Aller au contenu

Refroisisseur de Cannette 2

Introduction

The goal of this project is to create a can cooler using a Peltier module. It was inspired by project ideas from the EIRLab website. Our aim is to build a low-cost electric cooler, unlike the expensive ones found on sites like Amazon (over $100). The Peltier module works by cooling one side while heating the other, allowing us to create a temperature difference for cooling

Materials :

2 Peltier module

1 MDF BOARD 6mm

4 K type thermocouple

4 MAX6675

2 MOSFET IRL510 5A

Aluminium

Thermal paste

Wood glue

8 heatsink  30x30x30 mm

2 heatsinks for mosfet 25x34x12 mm

2 Arduino uno R4 minima

2 sunon fans

2 breadboard

2 perfboards

Cables

Resistances

Eletric circuit

For the electric circuit, we created two identical units. Each unit consists of a breadboard with two MAX6675 modules connected to an Arduino board. Each Arduino is linked to a perfboard where a MOSFET is placed. Each unit is also connected to a fan, as shown in the following figure created with Tinkercad. In the diagram, the fans are represented with blue LEDs, the Peltier modules with red LEDs, and the MAX6675 with a motor having six teeth instead of five. We would like to thank Mr. Boussicault for his help with the circuit design.

Code

For the Arduino code: we start with a small PWM (5%) and increase it gradually every second until it reaches 100%. If the temperature difference between the two sides of the Peltier exceeds MAX_DELTA (60 °C in our code), we immediately set PWM to 0. Then, we decrease the difference until it drops below MAX_DELTA / 3 (20 °C), before resuming. The code is at the end of the document.

Model

Due to time constraints, we opted for the simplest model that came to mind for the first iteration. It’s not very effective and doesn’t fully utilize the available materials. The model was created on Onshape, and the link can be found at the end of this document.

In this model, we place the can horizontally in the upper compartiment, where it is in contact with both a support and an aluminum base, created by Mr. Allali. These aluminum components are also useful because they have small holes for inserting the thermocouple to measure temperature. Without this, the thermocouple would need to be placed between the heatsink and the Peltier module, which isn’t ideal, as we want the contact between the can and the Peltier module to be as perfect as possible.

Assembly

For the project, we used the laser cutter because it’s fast and easy to use. You can operate it through the laser cut feature in Onshape.

We used fans for the hot side of the Peltier module to speed up heat transfer, and we made large holes to enhance heat dissipation to the external environment. The fan was added also for the Mosfet also because based on our calculations, the heatsink alone is not enough.

Demonstration

Problems

This project had many challenges, with two main issues. First, as a computer science student, I had never worked with most of the components, making it extremely hard to learn and manage everything at once. Second, I worked on the project alone, which I found very difficult. I had to handle both the mechanical and electrical parts simultaneously, which would have been much easier with two people. I ended up rushing through the mechanical part and leaving it ugly because I realized too late that I didn’t have enough materials, like the aluminum gadgets from Mr. Allali, and should have requested them earlier. The thermal paste was also tricky to use, as it caused components to slip, and I had to improvise with screws instead of properly securing everything. Additionally, I didn’t account for the 0.1mm allowance in the laser cuts on Onshape, which should have been at least 0.2mm, leading me to manually enlarge cuts and making the result look worse. Moreover, even though there were holes to hold the thermocouples, they were not sufficient to keep them in place, so they came loose due to the constant movement of the box. It was impossible to put them back after assembly. Lastly, due to our lack of experience in the electrical field, we didn’t use resistors for the temperature sensors, which led to their failure—since the Arduinos supplied more current than they could handle.

Enhancements

There are several things that can be done to improve the final product, such as:

  • Using PMMA instead of MDF: We used MDF because we were a bit behind schedule, and it’s easier to work with for a first iteration. However, MDF isn’t insulating and allows cold to escape, unlike PMMA, which is a very strong insulating material.
  • Adding buttons for better control: Adding buttons to start/stop the cooling process and to adjust the cooling speed would make the system more user-friendly and efficient.
  • Using plaques à souder (perfboards) instead of breadboards, as they provide better conductivity and the wires stay securely in place.
  • Building personalized components early: For example, we created an idea that could be useful for your project in the future, as shown in the figure below. This idea can also be improved upon.

Conclusion

The project was a bit intimidating at first because it was very different from what we were used to during the last two years at ENSEIRB-MATMECA. It was hard to build the confidence to start, and even something simple like soldering could have taken us a week. But once we got the hang of it, we began doing things more intuitively. It was a very enriching experience, even though I wasn’t able to perfect the techniques due to lack of time and it being our first exposure to them.

Onshape:

https://cad.onshape.com/documents/ec94913537389b14a9294c71/w/bbe9454d9dfb7bb9dda3714e/e/5460d1a8c18dfa08cc46e81c?renderMode=0&uiState=6826cd32f810f654e9dd991d

Code :

https://github.com/simoomar23/arduin/blob/main/project.ino