In the winter, a fireplace is the coziest place in the house when we need some thermal comfort. It is probably something hard to remove from our living standards and our culture (it is supposed to be the only way Santa comes into your house). But is the concept of fireplace -- an ancient way of warming up a house -- really a good idea today when the entire house is heated by a modern distributed heating system? In terms of energy efficiency, the advice from science is that it probably isn't.
 |
| Figure 1. A fire is lit in the fireplace. |
When the wood burns, a fireplace creates an updraft force that draws the warm air from the house to the outside through the chimney. This creates a "negative pressure" that draws the cold air from the outside into the house through small cracks in the building envelope. This is called
the stack effect. So while you are getting radiation heat from the fireplace, you are also losing heat in the house at a faster rate through convection. As a result, your furnace has to work harder to keep other parts of your house warm.
 |
| Figure 2. No fire. |
Our
Energy2D tool can be used to investigate this because it can simulate both the stack effect and thermostats. Let's just create a house heated by a heating board on the floor as shown in the figures in this article. The heating board is controlled by a thermostat whose temperature sensor is positioned in the middle of the house. A few cracks were purposely created in the wall on the right side to let the cold air from the outside in. Their sizes were exaggerated in this simulation.
Figure 1 shows the duty cycles of the heating board within two hours when the house was heated from 0 °C to 20 °C with a fire lit in the fireplace. A heating run is a segment of the temperature curve in which the temperature increases, indicating the house is being heated. In our simulation, the duration of a heating run is approximately the same under different conditions. The difference is in the durations of the cooling runs. A more drafty house tends to have shorter cooling runs as it loses energy more quickly. Let's just count those heating runs. Figure 1 shows that 15 heating runs were recorded in this case.
Figure 2 shows the case when there was no fire in the fireplace and the fireplace door was closed. 13 heating runs were recorded in this case.
What does this result mean? This means that, in order to keep the house at 20 °C, you actually need to spend a bit more on your energy bill when the fireplace is burning. This is kind of counter-intuitive, but it may be true, especially when you have a large drafty house.
 |
| Figure 3. In a house without cracks... |
How do we know that the increased energy loss is due to the cracks? Easy. We can just nudge the window and the wall on the right to close the gaps. Now we have a tight house. Re-run the simulation shows that only 11 heating runs were recorded (Figure 3). In this case, you can see in Figure 3 that the cooling runs lasted longer, indicating that the rate of heat loss decreased.
Note that this Energy2D simulation is only an approximation. It does not consider the radiation heat gain from the fireplace. And it assumes that the fire would burn irrespective of air supply. But still, it illustrates the point.
This example demonstrates how useful Energy2D may be for all precollege students. In creating this simulation, all I did is to drag and drop, change some parameters, run the simulation, and then count the heating runs. As simple as that, this tool could be a game changer in science and engineering education in high schools or even middle schools. It really creates an abundance of learning opportunities for students to experiment with concepts and designs that would otherwise be inaccessible. Similar experiences are currently only possible at college level with expensive professional software that typically cost hundreds or even thousands of dollars for just a single license. Yet,
according to some of our users, our Energy2D rivals those expensive tools to some extent (I would never claim that myself, though).
Don't get me wrong. JavaScript is an excellent choice for creating interactive Web experiences. I use JavaScript extensively to create Web interfaces for interacting with the Energy2D applet. But I think it is in general healthy for the developer community if we are given more options. Recognizing the weaknesses of JavaScript, the community has already created CoffeeScript and TypeScript (supersets of JavaScript that strips off unproductive features of JavaScript) that also require compilation into native JavaScript. Dart is Google's solution to these problems that should be welcomed. To a Java developer like me, Dart provides a much better option because it returns the power of class-based object-oriented programming to developers who must create Web-based front ends. What is even sweeter is that its SDK provides a familiar Eclipse-based programming platform that makes many developers feel at home.
Excited about the potential of this new language (plus it is from Google and will be highly performant on Chrome), I am announcing the development of the Dart versions of our Energy2D and Quantum Workbench software. These software are based on complex mathematical solutions of extremely complex partial differential equations and will hopefully provide some showcases to anyone interested in Dart. This is not to say the development of the Java versions will cease. We are committed to develop and maintain both Dart and Java versions. 
