Lutan-Energy
WEEK 2
I currently have a small project that I'm very passionate about and have been working on: an embodied intelligence/edge AI project.
It's similar to a companion-type desktop robot. My main goal with this project is to build and manage a complete system encompassing perception, cognition, decision-making, and execution.
Below are the hardware circuits I've already built, along with some of my structural plans.
This is the current hardware circuit design, and it's unlikely to undergo significant modifications in the future. The speaker, servo motors (which are inductive loads), and multiple power-consuming sensors may generate various frequencies of noise, as well as voltage drops or instability issues when operating simultaneously (especially the servo motors; I didn't use a driver board due to size constraints). Therefore, the circuit primarily utilizes filtering circuits composed of small-capacity ceramic and electrolytic capacitors (multi-stage decoupling), as well as large-capacity electrolytic capacitors to stabilize the voltage.
In terms of signal processing, due to the specific characteristics of the I2S digital microphone and the power amplifier module driving the speaker, I mainly employed damping resistors and AC coupling circuits.
In addition, there are self-resetting fuses to protect the servo motors, as well as small surface-mount components for electrostatic discharge and surge protection.
This link contains my structural framework and the perception module, which I have mostly completed.
For the project ideas for this course, I mainly have two ideas, both revolving around the main project I'm currently working on.
First, my first idea is to build a solar power station for my small robot to provide auxiliary power to the system. I hope this can save some power consumption and allow for quantifiable monitoring of how much electricity is generated and saved each day. I think this is quite interesting and meaningful.
My second idea is that the initial design of this small robot was to serve as a companion/assistant. Therefore, it should be able to obtain some of my biological information, such as heart rate, blood oxygen levels, and other basic biochemical information. So I'm wondering if I can create a small device that generates electricity using my body heat, collects my biochemical information, and then sends it to the robot. This information would then be used as part of its PERCEPTION STATE to help it build its COGNITIVE abilities and react according to my different states.
I have already purchased an LTC3108 power manager to receive and manage the generated ultra-low voltage electricity. I have also purchased dozens of solar panels that can be freely soldered with positive and negative terminals and connected in series or parallel. I also have bismuth telluride semiconductor thermoelectric generators. For energy storage, I have lithium batteries and am considering purchasing supercapacitors.
WEEK4
My idea was to use water energy to create a light-up lure. This video gave me a lot of inspiration.
So I chose to use a principle similar to that of a homemade generator, where a turbine drives a magnet to rotate, and the magnetic field continuously cuts through the coil to generate electricity.
So I created the first version of the model.
He could probably use AWG 36 enameled copper wire, wound about 200 turns, with a resistance of about 10 ohms.
Then, the nightmare began. Even with the turbine running at full speed through the air—I think the turbine speed should be close to 1000 RPM—the LED light didn't respond at all. I used a multimeter to measure its AC voltage, and it only reached a maximum of about 0.002 millivolts.
At this point, through searching, I roughly located the following problems. First, the number of coil turns was too small. Second, the magnetic field strength and its variation were not large enough, resulting in a small change in magnetic flux. Finally, I should use a red LED with the lowest forward voltage.
So I started testing. Taking into account the buoyancy of the lure, I wound two coils, one with 1000 turns and the other with 500 turns of varying thickness.
I tried various methods, including adding an iron core and back iron to the coil to reduce magnetic field dissipation. I also used alternating magnets to create rapid changes in the magnetic field (I used four, alternating between north and south magnets). However, all these attempts failed. The maximum voltage I achieved was 0.03V, which, while far exceeding the initial 0.002mV, was still a long way from actually turning on the LED. (I should add that, to simulate the actual rotational speed of a propeller in water, the coil I used here rotated at approximately 300-500 RPM, which is low speed.)
I can no longer increase the number of coil turns or the size and number of magnets. I studied the internal structure of the motor and originally planned to use a stator + 8/12 pole magnets, but if I want it to float, it would require too large an empty chamber, which completely deviates from my original intention of making a practical, real-world lure, turning it into a demonstration model, which I don't want.
I did consider building a motor directly in. I know that as a mature product, the motor's construction and energy conversion efficiency are much higher than the simple coil I made myself. However, to ensure the size and weight, I could only choose a small motor like the N20. At low speeds, its power generation capacity is not large, at most 0.2V, which is still far from enough.
I also considered a built-in energy harvesting module to accumulate energy. But I immediately rejected that idea because my goal was a product. Of course, I know it's not a true product; it's far from it. But I feel that constantly expanding its boundaries and content is a cumbersome thing to do. It should be simple and compact, not just a constant stacking of technologies.
So I designed a second version:
In this system where space, weight, and rotation speed are all limited, I think energy can no longer be easily collected and used. Therefore, I think it is reasonable to use this energy as a trigger signal to convert the kinetic energy of water into an electrical signal.
Furthermore, I believe there is significant room for improvement.
The battery can be replaced with this ultra-miniature/button battery. A major reason my current lure is so large is that the reed switch is 14mm long. I have to ensure it's on one side of the cross-section and not too close to the center, otherwise it will be triggered frequently. Therefore, the diameter of the lure's cross-section at the reed switch location is 30mm, which dictates that the lure's size won't be small. I've noticed that some patch reed switches are only 4/5mm long, which can greatly reduce space requirements.So although the process was tortuous and even deviated somewhat from the original intention of hydropower generation, I am still quite satisfied. I believe the approach is simple enough and reliable enough.
These battery-powered, glowing lures for fishing do exist in real products; this isn't just my conjecture.
Finally, I strongly doubt the authenticity of this video. After my painful experimentation, I don't believe that a coil of this size, along with a magnet, could generate a voltage of over 1V when rotating at low speed.😢😭😭😩
Finally, I strongly doubt the authenticity of this video. After my painful experimentation, I don't believe that a coil of this size, along with a magnet, could generate a voltage of over 1V when rotating at low speed.😢😭😭😩
WEEK9
For my final project, I plan to create a small device—which I’ve named "Distant Tides." My goal is for this device to display the real-time tidal conditions of my hometown (a coastal city), conveying this information through visual cues such as varying light intensity and rhythmic pulsing. The device will be powered entirely by solar energy collected here on my end.
I began by performing a series of calculations to assess feasibility. Crucially—and this is a very important point—I had to select the appropriate component models. The solar panel, charging module, battery, and voltage regulator module all required careful selection, as each has its own specific operating range for voltage and current.
I encountered two primary issues: first, a high-pitched whining sound from the voltage regulator module, accompanied by output voltage instability; and second, severe overheating of the ESP32 board—which I attribute to energy loss—resulting from the frequent voltage conversions involved in the 3.7V-to-5V-to-3.3V power path.
The final issue is that the GPIO pin cannot be connected directly to the transistor's base. This places an excessive current load on the pin, causing my board to become extremely hot to the touch.
Comments
Post a Comment