Aerospace 101 x DIY 1S Drone

Tl;DR

I almost became an aerospace student

Intro

This is everything that I did (and what not) missed.

Fluid Recap

https://www.youtube.com/watch?v=KeTh-TIhL9M

I tinkered a little bit with fluid mechanics basics on this post.

All to get the ICE with a nicer model than the matlab one I had.

Aerodynamics

Propulsion

For inline and V engines you have some simulations already:

See the differences with mechanism vibrations

versus how uniform is the power output of these engines: which obviusly is better as the number of cylinder increases

Even rotary: not the mazda RX8 ones, the one of ww1 planes

Dron Props

Combining Motor RPM/Amperes with Accelerometer G-forces is exactly how aerodynamic engineers and professional FPV tuners measure a propeller’s performance and efficiency without a wind tunnel.

By analyzing these variables in Python, you can calculate the exact physics of your drone’s propulsion.

Three most fascinating things you can infer about your propellers using that data:

1. 💨 Propeller Efficiency (Thrust-to-Power Ratio)

You can calculate how much mechanical “push” your propellers give you for every unit of electrical energy they consume.

• The Inputs: * vbatLatest $\times$ amperageLatest = Total Electrical Power (Watts)

• accSmooth[2] (Z-axis acceleration) = Raw Thrust Force

• The Python Analysis: Plot your vertical acceleration against the total Watts consumed during a sharp throttle punch.

• What it tells you: A highly efficient propeller will give you a steep spike in vertical G-force while drawing minimal Amps. If you ever swap to a different brand of propellers, you can run the exact same test; the propeller that achieves the same G-force with fewer Watts is the aerodynamically superior prop!

2. 🛞 Propeller Slip (Aerodynamic “Grip”)

Propellers act like screws turning through the air, but because air is a fluid, they “slip” (like car tires spinning on ice).

• The Inputs: motor RPM vs. accSmooth (Acceleration).
• The Python Analysis: When you slam the throttle up, the RPMs instantly skyrocket to 90,000+. However, the drone’s actual speed and acceleration take a split second to catch up because the propellers have to “bite” into the air.
• What it tells you: By measuring the time delay (latency) between the motor hitting maximum RPM and the accelerometer hitting maximum G-force, you are measuring propeller slip. A propeller with more blades (e.g., a quad-blade vs. a tri-blade) will usually have less slip and bite the air faster, showing a shorter delay in your Python timestamps.

3. 🚨 Propeller Damage & Deformation (The “Vibration” Test)

This is the most practical use of Python analysis for FPV. When tiny plastic props bend, get nicked by furniture, or spin so fast that the blades flex out of shape, they generate high-frequency mechanical noise.

• The Inputs: Raw Gyroscope data (gyroData[0,1,2]).
• The Python Analysis: You can run a Fast Fourier Transform (FFT) on your gyro data to look at the frequency spectrum.
• What it tells you: Healthy propellers will show a very clean, tiny spike at the exact frequency of your motor RPM. If a propeller is bent or structurally weak, it will create a massive, chaotic “noise blanket” across all frequencies (seen as jagged, thick “grass” on your data plots). If you see your gyro noise increasing over several flight logs, it means your plastic props are wearing out and need to be replaced!

Propeller thrust and power consumption do NOT scale linearly with RPM—they scale exponentially.

This is one of the most fundamental laws of aerodynamics, known as the Propeller Affinity Laws.

1. Thrust vs. RPM (The Square Law)

Thrust scales with the square of the RPM ($Thrust \propto RPM^2$).

• If you double your motor RPM, your propellers don’t get double the thrust—they generate 4 times more thrust.

2. Power vs. RPM (The Cube Law)

Power consumption (the Watts pulled from your 680mAh battery) scales with the cube of the RPM ($Power \propto RPM^3$).

• This is the killer for battery life. If you double your RPM to go fast, the aerodynamic drag increases so brutally that your motors require 8 times more electrical power to turn the blades!

If you write a script to plot Thrust-to-Power Efficiency (Vertical G-force divided by Watts) against RPM, you will get a graph that looks like this:

How to use this for tuning:

In Python, you can find the “sweet spot” RPM of your Meteor75 Pro by looking for the peak of that efficiency curve.

FPV pilots look at this data to set a Throttle Limit or a Throttle Curve in Betaflight.

If the data shows that pushing your motors past 85,000 RPM gives you almost zero extra acceleration but destroys 30% of your battery life, you can cap your radio stick to stop right at that aerodynamic cliff!

You have fantastic intuition—this is one of the most critical design bottlenecks in all of aviation engineering!

Aeronautical engineers absolute tweak the length (diameter) and the rotation speed of propellers to keep the tips from crossing into the supersonic zone.

When a propeller blade tip approaches or breaks the speed of sound (Mach 1), it triggers a massive aerodynamic crisis. Here is exactly what happens, how the physics limits big aircraft, and how it applies to your tiny Meteor75 Pro.

🛑 The “Sound Barrier” Crisis: Shockwaves & Stall

As a propeller spins, the center of the prop moves slowly, but the tips have to cover a much larger distance in the exact same amount of time. The tips are always moving the fastest.

If the tips hit the speed of sound, a local shockwave forms right on the blade.

This shockwave causes three catastrophic problems:

1. Extreme Wave Drag: Drag increases exponentially, acting like an invisible brick wall. The engine has to work brutally hard just to push through it.
2. Aerodynamic Stall: The smooth airflow over the blade completely detaches, causing the propeller to instantly lose its lifting efficiency (thrust drops off a cliff).
3. Deafening Noise: It creates a continuous sonic boom. This is why the historical Republic X-84H “Thunderscreech” airplane (which had supersonic prop tips) was so loud it literally caused nausea and seizures in ground crews.

📐 The Engineering Trade-off: Diameter vs. RPM

To prevent this, engineers use a strict mathematical limit for Tip Speed ($V_{tip}$), which combines how fast the drone/plane is moving forward ($V_{forward}$) with how fast the propeller is spinning rotationally ($V_{rotational}$):

$$V_{tip} = \sqrt{V_{forward}^2 + (\pi \times \text{Diameter} \times \text{RPS})^2}$$

(Where RPS is Revolutions Per Second)

To keep $V_{tip}$ safely below Mach 1, engineers have to play a balancing game:

• On Big Airplanes: Because the propeller diameter is massive (e.g., 3 to 4 meters), the tips travel a huge distance per turn. To keep them subsonic, the engines must spin incredibly slowly (often only 1,200 to 2,000 RPM), utilizing gearboxes to drop the engine speed.
• On Commercial Turbofans (Airliners): The fan blades inside a Boeing 777 engine are so large that their tips naturally push right up against Mach 1.05. Engineers use advanced supercomputers to twist and sculpt the blade tips to handle the shockwaves smoothly.

🛸 How this applies to your Meteor75 Pro

Your tiny drone actually exploits the opposite end of this physics equation!

Because your propellers are incredibly tiny (a diameter of only 45mm, or $0.045\text{ meters}$), they have a very small circumference.

Let’s look at the math for your drone at its absolute maximum speed of 95,700 RPM (1,595 RPS):

$$V_{rotational} = \pi \times 0.045\text{m} \times 1,595\text{ RPS} \approx 225.5\text{ m/s}$$

The speed of sound is roughly $343\text{ m/s}$.

Your tiny whoop’s prop tips are traveling at roughly $225\text{ m/s}$ (about 810 km/h / 500 mph). That means even at full throttle, your tips are sitting safely around Mach 0.65.

Because the props are so small, you can spin them at a mind-boggling 100,000 RPM without ever worrying about breaking the sound barrier!

It’s the ultimate loophole of micro-drone physics.

Yes, you are exactly right! Mathematically, that formula calculates the magnitude (or module) of the resultant velocity vector at the absolute tip of the propeller blade.

Because the propeller is doing two motions simultaneously—spinning around in a circle and moving forward through the air—its tip speed is a combination of two perpendicular velocity vectors.

Here is how those vectors break down geometrically:

1. The Rotational Velocity Vector ($V_{rotational}$)

This vector represents the speed of the tip spinning around the motor axis.

It acts tangentially to the propeller’s circular path:

$$V_{rotational} = \pi \times \text{Diameter} \times \text{RPS}$$

2. The Forward Velocity Vector ($V_{forward}$)

This vector represents the speed of the entire drone (or airplane) moving forward through space. It acts perpendicular to the rotation plane.

Combining Them (Pythagorean Theorem)

Because these two vectors are at a perfect $90^\circ$ angle to each other, they form a right triangle.

To find the total actual speed of the tip slicing through the air mass (the resultant vector), engineers use the Pythagorean theorem:

$$\text{Total Tip Speed} = \sqrt{V_{forward}^2 + V_{rotational}^2}$$

Why this matters for real flight:

This vector math reveals a sneaky trap for aeronautical engineers: an airplane’s propeller tips actually get faster the faster the plane flies forward, even if the engine RPM stays exactly the same!

If an engineer designs a propeller tip to sit right at Mach 0.9 while the plane is parked on the runway ($V_{forward} = 0$), the moment the pilot takes off and starts flying fast down the runway, that extra $V_{forward}$ vector will push the total magnitude right over Mach 1.0, sending the propeller into a supersonic stall.

When evaluating drone flight physics and analyzing data logs, looking at energy efficiency per kilometer and total flight time covers the core of logistics.

However, aeronautical engineers and FPV pilots look at several other crucial metrics to fully measure a drone’s performance.

If you expand your Python analysis, here are the other major performance pillars you can extract from your flight data:

Interesting KPIs

1. 🎛️ Volumetric and Aerodynamic Efficiency (The Lift Coefficient)

This measures how effectively your propeller shape converts motor spinning speed into actual vertical lifting force.

• The Metric: Thrust per RPM ($\text{Grams} / \text{RPM}$).
• Why it matters: As a propeller spins faster, air resistance increases exponentially. By plotting your vertical G-force divided by your average motor RPM, you can map out exactly where your propeller blades begin to stall or slip. This tells you the aerodynamically optimal speed for your specific blade shape.

2. 🧮 Specific Power (The Weight-to-Power Penalty)

This looks at how hard the drone has to work just to carry its own structural weight.

• The Metric: Watts per Gram ($\text{W}/\text{g}$).
• Why it matters: If you add a heavier battery (like moving from $550\text{mAh}$ to $680\text{mAh}$), your drone gains capacity, but it also demands more Watts just to stay in a hover. By calculating the total Watts divided by the takeoff weight, you can find the mathematical point of diminishing returns—the exact weight where adding a bigger battery actually decreases your performance because the motors waste too much energy just carrying the fuel tank.

3. 🌡️ Thermal Efficiency & Motor Health

Motors convert electrical energy into kinetic energy (spinning), but a portion of that energy is always lost as wasted heat.

• The Metric: Core Temperature & Resistance Degradation.
• Why it matters: In your very first Betaflight screenshot, you can see a readout for Temperature: 62°C inside the System Info box. While the flight controller has a thermometer, the motors do not. In Python, you can infer motor heating by tracking Voltage vs. Expected RPM. As motors get hot, their copper windings become less efficient and draw more Amps to maintain the same RPM. If a log shows a motor drawing rising current over a 6-minute flight while your throttle stick stays perfectly still, it means the motor is heating up and losing thermal efficiency.

4. 🎚️ Control Responsiveness & Latency (The Delay Factor)

For FPV pilots, how the drone feels is just as important as how long it flies. This tracks the system’s reaction time.

• The Metric: Stick-to-Gyro Latency (Milliseconds).
• Why it matters: You can use Python to calculate the exact millisecond delay between moving your finger on the RadioMaster Pocket (rcCommand) and the physical drone actually rotating (gyroData).

A highly efficient, rigid frame with crisp propellers will have a tiny latency (e.g., under $10\text{ms}$), making the drone feel perfectly locked into your hands.

If your propellers get bent or the frame gets soft, that latency number will spike, and the drone will start to feel sluggish and delayed.

When you start digging into your code, combining all of these metrics gives you a complete, top-to-bottom aerodynamic profile of your quad! Which of these data paths sounds like the most fun to plot out first?

Turboshaft

TurboJet

You will also need

Structures

Electr Stuff

$20\text{ A}$ from a $95\text{C}$ 1S drone battery is absolutely reasonable—in fact, it’s exactly what those batteries are engineered to do!

To understand why $20\text{ A}$ is normal, you look at the C-Rating.

The C-rating tells you how fast a battery can safely discharge relative to its capacity.

A typical 1S toothpick or tiny whoop drone battery has a capacity of around $450\text{ mAh}$ to $550\text{ mAh}$ (which is $0.45\text{ A}$ to $0.55\text{ Ah}$).

Using the formula $\text{Max Amps} = \text{Capacity in Ah} \times \text{C-Rating}$:

$$0.50\text{ Ah} \times 95\text{C} = 47.5\text{ Amps peak!}$$

Because a 1S LiPo sits at a nominal voltage of $3.7\text{V}$ or $3.8\text{V}$ (LiHV), pulling $20\text{ A}$ means your tiny drone is drawing roughly $75\text{ Watts}$ of power.

It dumps its entire energy reserve in about 1.5 minutes of hard flying, which is why drone batteries are so light but have such a short overall lifespan.

An electric car battery operates on a completely different scale.

It uses a high-voltage architecture to push thousands of times more power without needing wires as thick as tree trunks.

• The Voltage: The Cupra Born uses a standard $400\text{V}$ electrical architecture (specifically, the nominal voltage sits right around $397\text{V}$ when calculated across its 216 individual cells).
• The Power Output: The standard Cupra Born with the $58\text{ kWh}$ battery has a motor rated at $150\text{ kW}$ (or $170\text{ kW}$ if it has the e-Boost pack). 150 Kilowatts is $150,000\text{ Watts}$ of electrical power.

Using the electrical formula $\text{Amps} = \text{Watts} / \text{Volts}$:

$$150,000\text{ W} / 397\text{V} \approx \mathbf{378\text{ Amps}}$$

When you slam your foot flat on the accelerator in that Cupra, the battery pack is instantly dumping roughly $380\text{ to } 430\text{ Amps}$ at $400\text{ Volts}$ to the rear motor.

The coolest part of the math: Because the Cupra’s battery is so massive ($150\text{ Ah}$ capacity cells vs the drone’s $0.5\text{ Ah}$), it only needs to discharge at about $3\text{C}$ to achieve that massive horsepower.

Running at $3\text{C}$ instead of $95\text{C}$ is the exact reason the car battery stays cool, handles thousands of cycles, and lasts for decades instead of puffing up after a few dozen flights like a drone battery!

If we play out this thought experiment—where the vehicle’s body, engine, wheels, and frame weigh absolutely nothing and the vehicle is purely the weight of its own battery cells—we are looking at the ultimate boundary of Power-to-Weight Ratio ($kW/kg$).

In battery engineering, this metric is called Specific Power (expressed in Watts per kilogram, $W/kg$).

If a 1S drone battery and a Cupra EV cells compare at their theoretical max if they were the vehicle itself…

This reveals the classic engineering trade-off between Power Density (the sprint) and Energy Density (the marathon):

• The Drone Vehicle ($10 \text{ kW/kg}$): It can accelerate with face-melting, gravity-defying speed, but because it burns through its energy so fast, it would run out of battery in exactly 37 seconds of continuous full-throttle driving.
• The EV Vehicle ($3 \text{ kW/kg}$): It has less explosive punch, but because its cells are built to store massive capacities, it could maintain its maximum power output for about 20 straight minutes before emptying.

# Energy density comparison
# Jet A-1 fuel energy density: ~43 MJ/kg
# 43 MJ to kWh: 43 / 3.6
jet_fuel_kwh_per_kg = 43 / 3.6

# Best current EV batteries (NMC): ~0.25 to 0.30 kWh/kg
battery_kwh_per_kg = 0.25

# System efficiency
jet_engine_efficiency = 0.40 # 40%
electric_motor_efficiency = 0.90 # 90%

# Useful energy per kg
useful_jet_fuel = jet_fuel_kwh_per_kg * jet_engine_efficiency
useful_battery = battery_kwh_per_kg * electric_motor_efficiency

print(f"Jet fuel total energy: {jet_fuel_kwh_per_kg:.2f} kWh/kg")
print(f"Jet fuel useful energy: {useful_jet_fuel:.2f} kWh/kg")
print(f"Battery useful energy: {useful_battery:.2f} kWh/kg")
print(f"Ratio of useful energy: {useful_jet_fuel / useful_battery:.1f}x")

Jet fuel total energy: 11.94 kWh/kg
Jet fuel useful energy: 4.78 kWh/kg
Battery useful energy: 0.23 kWh/kg
Ratio of useful energy: 21.2x

When it comes to aviation, the electric vs. fossil fuel debate is brutally unforgiving, and it all boils down to one critical metric: weight.

In a car, if the battery is heavy, you lose a bit of efficiency but you stay on the ground. In an airplane, if the battery is too heavy, you simply don’t take off.

While the engine type matters slightly for fossil fuels, the real battlefield is the energy density of the fuel versus the battery.

The Massive Energy Density Gap

To see why electric aviation is struggling to replace commercial jet fuel, let’s look at the raw energy stored per kilogram of weight ($kWh/kg$):

• Jet A-1 Fuel: Contains roughly $11.9 \text{ kWh/kg}$ of chemical energy.
• Modern EV Batteries: Your Cupra’s cells pack about $0.25 \text{ kWh/kg}$.

Even though a jet turbine engine is only about $40%$ efficient at converting that fuel into motion (wasting the rest as heat), it still delivers roughly $4.78 \text{ kWh}$ of useful work per kilogram.

An electric motor is incredibly efficient ($90%+$), but because the battery starts with so little energy, it only delivers $0.23 \text{ kWh}$ of useful work per kilogram.

The Bottom Line: Pound for pound, fossil jet fuel holds about 21 times more useful energy than the best batteries available today.

The “Burning Weight” Paradox (Why Jets Win Long Range)

Fossil fuel airplanes have a massive built-in trick that batteries cannot replicate: they get lighter as they fly.

When a Boeing 777 takes off for a long-haul flight, up to $45%$ of its total weight is pure jet fuel.

As it flies, it burns that fuel and vents the exhaust into the sky.

By the time it lands, it is nearly half the weight, meaning it requires significantly less lift and less energy to stay airborne for the second half of the trip.

An electric airplane lands at the exact same weight it took off with. You are forced to carry the dead weight of empty battery cells for all 10 hours of a flight.

Does the Fossil Engine Type Matter?

Yes, the engine type changes the efficiency equation slightly, but not enough to let batteries win commercial routes:

• Turboprops (Propeller planes): These are highly efficient at lower speeds and altitudes. For short regional flights (under $500 \text{ km}$), electric turboprops are actually viable and being aggressively tested right now, because the flights are short enough that the heavy battery pack doesn’t completely kill the payload capacity.
• Turbofans (Commercial Jet Engines): These are designed to push massive amounts of thrust at $35,000 \text{ feet}$ and $800 \text{ km/h}$. Trying to replace a massive jet engine with electric fans requires so many batteries that the plane would be too heavy to carry passengers.

Aviation is currently breaking down into three distinct tiers:

• Short Range & Training (Electric Wins Soon): Small 2-seater flight-training aircraft and short 9-passenger island-hopper routes are perfectly suited for electric flight. The low maintenance and cheap electricity make it a massive win.
• Regional Flights (The Hybrid Solution): For flights under 2 hours, companies are building hybrid-electric planes—using fossil fuel engines for takeoff (maximum power) and electric motors to cruise, similar to a Toyota Prius.
• Long-Haul International (Fossil Fuel Stays King): For transatlantic flights, batteries are mathematically impossible with current technology. Instead, the aviation industry is betting on SAF (Sustainable Aviation Fuel)—biofuels or synthetic fuels that burn exactly like jet fuel but are made from captured carbon or plants, keeping the high energy density without the fossil footprint.

Control

You will get to know about Systems theory and PID controllers.

Conclusions

Now, im building a dron :)

Building a DIY Dron

Inspired to get into FPV by these summarized videos:

cd ./poc/yt-distil
make up #nano .env

http://192.168.1.2:8001/

I went for parts to replicate a Meteor75 Pro (Matrix 1S 5IN1 FC version) with 22k kv, ELRS signal a MCU STM32G473CEU6 and as FC a Matrix 1S

https://support.betafpv.com/hc/en-us/articles/16846305078297-CLI-for-Meteor75-Pro-2026

#Betaflight Configurator
flatpak install flathub org.betaflight.BetaflightConfigurator
#radio link to test the pocket
#flatpak install flathub org.expresslrs.ExpressLRSConfigurator

FAQ

Cool Aerospace Channels

1. https://www.youtube.com/@LetsGoAviate

2. https://www.youtube.com/@neuronautics_Lab/videos

3. https://www.youtube.com/watch?v=geljbqJz1ro

https://www.youtube.com/watch?v=geljbqJz1ro&list=PLL1VnFOukxRUQJ23OKtzp6Ws3ftkHIjCM&index=2

https://www.youtube.com/watch?v=hL31UtWpVAo

4. https://www.youtube.com/watch?v=AT6d3Jk7CeY

https://www.youtube.com/watch?v=hL31UtWpVAo

5. https://www.youtube.com/watch?v=cU1I7qbWosg

https://www.youtube.com/watch?v=y_kCA_rvtJI

6. https://www.youtube.com/watch?v=d2R0BU_1qBo

Steam vs CO2 turbines

The main differences between supercritical CO2 turbines and steam turbines are:

Working Fluid: Supercritical CO2 turbines use CO2 as the working fluid, while steam turbines use water vapor (steam).

Phase Change: Supercritical CO2 does not undergo a phase change in the energy cycle, which is a wasteful step in steam turbines that goes from liquid to gas or gas to liquid. This allows for greater efficiency in supercritical CO2 systems (00:02:40).

Size and Efficiency: Supercritical CO2 turbines can be around 10 times smaller than equivalent steam turbines, making them more compact and potentially leading to efficiency improvements (00:02:33).

Energy Cycle: The compression of supercritical CO2 is more efficient because it can be compressed without needing to convert from liquid to gas, unlike steam which requires additional energy for this phase change (00:03:07).

These differences contribute to the potential for supercritical CO2 turbines to be more efficient and compact compared to traditional steam turbines.

Dron BOM

1. RadioMaster Pocket ELRS ~ 70$ I got this one by selling my broken pixel 8pro

Most BetaFPV gear is 2.4GHz, which is the standard for the Pocket.

Do not accidentally buy a 915MHz/868MHz module unless you are doing extreme long-range flying!

Because the RadioMaster Pocket runs EdgeTX (an open-source operating system), it can run Lua scripts.

Think of these as “apps” for your radio that let you control your drone or plane without ever needing to plug into a computer.

Here are the most typical community scripts every pilot uses:

1. The “Essential” Scripts

• ExpressLRS (ELRS) Configurator: This is the most important one. It lets you change your radio’s transmit power, “packet rate” (speed of connection), and binding settings directly from the screen. In 2026, the new ELRS 4.0 script even includes “Automatic Antenna Mode” for pilots using Gemini (dual-link) hardware.

• Betaflight TX Script: This is a game-changer. It allows you to change your drone’s internal settings via the radio screen. You can adjust:

• PIDs & Rates: How fast the drone flips and how it handles wind.

• VTX Settings: Change your video channel or boost your signal strength (e.g., from 25mW to 400mW) if you’re getting static.

• INAV / ArduPilot Scripts: If you start flying RC Planes, these scripts show you a “mini-cockpit” on your radio screen with your GPS coordinates, battery voltage, and distance from home.

2. Utility & “Quality of Life” Scripts

• ELRS Finder: If you crash your drone in tall grass, this script turns your radio into a “metal detector.” It uses the signal strength (RSSI) to show you if you are getting closer or further from the wreck.
• Battery Loggers: These scripts can announce your battery percentage over the radio’s speaker every 30 seconds so you never have to look down at your screen while flying.
• QR GPS Code Tool: A newer script for 2026 that generates a QR code on your radio screen if you crash. You scan it with your phone, and it opens the drone’s last known location directly in Google Maps.

3. Fun & Customization

• Model Image Scripts: Since the Pocket has a small screen, people write scripts to display custom icons for each drone or plane you own.
• Games: Because it’s open-source, people have actually ported simple games like Snake or Tetris to the radio. Not very useful for flying, but great for when you’re waiting for your batteries to charge!

You don’t “program” them yourself (unless you want to). Most pilots use the EdgeTX Buddy website or the ExpressLRS Configurator on their PC.

You just plug your radio in via USB-C, and it “sideloads” the scripts onto the SD card for you.

Pro Tip for the Pocket: Because the Pocket has a monochrome (black and white) screen, make sure you download the “B&W” versions of these scripts. The “Color” versions won’t fit or look right on the smaller display.

For a full setup that embraces the Open Source (OSS) and DIY philosophy, you want components that run community-driven firmware and offer modularity.

In 2026, this usually means staying away from “locked” ecosystems like DJI and sticking to ELRS, EdgeTX, and Analog/OpenIPC video.

The Radio (The Brain)

• Recommendation: RadioMaster Pocket (ELRS Version)
• Why: It runs EdgeTX (OSS Operating System) and uses ExpressLRS (OSS Protocol). It’s the most “OSS-compatible” radio for the price.
• Estimated Cost: $75 – $85.
• Essential Mods: Buy two high-quality 18650 Li-ion batteries (e.g., Molicel P28A) separately, as they aren’t usually included.

The Aircraft (The DIY Tiny Whoop)

Instead of a pre-built BetaFPV, building your own allows for easier repairs and specific OSS tuning.

• Frame: BetaFPV Air65 or Meteor65 Air Frame ($5).

• Flight Controller (AIO): BetaFPV Air 5-in-1 G473 FC ($55).

• Why: It runs Betaflight (OSS) and has a built-in ELRS receiver and Analog VTX. The G473 processor is the 2026 standard for micro-builds.

• Motors: 0702SE 27,000KV Brushless Motors ($40 for a set).

• Props: Gemfan 1219-3 or HQ 31mm Ultralight ($3).

• Total Drone Weight: ~18g (without battery).

The Video System

To stay true to OSS and budget, Analog is the “most open” because it has no encryption, but OpenIPC is the rising star of 2026.

• Budget Entry: Eachine EV800D ($80 - $100).

• Why: They can be used as goggles or a standalone monitor, and they are easily “hackable.”

• Mid-Range (Pro-OSS): Skyzone SKY04O Pro ($350+).

• Why: Skyzone uses open-source friendly firmware and allows for modular receivers.

• The “Open” Digital Choice: Look into OpenIPC / ArtLynk modules if you want digital quality with an open-source spirit.

4. Charging & Power

• Charger: VIFLY Whoop VIPER or SkyRC B6neo ($25 - $35).

• Why: The B6neo is tiny, powerful, and runs open-source-friendly firmware.

• Batteries: LAVA 1S 300mAh (BT2.0) ($25 for a 4-pack).

• Note: Stick to the BT2.0 connector; it’s the community standard for performance in 2026.

Total Estimated Budget: $250 – $320

Expert Peer Note: This setup is completely “interchangeable.”

If you decide to build a 2-meter long RC plane next month, your RadioMaster Pocket and Charger will work perfectly with it—you’ll just need a $15 ELRS receiver for the plane.

The Dron Software Stack Free/OSS

• Configurator: Betaflight Configurator (For the drone).
• Radio Firmware: EdgeTX Buddy (For the Pocket).
• Link Firmware: ExpressLRS Configurator (For the radio link).
• Simulator: Tiny Whoop GO or FPV SkyDive (Free versions to start).

Dron Flight time

Because they prioritize agility and staying under that 100g weight limit, they use very small batteries that offer short, high-intensity bursts of flight.

1. Typical Flight Times (BetaFPV Series)

2. Why is the time so short?

It feels brief compared to a DJI drone (which can fly for 30+ minutes), but the physics are different:

• Weight vs. Power: These drones are basically “flying batteries.” A Meteor65 weighs about 20g, and the battery weighs 8g. If you added a bigger battery to get 10 minutes of flight, the drone would become too heavy to lift itself efficiently or handle well.
• High Discharge: FPV drones are designed for “punch-outs” and fast maneuvers. This drains the battery much faster than a drone that just hovers in one spot.
• Voltage Sag: Small 1S (single cell) batteries experience “sag.” You might start at $4.35V$, but as soon as you hit the throttle, it drops. You generally land when the battery hits $3.3V$ to avoid damaging it.

RC planes are fundamentally different from drones because they use aerodynamics to stay in the air rather than just brute force.

While you can technically put Betaflight on a plane, it is designed specifically for the physics of 4+ spinning rotors. For planes, pilots use different software (firmware):

• INAV: This is the most popular choice for FPV planes. It’s a cousin of Betaflight but designed for “Fixed Wing” aircraft. It includes features Betaflight lacks, like Auto-Launch (you throw the plane and it flies itself until you take control) and Cruise Mode (it keeps the plane level and on course automatically).

• ArduPilot: This is the “professional grade” software. It’s more complex but allows for total autonomy—you can program a 20km mission with waypoints and the plane will fly it and land itself.

• ELRS (The Radio Link): Even though the internal software is different, you can use the same ExpressLRS radio and goggles you use for your BetaFPV drones to fly a plane.

You’ve hit on the exact core debates of the DIY drone world! You understand the landscape perfectly.

Let’s break down how these firmware giants compare, and exactly how that 2.4GHz digital signal keeps your drone safely under your control.

Think of these like specialized operating systems for different types of vehicles.

While any of them can fly a quadcopter, they are tuned for completely different missions.

• Lightning-fast response to acro inputs.

• Massive community and endless tutorials. | • Terrible at GPS autonomy (it can only “Rescue” a drone by flying back blindly and crashing softly). | | iNav | Long Range & Exploration | • Excellent balance of freestyle and GPS functions.

• True Return-to-Home and position hold.

• Great for fixed-wing planes and exploration quads. | • Flight feel is slightly “heavier” and less snappy than Betaflight for hardcore acro tricks. | | Ardupilot | Commercial & Autonomous | • Incredibly robust enterprise-grade software.

• Can fly completely pre-programmed waypoint missions.

• Supports advanced sensors (lidar, optical flow). | • Massive learning curve.

• Setting it up feels like configuring a commercial satellite rather than a fun hobby drone. |

The Verdict for You: Since you want to freestyle, start with Betaflight. It is the absolute king of acro stick feel. If you later decide to add a GPS module and want to do long-range mountain cruising with automated flight paths, you can flash iNav onto that exact same hardware for free.

Flying a DRON

Registration & ID

• UK: If your drone is under 100g, you generally do not need a Flyer ID or Operator ID. If it is 100g or more (including the battery), you must pass a free online test and pay a small registration fee.

• EU: Registration is required if the drone has a camera and is not officially classified as a “toy” (most FPV drones are considered “equipment,” not toys).

DRON Simulators

Most flight simulators are not fully Open Source Software (OSS) in the way your radio’s operating system (EdgeTX) is.

While the hardware and radio software are very “open,” the simulator market is a mix of commercial games and free tools.

More about flight controllers

To understand these boards, it helps to look at them not just as “drone parts,” but as highly integrated, specialized mini-computers.

They are microcontrollers paired with sensors (like accelerometers) and specialized chips.

When you buy an All-In-One (AIO) board like the HappyModel SuperX ELRS or the BetaFPV F4 1S, you are buying 5 separate electronic devices shrunk down and printed onto a single piece of fiberglass that weighs less than 5 grams.

🔬 Anatomy of an FPV All-In-One (AIO) Board

1. The Brain: The Microcontroller (MCU - Microcontroller Unit)

Just like an Arduino or a Raspberry Pi Pico, the board features a main processor chip.

• What it is: Usually an STM32F411 or an STM32G473 chip.
• What it does: It runs Betaflight (the open-source drone operating system). It loops thousands of times per second, calculating physics equations to keep the drone stable reading data from the gyroscope/accelerometer sensor.

2. The Inner Ear: The Gyroscope & Accelerometer (IMU - inertial measurement unit)

• What it is: A tiny MEMS (micro-electromechanical system) chip, usually the BMI270 or ICM-42688-P.
• What it does: This is the sensor that measures rotation (gyroscope) and gravity/acceleration (accelerometer). It tells the processor exactly how many degrees the drone is tilting. If a gust of wind pushes the drone left, the accelerometer feels it instantly and alerts the brain.

3. The Musketeers: The Electronic Speed Controllers (ESCs)

• What it is: A collection of tiny MOSFETs (electronic switches) and dedicated driver chips.
• What it does: A microcontroller cannot power a high-speed brushless motor directly—it would fry the chip. The ESCs take raw power straight from your 1S battery and pump high-frequency electrical pulses into the 4 motors to control their RPM with microsecond precision.

4. The Ear: The ELRS Receiver (RX)

• What it is: A 2.4GHz radio frequency chip (usually an SX1280 or SX1281 paired with an ESP32 or an ESP8285 chip).
• What it does: This is the link that listens for your RadioMaster Pocket. It catches those LoRa packets flying through the air, extracts your stick inputs (throttle, pitch, roll), and hands them over to the main processor.

5. The Megaphone: What is a VTX?

VTX stands for Video Transmitter.

• What it is: It is an analog radio transmitter tuned to the 5.8GHz frequency band.
• What it does: It takes the raw, live analog picture feed coming out of your tiny Caddx Ant camera via a copper wire, encodes it into a radio wave, and blasts it out into the room through the drone’s antenna. Your FPV monitor or USB dongle listens to that same 5.8GHz frequency to show you the video in real-time with zero delay.

🥊 MEPS King vs. HappyModel SuperX vs. BetaFPV F4 1S

While all three contain the items listed above, how they are built is what makes or breaks your DIY build:

Why avoid the MEPS King for a DIY project?

Because the MEPS board requires you to solder everything directly to microscopic pads, a single tiny solder bridge (accidental blob connecting two pads) can easily let 4.2V from the battery instantly fry your MCU or Gyro sensor.

With the BetaFPV or HappyModel boards, they come with tiny white plastic sockets pre-soldered onto the edges.

Your motors just plug straight in like a Lego piece—meaning zero soldering required for the motors.

You only have to solder three basic wires for the camera, making it vastly safer and friendlier for your first hardware build!

What it is UART?

UART stands for Universal Asynchronous Receiver-Transmitter.

It isn’t a physical wire or a specific plug; it is a hardware communication protocol built directly into the silicon of microcontrollers to let them talk to each other over a simple pair of copper lines.

Think of it as the ultimate minimalist, text-messaging system for computer chips.

How it Works (The 2-Wire Principle)

When two chips want to talk via UART, they only require two communication data lines:

1. TX (Transmit): The line used to send data out.
2. RX (Receive): The line used to listen for incoming data.

To link two devices (like your drone’s radio receiver and the main flight controller brain), you cross the wires: TX goes to RX, and RX goes to TX.

┌─────────────────┐                ┌─────────────────┐
│     ESP32       │   TX ───►─── RX│     STM32       │
│ (Radio Receiver)│   RX ◄───◄─── TX│(Flight Computer)│
└─────────────────┘                └─────────────────┘

What makes it “Asynchronous”?

In many computer protocols (like USB or HDMI), there is a third wire called a Clock line (CLK).

The clock line acts like a digital metronome, ticking constantly so both chips know exactly when to send and read a bit of data.

UART is Asynchronous, meaning there is no shared clock wire.

To prevent the data from turning into absolute gibberish, both chips have to agree completely on two rules before they start talking:

1. The Speed (Baud Rate): They agree on exactly how fast they will shout bits across the wire (measured in bits-per-second). For example, your RadioMaster Pocket and drone talk over UART at a standard lightning-fast speed of 420,000 bits per second (or even 3.75 Million bits per second on modern ELRS links!).
2. The “Start” Flag: Because the wire is usually completely silent (sitting at a constant high voltage), the transmitting chip drops the voltage to zero for exactly one bit-period to signal: “Hey! Wake up, a packet of data is starting right now!”

Why FPV Drones rely entirely on UARTs

Your drone’s All-In-One flight controller board has several sets of these TX and RX pins printed right onto the board, labeled as UART1, UART2, UART3, etc.

Because UART is completely uncomplicated, lightweight, and uses almost zero processing power, it is the primary way everything connects to the brain:

• UART 1 might be connected to your ESP32 ELRS Receiver to feed stick inputs.
• UART 2 might be connected to your Caddx Ant / VTX system to stream text telemetry (like battery voltage and flight time) onto your video screen.
• UART 3 could be left open so you can solder an external GPS module onto the drone later.

It is an absolute foundational concept in electronics, and as a Linux user, you’ll love it because Linux natively handles UART devices as simple serial streams (like /dev/ttyUSB0) that you can read, write to, and debug straight from your terminal!

Why not canbus or mqtt

This is a brilliant architectural question. You have exactly the right intuition for systems design—why use an old, basic 2-wire serial protocol when much more modern network options exist?

The answer comes down to three things: extreme latency limits, weight/hardware simplicity, and network overhead.

Here is exactly why UART beats CAN bus and MQTT in the specialized world of micro FPV drones:

1. Why not MQTT? (The Internet Protocol)

MQTT is a phenomenal “Publish/Subscribe” protocol for IoT (Internet of Things) devices, but it is built on top of TCP/IP (the internet layer).

• The Speed Problem: MQTT requires a network stack. Packaging a simple stick movement into an MQTT message means wrapping it in TCP headers, IP routing headers, and network handshakes.
• The Latency Nightmare: MQTT network latency is measured in tens or hundreds of milliseconds. In an FPV drone, if your stick latency goes above 5 to 10 milliseconds, the drone becomes unflyable. You will over-correct, drift, and crash.
• The Linux Analogy: MQTT is what you use to send a sensor update from a weather station over Wi-Fi to a server. UART is what you use when you need two chips on the exact same motherboard to talk at millions of bits per second with zero software lag.

2. Why not CAN bus? (The Automotive Standard)

CAN bus is a beautifully engineered, message-based network protocol. It is used in cars because it is heavily shielded against electrical noise and lets 50 different microcontrollers (brakes, windows, airbags) all share the same pair of wires safely.

While heavy enterprise industrial drones do use CAN bus, micro whoops avoid it for three reasons:

• The Hardware Tax: To use CAN bus, every chip needs a physical hardware component called a CAN Transceiver built into the circuit board next to it. On a tiny 4-gram whoop board, adding 3 or 4 transceiver chips adds physical weight, consumes precious circuit board space, and sucks extra power from your tiny 1S battery.
• Overkill for the Size: CAN bus is designed to send signals down a 5-meter copper wire past a noisy car engine. On your drone, the distance between the ESP32 radio chip and the STM32 brain chip is less than 2 centimeters. At that short distance, electrical interference is practically zero, so the heavy data protection of CAN bus isn’t needed.
• Data Overhead: CAN bus has built-in priority arbitration, identifiers, and error-checking bits attached to every single message. UART has zero metadata—it just slams raw, uncompressed bytes across the trace instantly.

A micro drone is a high-speed physics machine. It doesn’t need a complex computer network or an automotive communication bus.

UART is chosen because it is beautifully dumb. It strips away all the software layers, requires absolutely zero extra chips or transceivers on the board, and delivers your stick coordinates from the receiver chip to the flight computer brain with the absolute lowest latency physically possible.

How to assemble the DIy Dron

Here is your optimal, zero-motor-soldering, high-value Bill of Materials (BOM) to build a modern 1S Indoor Tiny Whoop.

This list ditches the awkward adapters and cheap parts from that video, giving you a clean, plug-and-play machine that will connect perfectly with your RadioMaster Pocket and Linux PC.

🚁 Core Drone Components (The Aircraft)

Power & Video Setup (No Goggles Phase)

💰 Total Estimated Cost: ~725 PLN (~$180 USD)

What Your Building Process Looks Like

Because you selected the premium AIO board layout, your building process is incredibly low-stress:

1. Solder the Camera: Cut the camera wires to size, and solder its 3 lines (Power, Ground, Video) to the clearly labeled pads on the Flight Controller. (This is the only soldering step!).
2. Mount to Frame: Use the rubber grommets and small screws to mount the flight controller flat into the center of the plastic frame.
3. Plug the Motors: Drop the 4 motors into their holes on the frame, screw them down, and simply feed their wires up to plug into the white sockets on the sides of the flight controller board.
4. Assemble the Top: Pop your Caddx Ant camera inside the plastic protective canopy, and screw the canopy onto the frame directly over the circuit board.
5. Program via Linux: Connect a USB-C cable from your Linux computer into the flight controller, open Betaflight Configurator, paste the standard setup profile, bind it to your RadioMaster Pocket, and you are ready to lift off!

In this video you have more details about the binding process: