Details : Power supply¶
Week 1¶
Week 2¶
Week 3¶
Week 4¶
Week 5¶
Regulator x 9 V solar panel¶
The previously purchased buck regulator requires a minimum input voltage of 9 V, while a 9 V Voc solar panel typically delivers a lower voltage under normal operating conditions. As a result, it must be replaced with the component that supports a wider input range, specifically from 5 V to 36 V.
REVIEW - Picture of the test
Low dropout linear regulator (LDO) vs buck¶
Both produces a lower regulated voltage from a higher input. However :
-
LDO : Behaves like a smart variable resistor. "Throws away" extra voltage by turning it into heat.
\[ P_{heat} \approx (V_{in} - V_{out}) \times I \]Example:
\[ V_{in} = 9\ \text{V} \]\[ V_{out} = 3.3\ \text{V} \]\[ I = 0.2\ \text{A} \]Voltage dropped across the regulator:
\[ V_{drop} = 9 - 3.3 = 5.7\ \text{V} \]Heat dissipated:
\[ P_{heat} \approx 5.7 \times 0.2 = 1.14\ \text{W} \]For a smaller load current:
\[ I = 0.02\ \text{A} \]\[ P_{heat} \approx 5.7 \times 0.02 = 0.114\ \text{W} \]The heat is 1.14 W here and so is a lot for a tiny chip.
-
Buck
- Switching regulator. Continuously burning extra voltage as heat and rapidly switches current through an inductor and capacitors.
Battery pack¶
It is known that servos can create short spikes which could reach even 4 A. Therefore a pack that is comfotable to reach 5 A class output (i.e. compare to the 2.5 -3A) gives more margin and hopefully for other sensors and devices.
Pass-through and UPS¶
Video of the experiment
The previous battery pack appeared to require unplugging and reconnecting the charging input before charging could resume properly. In contrast, the replacement UPS-style (i.e continues seamlessly without physical manual reconnection from the user) battery pack supports pass-through charging (i.e. charge while using) - allowing the system to remain powered while simultaneously charging from the solar regulator. This also helps maintain uninterrupted 5 V output when external input power fluctuates or is temporarily disconnected
Week 6¶
-
Week 7¶
MOSFET could be used as an amplifier or electronically controlled switch. In this project, it appears that the latter is more of interest.
- MOSFET is a voltage-controlled switch between drain and source (V_Gs)
- It acts like a small capacitor between gate and source
-
Two channel types of MOSFETs :
-
N-channel
- usually low-side switching (interrupting the low side of a load or source near GND)
- Turns on when VGs is positive (gate is more positive than source)
- At a similar price and package size, N-channel often has lower R_Ds (on) than P-channel. Lower R_DS(on) means less voltage dropped inside the transistor and less heat at the same current
-
P-channel
- usually high-side switching (interrupting the high side of a load or source near +5V)
- Turns on when VGs is negative (gate is more negative than source)
- Packages
- SOT-23 is for small/signal jobs
- TO-220 is for amps and heat, often with a metal tab for heat sinking
-
-
Important parameters at the datasheet :
- V_GS max (gate–source): don’t exceed it ( ±20 V is common)
- R_DS(on) at the real condition V_GS: for ESP32 3.3 V drive, care about R_DS(on) @ V_GS = 4.5 V or 3.3 V lines—not only the 10 V line.
- V_GS(th) threshold: tells when it starts to turn on
P-channel and N-channel MOSFETs can be combined to form high-side switching circuits that provide a strong low-impedance path to either the supply rail or ground, allowing the system to switch on and off cleanly and reliably. In contrast, resistors mainly provide a weak default logic state or “hint” to a signal line rather than actively driving it. It os usually simpler and cheaper and so suitable for pull up and pull down task.
Week 8¶
Capacitor stores energy in an electric field between two conductive plates.
-
Types :
- Ceramic MLCC (X7R/X5R/C0G, etc)
- Non-polarized
- Typically used for smaller capacitance values and high-frequency decoupling applications (i.e. stop fast spikes)
- Usually physically compact
- Typically exhibit lower parasitic resistance (ESR) and parasitic inductance (ESL) compared to aluminum electrolytic capacitors
- ESR (Equivalent Series Resistance) refers to the small internal resistance present inside a real capacitorwhich causes energy loss and heat generation when current flows.
- ESL (Equivalent Series Inductance) represents the small inductive behavior caused by the capacitor’s internal structure and leads, which can limit performance at high frequencies.
- Aluminum electrolytic ("can" caps)
- Polarized (must be connected accordingly)
- Commonly used where larger energy storage or bulk capacitance is required
- Typically larger in physical size
- Series vs parallel
- Parallel :
$$
C_{total}=C_1+C_2+\cdots
$$
- Increasing total capacitance
- Providing more local energy storage
-
Series : $$ \frac{1}{C_{total}}=\frac{1}{C_1}+\frac{1}{C_2}+\cdots $$
- AC coupling / DC blocking applications
- High-voltage capacitor stacks (often requiring balancing resistors or additional protection circuitry)
- Parallel :
$$
C_{total}=C_1+C_2+\cdots
$$
- Ceramic MLCC (X7R/X5R/C0G, etc)
At this stage, the detailed physics governing the differences in capacitor size and construction are not explored further.
Designers often begin with known-good capacitor configurations that repeatedly appear in reference designs and datasheets. For many general-purpose embedded systems, exact capacitor selection is not derived from detailed signal integrity calculations, but instead follows practical and proven design conventions that provide stable operation and a reasonable safety margin.
Unless the system involves sensitive analog signal processing, high-speed communication, or high-power switching where waveform quality becomes critical, these standard recommendations are usually sufficient without requiring detailed frequency-domain analysis.
For example, a common starting configuration for regulators such as the AMS1117 is the use of a 10 µF ceramic capacitor on the output together with a nearby 100 nF bypass capacitor. The larger capacitor helps support regulator stability and transient load response, while the smaller capacitor provides local high-frequency decoupling close to the regulator output node.