Refrigerated Trailer Cooling in Ansys IcePak
Hi, this is Mingyao from Ozen Engineering. In this video, I'll demonstrate how to set up a refrigerated container simulation using Ansys IcePak. IcePak is an excellent tool for thermal analysis, combining computational fluid dynamics with heat transfer, turbulence, and even mixture and humidity modeling. It's suitable for simulating anything from small devices like a single computer chip to entire data centers. Here, we're focusing on a refrigerated truck.
Initial Setup
We start with the container model, which we'll use for our initial analysis. For this simulation, we'll use imperial units, specifically feet, to simplify the setup:
- X-axis: 20 feet
- Y-axis: 8 feet
- Z-axis: 8.6 feet
This setup reflects the dimensions of our container. Next, we change the wall type of the cabinet to walls on each side and adjust the background to a dark color for better visibility.
Material Properties
We specify the wall thickness as 6 inches and use polystyrene foam as the solid material. This is a basic model, and we can adjust the effective thickness and material properties as needed. The external material is set for radiation purposes.
Heat Transfer Coefficient
We model the container as exposed to ambient air, with a heat transfer coefficient of 5 watts per Kelvin per meter squared for five sides, assuming the bottom is insulated.
Ambient Conditions
We set the ambient temperature to 110 degrees Fahrenheit, simulating a hot day outside. We then set up an HVAC system and a payload:
- Payload: Modeled as solid water with specific heat, density, and conductivity properties.
- Cooling System: Positioned on the ceiling for optimal airflow.
Simulation Parameters
We specify a volumetric flow rate of 1000 cubic feet per minute and extract energy at -10,000 watts to maintain cooling. The simulation includes:
- Velocity, pressure, and temperature analysis
- Buoyancy-driven flow with gravity on the minus Y-axis
- Turbulent flow with standard K-epsilon model
- Solar radiation model for realistic conditions
Running the Simulation
We run a steady-state simulation with a few hundred iterations, using five cores on a simple laptop. We monitor temperature at key points, such as the opening and the block's center. The mesh is a fully structured hex mesh, representing the geometry size.
Results and Adjustments
Initial results showed excessive cooling, so we adjusted the cooling power to 1000 watts. We can save multiple results and adjust parameters for optimal cooling. The simulation shows how temperature changes over time, with options for more detailed analysis.
Transient Simulation
We set up a transient simulation by opening the door at time zero, maintaining previous conditions. This simulation uses automatic time stepping for efficiency. Results show temperature changes and airflow dynamics over time.
Conclusion
This demonstration highlights how quickly you can set up a thermal simulation for a trailer-sized model in IcePak. You can easily adjust equipment, analyze steady-state or transient conditions, and explore temperature changes throughout the system.
If you have any questions, feel free to reach out to us at ozeninc.com. If you enjoyed this video, please subscribe and like on YouTube. Thank you and have a great day!
Hi, this is Mingyao from Ozen Engineering and in this video I'll be looking at how to set up a refrigerated container simulation in Ansys Icepak. Icepak is a great tool for modeling thermal analysis.
It combines computational fluid dynamics as well as heat transfer, turbulence, and even mixture and humidity, among other things. So it's pretty good for modeling a refrigerated truck. It can be used to simulate small devices on the scale of a single computer chip or entire data centers.
We'll be somewhere in the middle here. We always start with the cabinet or, in this case, the container in the model and we're going to use this as our initial analysis. I often change the units to imperial units. So we'll use the imperial units and we'll use the units that are in the model.
We'll use feet, making setting this up a little bit easier. So on the x side, I'll set this up to 20 feet, y is 8 feet, and z is 8.6 feet. Now our cabinet is reflective of our container. We want to go to the cabinet and change the wall type. They'll all become walls on each side.
You can adjust the background too of Icepak and usually a dark background works better because most of the colors are fairly light. I have all six sides to my cabinet and we can put in some material properties for that. I'm going to select all, edit the properties, and specify a wall thickness.
Let's change this to maybe 6 inches and the solid material will be a polystyrene foam. This is a basic model. Let's edit this again. Let's make this effective thickness. So that means we're not physically modeling the property of that. Now, I have to manually adjust each one of these. 6 inches.
Oh, this external material, this is for radiation. We want to use polystyrene. Effective thickness. I should be able to edit all of these at the same time. Okay, now it's editing multiple of them. That will save me some work. 6 inch effective thickness.
Now, I want to select all of them and specify heat transfer coefficient. So right now, we're modeling as if this is exposed to ambient air. The bottom shouldn't really have any heat loss. So maybe I should not select the minimum Y. We'll assume that's insulated. So I will select these five sides.
I'll say this will have a heat transfer coefficient of 5 watts per Kelvin per meter squared. Now, we're going to model this as if we're trying to keep something cool in a warm place. So, going to basic parameters, defaults, we're going to set this to, let's say, 110 degrees Fahrenheit.
So it's going to be a hot day outside. That's the ambient temperature that we will have in the system. And then we can set up some HVAC system, maybe some payload. So, let's start with the payload. We have a block here. We can set it to, I don't know, let's say, suggest a Y first.
So maybe we'll have this on the bottom, sitting in the trailer, and model this as if it's fairly well loaded. So X will start here and end somewhere back here. And then, the Z axis will be kind of... Okay. Let's set up the material property for this block. I'm going to make it water.
So it's going to be solid water. So let's get the material properties. Water is a liquid. So we first select the water material. And you can see it prints out our specific heat, density, conductivity. So then we're going to make a new material using this button here.
This new material, I'll call it water solid. And it's going to be a solid material. Density will be 1000 kilograms per meter cube. Specific heat. We can choose to make this linear or temperature dependent. In this case, I'm going to just make it a constant.
So this is joules per SI unit is 4180. And conductivity will be 0. 6. One thing that's helpful to point out is that right now I just made the walls of this trailer polystyrene. But we can also make it a composite material.
We can change any of these values from isotropic to orthotropic and isotropic by axial or silo orthotropic. So you can see that conductivity type can easily be made into orthotropic. And we can specify different types of material as needed.
But for now, let's just go ahead and make this out of my water solid material. And it reflects back on the other material properties I put in. And it all looks the same. So I think we're good to go. Now, I have a big blob of material in there and I'm going to put in an HVAC system.
So this is cooling. We probably want to put the cooling on top, let it drop down and then we suck it back out somewhere. So let's put in an opening.
One of the nice things about openings in Icepak is that we can make this a recirculation opening, which means whatever we extract a certain amount of energy from the system during the during the simulation process. I want to make this XZ because we want to put this on the ceiling.
And let's go ahead and put it on the ceiling here right in the center. Maybe we want to move it more to the to the back here. So we're going to move this in the X axis, maybe the back here. Maybe... Maybe we can make this three foot long, something like that. Then that's a supply.
So that's where the air, the cold air will come in. Going to switch this to extract and I'll put this on the bottom somewhere down, maybe over here to suck away the the the cold air. Or maybe it should be on the bottom of the wall. Maybe should be on the wall.
We can play with different locations, but let's let's let's put it on this wall for now and move it a little bit over here. And maybe a bit lower. So we start down here somewhere and up somewhere here. OK. OK. So we can specify a temperature change or heat input or extracted.
We can specify a volumetric flow rate in cubic feet per minute. I want to make up some values here. I'll just put a thousand cubic feet per minute of flow rate and we can extract energy from this. So let's do watts, maybe minus ten thousand. Watts.
We're going to try to keep this thing cool and we'll see what what temperature in that. So that's all there is to a basic analysis. We can also import geometry and and have more detailed models of what's inside.
But here is going to give us a simplified view of maybe depending on how full this truck is, what the temperature will be. Could go through the the automatic setup. So we're going to have velocity, pressure and temperature. We can add a humidity later on if needed.
We're going to have this will be a buoyancy driven flow just so we can specify gravity minus Y axis. That also looks just fine. We'll set this as turbulent flow with standard K epsilon. We can include heat transfer.
Probably don't need it for this case, but we can do use a discrete ordinate model here. Standard radiation modeling is included in Icepak. We want to include a solar radiation model so we can specify the solar flux direction or have a calculator. So maybe we want this to be in June at noon here.
OK, so we want to be maybe July July 1st, maybe at three in the afternoon so that will be. Fifteen hundred. And latitude, we're going to do thirty seven point three eight. And minus one to one point nine nine. So this is. This is this is a Sunnyvale location.
And we'll do minus seven to correct for Pacific time. And that's going to put in the solar radiation here. We're going to start with a steady state simulation and that's it. So we're ready to go ahead and do the analysis. Typically, let's let's run a couple of hundred iterations.
Parallel setting, we typically run I have run this on a simple laptop, so I'm going to use five cores. And typically it's a good policy to monitor some locations so we can create monitor surface. Here. And we're going to monitor the temperature of this opening to we can select. This is the opening.
And... That's good. We can all we should probably also monitor the temperature of this block too. And we can just do a monitor point right in the middle of the block. And now we're monitoring the point of the block. So that's ready for simulation.
We can go through the process of generating a mesh here. The model is super simple. So the mesh is extremely small. We can take a look at the mesh here. That's a fully structured hex mesh that represents the size of the geometry. So I'm going to go ahead and run the simulation.
We can make it a bit more refined later on. And we can see that. Oh, doesn't like the minus value. For my longitude. We can step through this. Here and... I think we can just change this to a positive north and west. There you go. Okay. Try that to run this again.
We can enable pressure velocity couple formulation. This sometimes goes a little bit faster. There's an option to use the pseudo transient. Or just let us run by default. Now, it looks like I was way too aggressive with my cooling here. It's going down to over minus 200 degrees Celsius.
So let's dial that back to maybe a 1000 watt. Run it again. And we can run this at save multiple results if needed. Okay. So we can see that we can run this. And we can see that we can run this. And so let's crank up the temperature a little bit. I mean crank up the cooling system a little bit.
Maybe double this. Maybe triple it. You can see how fast that this can be run. And here is let me save this model. So we want to. Save this in demos. Icepak. We'll call this trailer. Okay. So now we can do a series of simulations. So this will be at three kilowatt of cooling. Okay. Okay.
So we're going to run this. Okay. Running this for 200 iterations. I can step it out for a little while. But it looks like we get to minus 16 degrees Celsius here. At the minimum. There are some areas that is a little bit hotter. But it looks like we're doing a good job of freezing. 3000 watts.
One of the nice things about Fluent and Icepak is the ability to continue the analysis to see what happens in transient. So now I'm going to find this door here which is the minus Z and I'm going to go to the cabinet, minus X and turn this into opening.
So suddenly now we've opened the door at time equals zero. But the inside is already cool to a certain extent, so let's go ahead and set up a transient simulation based on this. Did I miss it? No.
So we'll keep everything the same, including radiation, but now we're going to do a transient simulation. So let's set everything to seconds, and then we're going to run this. Starting time is zero, we're going to run this for ten seconds. And you can use variable time step, initial time step.
Let's have it at hundreds of a second. We'll let it finish here. There's actually an interesting capability for transient simulation that wasn't in the list there, and that is this automatic transient. So here we can say initial time step, we're going to have it in hundreds of a second.
And minimum time step is that, hundreds of a second. And maximum time step is one second. So this way ANSYS will automatically adjust the time stepping. Okay. And we're going to save the solution every ten steps. ANSYS will adjust the time stepping depending on how well the convergence is.
So we don't, so it makes it easy to set up these simulations. And I'm going to say transient door open. To make use of previous results, we grab that. So we're going to interpolate the results from my steady state simulation here to my transient analysis. And that would be the starting point.
So let's go ahead and run the simulation. Okay. The simulation has finished. Looks like maybe I could have used a few more steps in here. But you can see that we have from very low time step, we have a very low time step. So we're going to go ahead and run the simulation. Okay.
So we have from very low temperature, sudden heat up of the air. And then the temperature dropping somewhat, which is an interesting phenomenon. So maybe this is the initial rush of hot air and then the cooling system kind of kicking in to once the flow stabilized somewhat.
But you can see the temperature of the actual cargo inside, at least in the center, didn't change much. Let's go ahead and say done. And we can now look at some results. So this is initial condition, suddenly opening the door. Maybe we should extend our... Let's make it a bit thicker. And...
Oh, it's not lighting. Okay. So we're going to extend the streamlines. Well, we can see how the hot air comes rushing in after just a few milliseconds. So this is now after about 10 seconds. Maybe smaller time steps would have been beneficial here. I probably...
And I also only save results every 10 steps. So those all could be results of what happened here. But we can see the temperature changing and rising up while the cool air kind of tries to fill in on the bottom here. So a little transient simulation.
There are a lot more things we can do with Icepak here. We can solve for individual species. In which case, we can start looking at... Different species analysis. For example, here we have air and H2O.
Now we can look at relative humidity of the inside of the block of our trailer and look at how that changes as a function of temperature and flow. We can add dehumidifiers into the simulation. So a lot of different possibilities.
But this is a quick demonstration of how you can very quickly set up a thermal simulation of something like the size of a trailer. In Icepak very easily. You can easily move your equipment around.
Look at steady state or transient simulation and investigate how temperature changes in the entire system. Hope you found this useful. If you have any questions, feel free to reach out to us at ozeninc.com. And if you like these videos, please subscribe and like on YouTube.
Thank you and have a great day.

