Thermal Simulation

A project log for Modular Continuous Track System

A inexpensive modular tank track system that can be used for robotic and radio controlled vehicles.

jupdykejupdyke 09/10/2015 at 05:400 Comments

I tend to over do projects. Especially when I have access to the tools already. One of those tools is the simulation package. I have don't other simulations when I was in graduate school and over the years I have done several stress simulations to design better parts from a physical loading perspective. But I had never done a thermal simulation on a real part. Just the examples.

I would like to point out that I probably didn't need to do this. The mold is made of aluminum and conducts heat quite well. I could have guessed at the size heater cartridge and probably been fine. The resin I am working with is not does not need precise temperature control either. In fact the data sheet reads:

"Cure time can be reduced with mild heat or by adding Smooth-On “Kick-It®” Cure Accelerator"

I emailed them and asked them what 'mild heat' meant and the response was:

"I would describe mild heat as being anywhere between 100-120°F."

Hardly very scientific. The data sheet mentioned that you could post cure and had a little more information about that.

"After rubber has cured at room temperature, heating the rubber to 150°F (65°C) for 4 to 8 hours will increase physical properties and performance. "

But if you are gonna do something, you might as well over do it. Plus this was a chance for me to learn a little more. I set up the simulation and ran a few test cases. The heater cartridges I was looking at were rated for 250 watts. So I could assume that most of that energy was being converted directly to heat. I ran the simulation with 500 watts of heating and free convection for the whole outside of the mold.

It showed that the mold could get up to 700 degree F. WOW. So I probably don't have any problem heating up the mold. I could use smaller heating elements if I wanted, but these were only $5.89 each and fit the size of the mold well.

Here is a link: (looks like the price went up a few dollars, they are now $7.16)

I don't need or want the mold to be 700 degrees. But it give you an idea for how quickly the mold can heat up to 150 F. I am more concerned with how even the mold temperature at steady state. I have two heating elements and one thermocouple. If the center of the mold is significantly higher temp than the rest of the mold the thermocouple readings will not be accurate. For this simulation I set the heaters to a consistent 150 F and looked at the mold at steady state. This is not 100% accurate. Technically my control is when the thermocouple is at 150 F not the heater elements. But the thermocouple is in between the elements, and the mold is made of aluminum so the difference will be very small.

That is the thing about simulations, they simulate exactly what you tell them to. You have to make good assumptions. In this situation I am assuming there is no difference between the thermocouple temperature and the heaters temperature. Below is a plot of the steady state simulation.

At first glance it might look like there is a big difference between the center of the mold and the outside of it. But look closely the color map. The dark blue is 147.159 degrees and the bright red is 150.003 degrees. So there is only about a 3 degree difference across the whole mold, and probably only about 1.5 to 2 degrees where my actual part are.

Based on this I am pretty happy with the size of the heaters and the placement.