Your DVR shows horizontal lines that pulse with throttle. Your RSSI drops 20dB when the VTX powers up. Your blackbox gyro trace looks like a seismograph during an earthquake. These aren’t separate problems — they’re all noise, and they share a common electrical root that most pilots misdiagnose on the first attempt.
The Three Noise Types and Their Symptoms
Noise in an FPV drone is electrical interference that corrupts signals. Three systems are vulnerable, and each produces distinct symptoms.
Video Noise (VTX/OSD Corruption)
Symptom: Horizontal bands that move or pulse with throttle. Brightness changes with motor RPM. OSD elements flicker or ghost.
The camera sends an analog signal to the OSD chip, which overlays text, then routes to the VTX. Noise enters at the power rail. When ESCs draw current, they create voltage ripple — rapid fluctuations on the VBAT line. If that ripple reaches the camera or OSD chip’s power input, it modulates the video signal.
RC Link Noise (RSSI/LQ Drop)
Symptom: RSSI drops 15-25dB when VTX is above 200mW. LQ fluctuates despite close range. Failsafes at distances where other builds fly fine.
The VTX power amplifier generates harmonics — multiples of the transmission frequency. A 5.8GHz VTX at 800mW produces noise at 11.6GHz, 17.4GHz, and importantly, spreads broadband noise in the 2.4GHz and 900MHz bands. If the receiver antenna sits too close to the VTX antenna or the VTX PCB, that noise drowns out the RC transmitter’s signal.
Gyro Noise (Vibration-Induced PID Instability)
Symptom: Motors run hot after 30 seconds of hover. Blackbox shows gyro trace at 200-400Hz with 50+ amplitude. PID traces oscillate at the same frequency. Quad twitches in forward flight.
The gyro chip detects angular velocity. It can’t distinguish between the quad rotating and the frame vibrating. Motor vibrations travel through the frame into the FC. If those vibrations match the gyro’s sampling frequency or the PID loop harmonics, the FC responds to phantom movement, creating a feedback loop.
Step-by-Step: Diagnosing and Fixing Each Noise Type
Video Noise — The Capacitor + LC Filter Approach
Step 1: Install a Low-ESR Capacitor
Solder a 35V 1000µF low-ESR electrolytic capacitor across the main battery pads on the ESC or PDB. Polarity matters — the stripe is negative. Position it as close to the ESC power input as physically possible. Each millimeter of wire between the capacitor and the pads reduces effectiveness.
A 4S build needs minimum 35V rating. 6S needs 50V. The Panasonic FR or FM series are the industry standard — Nichicon UHW if you want the absolute lowest ESR.
Step 2: Check for Ground Loops
If the capacitor alone doesn’t fix it, you have a ground loop. Measure continuity between every ground pad: ESC ground to FC ground to camera ground to VTX ground. All should be 0 ohms. If the camera gets ground from the VTX and the VTX gets ground from the FC, current flows through the signal wire’s ground reference instead of the main ground pad. Solution: run a dedicated ground wire from every component back to the FC’s main ground pad. Star topology, not daisy-chain.
Step 3: Add an LC Filter on the Camera Power Line
If the capacitor and ground fix don’t clean video, the camera power rail is contaminated. Solder an LC filter inline on the camera’s positive wire: inductor on the positive line, capacitor to ground after the inductor. Values: 100µH inductor, 100µF 25V capacitor. The inductor blocks AC ripple. The capacitor shunts remaining ripple to ground. This is a low-pass filter — DC passes, noise above ~1.6kHz is attenuated.
RC Link Noise — Antenna Separation and VTX Shielding
Step 1: Measure the Noise Floor
Arm the quad in a quiet environment. Walk 50 meters away. Have a friend (or a camera on a tripod) watch the RSSI/LQ on your radio. Power the VTX up through 25mW → 200mW → 800mW. If RSSI drops more than 5dB between 25mW and 800mW at the same distance, VTX noise is coupling into the receiver.
Step 2: Separate the Antennas
The receiver antenna and VTX antenna need physical distance. Minimum: 15cm between the receiver’s active element and the VTX antenna. More is better. On a 5-inch frame, mount the RX antenna on a rear arm with a zip-tie and heat shrink — as far from the VTX antenna (typically rear-mounted on the top plate) as geometry allows.
Step 3: Shield the VTX
Wrap the VTX PCB in copper tape, then cover with Kapton tape (copper is conductive — don’t short anything). Ground the copper tape to the main battery ground. This creates a Faraday cage around the VTX, containing radiated noise. Not all VTX models benefit — ones with metal housing (TBS Unify, Rush Tank) are already shielded. Plastic-housed budget VTX modules improve dramatically.
Gyro Noise — Mechanical Filtering
Step 1: Soft Mount the FC
If your FC has rubber grommets, use them. If not, install nylon standoffs with O-rings on both sides of the board. The FC should float — press on it with a finger and it should move slightly. Overtightening the stack screws defeats soft mounting. Tighten until the nut just contacts the O-ring, then stop.
Step 2: Balance or Replace Props
Bent or unbalanced props are the #1 source of gyro noise. A prop strike that looks cosmetic bends a blade by 0.5mm, which creates massive vibration at 25,000 RPM. Method: remove props, install new ones, fly, check blackbox. If noise drops sharply, the old props were unbalanced. Props are $3 a set — don’t bother balancing them. Replace.
Step 3: Check Frame Resonance
Different frame materials resonate at different frequencies. Carbon fiber arms have a natural resonance around 200-300Hz — right in the gyro noise band. If your build has 200-300Hz noise that doesn’t come from props or motors, your frame is resonating. Solutions: thicker arms (6mm vs 4mm), arm bracing, or different frame. Some frames are simply noisier than others — Source One V5 is quieter than Martian II across the board.
Parameter Table: Noise Sources and Solutions
| Noise Type | Frequency Range | Primary Cause | First Fix | Second Fix | Third Fix |
|---|---|---|---|---|---|
| Video bands | 50-200Hz (visible) | ESC PWM ripple | 1000µF cap on battery pads | Star ground topology | LC filter on camera power |
| RC link LQ drop | Broadband (2.4G/900M) | VTX harmonic radiation | Separate antennas 15cm+ | Copper foil shield VTX | Lower VTX power |
| Gyro noise | 200-400Hz | Unbalanced props | Replace props | Soft mount FC with O-rings | Betaflight notch filter |
| Motor noise | Matches motor RPM | Bent motor bell/shaft | Replace motor bell | Check bearing runout | Replace motor |
| Frame resonance | 200-300Hz | Frame arm thickness | 6mm+ arms | Arm bracing | Different frame |
| ESC noise | 24-48kHz (PWM freq) | No capacitor | 1000µF Low-ESR cap | Second cap on ESC pads | PWM frequency change |
Common Mistakes & How to Avoid Them
Mistake 1: Solving Video Noise With a Capacitor, Ignoring Ground Layout
Adding a 1000µF cap to a quad with daisy-chained grounds only helps 50%. The remaining noise travels through ground potential differences. Fix: dedicate 30 minutes to running individual ground wires in a star pattern. It’s tedious. It works every time.
Mistake 2: Raising VTX Power to Fix Video Breakup
Video breakup at range isn’t always VTX power — it’s often noise on the VTX input signal. Raising power amplifies the noise along with the signal. If your DVR shows lines that get more visible at 800mW vs 200mW, your video issue is upstream of the VTX. Fix the noise source, not the symptom.
Mistake 3: Using Blackbox While Blind to Frame Issues
Blackbox log analysis shows gyro noise, but interpreting it requires context. A 300Hz peak could be a bent prop, a resonant frame, or a dying bearing. The fast diagnostic: swap props. Noise gone? Props. Noise unchanged? Check bearings. Still there? Frame resonance.
Mistake 4: Ignoring the ELRS Dynamic Power Interaction
ExpressLRS dynamically adjusts output power based on received link quality. If VTX noise corrupts the telemetry back-channel, the TX sees poor LQ and cranks power to maximum — which doesn’t help because the problem is RX-side noise, not signal strength. Disable dynamic power temporarily when diagnosing. If RSSI stabilizes at a fixed 100mW but fluctuates at 1W dynamic, your VTX is interfering with the RX.
Mistake 5: Overtightening the Stack
Nylon standoffs tightened until the frame compresses defeat soft mounting completely. The vibration path goes from frame → standoff → FC with no rubber in between. Tighten until the O-ring just touches both surfaces, then stop. Use Loctite on the screws if you’re worried about loosening — don’t compensate with torque.
⚠️ Regulatory Notice: The flight recommendations in this article should be followed in accordance with the latest 2026 drone regulations in your country or region. VTX power output limits vary by jurisdiction — 25mW is the maximum unlicensed limit in many countries without an amateur radio license. Always verify local RF power regulations before operating FPV transmitters. Regulations vary significantly between the FAA (US), EASA (EU), CAA (UK), CAAC (China), and other authorities.
Our capacitor selection and power filtering guide covers the full science of capacitor ESR, ripple current ratings, and proper installation technique — pair it with this guide for a comprehensive noise elimination workflow. For the gyro noise side, our Betaflight RPM filtering setup guide walks through the software filtering chain that catches what mechanical fixes can’t.
When noise corrupts your blackbox data, making it unreadable for Betaflight log analysis, start with our Blackbox log analysis reading guide to extract useful information from even the noisiest logs.
The Panasonic FR 1000µF 35V low-ESR capacitor is the one I put on every build. At $1.50 each, I buy them in 10-packs. The difference between a quad with and without that cap on the battery pads is night and day — clean video, stable gyro, no mid-throttle oscillations. If you install one thing on a new build that isn’t strictly functional, make it this capacitor.