: Not Found. Where can I find the patch? When I visited your site, I show that you have patches for different versions of OpenFoam, but not for v1912. If I download the most recent one, "OpenFOAM-7-0ebbff061.patch" and execute "git apply OpenFOAM-7-0ebbff061.patch" instead, do you think it will be OK? |
Filippo Maria Denaro added an answer December 7, 2017 Lalit the Nyquist theorem says that for a step sampling dt you can describe the smallest wavelenght 2*dt (three samples describe a sine). For a given period lenght T, the ratio T/(2*dt) gives the maximum wavenunber you can represent |
In this post, I’ll give a simple example of how to create curves in blockMesh. For this example, we’ll look at the following basic setup:
As you can see, we’ll be simulating the flow over a bump defined by the curve:
First, let’s look at the basic blockMeshDict for this blocking layout WITHOUT any curves defined:
/*--------------------------------*- C++ -*----------------------------------*\
========= |
\\ / F ield | OpenFOAM: The Open Source CFD Toolbox
\\ / O peration | Website: https://openfoam.org
\\ / A nd | Version: 6
\\/ M anipulation |
\*---------------------------------------------------------------------------*/
FoamFile
{
version 2.0;
format ascii;
class dictionary;
object blockMeshDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
convertToMeters 1;
vertices
(
(-1 0 0) // 0
(0 0 0) // 1
(1 0 0) // 2
(2 0 0) // 3
(-1 2 0) // 4
(0 2 0) // 5
(1 2 0) // 6
(2 2 0) // 7
(-1 0 1) // 8
(0 0 1) // 9
(1 0 1) // 10
(2 0 1) // 11
(-1 2 1) // 12
(0 2 1) // 13
(1 2 1) // 14
(2 2 1) // 15
);
blocks
(
hex (0 1 5 4 8 9 13 12) (20 100 1) simpleGrading (0.1 10 1)
hex (1 2 6 5 9 10 14 13) (80 100 1) simpleGrading (1 10 1)
hex (2 3 7 6 10 11 15 14) (20 100 1) simpleGrading (10 10 1)
);
edges
(
);
boundary
(
inlet
{
type patch;
faces
(
(0 8 12 4)
);
}
outlet
{
type patch;
faces
(
(3 7 15 11)
);
}
lowerWall
{
type wall;
faces
(
(0 1 9 8)
(1 2 10 9)
(2 3 11 10)
);
}
upperWall
{
type patch;
faces
(
(4 12 13 5)
(5 13 14 6)
(6 14 15 7)
);
}
frontAndBack
{
type empty;
faces
(
(8 9 13 12)
(9 10 14 13)
(10 11 15 14)
(1 0 4 5)
(2 1 5 6)
(3 2 6 7)
);
}
);
// ************************************************************************* //
This blockMeshDict produces the following grid:
It is best practice in my opinion to first make your blockMesh without any edges. This lets you see if there are any major errors resulting from the block topology itself. From the results above, we can see we’re ready to move on!
So now we need to define the curve. In blockMesh, curves are added using the edges sub-dictionary. This is a simple sub dictionary that is just a list of interpolation points:
edges
(
polyLine 1 2
(
(0 0 0)
(0.1 0.0309016994 0)
(0.2 0.0587785252 0)
(0.3 0.0809016994 0)
(0.4 0.0951056516 0)
(0.5 0.1 0)
(0.6 0.0951056516 0)
(0.7 0.0809016994 0)
(0.8 0.0587785252 0)
(0.9 0.0309016994 0)
(1 0 0)
)
polyLine 9 10
(
(0 0 1)
(0.1 0.0309016994 1)
(0.2 0.0587785252 1)
(0.3 0.0809016994 1)
(0.4 0.0951056516 1)
(0.5 0.1 1)
(0.6 0.0951056516 1)
(0.7 0.0809016994 1)
(0.8 0.0587785252 1)
(0.9 0.0309016994 1)
(1 0 1)
)
);
The sub-dictionary above is just a list of points on the curve . The interpolation method is polyLine (straight lines between interpolation points). An alternative interpolation method could be spline.
The following mesh is produced:
Hopefully this simple example will help some people looking to incorporate curved edges into their blockMeshing!
Cheers.
This offering is not approved or endorsed by OpenCFD Limited, producer and distributor of the OpenFOAM software via http://www.openfoam.com, and owner of theOPENFOAM® andOpenCFD® trademarks.
Experimentally visualizing high-speed flow was a serious challenge for decades. Before the advent of modern laser diagnostics and velocimetry, the only real techniques for visualizing high speed flow fields were the optical techniques of Schlieren and Shadowgraph.
Today, Schlieren and Shadowgraph remain an extremely popular means to visualize high-speed flows. In particular, Schlieren and Shadowgraph allow us to visualize complex flow phenomena such as shockwaves, expansion waves, slip lines, and shear layers very effectively.
In CFD there are many reasons to recreate these types of images. First, they look awesome. Second, if you are doing a study comparing to experiments, occasionally the only full-field data you have could be experimental images in the form of Schlieren and Shadowgraph.
Without going into detail about Schlieren and Shadowgraph themselves, primarily you just need to understand that Schlieren and Shadowgraph represent visualizations of the first and second derivatives of the flow field refractive index (which is directly related to density).
In Schlieren, a knife-edge is used to selectively cut off light that has been refracted. As a result you get a visualization of the first derivative of the refractive index in the direction normal to the knife edge. So for example, if an experiment used a horizontal knife edge, you would see the vertical derivative of the refractive index, and hence the density.
For Shadowgraph, no knife edge is used, and the images are a visualization of the second derivative of the refractive index. Unlike the Schlieren images, shadowgraph has no direction and shows you the laplacian of the refractive index field (or density field).
In this post, I’ll use a simple case I did previously (https://curiosityfluids.com/2016/03/28/mach-1-5-flow-over-23-degree-wedge-rhocentralfoam/) as an example and produce some synthetic Schlieren and Shadowgraph images using the data.
Well as you might expect, from the introduction, we simply do this by visualizing the gradients of the density field.
In ParaView the necessary tool for this is:
Gradient of Unstructured DataSet:
Once you’ve selected this, we then need to set the properties so that we are going to operate on the density field:
To do this, simply set the “Scalar Array” to the density field (rho), and change the name of the result Array name to SyntheticSchlieren. Now you should see something like this:
There are a few problems with the above image (1) Schlieren images are directional and this is a magnitude (2) Schlieren and Shadowgraph images are black and white. So if you really want your Schlieren images to look like the real thing, you should change to black and white. ALTHOUGH, Cold and Hot, Black-Body radiation, and Rainbow Desatured all look pretty amazing.
To fix these, you should only visualize one component of the Synthetic Schlieren array at a time, and you should visualize using the X-ray color preset:
The results look pretty realistic:
The process of computing the shadowgraph field is very similar. However, recall that shadowgraph visualizes the Laplacian of the density field. BUT THERE IS NO LAPLACIAN CALCULATOR IN PARAVIEW!?! Haha no big deal. Just remember the basic vector calculus identity:
Therefore, in order for us to get the Shadowgraph image, we just need to take the Divergence of the Synthetic Schlieren vector field!
To do this, we just have to use the Gradient of Unstructured DataSet tool again:
This time, Deselect “Compute Gradient” and the select “Compute Divergence” and change the Divergence array name to Shadowgraph.
Visualized in black and white, we get a very realistic looking synthetic Shadowgraph image:
Now this is an important question, but a simple one to answer. And the answer is…. not much. Physically, we know exactly what these mean, these are: Schlieren is the gradient of the density field in one direction and Shadowgraph is the laplacian of the density field. But what you need to remember is that both Schlieren and Shadowgraph are qualitative images. The position of the knife edge, brightness of the light etc. all affect how a real experimental Schlieren or Shadowgraph image will look.
This means, very often, in order to get the synthetic Schlieren to closely match an experiment, you will likely have to change the scale of your synthetic images. In the end though, you can end up with extremely realistic and accurate synthetic Schlieren images.
Hopefully this post will be helpful to some of you out there. Cheers!
Sutherland’s equation is a useful model for the temperature dependence of the viscosity of gases. I give a few details about it in this post: https://curiosityfluids.com/2019/02/15/sutherlands-law/
The law given by:
It is also often simplified (as it is in OpenFOAM) to:
In order to use these equations, obviously, you need to know the coefficients. Here, I’m going to show you how you can simply create your own Sutherland coefficients using least-squares fitting Python 3.
So why would you do this? Basically, there are two main reasons for this. First, if you are not using air, the Sutherland coefficients can be hard to find. If you happen to find them, they can be hard to reference, and you may not know how accurate they are. So creating your own Sutherland coefficients makes a ton of sense from an academic point of view. In your thesis or paper, you can say that you created them yourself, and not only that you can give an exact number for the error in the temperature range you are investigating.
So let’s say we are looking for a viscosity model of Nitrogen N2 – and we can’t find the coefficients anywhere – or for the second reason above, you’ve decided its best to create your own.
By far the simplest way to achieve this is using Python and the Scipy.optimize package.
Step 1: Get Data
The first step is to find some well known, and easily cited, source for viscosity data. I usually use the NIST webbook (
https://webbook.nist.gov/), but occasionally the temperatures there aren’t high enough. So you could also pull the data out of a publication somewhere. Here I’ll use the following data from NIST:
Temparature (K) | Viscosity (Pa.s) |
200 |
0.000012924 |
400 | 0.000022217 |
600 | 0.000029602 |
800 | 0.000035932 |
1000 | 0.000041597 |
1200 | 0.000046812 |
1400 | 0.000051704 |
1600 | 0.000056357 |
1800 | 0.000060829 |
2000 | 0.000065162 |
This data is the dynamics viscosity of nitrogen N2 pulled from the NIST database for 0.101 MPa. (Note that in these ranges viscosity should be only temperature dependent).
Step 2: Use python to fit the data
If you are unfamiliar with Python, this may seem a little foreign to you, but python is extremely simple.
First, we need to load the necessary packages (here, we’ll load numpy, scipy.optimize, and matplotlib):
import numpy as np
import matplotlib.pyplot as plt
from scipy.optimize import curve_fit
Now we define the sutherland function:
def sutherland(T, As, Ts):
return As*T**(3/2)/(Ts+T)
Next we input the data:
T=[200,
400,
600,
800,
1000,
1200,
1400,
1600,
1800,
2000]
mu=[0.000012924,
0.000022217,
0.000029602,
0.000035932,
0.000041597,
0.000046812,
0.000051704,
0.000056357,
0.000060829,
0.000065162]
Then we fit the data using the curve_fit function from scipy.optimize. This function uses a least squares minimization to solve for the unknown coefficients. The output variable popt is an array that contains our desired variables As and Ts.
popt = curve_fit(sutherland, T, mu)
As=popt[0]
Ts=popt[1]
Now we can just output our data to the screen and plot the results if we so wish:
print('As = '+str(popt[0])+'\n')
print('Ts = '+str(popt[1])+'\n')
xplot=np.linspace(200,2000,100)
yplot=sutherland(xplot,As,Ts)
plt.plot(T,mu,'ok',xplot,yplot,'-r')
plt.xlabel('Temperature (K)')
plt.ylabel('Dynamic Viscosity (Pa.s)')
plt.legend(['NIST Data', 'Sutherland'])
plt.show()
Overall the entire code looks like this:
import numpy as np
import matplotlib.pyplot as plt
from scipy.optimize import curve_fit
def sutherland(T, As, Ts):
return As*T**(3/2)/(Ts+T)
T=[200, 400, 600,
800,
1000,
1200,
1400,
1600,
1800,
2000]
mu=[0.000012924,
0.000022217,
0.000029602,
0.000035932,
0.000041597,
0.000046812,
0.000051704,
0.000056357,
0.000060829,
0.000065162]
popt, pcov = curve_fit(sutherland, T, mu)
As=popt[0]
Ts=popt[1]
print('As = '+str(popt[0])+'\n')
print('Ts = '+str(popt[1])+'\n')
xplot=np.linspace(200,2000,100)
yplot=sutherland(xplot,As,Ts)
plt.plot(T,mu,'ok',xplot,yplot,'-r')
plt.xlabel('Temperature (K)')
plt.ylabel('Dynamic Viscosity (Pa.s)')
plt.legend(['NIST Data', 'Sutherland'])
plt.show()
And the results for nitrogen gas in this range are As=1.55902E-6, and Ts=168.766 K. Now we have our own coefficients that we can quantify the error on and use in our academic research! Wahoo!
In this post, we looked at how we can simply use a database of viscosity-temperature data and use the python package scipy to solve for our unknown Sutherland viscosity coefficients. This NIST database was used to grab some data, and the data was then loaded into Python and curve-fit using scipy.optimize curve_fit function.
This task could also easily be accomplished using the Matlab curve-fitting toolbox, or perhaps in excel. However, I have not had good success using the excel solver to solve for unknown coefficients.
The most common complaint I hear, and the most common problem I observe with OpenFOAM is its supposed “steep learning curve”. I would argue however, that for those who want to practice CFD effectively, the learning curve is equally as steep as any other software.
There is a distinction that should be made between “user friendliness” and the learning curve required to do good CFD.
While I concede that other commercial programs have better basic user friendliness (a nice graphical interface, drop down menus, point and click options etc), it is equally as likely (if not more likely) that you will get bad results in those programs as with OpenFOAM. In fact, to some extent, the high user friendliness of commercial software can encourage a level of ignorance that can be dangerous. Additionally, once you are comfortable operating in the OpenFOAM world, the possibilities become endless and things like code modification, and bash and python scripting can make OpenFOAM worklows EXTREMELY efficient and powerful.
Anyway, here are a few tips to more easily tackle the OpenFOAM learning curve:
(1) Understand CFD
This may seem obvious… but its not to some. Troubleshooting bad simulation results or unstable simulations that crash is impossible if you don’t have at least a basic understanding of what is happening under the hood. My favorite books on CFD are:
(a) The Finite Volume Method in Computational Fluid Dynamics: An Advanced Introduction with OpenFOAM® and Matlab by
F. Moukalled, L. Mangani, and M. Darwish
(b) An introduction to computational fluid dynamics – the finite volume method – by H K Versteeg and W Malalasekera
(c) Computational fluid dynamics – the basics with applications – By John D. Anderson
(2) Understand fluid dynamics
Again, this may seem obvious and not very insightful. But if you are going to assess the quality of your results, and understand and appreciate the limitations of the various assumptions you are making – you need to understand fluid dynamics. In particular, you should familiarize yourself with the fundamentals of turbulence, and turbulence modeling.
(3) Avoid building cases from scratch
Whenever I start a new case, I find the tutorial case that most closely matches what I am trying to accomplish. This greatly speeds things up. It will take you a super long time to set up any case from scratch – and you’ll probably make a bunch of mistakes, forget key variable entries etc. The OpenFOAM developers have done a lot of work setting up the tutorial cases for you, so use them!
As you continue to work in OpenFOAM on different projects, you should be compiling a library of your own templates based on previous work.
(4) Using Ubuntu makes things much easier
This is strictly my opinion. But I have found this to be true. Yes its true that Ubuntu has its own learning curve, but I have found that OpenFOAM works seamlessly in the Ubuntu or any Ubuntu-like linux environment. OpenFOAM now has Windows flavors using docker and the like- but I can’t really speak to how well they work – mostly because I’ve never bothered. Once you unlock the power of Linux – the only reason to use Windows is for Microsoft Office (I guess unless you’re a gamer – and even then more and more games are now on Linux). Not only that- but the VAST majority of forums and troubleshooting associated with OpenFOAM you’ll find on the internet are from Ubuntu users.
I much prefer to use Ubuntu with a virtual Windows environment inside it. My current office setup is my primary desktop running Ubuntu – plus a windows VirtualBox, plus a laptop running windows that I use for traditional windows type stuff. Dual booting is another option, but seamlessly moving between the environments is easier.
(5) If you’re struggling, simplify
Unless you know exactly what you are doing, you probably shouldn’t dive into the most complicated version of whatever you are trying to solve/study. It is best to start simple, and layer the complexity on top. This way, when something goes wrong, it is much easier to figure out where the problem is coming from.
(6) Familiarize yourself with the cfd-online forum
If you are having trouble, the cfd-online forum is super helpful. Most likely, someone else is has had the same problem you have. If not, the people there are extremely helpful and overall the forum is an extremely positive environment for working out the kinks with your simulations.
(7) The results from checkMesh matter
If you run checkMesh and your mesh fails – fix your mesh. This is important. Especially if you are not planning on familiarizing yourself with the available numerical schemes in OpenFOAM, you should at least have a beautiful mesh. In particular, if your mesh is highly non-orthogonal, you will have serious problems. If you insist on using a bad mesh, you will probably need to manipulate the numerical schemes. A great source for how schemes should be manipulated based on mesh non-orthogonality is:
http://www.wolfdynamics.com/wiki/OFtipsandtricks.pdf
(8) CFL Number Matters
If you are running a transient case, the Courant-Freidrechs-Lewis (CFL) number matters… a lot. Not just for accuracy (if you are trying to capture a transient event) but for stability. If your time-step is too large you are going to have problems. There is a solid mathematical basis for this stability criteria for advection-diffusion problems. Additionally the Navier-Stokes equations are very non-linear and the complexity of the problem and the quality of your grid etc can make the simulation even less stable. When I have a transient simulation crash, if I know my mesh is OK, I decrease the timestep by a factor of 2. More often than not, this solves the problem.
For large time stepping, you can add outer loops to solvers based on the pimple algorithm, but you may end up losing important transient information. Excellent explanation of how to do this is given in the book by T. Holzmann:
https://holzmann-cfd.de/publications/mathematics-numerics-derivations-and-openfoam
For the record, this points falls into point (1) of Understanding CFD.
(9) Work through the OpenFOAM Wiki “3 Week” Series
If you are starting OpenFOAM for the first time, it is worth it to work through an organized program of learning. One such example (and there are others) is the “3 Weeks Series” on the OpenFOAM wiki:
https://wiki.openfoam.com/%223_weeks%22_series
If you are a graduate student, and have no job to do other than learn OpenFOAM, it will not take 3 weeks. This touches on all the necessary points you need to get started.
(10) OpenFOAM is not a second-tier software – it is top tier
I know some people who have started out with the attitude from the get-go that they should be using a different software. They think somehow Open-Source means that it is not good. This is a pretty silly attitude. Many top researchers around the world are now using OpenFOAM or some other open source package. The number of OpenFOAM citations has grown every year consistently (
https://www.linkedin.com/feed/update/urn:li:groupPost:1920608-6518408864084299776/?commentUrn=urn%3Ali%3Acomment%3A%28groupPost%3A1920608-6518408864084299776%2C6518932944235610112%29&replyUrn=urn%3Ali%3Acomment%3A%28groupPost%3A1920608-6518408864084299776%2C6518956058403172352%29).
In my opinion, the only place where mainstream commercial CFD packages will persist is in industry labs where cost is no concern, and changing software is more trouble than its worth. OpenFOAM has been widely benchmarked, and widely validated from fundamental flows to hypersonics (see any of my 17 publications using it for this). If your results aren’t good, you are probably doing something wrong. If you have the attitude that you would rather be using something else, and are bitter that your supervisor wants you to use OpenFOAM, when something goes wrong you will immediately think there is something wrong with the program… which is silly – and you may quit.
(11) Meshing… Ugh Meshing
For the record, meshing is an art in any software. But meshing is the only area where I will concede any limitation in OpenFOAM. HOWEVER, as I have outlined in my previous post (https://curiosityfluids.com/2019/02/14/high-level-overview-of-meshing-for-openfoam/) most things can be accomplished in OpenFOAM, and there are enough third party meshing programs out there that you should have no problem.
Basically, if you are starting out in CFD or OpenFOAM, you need to put in time. If you are expecting to be able to just sit down and produce magnificent results, you will be disappointed. You might quit. And frankly, thats a pretty stupid attitude. However, if you accept that CFD and fluid dynamics in general are massive fields under constant development, and are willing to get up to speed, there are few limits to what you can accomplish.
Please take the time! If you want to do CFD, learning OpenFOAM is worth it. Seriously worth it.
This offering is notapproved or endorsed by OpenCFD Limited, producer and distributorof the OpenFOAM software via http://www.openfoam.com, and owner of theOPENFOAM® andOpenCFD® trade marks.
Here I will present something I’ve been experimenting with regarding a simplified workflow for meshing airfoils in OpenFOAM. If you’re like me, (who knows if you are) I simulate a lot of airfoils. Partly because of my involvement in various UAV projects, partly through consulting projects, and also for testing and benchmarking OpenFOAM.
Because there is so much data out there on airfoils, they are a good way to test your setups and benchmark solver accuracy. But going from an airfoil .dat coordinate file to a mesh can be a bit of pain. Especially if you are starting from scratch.
The two main ways that I have meshed airfoils to date has been:
(a) Mesh it in a C or O grid in blockMesh (I have a few templates kicking around for this
(b) Generate a “ribbon” geometry and mesh it with cfMesh
(c) Or back in the day when I was a PhD student I could use Pointwise – oh how I miss it.
But getting the mesh to look good was always sort of tedious. So I attempted to come up with a python script that takes the airfoil data file, minimal inputs and outputs a blockMeshDict file that you just have to run.
The goals were as follows:
(a) Create a C-Grid domain
(b) be able to specify boundary layer growth rate
(c) be able to set the first layer wall thickness
(e) be mostly automatic (few user inputs)
(f) have good mesh quality – pass all checkMesh tests
(g) Quality is consistent – meaning when I make the mesh finer, the quality stays the same or gets better
(h) be able to do both closed and open trailing edges
(i) be able to handle most airfoils (up to high cambers)
(j) automatically handle hinge and flap deflections
In Rev 1 of this script, I believe I have accomplished (a) thru (g). Presently, it can only hand airfoils with closed trailing edge. Hinge and flap deflections are not possible, and highly cambered airfoils do not give very satisfactory results.
There are existing tools and scripts for automatically meshing airfoils, but I found personally that I wasn’t happy with the results. I also thought this would be a good opportunity to illustrate one of the ways python can be used to interface with OpenFOAM. So please view this as both a potentially useful script, but also something you can dissect to learn how to use python with OpenFOAM. This first version of the script leaves a lot open for improvement, so some may take it and be able to tailor it to their needs!
Hopefully, this is useful to some of you out there!
You can download the script here:
https://github.com/curiosityFluids/curiosityFluidsAirfoilMesher
Here you will also find a template based on the airfoil2D OpenFOAM tutorial.
(1) Copy curiosityFluidsAirfoilMesher.py to the root directory of your simulation case.
(2) Copy your airfoil coordinates in Selig .dat format into the same folder location.
(3) Modify curiosityFluidsAirfoilMesher.py to your desired values. Specifically, make sure that the string variable airfoilFile is referring to the right .dat file
(4) In the terminal run: python3 curiosityFluidsAirfoilMesher.py
(5) If no errors – run blockMesh
PS
You need to run this with python 3, and you need to have numpy installed
The inputs for the script are very simple:
ChordLength: This is simply the airfoil chord length if not equal to 1. The airfoil dat file should have a chordlength of 1. This variable allows you to scale the domain to a different size.
airfoilfile: This is a string with the name of the airfoil dat file. It should be in the same folder as the python script, and both should be in the root folder of your simulation directory. The script writes a blockMeshDict to the system folder.
DomainHeight: This is the height of the domain in multiples of chords.
WakeLength: Length of the wake domain in multiples of chords
firstLayerHeight: This is the height of the first layer. To estimate the requirement for this size, you can use the curiosityFluids y+ calculator
growthRate: Boundary layer growth rate
MaxCellSize: This is the max cell size along the centerline from the leading edge of the airfoil. Some cells will be larger than this depending on the gradings used.
The following inputs are used to improve the quality of the mesh. I have had pretty good results messing around with these to get checkMesh compliant grids.
BLHeight: This is the height of the boundary layer block off of the surfaces of the airfoil
LeadingEdgeGrading: Grading from the 1/4 chord position to the leading edge
TrailingEdgeGrading: Grading from the 1/4 chord position to the trailing edge
inletGradingFactor: This is a grading factor that modifies the the grading along the inlet as a multiple of the leading edge grading and can help improve mesh uniformity
trailingBlockAngle: This is an angle in degrees that expresses the angles of the trailing edge blocks. This can reduce the aspect ratio of the boundary cells at the top and bottom of the domain, but can make other mesh parameters worse.
Inputs:
With the above inputs, the grid looks like this:
Mesh Quality:
These are some pretty good mesh statistics. We can also view them in paraView:
The clark-y has some camber, so I thought it would be a logical next test to the previous symmetric one. The inputs I used are basically the same as the previous airfoil:
With these inputs, the result looks like this:
Mesh Quality:
Visualizing the mesh quality:
Here is an example of a flying with airfoil (tested since the trailing edge is tilted upwards).
Inputs:
Again, these are basically the same as the others. I have found that with these settings, I get pretty consistently good results. When you change the MaxCellSize, firstLayerHeight, and Grading some modification may be required. However, if you just half the maxCell, and half the firstLayerHeight, you “should” get a similar grid quality just much finer.
Grid Quality:
Visualizing the grid quality
Hopefully some of you find this tool useful! I plan to release a Rev 2 soon that will have the ability to handle highly cambered airfoils, and open trailing edges, as well as control surface hinges etc.
The long term goal will be an automatic mesher with an H-grid in the spanwise direction so that the readers of my blog can easily create semi-span wing models extremely quickly!
Comments and bug reporting encouraged!
DISCLAIMER: This script is intended as an educational and productivity tool and starting point. You may use and modify how you wish. But I make no guarantee of its accuracy, reliability, or suitability for any use. This offering is not approved or endorsed by OpenCFD Limited, producer and distributor of the OpenFOAM software via http://www.openfoam.com, and owner of the OPENFOAM® and OpenCFD® trademarks.
Here is a useful little tool for calculating the properties across a normal shock.
If you found this useful, and have the need for more, visit www.stfsol.com. One of STF Solutions specialties is providing our clients with custom software developed for their needs. Ranging from custom CFD codes to simpler targeted codes, scripts, macros and GUIs for a wide range of specific engineering purposes such as pipe sizing, pressure loss calculations, heat transfer calculations, 1D flow transients, optimization and more. Visit STF Solutions at www.stfsol.com for more information!
Disclaimer: This calculator is for educational purposes and is free to use. STF Solutions and curiosityFluids makes no guarantee of the accuracy of the results, or suitability, or outcome for any given purpose.
Happy 2021!
The year of 2020 will be remembered in history more than the year of 1918, when the last great pandemic hit the globe. As we speak, daily new cases in the US are on the order of 200,000, while the daily death toll oscillates around 3,000. According to many infectious disease experts, the darkest days may still be to come. In the next three months, we all need to do our very best by wearing a mask, practicing social distancing and washing our hands. We are also seeing a glimmer of hope with several recently approved COVID vaccines.
2020 will be remembered more for what Trump tried and is still trying to do, to overturn the results of a fair election. His accusations of wide-spread election fraud were proven wrong in Georgia and Wisconsin through multiple hand recounts. If there was any truth to the accusations, the paper recounts would have uncovered the fraud because computer hackers or software cannot change paper votes.
Trump's dictatorial habits were there for the world to see in the last four years. Given another 4-year term, he might just turn a democracy into a Trump dictatorship. That's precisely why so many voted in the middle of a pandemic. Biden won the popular vote by over 7 million, and won the electoral college in a landslide. Many churchgoers support Trump because they dislike Democrats' stances on abortion, LGBT rights, et al. However, if a Trump dictatorship becomes reality, religious freedom may not exist any more in the US.
Is the darkest day going to be January 6th, 2021, when Trump will make a last-ditch effort to overturn the election results in the Electoral College certification process? Everybody knows it is futile, but it will give Trump another opportunity to extort money from his supporters.
But, the dawn will always come. Biden will be the president on January 20, 2021, and the pandemic will be over, perhaps as soon as 2021.
The future of CFD is, however, as bright as ever. On the front of large eddy simulation (LES), high-order methods and GPU computing are making LES more efficient and affordable. See a recent story from GE.
![]() |
Figure 1. Various discretization stencils for the red point |
![]() ![]() |
p = 1 |
![]() ![]() |
p = 2 |
![]() ![]() |
p = 3 |
CL
| CD
| |
p = 1 | 2.020 | 0.293 |
p = 2 | 2.411 | 0.282 |
p = 3 | 2.413 | 0.283 |
Experiment | 2.479 | 0.252 |
We’ve reached the end of 2020, and I think it’s fair to say this year did not go as planned. The coronavirus pandemic disrupted our lives and brought on unexpected challenges and hardships. However, this difficult time has also highlighted the resiliency of people all around the globe—we have adapted and innovated to meet these challenges head on. At Convergent Science, that meant finding new ways to communicate and collaborate to ensure we could continue to deliver the best possible software and support to our users, all while keeping our employees safe.
Despite the pandemic, we experienced exciting opportunities, advancements, and milestones at Convergent Science this past year. We hosted two virtual conferences, continued to expand into new markets and new application areas, began new collaborations, increased our employee count, and, of course, continued to improve and develop CONVERGE.
We have spent much of 2020 developing the next major release of our CONVERGE CFD software: version 3.1. There’s a lot to look forward to in CONVERGE 3.1, which will be released next year. In CONVERGE 3.0, we added the ability to incorporate stationary inlaid meshes into a simulation. In 3.1, these inlaid meshes will be able to move within the underlying Cartesian grid. For example, you will be able to create an inlaid mesh around each of the intake valves in an IC engine simulation, and the mesh will move with the valve as it opens and closes. With this method, you can achieve high grid resolution normal to the valve surface using significantly fewer cells than with traditional fixed embedding.
Another enhancement will allow you to use different solvers, meshes, physical models, and chemical mechanisms for different streams (i.e., portions of the domain). This means you will be able to tailor your simulation settings to each stream, which will improve solver speed and numerical performance. CONVERGE 3.1 will also feature new sealing capabilities that enable you to have any objects come into contact with one another in your simulation or have objects enter or leave your simulation.
Furthermore, CONVERGE 3.1 will support solid- and gas-phase parcels in addition to the traditional liquid-phase parcels. This can be useful when modeling, for example, soot or injectors operating at flash-boiling conditions. CONVERGE 3.1 will also feature an improved steady-state solver that will provide significant improvements in speed, and we have enhanced our fluid-structure interaction, volume of fluid, combustion, and emissions modeling capabilities. There are many more exciting features and enhancements coming in 3.1, so stay tuned for more information!
Improving the scalability of CONVERGE continues to be a strong focus of our development efforts. We work with several companies and institutions, testing CONVERGE on different high-performance computing (HPC) architectures and optimizing our software to ensure good scaling. To that end, we were thrilled to begin a new collaboration this year with Oracle, a leader in cloud computing and enterprise software. In our benchmark testing, we have seen near perfect scaling of CONVERGE on Oracle Cloud Infrastructure on thousands of cores. This collaboration presents a great opportunity for CONVERGE users to take advantage of Oracle’s advanced HPC resources to efficiently run large-scale simulations in the cloud.
For the second year in a row, we were honored to win an HPCwire award for research performed with our colleagues at Aramco Research Center–Detroit and Argonne National Laboratory. This year, we received the HPCwire Readers’ Choice Award for Best Use of HPC in Industry for our work using HPC and machine learning to accelerate injector design optimization for next-generation high-efficiency, low-emissions engines. Our collaborative work is forging the way to leverage HPC, novel experimental measurements, and CFD to perform rapid optimization studies and reduce our carbon footprint from transportation.
In another collaborative effort, the Computational Chemistry Consortium (C3) made significant progress in 2020. Co-founded by Convergent Science, C3 is working to create the most accurate and comprehensive chemical reaction mechanism for automotive fuels that includes NOx and PAH chemistry to model emissions. The first version of the mechanism was completed last year and is currently available to C3’s industry sponsors. Once the mechanism is published, it will be released to the public on fuelmech.org. This past year, C3 has continued to refine the mechanism, which has now reached version 2.1. The results of these efforts have been rewarding—we’ve seen a significant decrease in error in selected validation cases. The next year of the consortium will focus on increasing the accuracy of the NOx and PAH chemistry. To that end, C3 welcomed a new member this year, Dr. Stephen Klippenstein from Argonne National Laboratory. Dr. Klippenstein will perform high-level ab initio calculations of rate constants in NOx chemistry. Ultimately, the C3 mechanism is expected to be the first publicly available mechanism that includes everything from hydrogen chemistry all the way up to PAH chemistry in a single high-fidelity mechanism.
In 2020, we celebrated our 10-year anniversary of collaboration with Argonne National Laboratory. Over the past decade, this collaboration has helped us extend CONVERGE’s capabilities and broach new application areas. We have performed cutting-edge research in the transportation field, developing new methods and models that are proving to be instrumental in designing the next generation of engines. In the aerospace field, we’ve broken ground in applying CFD to gas turbines, rotating detonation engines, drones, and more. We’ve made great strides in the last ten years, and we’re looking forward to the next decade of collaboration!
Every year, we look forward to getting together with our users, discussing the latest exciting CONVERGE research and having some fun at our user conferences. When the pandemic struck and countries began locking down earlier this year, we were determined to still hold our 2020 CONVERGE User Conference–Europe, even if it looked a bit different. Our conference was scheduled for the end of March, so we didn’t have much time to transition from an in-person to an online event, but our team was up for the challenge. In less than three weeks, we planned a whole new event and successfully held one of the first pandemic-era virtual conferences. We were so pleased with the result! More than 400 attendees from around the world tuned in for an excellent lineup of technical presentations, which spanned topics from IC engines to compressors to electric motors and battery packs.
While we hoped to hold our North American user conference in Detroit later in the year, the continued pandemic made that impossible. Once again, we took to the internet. We incorporated some more networking opportunities, including various social groups and discussion topics, and created some fun polls to help attendees get to know one another. We were also able to offer our usual slate of conference-week CONVERGE training and virtual exhibit booths for our sponsors. The presentations at this conference showcased the breadth and diversity of applications for which CONVERGE is suited, with speakers discussing rockets, gas turbines, exhaust aftertreatment, biomedical applications, renewable energy, and electromobility in addition to a host of IC engine-related topics.
It’s hard to know what 2021 will look like, but rest assured we will be hosting more conferences, virtual or otherwise. We’re looking forward to the day we can get together in person once again!
Even with the pandemic, 2020 was an exciting and productive year for Convergent Science around the globe. We gained nearly a dozen new employees, including bringing on team members in newly created roles to help expand our relationships with universities and to increase our in-house CAD design capabilities. We also continued to find new markets for CONVERGE as we entered the emobility, rocket, and burner industries.
Our Indian office flourished in 2020. Since its creation three years ago, Convergent Science India has grown to more than 20 employees, adding nine new team members this year alone. To accommodate our growing team, we moved to a spacious new building in Pune. Our team in India expanded our global reach, bringing new academic and industry clients on board. In addition, we continued to work on growing our presence in new applications such as gas turbines, aftertreatment, motor cooling, battery failure, oil churning, and spray painting.
In Europe, despite the challenging circumstances, we increased our client base and our license sales considerably, and we were able to successfully and seamlessly support our customers to help them achieve their CFD goals. In addition to moving our European CONVERGE user conference online in record time, we attended and exhibited at many virtual tradeshows and events and are looking forward to attending in-person conferences as soon as it is safe to do so.
Our partners at IDAJ continued to do excellent work supporting our customers in Japan, China, and Korea. Due to the pandemic, they held their first-ever IDAJ Conference Online 2020, where they had both live lectures and Q&A sessions as well as on-demand streaming content. While they support many IC engine clients, they are also supporting clients working on other applications such as motor cooling, battery failure, oil churning, and spray painting.
2020 was a difficult year for many of us, but I am impressed and inspired by the way the CFD community and beyond has come together to make the most of a challenging situation. And the future looks bright! We’re looking forward to releasing CONVERGE 3.1 and helping our users take advantage of the increased functionality and new features that will be available. We’re excited to expand our presence in electromobility, renewable energy, aerospace, and other new fields. In the upcoming year, we look forward to forming new collaborations and strengthening existing partnerships to promote innovation and keep CONVERGE on the cutting-edge of CFD software.
Can we help you meet your 2021 CFD goals? Contact us today!
In my first year of graduate school, a friend always filled up her water bottle, dropped some ice cubes into it, and then shook it up in order to cool the water faster. If she had added the ice cubes and let the water bottle sit, eventually all the water would equilibrate to the same temperature, but that would take a while without any movement—the water next to the ice cubes would cool down quickly, but the water farther away would cool down at a much slower rate. By shaking it up, she agitated the water and ice so that the ice came into contact with more of the warm water that needed to be cooled. This “cocktail shaker effect,” I would later find out, also applies to cooling engines.
Combustion in an internal combustion (IC) engine occurs on top of the piston, which means that there is an extraordinary amount of heat generated on the piston crown. If left unmediated, this heat can cause the piston to break. The threat of piston damage is particularly high in diesel engines because more heat is generated in the cylinder than in a traditional gasoline engine. Unlike a bottle of warm water, though, we can’t just drop a few ice cubes into the cylinder to act as a heat sink.
Here we see how engineers can use CONVERGE to efficiently solve the problem of cooling the piston so that it isn’t damaged by heat. The idea is simple—use engine oil as a heat sink—but the implementation is complex since the piston is constantly moving and nothing can be in contact with the piston crown inside the cylinder.
Since the heat sink can’t be inside the cylinder on the piston crown, there is an oil gallery in contact with the undercrown of the piston, as shown in Figure 1. Engine oil is taken through a pump, pressurized, and constantly sprayed at the oil gallery inlet hole. In the video below, you will see how the oil enters the gallery, and, as the piston motion continues, the oil sloshes inside the oil gallery, absorbing heat from the piston before exiting the outlet hole on the other side of the gallery.
There are several factors that are important to consider when designing this type of cooling system, all of which CONVERGE is well-equipped to handle. What size and shape should the inlet and outlet holes be to capture the stream of oil? How much oil will enter the gallery compared to how much was sprayed (i.e., capture ratio)? What is the best design of the gallery so that the oil effectively absorbs heat from the piston? What ratio of the gallery volume should be occupied (i.e., fill ratio) to ensure that the oil can move and absorb heat efficiently? CONVERGE provides answers to these questions and others through a volume of fluid (VOF) simulation.
Because a simple boundary condition is not predictive of the heat transfer throughout the entire piston, we use conjugate heat transfer (CHT) to more accurately predict the piston cooling by solving the heat distribution inside the piston. Understanding how heat transfer affects the whole piston is an essential step toward designing a geometry that will effectively cool more than just the piston surface. While CHT can be computationally expensive due to the difference in time-scales of heat transfer in the solid and fluid regions, CONVERGE provides the option to use super-cycling, which can significantly reduce the computational cost of this type of simulation.
In the video below, you will see how the above factors have been optimized to dissipate heat from the piston crown and throughout the piston as a whole. In the video on the left, you can watch the temperature contours change during the simulation as heat dissipates. The second view shows how CONVERGE’s Adaptive Mesh Refinement (AMR) is in action throughout the simulation, providing increased grid resolution near the inlet and around the oil gallery, where it is needed most.
Ready to run your own simulations to optimize oil jet piston cooling? Contact us today!
From the Argonne National Laboratory + Convergent Science Blog Series
Through the collaboration between Argonne National Laboratory and Convergent Science, we provide fundamental research that enables manufacturers to design cleaner and more efficient engines by optimizing combustion.
–Doug Longman, Manager of Engine Research at Argonne National Laboratory
The internal combustion engine has come a long way since its inception—the engine in your car today is significantly quieter, cleaner, and more efficient than its 1800s-era counterpart. For many years, the primary means of achieving these advances was experimentation. Indeed, we have experiments to thank for a myriad of innovations, from fuel injection systems to turbocharging to Wankel engines.
More recently, a new tool was added to the engine designer’s toolbox: simulation. Beginning in the 1970s and ‘80s, computational fluid dynamics (CFD) opened the door to a new level of refinement and optimization.
“One of the really cool things about simulation is that you can look at physics that cannot be easily captured in an experiment—details of the flow that might be blocked from view, for example,” says Eric Pomraning, Co-Owner of Convergent Science.
Of course, experiments remain vitally important to engine research, since CFD simulations model physical processes, and experiments are necessary to validate your results and ground your simulations in reality.
Argonne National Laboratory and Convergent Science combine these two approaches—experiments and simulation—to further improve the internal combustion engine. Two of the main levers we have to control the efficiency and emissions of an engine are the fuel injection system and the ignition system, both of which have been significant areas of focus during the collaboration.
The combustion process in an internal combustion engine really begins with fuel injection. The physics of injection determine how the fuel and air in the cylinder will mix, ignite, and ultimately combust.
Argonne National Laboratory is home to the Advanced Photon Source (APS), a DOE Office of Science User Facility. The APS provides a unique opportunity to characterize the internal passages of injector nozzles with incredibly high spatial resolution through the use of high-energy x-rays. This data is invaluable for developing accurate CFD models that manufacturers can use in their design processes.
Early on in the collaboration, Christopher Powell, Principal Engine Research Scientist at Argonne, and his team leveraged the APS to investigate needle motion in an injector.
“Injector manufacturers had long suspected that off-axis motion of the injector valve could be present. But they never had a way to measure it before, so they weren’t sure how it impacted fuel injection,” says Chris.
The x-ray studies performed at the APS were the first in the world to confirm that some injector needles do exhibit radial motion in addition to the intended axial motion, a phenomenon dubbed “needle wobble.” Argonne and Convergent Science engineers simulated this experimental data in CONVERGE, prescribing radial motion to the injector needle. They found that needle wobble can substantially impact the fuel distribution as it exits the injector. Manufacturers were able to apply the results of this research to design injectors with a more predictable spray pattern, which, in turn, leads to a more predictable combustion event.
More recently, researchers at Argonne have used the APS to investigate the shape of fuel injector flow passages and characterize surface roughness. Imperfections in the geometry can influence the spray and the subsequent downstream engine processes.
“If we use a CAD geometry, which is smooth, we will miss out on some of the physics, like cavitation, that can be triggered by surface imperfections,” says Sameera Wijeyakulasuriya, Senior Principal Engineer at Convergent Science. “But if we use the x-ray scanned geometry, we can incorporate those surface imperfections into our numerical models, so we can see how the flow field behaves and responds.”
Argonne and Convergent Science engineers performed internal nozzle flow simulations that used the real injector geometries and that incorporated real needle motion.1 Using the one-way coupling approach in CONVERGE, they mapped the results of the internal flow simulations to the exit of each injector orifice to initialize a multi-plume Lagrangian spray simulation. As you can see in Figure 1, the surface roughness and needle motion significantly impact the spray plume—the one-way coupling approach captures features that the standard rate of injection (ROI) method could not. In addition, the real injector parameters introduce orifice-to-orifice variability, which affects the combustion behavior down the line.
The real injector geometries not only allow for more accurate computational simulations, but they also can serve as a diagnostic tool for manufacturers to assess how well their manufacturing processes are producing the desired nozzle shape and size.
Accurately characterizing fuel injection sets the stage for the next lever we can optimize in our engine: ignition. In spark-ignition engines, the ignition event initiates the formation of the flame kernel, the growth of the flame kernel, and the flame propagation mechanism.
“In the past, ignition was just modeled as a hot source—dumping an amount of energy in a small region and hoping it transitions to a flame. The amount of physics in the process was very limited,” says Sibendu Som, Manager of the Computational Multi-Physics Section at Argonne.
These simplified models are adequate for most stable engine conditions, but you can run into trouble when you start simulating more advanced combustion concepts. In these scenarios, the simplified ignition models fall short in replicating experimental data. Over the course of their collaboration, Argonne and Convergent Science have incorporated more physics into ignition models to make them robust for a variety of engine conditions.
For example, high-performance spark-ignition engines often feature high levels of dilution and increased levels of turbulence. These conditions can have a significant impact on the ignition process, which consequently affects combustion stability and cycle-to-cycle variation (CCV). To capture the elongation and stretch experienced by the spark channel under highly turbulent conditions, Argonne and Convergent Science engineers developed a new ignition model, the hybrid Lagrangian-Eulerian spark-ignition (LESI) model.
In Figure 2, you can see that the LESI model more accurately captures the behavior of the spark under turbulent conditions compared to a commonly used energy deposition model.2 The LESI model will be available in future versions of CONVERGE, accessible to manufacturers to help them better understand ignition and mitigate CCV.
Ideally, every cycle of an internal combustion engine would be exactly identical to ensure smooth operation. In real engines, variability in the injection, ignition, and combustion means that not every cycle will be the same. Cyclic variability is especially prevalent in high-efficiency engines that push the limits of combustion stability. Extreme cycles can cause engine knock and misfires—and they can influence emissions.
“Not every engine cycle generates significant emissions. Often they’re primarily formed only during rare cycles—maybe one or two out of a hundred,” says Keith Richards, Co-Owner of Convergent Science. “Being able to capture cyclic variability will ultimately allow us to improve our predictive capabilities for emissions.”
Modeling CCV requires simulating numerous engine cycles, which is a highly (and at times prohibitively) time-consuming process. Several years ago, Keith suggested a potential solution—starting several engine cycles concurrently, each with a small perturbation to the flow field, which allows each simulation to develop into a unique solution.
Argonne and Convergent Science compared this approach—called the concurrent perturbation method (CPM)—to the traditional approach of simulating engine cycles consecutively. Figure 3 shows CCV results obtained using CPM compared to concurrently run cycles, which you can see match very well.3 This means that with sufficient computational resources, you can predict CCV in the amount of time it takes to run a single engine cycle.
The study described above, and the vast majority of all CCV simulation studies, use large eddy simulations (LES), because LES allows you to resolve some of the turbulence scales that lead to cyclic variability. Reynolds-Averaged Navier-Stokes (RANS), on the other hand, provides an ensemble average that theoretically damps out variations between cycles. At least this was the consensus among the engine modeling community until Riccardo Scarcelli, a Research Scientist at Argonne, noticed something strange.
“I was running consecutive engine cycle simulations to move away from the initial boundary conditions, and I realized that the cycles were never converged to an average solution—the cycles were never like the cycle before or the cycle after,” Riccardo says. “And that was strange because I was using RANS, not LES.”
Argonne and Convergent Science worked together to untangle this mystery, and they discovered that RANS is able to capture the deterministic component of CCV. RANS has long been the predominant turbulence model used in engine simulations, so how had this phenomenon gone unnoticed? In the past, most engine simulations modeled conventional combustion, which shows little cyclic variability in practice in either diesel or gasoline engines. The more complex combustion regimes simulated today—along with the use of finer grids and more accurate numerics—allows RANS to pick up on some of the cycle-to-cycle variations that these engines exhibit in the real world. While RANS will not provide as accurate a picture as LES, it can be a useful tool to capture CCV trends. Additionally, RANS can be run on a much coarser mesh than LES, so you can get a faster turnaround on an inherently expensive problem, making CCV studies more practical for industry timelines.
The gains in understanding and improved models developed during the Argonne and Convergent Science collaboration provide great benefit to the engine community. One of the primary missions of Argonne National Laboratory is to transfer knowledge and technology to industry. To that end, the models developed during the collaboration will continue to be implemented in CONVERGE, putting the technology in the hands of manufacturers, so they can create better engines.
What can we look forward to in the future? There will continue to be a strong focus on developing high fidelity numerics, expanding and improving chemistry tools and mechanisms, integrating machine learning into the simulation process, and speeding up CFD simulations—establishing more efficient models and further increasing the scalability of CONVERGE to take advantage of the latest computational resources. Moreover, we can look forward to seeing the innovations of the last decade of collaboration incorporated into the engines of the next decade, bringing us closer to a clean transportation future.
In case you missed the other posts in the series, you can find them here:
[1] Torelli, R., Matusik, K.E., Nelli, K.C., Kastengren, A.L., Fezzaa, K., Powell, C.F., Som, S., Pei, Y., Tzanetakis, T., Zhang, Y., Traver, M., and Cleary, D.J., “Evaluation of Shot-to-Shot In-Nozzle Flow Variations in a Heavy-Duty Diesel Injector Using Real Nozzle Geometry,” SAE Paper 2018-01-0303, 2018. DOI: 10.4271/2018-01-0303
[2] Scarcelli, R., Zhang, A., Wallner, T., Som, S., Huang, J., Wijeyakulasuriya, S., Mao, Y., Zhu, X., and Lee, S.-Y., “Development of a Hybrid Lagrangian–Eulerian Model to Describe Spark-Ignition Processes at Engine-Like Turbulent Flow Conditions,” Journal of Engineering for Gas Turbines and Power, 141(9), 2019. DOI: 10.1115/1.4043397
[3] Probst, D., Wijeyakulasuriya, S., Pomraning, E., Kodavasal, J., Scarcelli, R., and Som, S., “Predicting Cycle-to-Cycle Variation With Concurrent Cycles In A Gasoline Direct Injected Engine With Large Eddy Simulations”, Journal of Energy Resources Technology, 142(4), 2020. DOI: 10.1115/1.4044766
Renewable energy is being generated at unprecedented levels in the United States, and those levels will only continue to rise. The growth in renewable energy has been driven largely by wind power—over the last decade, wind energy generation in the U.S. has increased by 400% 1. It’s easy to see why wind power is appealing. It’s sustainable, cost-effective, and offers the opportunity for domestic energy production. But, like all energy sources, wind power doesn’t come without drawbacks. Concerns have been raised about land use, noise, consequences to wildlife habitats, and the aesthetic impact of wind turbines on the landscape 2.
However, there is a potential solution to many of these issues: what if you move wind turbines offshore? In addition to mitigating concerns over land use, noise, and visual impact, offshore wind turbines offer several other advantages. Compared to onshore, wind speeds offshore tend to be higher and steadier, leading to large gains in energy production. Also, in the U.S., a large portion of the population lives near the coasts or in the Great Lakes region, which minimizes problems associated with transporting wind-generated electricity. But despite these advantages, only 0.03% of the U.S. wind-generating capacity in 2018 came from offshore wind plants 1. So why hasn’t offshore wind energy become more prevalent? Well, one of the major challenges with offshore wind energy is a problem of engineering—wind turbine support structures must be designed to withstand the significant wind and wave loads offshore.
Today, there are computational tools that engineers can use to help design optimized support structures for offshore wind turbines. Namely, computational fluid dynamics (CFD) simulations can offer valuable insight into the interaction between waves and the wind turbine support structures.
Hannah Johlas is an NSF Graduate Research Fellow in Dr. David Schmidt’s lab at the University of Massachusetts Amherst. Hannah uses CFD to study fixed-bottom offshore wind turbines at shallow-to-intermediate water depths (up to approximately 50 meters deep). Turbines located at these depths are of particular interest because of a phenomenon called breaking waves. As waves move from deeper to shallower water, the wavelength decreases and the wave height increases in a process called shoaling. If a wave becomes steep enough, the crest can overturn and topple forward, creating a breaking wave. Breaking waves can impart substantial forces onto turbine support structures, so if you’re planning to build a wind turbine in shallower water, it’s important to know if that turbine might experience breaking waves.
Hannah uses CONVERGE CFD software to predict if waves are likely to break for ocean characteristics common to potential offshore wind turbine sites along the east coast of the U.S. She also predicts the forces from breaking waves slamming into the wind turbine support structures. The results of the CONVERGE simulations are then used to evaluate the accuracy of simplified engineering models to determine which models best capture wave behavior and wave forces and, thus, which ones should be used when designing wind turbines.
In this study, Hannah simulated 39 different wave trains in CONVERGE using a two-phase finite volume CFD model 3. She leveraged the volume of fluid (VOF) method with the Piecewise Linear Interface Calculation scheme to capture the air-water interface. Additionally, automated meshing and Adaptive Mesh Refinement ensured accurate results while minimizing the time to set up and run the simulations.
“CONVERGE’s adaptive meshing helps simulate fluid interfaces at reduced computational cost,” Hannah says. “This feature is particularly useful for resolving the complex air-water interface in breaking wave simulations.”
Some of the breaking waves were then simulated slamming into monopiles, the large cylinders used as support structures for offshore wind turbines in shallow water. The results of these CONVERGE simulations were validated against experimental data before being used to evaluate the simplified engineering models.
Four common models for predicting whether a wave will break (McCowan, Miche, Battjes, and Goda) were assessed. The models were evaluated by how frequently they produced false positives (i.e., the model predicts a wave should break, but the simulated wave does not break) and false negatives (i.e., the model predicts a wave should not break, but the simulated wave does break) and how well they predicted the steepness of the breaking waves. False positives are preferable to false negatives when designing a conservative support structure, since breaking wave loads are usually higher than non-breaking waves.
The study results indicate that none of the models perform well under all conditions, and instead which model you should use depends on the characteristics of the ocean at the site you’re considering.
“For sites with low seafloor slopes, the Goda model is the best at conservatively predicting whether a given wave will break,” Hannah says. “For higher seafloor slopes, the Battjes model is preferred.”
Four slam force models were also evaluated: Goda, Campbell-Weynberg, Cointe-Armand, and Wienke-Oumerachi. The slam models and the simulated CFD wave forces were compared for their peak total force, their force time history, and breaking wave shape.
The results show that all four slam models are conservative (i.e., predict higher peak forces than the simulated waves) and assume the worst-case shape for the breaking wave during impact. The Goda slam model is the least conservative, while the Cointe-Armand and Wienke-Oumerachi slam models are the most conservative. All four models neglect the effects of runup on the monopiles, which was present in the CFD simulations. This could explain some of the discrepancies between the forces predicted by the engineering models and the CFD simulations.
Offshore wind energy is a promising technology for clean energy production, but to gain traction in the industry, there needs to be sound engineering models to use when designing the turbines. Hannah’s research provides guidelines on which engineering models should be used for a given set of ocean characteristics. Her results also highlight the areas that could be improved upon.
“The slam force models don’t account for variety in wave shape at impact or for wave runup on the monopiles,” Hannah says. “Future studies should focus on incorporating these factors into the engineering models to improve their predictive capabilities.”
CFD has a fundamental role to play in the development of renewable energy. CONVERGE’s combination of autonomous meshing, high-fidelity physical models, and ability to easily handle complex, moving geometries make it particularly well suited to the task. Whether it’s studying the interaction of waves with offshore turbines, optimizing the design of onshore wind farms, or predicting wind loads on solar panels, CONVERGE has the tools you need to help bring about the next generation of energy production.
Interested in learning more about Hannah’s research? Check out her paper here.
[1] Marcy, C., “U.S. renewable electricity generation has doubled since 2008,” https://www.eia.gov/todayinenergy/detail.php?id=38752, accessed on Nov 11, 2016.
[2] Center for Sustainable Systems, University of Michigan, “U.S. Renewable Energy Factsheet”, http://css.umich.edu/factsheets/us-renewable-energy-factsheet, accessed on Nov 11, 2016.
[3] Johlas, H.M., Hallowell, S., Xie, S., Lomonaco, P., Lackner, M.A., Arwade, S.A., Myers, A.T., and Schmidt, D.P., “Modeling Breaking Waves for Fixed-Bottom Support Structures for Offshore Wind Turbines,” ASME 2018 1st International Offshore Wind Technical Conference, IOWTC2018-1095, San Francisco, CA, United States, Nov 4–7, 2018. DOI: 10.1115/IOWTC2018-1095
Across industries, manufacturers share many of the same goals: create quality products, boost productivity, and reduce expenses. In the pumps and compressors business, manufacturers must contend with the complexity of the machines themselves in order to reach these goals. Given the intricate geometries, moving components, and tight clearances between parts, designing pumps and compressors to be efficient and reliable is no trivial matter.
First, assessing the device’s performance by building and testing a prototype can be time-consuming and costly. And when you’re performing a design study, machining and switching out various components further compounds your expenses. There are also limitations in how many instruments you can place inside the device and where you can place them, which can make fully characterizing the machine difficult. New methods for testing and manufacturing can help streamline this process, but there remains room for alternative approaches.
Computational fluid dynamics (CFD) offers significant advantages for designing pumps and compressors. Through CFD simulations, you can obtain valuable insight into the behavior of the fluid inside your machine and the interactions between the fluid and solid components—and CONVERGE CFD software is well suited for the task.
Designed to model three-dimensional fluid flows in systems with complex geometries and moving boundaries, CONVERGE is equipped to simulate any positive displacement or dynamic pump or compressor. And with a suite of advanced models, CONVERGE allows you to computationally study the physical phenomena that affect efficiency and reliability—such as surge, pressure pulsations, cavitation, and vibration—to design an optimal machine.
CFD provides a unique opportunity to visualize the inner workings of your machine during operation, generating data on pressures, temperatures, velocities, and fluid properties without the limitations of physical measurements. The entire flow field can be analyzed with CFD, including areas that are difficult or impossible to measure experimentally. This additional data allows you to comprehensively characterize your pump or compressor and pinpoint areas for improvement.
Since CONVERGE leads the way in predictive CFD technology, you can analyze pump and compressor designs that have not yet been built and still be confident in your results. Compared to building and testing prototypes, simulations are fast and inexpensive, and altering a computer-modeled geometry is trivial. Iterating through designs virtually and building only the most promising candidates reduces the expenses associated with the design process.
While three-dimensional CFD is fast compared to experimental methods, it is typically slower than one- or two-dimensional analysis tools, which are often incorporated into the design process. However, 1D and 2D methods are inherently limited in their ability to capture the 3D nature of physical flows, and thus can miss important flow phenomena that may negatively affect performance.
CONVERGE drastically reduces the time required to set up a 3D pump or compressor simulation with its autonomous meshing capabilities. Creating a mesh by hand—which is standard practice in many CFD programs—can be a weeks-long process, particularly for cases with complex moving geometries such as pumps and compressors. With autonomous meshing, CONVERGE automatically generates an optimized Cartesian mesh based on a few simple user-defined parameters, effectively eliminating all user meshing time.
In addition, the increased computational resources available today can greatly reduce the time requirements to run CFD simulations. CONVERGE is specifically designed to enable highly parallel simulations to run on many processors and demonstrates excellent scaling on thousands of cores. Additionally, Convergent Science partners with cloud service providers, who offer affordable on-demand access to the latest computing resources, making it simple to speed up your simulations.
Accurately capturing real-world physical phenomena is critical to obtaining useful simulation results. CONVERGE features robust fluid-structure interaction (FSI) modeling capabilities. For example, you can simulate the interaction between the bulk flow and the valves to predict impact velocity, fatigue, and failure points. CONVERGE also features a conjugate heat transfer (CHT) model to resolve spatially varying surface temperature distributions, and a multi-phase model to study cavitation, oil splashing, and other free surface flows of interest.
CONVERGE has been validated on numerous types of compressors and pumps1-10, and we will discuss two common applications below.
Scroll compressors are often used in air conditioning systems, and the major design goals for these machines today are reducing noise and improving efficiency. Scroll compressors consist of a stationary scroll and an orbiting scroll, which create a complex system that can be challenging to model. Some codes use a moving mesh to simulate moving boundaries, but this can introduce diffusive error that lowers the accuracy of your results. CONVERGE automatically generates a stationary mesh at each time-step to accommodate moving boundaries, which provides high numerical accuracy. In addition, CONVERGE employs a unique Cartesian cut-cell approach to perfectly represent your compressor geometry, no matter how complex.
In this study1, CONVERGE was used to simulate a scroll compressor with a deforming reed valve. An FSI model was used to capture the motion of the discharge reed valve. Figure 1 shows the CFD-predicted mass flow rate through the scroll compressor compared to experimental values. As you can see, there is good agreement between the simulation and experiment.
This method is particularly useful for the optimization phase of design, as parametric changes to the geometry can be easily incorporated. In addition, Adaptive Mesh Refinement (AMR) allows you to accurately capture the physical phenomena of interest while maintaining a reasonable computational expense.
Next, we will look at a twin screw compressor. These compressors have two helical screws that rotate in opposite directions, and are frequently used in industrial, manufacturing, and refrigeration applications. A common challenge for designing screw compressors—and many other pumps and compressors—is the tight clearances between parts. Inevitably, there will be some leakage flow between chambers, which will affect the device’s performance.
CONVERGE offers several methods for capturing the fluid behavior in these small gaps. Using local mesh embedding and AMR, you can directly resolve the gaps. This method is highly accurate, but it can come with a high computational price tag. An alternative approach is to use one of CONVERGE’s gap models to account for the leakage flows without fully resolving the gaps. This method balances accuracy and time costs, so you can get the results you need when you need them.
Another factor that must be taken into account when designing screw compressors is thermal expansion. Heat transfer between the fluid and the solid walls means the clearances will vary down the length of the rotors. CONVERGE’s CHT model can capture the heat transfer between the solid and the fluid to account for this phenomenon.
This study2 of a dry twin screw compressor employs a gap model to account for leakage flows, CHT modeling to capture heat transfer, and AMR to resolve large-scale flow structures. Mass flow rate, power, and discharge temperature were predicted with CONVERGE and compared to experimentally measured values. This study also investigated the effects of the base grid size on the accuracy of the results. In Figure 2, you can see there is good agreement between the experimental and simulated data, particularly for the most refined grid. The method used in this study provides accurate results in a turn-around time that is practical for engineering applications.
The benefits CONVERGE offers for designing pumps and compressors directly translate to a tangible competitive advantage. CFD benefits your business by reducing costs and enabling you to bring your product to market faster, and CONVERGE features tools to help you optimize your designs and produce high-quality products for your customers. To find out how CONVERGE can benefit you, contact us today!
[1] Rowinski, D., Pham, H.-D., and Brandt, T., “Modeling a Scroll Compressor Using a Cartesian Cut-Cell Based CFD Methodology with Automatic Adaptive Meshing,” 24th International Compressor Engineering Conference at Purdue, 1252, West Lafayette, IN, United States, Jul 9–12, 2018.
[2] Rowinski, D., Li, Y., and Bansal, K., “Investigations of Automatic Meshing in Modeling a Dry Twin Screw Compressor,” 24th International Compressor Engineering Conference at Purdue, 1528, West Lafayette, IN, United States, Jul 9–12, 2018.
[3] Rowinski, D., Sadique, J., Oliveira, S., and Real, M., “Modeling a Reciprocating Compressor Using a Two-Way Coupled Fluid and Solid Solver with Automatic Grid Generation and Adaptive Mesh Refinement,” 24th International Compressor Engineering Conference at Purdue, 1587, West Lafayette, IN, United States, Jul 9–12, 2018.
[4] Rowinski, D.H., Nikolov, A., and Brümmer, A., “Modeling a Dry Running Twin-Screw Expander using a Coupled Thermal-Fluid Solver with Automatic Mesh Generation,” 10th International Conference on Screw Machines, Dortmund, Germany, Sep 18–19, 2018.
[5] da Silva, L.R., Dutra, T., Deschamps, C.J., and Rodrigues, T.T., “A New Modeling Strategy to Simulation the Compression Cycle of Reciprocating Compressors,” IIR Conference on Compressors, 0226, Bratislava, Slovakia, Sep 6–8, 2017. DOI: 10.18462/iir.compr.2017.0226
[6] Willie, J., “Analytical and Numerical Prediction of the Flow and Performance in a Claw Vacuum Pump,” 10th International Conference on Screw Machines, Dortmund, Germany, Sep 18–19, 2018. DOI: 10.1088/1757-899X/425/1/012026
[7] Jhun, C., Siedlecki, C., Xu, L., Lukic, B., Newswanger, R., Yeager, E., Reibson, J., Cysyk, J., Weiss, W., and Rosenberg, G., “Stress and Exposure Time on Von Willebrand Factor Degradation,” Artificial Organs, 2018. DOI: 10.1111/aor.13323
[8] Rowinski, D.H., “New Applications in Multi-Phase Flow Modeling With CONVERGE: Gerotor Pumps, Fuel Tank Sloshing, and Gear Churning,” 2018 CONVERGE User Conference–Europe, Bologna, Italy, Mar 19–23, 2018. https://api.convergecfd.com/wp-content/uploads/David-Rowinski_Multiphase-Modeling-Gearbox-Power-Losses-Oil-Pump-Cavitation-and-Fuel-Tank-Sloshing.pdf
[9] Willie, J., “Simulation and Optimization of Flow Inside Claw Vacuum Pumps,” 2018 CONVERGE User Conference–Europe, Bologna, Italy, Mar 19–23, 2018. https://api.convergecfd.com/wp-content/uploads/james-willie-simulation-and-optimization-of-flow-inside-claw-vacuum-pumps.pdf
[10] Scheib, C.M., Newswanger, R.K., Cysyk, J.P., Reibson, J.D., Lukic, B., Doxtater, B., Yeager, E., Leibich, P., Bletcher, K., Siedlecki, C.A., Weiss, W.J., Rosenberg, G., and Jhun, C., “LVAD Redesign: Pump Variation for Minimizing Thrombus Susceptibility Potential,” ASAIO 65th Annual Conference, San Francisco, CA, United States, Jun 26–29, 2019.
In a competitive market, predictive computational fluid dynamics (CFD) can give you an edge when it comes to product design and development. Not only can you predict problem areas in your product before manufacturing, but you can also optimize your design computationally and devote fewer resources to testing physical models. To get accurate predictions in CFD, you need to have high-resolution grid-convergent meshes, detailed physical models, high-order numerics, and robust chemistry—all of which are computationally expensive. Using simulation to expedite product design works only if you can run your simulations in a reasonable amount of time.
The introduction of high-performance computing (HPC) drastically furthered our ability to obtain accurate results in shorter periods of time. By running simulations in parallel on multiple cores, we can now solve cases with millions of cells and complicated physics that otherwise would have taken a prohibitively long time to complete.
However, simply running cases on more cores doesn’t necessarily lead to a significant speedup. The speedup from HPC is only as good as your code’s parallelization algorithm. Hence, to get a faster turnaround on product development, we need to improve our parallelization algorithm.
Breaking a problem into parts and solving these parts simultaneously on multiple interlinked processors is known as parallelization. An ideally parallelized problem will scale inversely with the number of cores—twice the number of cores, half the runtime.
A common task in HPC is measuring the scalability, also referred to as scaling efficiency, of an application. Scalability is the study of how the simulation runtime is affected by changing the number of cores or processors. The scaling trend can be visualized by plotting the speedup against the number of cores.
In CONVERGE versions 2.4 and earlier, parallelization is performed by partitioning the solution domain into parallel blocks, which are coarser than the base grid. CONVERGE distributes the blocks to the interlinked processors and then performs a load balance. Load balancing redistributes these parallel blocks such that each processor is assigned roughly the same number of cells.
This parallel-block technique works well unless a simulation contains high levels of embedding (regions in which the base grid is refined to a finer mesh) in the calculation domain. These cases lead to poor parallelization because the cells of a single parallel block cannot be split between multiple processors.
Figure 1 shows an example of parallel block load balancing for a test case in CONVERGE 2.4. The colors of the contour represent the cells owned by each processor. As you can see, the highly embedded region at the center is covered by only a few blocks, leading to a disproportionately high number of cells in those blocks. As a result, the cell distribution across processors is skewed. This phenomenon imposes a practical limit on the number of levels of embedding you can have in earlier versions of CONVERGE while still maintaining a reasonable load balance.
In CONVERGE 3.0, instead of generating parallel blocks, parallelization is accomplished via cell-based load balancing, i.e., on a cell-by-cell basis. Because each cell can belong to any processor, there is much more flexibility in how the cells are distributed, and we no longer need to worry about our embedding levels.
Figure 2 shows the cell distribution among processors using cell-based load balancing in CONVERGE 3.0 for the same test case shown in Figure 1. You can see that without the restrictions of the parallel blocks, the cells in the highly embedded region are divided between many processors, ensuring an (approximately) equal distribution of cells.
The cell-based load balancing technique demonstrates significant improvements in scaling, even for large numbers of cores. And unlike previous versions, the load balancing itself in CONVERGE 3.0 is performed in parallel, accelerating the simulation start-up.
In order to see how well the cell-based parallelization works, we have performed strong scaling studies for a number of cases. The term strong scaling means that we ran the exact same simulation (i.e., we kept the number of cells, setup parameters, etc. constant) on different core counts.
Figure 3 shows scaling results for a typical SI8 port fuel injection (PFI) engine case in CONVERGE 3.0. The case was run for one full engine cycle, and the core count varied from 56 to 448. The plot compares the speedup obtained running the case in CONVERGE 3.0 with the ideal speedup. With enough CPU resources, in this case 448 cores, you can simulate one engine cycle with detailed chemistry in under two hours—which is three times faster than CONVERGE 2.4!
Cores | Time (h) | Speedup | Efficiency | Cells per core | Engine cycles per day |
---|---|---|---|---|---|
56 | 11.51 | 1 | 100% | 12,500 | 2.1 |
112 | 5.75 | 2 | 100% | 6,200 | 4.2 |
224 | 3.08 | 3.74 | 93% | 3,100 | 7.8 |
448 | 1.91 | 6.67 | 75% | 1,600 | 12.5 |
If the speedup of the SI8 PFI engine simulation impressed you, then just wait until you see the scaling study for the Sandia Flame D case! Figure 4 shows the results of a strong scaling study performed for the Sandia Flame D case, in which we simulated a methane flame jet using 170 million cells. The case was run on the Blue Waters supercomputer at the National Center for Supercomputing Applications (NCSA), and the core counts vary from 500 to 8,000. CONVERGE 3.0 demonstrates impressive near-linear scaling even on thousands of cores.
Although earlier versions of CONVERGE show good runtime improvements with increasing core counts, speedup is limited for cases with significant local embeddings. CONVERGE 3.0 has been specifically developed to run efficiently on modern hardware configurations that have a high number of cores per node.
With CONVERGE 3.0, we have observed an increase in speedup in simulations with as few as approximately 1,500 cells per core. With its improved scaling efficiency, this new version empowers you to obtain simulation results quickly, even for massive cases, so you can reduce the time it takes to bring your product to market.
Contact us to learn how you can accelerate your simulations with CONVERGE 3.0.
[1] The National Center for Supercomputing Applications (NCSA) at the University of Illinois at Urbana-Champaign provides supercomputing and advanced digital resources for the nation’s science enterprise. At NCSA, University of Illinois faculty, staff, students, and collaborators from around the globe use advanced digital resources to address research grand challenges for the benefit of science and society. The NCSA Industry Program is the largest Industrial HPC outreach in the world, and it has been advancing one third of the Fortune 50® for more than 30 years by bringing industry, researchers, and students together to solve grand computational problems at rapid speed and scale. The CONVERGE simulations were run on NCSA’s Blue Waters supercomputer, which is one of the fastest supercomputers on a university campus. Blue Waters is supported by the National Science Foundation through awards ACI-0725070 and ACI-1238993.
Graphcore has used a range of technologies from Mentor, a Siemens business, to successfully design and verify its latest M2000 platform based on the Graphcore Colossus™ GC200 Intelligence Processing Unit (IPU) processor.
Simcenter™ FLOEFD™ software, a CAD-embedded computational fluid dynamics (CFD) tool is part of the Simcenter portfolio of simulation and test solutions that enables companies optimize designs and deliver innovations faster and with greater confidence. Simcenter FLOEFD helps engineers simulate fluid flow and thermal problems quickly and accurately within their preferred CAD environment including NX, Solid Edge, Creo or CATIA V5. With this release, Simcenter FLOEFD helps users create thermal models of electronics packages easily and quickly. Watch this short video to learn how.
Simcenter™ FLOEFD™ software, a CAD-embedded computational fluid dynamics (CFD) tool is part of the Simcenter portfolio of simulation and test solutions that enables companies optimize designs and deliver innovations faster and with greater confidence. Simcenter FLOEFD helps engineers simulate fluid flow and thermal problems quickly and accurately within their preferred CAD environment including NX, Solid Edge, Creo or CATIA V5. With this release, Simcenter FLOEFD allows users to add a component into a direct current (DC) electro-thermal calculation by the given component’s electrical resistance. The corresponding Joule heat is calculated and applied to the body as a heat source. Watch this short video to learn how.
Simcenter™ FLOEFD™ software, a CAD-embedded computational fluid dynamics (CFD) tool is part of the Simcenter portfolio of simulation and test solutions that enables companies optimize designs and deliver innovations faster and with greater confidence. Simcenter FLOEFD helps engineers simulate fluid flow and thermal problems quickly and accurately within their preferred CAD environment including NX, Solid Edge, Creo or CATIA V5. With this release, the software features a new battery model extraction capability that can be used to extract the Equivalent Circuit Model (ECM) input parameters from experimental data. This enables you to get to the required input parameters faster and easier. Watch this short video to learn how.
Simcenter™ FLOEFD™ software, a CAD-embedded computational fluid dynamics (CFD) tool is part of the Simcenter portfolio of simulation and test solutions that enables companies optimize designs and deliver innovations faster and with greater confidence. Simcenter FLOEFD helps engineers simulate fluid flow and thermal problems quickly and accurately within their preferred CAD environment including NX, Solid Edge, Creo or CATIA V5. With this release, Simcenter FLOEFD allows users to create a compact Reduced Order Model (ROM) that solves at a faster rate, while still maintaining a high level of accuracy. Watch this short video to learn how.
High semiconductor temperatures may lead to component degradation and ultimately failure. Proper semiconductor thermal management is key for design safety, reliability and mission critical applications.
Tecplot 360 2020 R2 has additional multi-threading for variable calculations. Under the Analyze>Calculate Variables dialog, all functions listed will be fully multi-threaded. In previous versions, multi-threading was used only if there were multiple zones. Multi-threading is now used within a zone. In the video example, the improvement is 8 times faster than with earlier versions of Tecplot 360. This test was done on an 8-core windows machine computing Q Criterion. The dataset was 8.6 million polyhedrals.
I’ve loaded the dataset, calculated Q criterion, and generated an isosurface. In Tecplot 360 2020 R2 you can see that all 8 cores are used for the computation. In Tecplot 360 2020 R1 only one CPU core is effectively being used. The full computation takes 284 seconds in 2020 R1, and by the time I finish this sentence the computation in 2020 R2 will already be done. And it took only 36 seconds.
The next calculation was tested on a 32-core Windows machine. It was over 11 times faster in Tecplot 360 2020 R2 compared to the previous release. The improvement was not as large with other datasets with multiple zones, because Tecplot 360 2020 R1 already uses multi-threading across zones. But multi-threading within a zone still results in a faster computation: 1.3 times faster for the OpenFOAM dataset, and over two times faster for the Plot3D data.
The post Multi-threaded Variable Calculations appeared first on Tecplot.
It may come as a shock to some of you, but here at Tecplot we have a thing for well-made plots and the presentations that they occupy. There is something eminently satisfying about looking at a set of data that has been thoughtfully reduced into the relevant facts necessary to enable sound engineering judgment. It is in service of that ideal that we have decided to put together a short blog series on how to make plots & presentations that tell a clear, concise, and convincing story.
Our goal isn’t to present anything profound – only to share practical reminders of the importance of effective communication in engineering. An expert engineer doesn’t simply need to understand the science of their discipline – they need to also know how to convey the relevant facts to their colleagues and stimulate productive discussion.
In this first post we’ll tackle something quite tangible – consistency. Consistency is important for comparisons between datasets (or between regions of the same dataset) because it enables the audience to identify significant differences more easily. If your audience is presented with plots that convey similar datasets but use varying format, scale, color, orientation, etc., it distracts them and takes the focus away from what really matters – the story behind the data.
The image at right is an example of two pressure coefficient distribution plots at discrete spanwise locations; we’ll dive into a few of the ways this plot uses consistency to enhance its readability.
How do you decide what to keep consistent and what to vary? That will depend on the data you are presenting, the type of plot you’re using, and the differences you wish to highlight.
In this example, the two Cp plots at different spanwise locations demonstrate the relative position of the pressure change due the lambda shock structure that is characteristic of the Onera M6 wing.
Receive Tips and Tricks like this directly to your inbox – Subscribe to Tecplot »
Line plots are critical for deriving actionable insights from most engineering analyses, but they aren’t the only types of plots that can be useful. The 2D contour plot at right is a great example of how consistency can be used to compare changes in a dataset over time.
This 2D contour plot compares the pressure of fluid flow around a rotating cylinder at two different time steps. Also included is a single streamline to demonstrate the change in the vortex shedding over time.
We have kept our axis labels, markers, and annotations consistent for easy readability. In addition, we have:
If the contour levels or the streamline location varied between the two plots, you could easily make some inaccurate assumptions about the fluid flow.
Our last example is a 3D contour plot using the Onera M6 wing. The image below is similar to the 2D contour plot in that we’ve kept our contour levels, colormap, and labels consistent between frames. However, when comparing 3D plots it’s important to consider how the 3D perspective can affect your ability to make an unbiased assessment.
The image shows a fixed perspective, which includes the pan, zoom, and rotation settings. You’ll notice that because the volume slices are at different locations the wing appears to move between frames. But the wing is fixed in place. With this approach, it is immediately clear that the plot is not comparing flow states at the same slice location for two different solutions.
Ultimately there are few hard and fast rules when it comes to formatting plots. And it is certainly worth taking the time to ensure that the critical insights from your analyses are not obscured by inconsistency across your plots. Consistency has the most impact when comparing two plots side by side. But don’t underestimate the value of maintaining consistency throughout your presentations – and even between different presentations. Consistency will establish a presentation style that your audience will have an easier time digesting.
Consistency is one tactic to making your plots and your presentations easier to understand. We will explore more plotting tips and tricks in future posts.
Stay tuned!
The post Consistency is Key: Visual Communication appeared first on Tecplot.
Bellevue, Washington (January 5, 2021) – Tecplot, Inc. today announced that Alan Klug is named Tecplot President. Alan will run day-to-day operations at Tecplot USA. In addition, he will oversee the FieldView CFD and Tecplot Europe business units. Tom Chan will continue as Tecplot CEO in an advisory role but will focus on managing the portfolio of acquisitions within the Vela Software Group.
Tecplot also announced that Charles Schnake, previously with Rolls Royce and Aerion Supersonic, is named Tecplot Director of Customer Development.
“As Tecplot’s Vice President of Customer Development, Alan has been instrumental in our transition and growth under Vela/CSI ownership. I am excited for Tecplot’s future under his direction as President. Please join me in congratulating Alan on his promotion,” says Tom Chan, Tecplot CEO.
“From aerospace to life sciences, our customers continue to make outstanding achievements, even during the difficult past year. I look forward to working closely with them to provide best-in-class solutions and find better ways to serve them,” says Alan Klug, Tecplot President. “And I welcome the experience and expertise Charles brings to Tecplot as he steps into the customer development leadership role.”
“It is my honor and pleasure to assume leadership of Tecplot’s Customer Development team,” says Charles Schnake, Tecplot Director of Customer Development. “Alan has left behind some big shoes to fill, but fortunately we will continue to benefit from his experience and guidance. I am excited to add my energy and perspective to an already successful team, and to forge relationships with our customers and partners.”
Read more about Tecplot Leadership.
An operating company of Toronto-based Constellation Software, Inc. (CSI), Tecplot is the leading independent developer of visualization and analysis software for engineers and scientists. CSI is a public company listed on the Toronto Stock Exchange (TSX:CSU). CSI acquires, manages, and builds software businesses that provide mission-critical solutions in specific vertical markets.
Tecplot visualization and analysis software allows customers using desktop computers and laptops to quickly analyze and understand information hidden in complex data and communicate their results to others via professional images and animations. The company’s products are used by more than 47,000 technical professionals around the world.
Margaret Connelly
Marketing Manager, Tecplot, Inc.
pr@tecplot.com
(425) 653-1200
The post Alan Klug Named Tecplot President appeared first on Tecplot.
A new capability to reference variables by name has been added to Tecplot 360 2020 R2.
This is important because sometimes you may have two simulation runs where one simulation exports more variables than the other. In this case macros and stylesheets may produce a plot that references incorrect variables.
To demonstrate this new capability, we’ll use two CONVERGE simulations. One where NOX is variable number 25 and the other where NOX is variable number 12. When we did our first simulation, we captured more variables than we needed. And for a later simulation we trimmed the output to only the required variables for our study.
If we save this frame style and apply it to our dataset where NOX is variable number 25, the resulting style is incorrect. This is because we saved by variable number, and variable number 12 is NOX for the first dataset but is Equivalence Ratio in the second.
Saving variable names, instead of variable by numbers, in the stylesheet will correct this issue. This capability is not on by default. It can be turned on by uncommenting a line in the Tecplot configuration file. Layouts, stylesheets and macros will now be saved using variable names instead of variable numbers.
When we save the frame style and apply it to our other dataset, that style is applied as expected. Note that if you have duplicate variable names, for example multiple sets of UVW vector variables, the variable name that is first numerically will be picked.
This concludes the tutorial for Referencing Variables by Name. Thank you for watching.
The post Referencing Variables by Name in Tecplot 360 appeared first on Tecplot.
Tecplot 360 2020 R2 Now Available
BELLEVUE, WA (December 9, 2020) – Tecplot, Inc. has announced the general availability of Tecplot 360 2020 Release 2.
Tecplot 360 2020 R2 represents our commitment to helping engineers and scientists analyze their results quickly and easily. The reintroduction of Tecplot Chorus ensures that you’re able to understand your simulated results on the whole rather than just as individual cases. We’ve also improved the speed of variable calculations, in some cases up to 6x faster, as well as making Tecplot 360 compatible with additional file formats.
The ability to reference variables by name makes layouts, stylesheets, and macros more flexible, easier to read, and more applicable to other, similar, datasets.
Finally, the ability to split zones that have distinct connected regions makes it easier to isolate data for which you need quantifiable results. For example, an exhaust manifold may have four distinct outlets, this new feature allows you to easily split those outlets into individual zones to compute quantities like averages and mass flow rates to ensure balanced flow.
See all updates in this release.
Tecplot 360 2020 Release 2 is available for download as Free Trial Software, or for customers through the MyTecplot Customer Portal.
Tecplot 360 is a suite of CFD visualization and analysis tools that can handle large data sets, automate workflows, and visualize parametric results. Three powerful modules in this one tool include:
Integrated XY, 2D and 3D plotting speeds data analysis and increases productivity. Easy-to-use, fast and memory efficient, Tecplot 360 produces visually powerful output to help engineers and scientists communicate CFD, other simulation and test data results to clients and stakeholders.
The Tecplot 360 suite of tools are available for customers on TecPLUS maintenance service and anyone who downloads a free trial of the software.
Special pricing is available for Academic users upon request, see Tecplot for Academics.
Tecplot, an operating company of Toronto-based Constellation Software, Inc. (CSI), is the leading independent developer of visualization and analysis software for engineers and scientists. CSI is a public company listed on the Toronto Stock Exchange (TSX:CSU). CSI acquires, manages, and builds software businesses that provide mission-critical solutions in specific vertical markets.
Tecplot visualization and analysis software allows customers using desktop computers and laptops to quickly analyze and understand (local or remote) information hidden in complex data and communicate their results to others via professional images and animations. The company’s products are used by more than 47,000 technical professionals around the world.
Margaret Connelly
Marketing Manager, Tecplot
pr@tecplot.com
(425) 653-1200
The post Tecplot 360 2020 R2 Now Available appeared first on Tecplot.
To access this dialog, select View > Key Frame Animation. From there, you can start creating animations.
Easily animate a smooth progression of two or more specified views. This tutorial shows you how this feature works.
I will start by appending my initial view, which is where I want my animation to start. Next, I could rotate, zoom, or translate my grid solution and append additional views. I could also move the key frames closer together if I wish to shorten up a movie and name them for easy reference. And now I could animate my key frame animation within the Tecplot RS interface to see what my movie looks like. And finally, I could export the movie into my preferred output format. The resulting output movie can then be shared with colleagues or embedded into a presentation.
The post Key Frame Animation in Tecplot RS appeared first on Tecplot.
And the consolidation continues. Cadence announced this morning that it will acquire NUMECA, maker of the OMNIS CAE framework (though you probably know it for its FINE/Design3D optimization solution and its FINE/Marine virtual towing tank/CFD naval architecture tools — that’s what I first learned about, years ago). I don’t follow Cadence in enough detail to know, so here’s what they said about the logic behind the deal:
The addition of NUMECA’s technologies and talent supports the Cadence Intelligent System Design strategy and broadens its system analysis portfolio with CFD solutions, servicing a fast-moving market segment [by which they mean CFD, I think] where accuracy, reliability and predictability are paramount concerns for high-fidelity modeling … The acquisition builds on the momentum of recent Cadence system innovation developments with the Clarity 3D Solver for electromagnetic (EM) simulation, the Clarity 3D Transient Solver for finite difference time domain (FDTD) system-level EM simulation and the Celsius Thermal Solver for electrical-thermal co-simulation product introductions. NUMECA’s technology will also contribute to Cadence best-in-class system analysis solutions for integrated circuits (ICs), electronic subsystems and full system designs.
Cadence President Anirudh Devgan, said, “The acquisition of NUMECA’s proven CFD technology and talented team complements Cadence’s finite element analysis and other system innovation technologies and is another successful step that will advance our customers’ ability to design the exciting products of tomorrow.”
Tom Beckley, SVP of the Custom IC & PCB Group at Cadence, added, “With the addition of NUMECA’s technology to the Cadence portfolio, we are broadening our system analysis capabilities and integrated design solutions, addressing critical customer challenges in areas such as internal and external flows, acoustics, heat transfer, fluid-structure interaction and optimization.”
Terms of the transaction were not disclosed. The acquisition is expected to close in the first quarter of 2021, subject to customary closing conditions.
It’s amazing that I just this week wrote a piece for NAFEMS on consolidation in CAE — and commented that these acquisitions will only foster more innovation since there’s an obvious path to a payout. I’ll need to add NUMECA+Cadence to that list.
As I said above, I haven’t followed Cadence to any level of detail. That’s because they’ve historically been EDA (electronic design automation)-focused, and that’s not my wheelhouse. But as they stretch into new areas of CAE add more 3-D solvers to their electromagnetic and thermal offerings (and the NUMEACA OMNIS platform becomes more widely promoted), Cadence can move further into marine, aerospace, automotive, and industrial markets. Definitely, something to watch.
I wrote the piece below in 2014 and can’t believe how much has happened since then — and, but, how little progress we’ve made in making the world a more equal place. Dr. King’s message was simple: all people are created equal and it’s our actions that make them unequal. In health care, in housing, in the job market, in educational opportunities, in all of the areas of life that many of us take for granted. His passion and eloquence were awesome and his words are more important today than ever.
Today the US stops (mostly) to remember and honor Martin Luther King, Jr. –known as MLK– for his work and legacy, redefining what it means to fight for the right to be equal. I was too little to remember Dr. King’s speeches from when he was alive but I do remember the aftermath of his death. My family moved to New York City in 1967; on April 4, 1968, Dr. King was shot while preparing to lead a protest march in Memphis, Tennessee. Either that same evening or the next day (news traveled a lot more slowly back then), the mayor of New York asked the city to stay calm and said that the city’s leaders would continue Dr. King’s work to end poverty. We saw, in school and on the nightly news, other cities in the US erupt in flames as grief and anger overcame Dr. King’s principles of nonviolence and working with the political system to create change. My parents wondered if I’d be safe, walking the few blocks to school — and I was. New York was far from perfect in the 1960s, but this was something to be proud of.
Unfortunately, it took his assassination to cause adults to talk to us about freedom, equality, poverty and nonviolence. But once those conversations start, they cannot be stopped and I’m sure they profoundly changed my views of the world, and those of the other little kids in the classrooms and living rooms around the US.
Dr. King’s legacy only grew after his death and, at least in the schools I attended school in New York, his work and speeches came front and center in the curriculum. If you have a few minutes, it’s worth your time to read his Letter from a Birmingham Jail and watch his most famous speech, I Have A Dream.
His words about how we are all connected, how every human being matters, are as true today as they were when he first wrote them.
Image credit: Nobel Prize, Nobel Media AB
First things first: the lack of financial news in the PLMish world is encouraging. If the December quarter (meaning October+November+December) had been far below expectations, we would have heard from companies issuing earnings pre-announcements. There’s only been one of those, from SAP, so, so far so good.
And even SAP’s wasn’t dire. SAP said last night that business improved sequentially –meaning that the fourth quarter was better than the third– “even as the COVID-19 crisis persisted and lockdowns were reintroduced in many regions”. We’ll get details when they release their full details on January 29, but here are two possible interpretations: SAP customers are learning how to work within these restrictions and are using more SAP kit to do it and/or SAP has learned to sell from afar. Of course, even that can’t help all parts of SAP’s business; the company said its business travel-related cloud service continues to struggle.
Overall, though, SAP says preliminary data for Q4 has software licenses revenue down 15% year over year (down 11% in constant currencies, cc). Cloud revenue was up 8% (up 13% cc), and total revenue was down 6% year over year to €7.54 billion (down 2% cc). That doesn’t sound great but is better that SAP (and its investors) had expected.
SAP CFO Luka Mucic said, “In a uniquely challenging environment, 2020 was a record year for cash flow in every single quarter and the full year. Our better-than-anticipated top line performance combined with our quick response on the cost side drove strong operating profit. SAP’s expedited shift to the cloud will drive long-term, sustainable growth while significantly increasing the resiliency and predictability of our business.”
We heard more upbeat news from AVEVA, which released a trading update this morning. The company said the December quarter was “strong [with] Organic constant currency revenue growth over 26%. This was driven by a significant number of scheduled subscription renewals, including a large three-year contract renewal in the Food sector. It also benefited from the early renewal of a large three-year EPC contract that had been scheduled for AVEVA’s Q4 [meaning, the March 2021 quarter], and the conversion of two large contracts in the Marine sector from annual fees to multi-year subscription, giving the customers more flexibility in a challenging Marine market environment.”
So much to unpack there. First, 26% cc growth is astonishing. Remember that the December 2019 quarter was still normal, pre-COVD, business-as-usual, so not obviously an easy comparable. Next, early renewals are awesome but rare and show how important AVEVA is to that customer’s business. Last, good news from the Marine sector! I can’t remember the last time that happened.
AVEVA said the strong December quarter boosted revenue growth to “approximately 1.5% in the nine months to 31 December 2020 on an organic constant currency basis.”
The company also gave a quick update on its acquisition of OSIsoft: The deal is a go, with only the US’ Committee on Foreign Investments (CFIUS) still to approve. AVEVA says it expects to receive that by early February, and that the transaction should close shortly afterwards.
What does it all mean? I think we’re in for a bumpy ride this earnings season. Good news when buyers and sellers find each other, even in a remote selling situation. Bad news when sales require a lot of in-person contact or on-site services. More good news where vendors have and buyers want cloud technology. And bad news when the inverse are true.
Based on two data points, though, business in the December quarter was better than in September quarter. Let’s hold on to that.
3D Systems announced today that it completed the sale of Cimatron and GibbsCAM to “a subsidiary of ST Acquisition Co., an affiliate of Battery Ventures” on January 1, 2021. 3D Systems said it netted around $64 million in the sale, part of which it used to pay down debt in order to improve its financial position and move along in its reorganization plan.
Jeffrey Graves, CEO of 3D Systems, said, “The divestiture of Cimatron and GibbsCAM, which were businesses focused on subtractive technologies, was an important step in our plans to refocus our company on our core mission – ‘to be the leader in enabling additive manufacturing solutions for applications in growing markets that demand high-reliability products.’ These divestitures strengthened our balance sheet, enabling us to both pay off our debt and terminate the ATM Program [a program under which it sold shares of its common stock to raise cash] much earlier than originally planned.”
This update is important to many people –employees, customers, partners, and friends of Cimatron and GibbsCAM, 3D Systems, and Battery Ventures, among others– but it isn’t the piece I intended to write today. I drafted and (digitally) tore up my thoughts on yesterday’s storming of the US Capitol numerous times. You don’t come here for politics, and I get that. But silence isn’t really an option for me.
Let’s just say that Winston Churchill was right: “Democracy is the worst form of government except for all the rest.” Those who enabled and riled up these rioters must be held accountable — whether by law or at the ballot box, next time they come up for election. But right now, we have so much work to do that we can’t let this temper tantrum sideshow distract us. On Wednesday. according to the Boston Globe, more than 3,000 people in the US lost their lives to the coronavirus. That means 3,000 more families got truly devastating news. And that’s just one of the many things that need our attention. Let’s cut the !@#$ and get to work.
Thank you to the police and National Guard who protected our elected officials and government workers.
I hope you had the opportunity to unwind and relax as 2020 came to a close. Let’s hope 2021 is a much kinder year!
To start off the new year, a quick update: While we were pondering New Year’s Eve cocktails, Sandvik announced that it completed its acquisition of CGTech, make of numerical control (NC/CNC) simulation, verification, and optimization solutions. You likely know them for their Vericut offering.
Happy New Year!