How to use this guide
Use the row matching your quad. Follow the steps in order. Change only the setting named in each step.
These are exact starting values for this workflow. The logs determine the final result.
Complete test order
- Mechanical check
- Back up Betaflight
- Apply receiver preset and enable bidirectional DShot
- Configure minimal Blackbox fields
- Fly a 40-50 second hover
- Set RPM-filter minimum in PID Tuna
- Set gyro filtering
- Set tuning rates and initial sliders
- Test damping 0.6, 0.8, 1.0, 1.2, 1.4
- Match pitch to roll
- Test master values from the size table
- Check D-term noise
- Set final I
- Test feedforward
- Leave D-Max off
- Perform three punch-outs
- Perform propwash and normal flight tests
- Record final settings
Starting settings
| Quad | Blackbox rate | Starting master | I-term Relax cutoff | I during P/D tests | Feedforward tests |
|---|---|---|---|---|---|
| 2.5-inch light open-prop | 4 kHz | 0.8 | 15 | 0.0 | 0.25, 0.5 |
| 2.5-inch cinewhoop / heavy | 4 kHz | 1.0 | 15 | 0.2 | 0.25, 0.5 |
| 3-3.5-inch light open-prop | 4 kHz | 0.8 | 15 | 0.0 | 0.25, 0.5, 0.75 |
| 3-3.5-inch cinewhoop / heavy camera build | 4 kHz first, 2 kHz later | 1.0 | 15 | 0.2 | 0.25, 0.5 |
| 5-inch freestyle | 2 kHz | 1.0 | 15 | 0.0 | 0.5, 1.0 |
| 6-7-inch | 2 kHz | 1.0 | 8 | 0.4 | 0.5, 1.0 |
| 8-inch and larger | 2 kHz | 1.2 | 5 | 0.4 | 0.5 |
| Large lifter | 1-2 kHz | 1.2 | 5 | 0.4 | 0.25, 0.5 |
For small quads, use 4 kHz for the first hover test. If PID Tuna shows nothing useful above 1 kHz, use 2 kHz for the remaining tests.
Step 1: Check the quad
Before connecting Betaflight:
- Fit undamaged props.
- Tighten prop nuts and frame screws.
- Check arms for cracks.
- Check motor bells and bearings.
- Secure the flight controller, camera, GPS, antenna, capacitor, and wires.
- Confirm no wire can touch a prop.
- Confirm motor order, motor direction, and prop direction.
- Confirm the quad arms and hovers normally.
Do not tune a damaged or loose quad.
Step 2: Back up Betaflight
Open CLI:
diff allSave the output.
Use:
PID Profile 1 = tuning
PID Profile 2 = original/defaultDo not change Profile 2.
Step 3: Basic Betaflight setup
Receiver
Apply the preset matching the real link rate:
ELRS 150 Hz
ELRS 250 Hz
ELRS 500 Hz
Crossfire 50 Hz
Crossfire 150 HzDo not change link rate during tuning.
Loop rate
Keep CPU load below about 50%.
8 kHz gyro: use 8K/4K if 8K/8K is too heavy
BMI270 / 3.2 kHz gyro: use the supported 3.2 kHz rateMotors
Motor protocol: DShot
Bidirectional DShot: ONDo not continue if RPM telemetry errors are high.
Step 4: Minimal Blackbox setup
Keep enabled
Setpoint
Filtered gyro
Unfiltered gyro
P, I, D and feedforward terms
Motor outputs
RPM telemetry
Battery voltageDisable
Accelerometer
Altitude
GPS
RSSI
Attitude
Servos
Raw RC channels
Debug fieldsSet:
Blackbox high resolution: OFF
Debug mode: NONEYou do not need debug fields for normal PID Tuna tuning.
Only for a separate dynamic-notch check:
Debug mode: FFT_FREQ
Debug fields: ONAfter that single diagnostic flight:
Debug mode: NONE
Debug fields: OFFErase Blackbox storage.
Step 5: Hover test
Use the Blackbox rate from the starting table.
- Take off.
- Hover 1-1.5 metres above the ground.
- Hold a steady hover for 40-50 seconds.
- Do not punch out.
- Do not flip or roll.
- Do not make fast stick movements.
- Land gently.
Save as:
01_hoverLoad the log in PID Tuna. Select only the stable hover section:
- Start after takeoff.
- End before landing.
- Exclude ground and disarmed data.
Step 6: Set RPM-filter minimum
In PID Tuna:
- Open Spectrum.
- Select Unfiltered Gyro or Gyro Prefilter.
- Show roll, pitch, and yaw.
- Enable RPM or harmonic overlays.
- Look at the first motor harmonic.
- Find the lowest frequency where the motor peak starts.
Calculate:
RPM-filter minimum = lowest motor frequency - 30 HzExamples:
190 Hz -> set 160 Hz
150 Hz -> set 120 Hz
110 Hz -> set 80 HzIn Betaflight:
RPM filter harmonics: 3
RPM filter minimum: calculated valueSave and reboot.
Step 7: Set gyro filters
2.5-inch and 3-3.5-inch
Gyro Lowpass 1: ON
Gyro Lowpass 2: ON
Dynamic notch: defaultDo not disable Gyro Lowpass 1 during the first tune.
5-inch and larger
Look above 500 Hz in the hover spectrum. If all axes are mostly below about -40 dB and there are no unexplained peaks:
Gyro Lowpass 1: OFF
Gyro Lowpass 2: ONOtherwise:
Gyro Lowpass 1: ON
Gyro Lowpass 2: ONLeave dynamic notch at default.
If a strong peak sits outside the RPM harmonic overlays, fix the prop, motor, arm, camera, antenna, or loose part first.
Step 8: Prepare wobble-test settings
Create a separate rate profile:
Rate type: linear
Roll centre sensitivity: 150
Roll maximum rate: 150 degrees/s
Pitch centre sensitivity: 150
Pitch maximum rate: 150 degrees/s
Angle limit: 30 degreesUse 20 degrees if the quad is difficult to control.
Set:
Feedforward / Stick Response: 0
Dynamic Damping / D-Max: 0
I-term Relax cutoff: use the starting table
I gain: use the starting table
Master multiplier: use the starting table
Damping: 0.6Step 9: Damping sweep
Fly the same wobble test at:
0.6
0.8
1.0
1.2
1.4For every test:
- Take off in Angle mode.
- Move roll and pitch quickly in random directions.
- Use large stick inputs.
- Mix roll and pitch.
- Make about 20-25 movements per axis.
- Do not leave long hover gaps.
- Land.
- Check every motor.
- Change only damping.
- Repeat.
Names:
02_pd_0.6
03_pd_0.8
04_pd_1.0
05_pd_1.2
06_pd_1.4Stop if a motor is too hot to hold, the quad oscillates hard, or a motor desyncs.
Step 10: Choose damping in PID Tuna
Load all damping logs. Open Step Response.
| Result | Action |
|---|---|
| Large overshoot | Use next higher damping |
| Overshoot and rebound | Use next higher damping |
| Repeated ringing | Use next higher damping |
| Small overshoot and clean settling | Keep this damping |
| Slow rounded response | Use previous lower damping |
Example:
1.0 overshoots
1.2 gives small overshoot and settles
1.4 is slower
Use 1.2Choose the lowest damping value that settles cleanly.
Step 11: Match pitch to roll
Compare:
Roll setpoint-to-gyro delay
Pitch setpoint-to-gyro delayIf pitch is more than 4 ms slower:
Pitch Tracking: 1.3
Pitch Damping: 1.3Repeat one wobble test. If pitch is still more than 4 ms slower:
Pitch Tracking: 1.4
Pitch Damping: 1.4Do not go above 1.4 in this guide. If the difference is 4 ms or less, leave both at 1.0.
Step 12: Master multiplier sweep
Fit a fresh battery. Use the damping selected in Step 10.
| Quad | Master tests |
|---|---|
| 2.5-inch light open-prop | 0.8, 1.0, 1.2 |
| 2.5-inch cinewhoop | 1.0, 1.2, 1.4 |
| 3-3.5-inch light open-prop | 0.8, 1.0, 1.2, 1.4 |
| 3-3.5-inch cinewhoop / heavy | 1.0, 1.2, 1.4 |
| 5-inch | 1.0, 1.2, 1.4, 1.6 |
| 6-7-inch | 1.0, 1.2, 1.4, 1.6 |
| 8-inch and larger | 1.2, 1.4, 1.6 |
| Large lifter | 1.2, 1.4, 1.6 |
For each value:
- Fly the same wobble test.
- Land.
- Check motors.
- Save the log.
- Change only master.
Step 13: Choose master in PID Tuna
Open:
Step Response
Setpoint-to-Gyro DelayChoose the highest master before one of these happens:
Delay improves by less than 1 ms
High-frequency oscillation appears
Motors become hot
D-term noise becomes excessive
Motor outputs become roughExample:
1.0 = 28 ms
1.2 = 25 ms
1.4 = 23 ms
1.6 = 22.5 ms
Use 1.4If oscillation starts at 1.4, use 1.2. Do not increase damping to fix oscillation caused by too much master.
Step 14: Check D-term noise
In PID Tuna:
- Open Spectrum.
- Select D Term.
- Compare master logs.
If most D-term noise is below about -10 dB:
Leave D-term filtering unchangedIf a large part is above -10 dB:
D-term filter multiplier: 0.8Repeat one wobble test. If noise remains high or motors are hot:
Restore the previous D-term filter value
Reduce master by one stepExamples:
1.4 -> 1.2
1.2 -> 1.0
1.0 -> 0.8Step 15: Set final I gain
| Quad | Final I slider |
|---|---|
| 2.5-inch light open-prop | 0.8 |
| 2.5-inch cinewhoop | 1.0 |
| 3-3.5-inch light open-prop | 0.8 |
| 3-3.5-inch cinewhoop / heavy | 1.0 |
| 5-inch | 1.0 |
| 6-7-inch | 1.0 |
| 8-inch and larger | 1.0 |
| Large lifter | 1.0 |
Fly one wobble test. In PID Tuna:
Response slopes downward after target -> increase I by 0.2
Slow broad overshoot -> reduce I by 0.2
Neither happens -> keep table valueStep 16: Feedforward tests
| Quad | Feedforward tests |
|---|---|
| 2.5-inch light open-prop | 0.25, 0.5 |
| 2.5-inch cinewhoop | 0.25, 0.5 |
| 3-3.5-inch light open-prop | 0.25, 0.5, 0.75 |
| 3-3.5-inch cinewhoop / heavy | 0.25, 0.5 |
| 5-inch | 0.5, 1.0 |
| 6-7-inch | 0.5, 1.0 |
| 8-inch and larger | 0.5 |
| Large lifter | 0.25, 0.5 |
For each value:
- Fly the same wobble test.
- Land.
- Check motors immediately.
- Save the log.
- Change only feedforward.
In PID Tuna open Raw Traces. Show:
Setpoint
Gyro
P term
Feedforward
Motor outputsChoose the highest value that does not cause:
Large gyro overshoot
Gyro leading setpoint
P term opposing movement
Hot motors
Large motor spikesFor 5-inch and larger, if 1.0 gives little improvement over 0.5, use 0.5. For cinewhoops and large lifters, do not exceed 0.5 in this workflow.
Step 17: Leave D-Max off
Set:
Dynamic Damping / D-Max: 0This simple workflow does not require D-Max.
Step 18: Punch-out test
Fit a fresh battery.
- Hover for 5 seconds.
- Perform a 1-second full-throttle punch-out.
- Return to normal throttle for 2-3 seconds.
- Repeat three times.
- Perform one fast forward pass.
- Perform one hard roll stop.
- Perform one hard pitch stop.
- Land immediately.
- Check every motor.
Save as:
final_punchoutIn PID Tuna inspect:
Gyro
D term
Motor outputs
RPMReduce master by one step if:
- High-frequency oscillation appears.
- Motors are too hot to hold.
- Motor outputs show repeated rapid spikes.
- A motor stays at maximum output.
- A motor desyncs.
If only one motor is hotter, inspect that motor, prop, arm, bearing, and ESC channel.
Step 19: Propwash and real-flight test
Perform:
Propwash descent
Hard 180-degree turn
Fast roll and stop
Fast pitch and stop
Throttle cut and recovery
Sustained carving turn
Normal cruising| Problem | Change |
|---|---|
| Fast oscillation | Reduce master one step |
| Motors hot | Reduce master one step |
| Bounceback after stop | Increase damping by 0.1 |
| Slow broad overshoot | Reduce I by 0.2 |
| Loose in long turns | Increase I by 0.2 |
| Sharp movement overshoots | Reduce feedforward one step |
| Stick response slow, no heat or overshoot | Increase feedforward one step |
| Propwash poor, motors cool | Increase damping by 0.1 |
| One motor hotter than others | Repair hardware |
Repeat only the test related to the setting changed.
Final settings record
Record your final values:
RPM filter minimum:
Gyro Lowpass 1:
D-term filter multiplier:
Damping:
Pitch Tracking:
Pitch Damping:
Master multiplier:
I gain:
I-term Relax cutoff:
Feedforward:
Dynamic Damping / D-Max: 0Keep the original PID profile until several normal flights are completed without hot motors, oscillation, desync, or unstable zero-throttle behaviour.