Basic Solving and Post-Processing in Icepak
The objective of this video is to understand the solution workflow in iSpec Classic. We will:
- Modify simple solver settings
- Create different monitor points
- Check solution convergence
- Verify mass and energy conservation
- Post-process the solution
- Examine the solution using a cut plane at the faces
- Analyze various plots
Model Overview
In the model, we have an electronic component inside a box with two fans blowing air over it. The chips generate heat, and we will observe the solution to determine the maximum temperature computed.
Initial Setup
We start by changing some preferences:
- Edit preference display to change the color legend format to float.
- Set the numerical display precision to four.
- Change the background style and object colors and widths.
Mesh Generation
We quickly generate a mesh with 66,562 nodes.
Problem Setup
- Go to problem setup and select basic parameters.
- We will solve for flow and temperature, with radiation off initially.
- Use the zero-equation turbulent model and do not solve for gravity.
Solution Settings
- Set the number of iterations to 400.
- Choose parallel processing with four processors.
- Create monitor points at specific locations, ensuring no more than 10 to maintain solution speed.
Solution Execution
Start the solution and observe the convergence. If temperatures are not constant after 400 iterations, modify the solver settings and continue from the last computed data point.
Results Verification
Check mass and volume flow rates, ensuring calculated values are close to specified values. Verify heat inputs for different cell zones and compute overall power.
Post-Processing
Temperature Plot
- Create a face object for temperature plots.
- Control-select different regions to display the temperature profile.
Vector Visualization
- Create vectors to show velocity and magnitude.
- Modify to include heat sink and observe non-uniform velocities.
Particle Traces
- Create particle traces using the fan group.
- Set temperature as the color variable and observe airflow through the system.
Cut Plane Analysis
- Create a cut plane to observe temperature at specific locations.
- Adjust plane position to view different regions.
Radiation Activation
Activate radiation using the ray-tracing model to observe its effect on temperature. Set flow and radiation iterations to 5 and observe reduced temperatures.
Fan Failure Simulation
Simulate fan failure to observe significant temperature increases, highlighting the importance of cooling efficiency.
Conclusion
The video demonstrates the importance of various settings and parameters in achieving accurate and efficient solutions in Icepak. Thank you for your time, and we will see you in the next video.
Note: Ozen Engineering, Inc. is referenced in this transcript.
The objective of this video is to understand the solution workflow in iSpec classic. We will modify simple solver settings and create different monitor points, check solution convergence, verify mass and energy conservation and post-process the solution.
We will look at the solution using a cut plane and various plots. In the model, we have electronics components inside a box with two fans blowing air on top. There are chips generating heat, and we will look into the solution to see the maximum temperature. We start by changing some preferences.
Edit preference display and change the color legend format to float, set the numerical display precision to four. Also, change the screen upside cover, targeting percent and rate, and adjust the cut value under one. We will look into this later in the video.
We have a white background using the positive Z direction. The background style can be changed; we will look at color 1. We will also try and color different objects differently. To do this, go to preference, then object types, and change the color and width of the object.
We will quickly generate a mesh. The mesh is complete and generated, and the number of nodes generated is 66562. We will change some basic parameters. Go to problem setup, basic parameters. We'll be solving the flow and temperature. Radiation is off at this time.
We'll be using a zero-equation turbulent model and not solve for gravity. We go to solution setting. The number of iterations we will run is 400, and this is the convergence criteria. Parallel setting. We can use serial or we can use parallel. This case, we'll use parallel and use four processors.
We will not use GPU computing. We can create monitor points for these points. Select all these points, remove the board, and drag and drop them into points. This will automatically create monitor points for these points. We will start the solution. The calculation is converging monotonously.
The rest will start falling. And on the right-hand side, you can see the computer temperature at different locations. The calculation is completed 400 iterations, but as we can see, the temperatures have still not become constant. So we will modify the solve.
Once we terminate the calculation, the solution overview will be displayed. We can check in the first section about the mass and volume flow rate. The calculated value for the fan left and right is good, and the overall calculation within the cabinet is the addition of the two.
The addition of these two should be close to the value that has been specified, and the mass conservation should be as close to zero as possible. We can post-process some data. First, we will create a face object based on the temperature plot.
Click on the face object and select different areas where you want to control select different regions. This will be displayed. We can further add more to it, like we missed these two. We will accept and update. Now you can see the full temperature profile.
You can also control and use your center mouse button to place the contours scale at different locations. Now let's create vectors. Click on the object face again. Name it as vector. Select the object. Select cabinet. Click on show vectors. Parameters.
Velocity and magnitude have been selected for color. Using this object apply and done. You can see how the vectors look like, but you also want to see the vectors in these regions. So let's select modify it a little bit and add the heat sink to it. Add it. Add it. Done.
You can immediately see that the heat sink is receiving non-uniform velocities. You have much higher velocity of flow through the heat sink on the upper side compared to on the lower side.
So essentially, this will immediately lead to a higher temperature on this side compared to the lower temperature on that side. Let's deactivate this. Let's create another one. This time we will be creating tracks or particle traces. Particle traces. Select the object.
Let's select a group which is the fan group that we created earlier. Accept it. Select temperature as the color variable. We want uniform and we want 50 points per input location. So there will be 50 tracks per fan. Use the local object and you can change the marker type.
In this case, we will be using marker cone and set the size to 2 and width to 3. Apply. You can see how the flow looks like as the fan pushes it through the system. And we can immediately see why one side of the heat sink is receiving more air flow compared to the other side.
Now let's deactivate this and create a cut plane. To create a cut plane, you will use the plane cut tool. You can create a cut plane just by clicking on different locations. So let's select horizontal screen select. Just position it properly. Select horizontal screen select and left click.
Once you have selected that location, activate show contours. Want to see the temperature. Set it to local level. And then once smooth contours. Apply. And you can see that how the temperature at that specific location looks like. Rescale it.
Of course, you can change the plane position by using this scale. So you can just move it down or move it up to see what the temperature looks like at that location. So let's just put it at 0.165 and we can see what the temperature looks like in the board region.
Of course, this is a two-dimensional contour plot. So we can change that to a three-dimensional control plot as well. So to do that, you can activate the show mesh object. Also, go to parameters and you can change the shading options to 3D. Scale it.
Increase the scale and we'll be using fixed values. Click online. We will be selecting line color. Line color is fixed to a color which is black. Line and done. So now you can see in terms of where the temperature is maximum. It will show you the trend in 3D plot.
So it gives you a very good visual representation of the magnitude of the temperature at different locations. Again, let's activate this. Now let's create another cut plane.
Set the cut plane to the Z axis and set it at 0. 12. Show vectors and the velocity by default, this option is set to and apply. Done. Let's increase the plane location a little bit so we can see how the velocities look like now. We can also cut out part of it.
So we can clip it by changing these values. And you can look at only the velocity vectors at this location. So this is a very handy tool if you want to just look at a specific location. You can essentially just clip it on an already existing solution and then you can just add it to the view.
So you can see that the velocity is the maximum and minimum temperature and the minimum temperature of the system. Next, you can create a report summary.
Go to report, summary report, set the name, the report face values, essentially to create objects you will have to just select new and select any one. Select the board, accept and look at the temperature on the board. It will automatically give you the solution order once you do that.
Click on write and it will give you a report summary data where you can see the mean temperature and max speed in the cabinet, minimum speed in the cabinet, the mean speed in the cabinet.
Now let's activate radiation because we see the temperatures are reaching almost 120 degrees centigrade so maybe it's a good idea to add radiation to it and see if radiation is present in the solution what will be the maximum temperature inside the cabinet.
So we go to basic parameters and activate radiation. We are going to use the rate-racing radiation model and we will use a coarse setting so let's keep the flow iteration at 5 maximum radiation iteration to 5. Set the phases to 20 and the resolution to 5 as well. Accept.
Now we go back to solve, change the name, we will interpret the previous data, results, write overview and results file and when the solution is finished. And start the calculation.
The calculation with radiation is computed and I can see how the temperatures are lower than the case when there was no radiation. Also, if we create a temperature contour, you and immediately see how the temperature has reduced quite a bit because of the radiation.
Now let's activate failure possibility so we go to fans properties since that it to fail early the next fan since it we also set our calculation laminar the initial velocity one setting did the pressure disk realized reason for dia force average week under relaxation factors for better stability solve.
You can see the solution has computed and because of the fan failures, the cooling efficiency has dropped significantly. And we can see how the temperatures have increased significantly in this case.
So you can see the maximum temperature for AGP has gone up by almost 150 degree centigrade to 269. Again, the DDR RAM temperatures have gone up by several notches, 250s range. The board temperature has gone up significantly.
So it kind of showcases that when there is a cooling failure in the system, you can have very high temperature, which will definitely lead to the failure of the whole component. With that, thank you for your time and we will see you again in the next video. Bye-bye.

