Diagnostics: VX3.5

Betaflight 4.5.2 (024f8e13d) STM32F405 · SPBE SPEEDYBEEF405MINI · log #1 · example.bbl
Recorded 110 s, in flight 107 s · log rate 1000 Hz · raw gyro: yes · motor RPM: yes
Critical: 1Warning: 4Info: 3OK: 7

Findings

CriticalRear motors are loaded 50% more than the front onesFrame & balance

Average command: rear 28.7%, others 19.2%. The pitch I-term is never zero (45) — it is compensating for the imbalance. The CG is shifted back (battery/camera), or there was a lot of fast forward flight.

What to do: Move the battery so the quad balances level on your fingers under the center of the frame. Extra load on one pair = less thrust headroom, more heat and worse propwash.

WarningRaw gyro noise: 4.1 deg/s (above 80 Hz)Filters & noise

roll 4.1, pitch 2.4, yaw 2.5 deg/s. Noisy — look for the cause in the props/motors/frame, not in the filters.

What to do: Mechanics first: fresh props, check the bearings, soft-mount the stack.

Warningroll: overshoot 36% (90%: 29%–44%)PID

The response overshoots the target (reaches 50% in 8 ms). The figure is already corrected for the method's own bias (2%); based on 165 windows of 372. Current roll P/I/D: [45, 80, 45], FF [120, 125, 120].

What to do: One step at a time: lower P by 10% or raise D (Damping slider). If the overshoot only shows on sharp stick moves, it's feedforward — lower it. Fly again and compare.

Warningpitch: overshoot 28% (90%: 23%–37%)PID

The response overshoots the target (reaches 50% in 9 ms). The figure is already corrected for the method's own bias (2%); based on 295 windows of 577. Current pitch P/I/D: [47, 84, 51], FF [120, 125, 120].

What to do: One step at a time: lower P by 10% or raise D (Damping slider). If the overshoot only shows on sharp stick moves, it's feedforward — lower it. Fly again and compare.

WarningHeavy voltage sag: 3.23 V/cellBattery

It lasted longer than 0.5 s — the pack is working at its limit. The sag happens on current peaks — the pack can't deliver enough current.

What to do: Use a pack with a higher C rating, less capacity/weight, or land earlier.

InfoLogging rate is below the motor frequencyLog

Motors spin up to 610 Hz, but the log is recorded at 1000 Hz (only visible up to 500 Hz). Higher frequencies fold back (aliasing). The per-motor vibration analysis accounts for this, but the noise spectrum and resonance frequencies are ambiguous.

What to do: For accurate diagnostics log at ~2 kHz (PID loop 4000 Hz → divider 1/2).

Paste into the Betaflight CLI (Configurator → CLI), then save
set blackbox_sample_rate = 1/2
save
InfoDid not start on a full packBattery

3.84 V/cell at start (full ≈ 4.2 V).

InfoImpacts/hard landings: 1Flight

109.5 s — 8.8 g (at the end of the log — landing or crash).

What to do: After impacts, check the props and bells — they are what causes imbalance later.

OK1× vibration per motor (deg/s, noise subtracted)Motors & props

M1: 1.1, M2: 0.5, M3: 1.8, M4: 1.9 (the more even, the better)

OKMotor RPMs are consistentMotors & props

Deviation from the others: M1 +1.3%, M2 -0.8%, M3 +1.4%, M4 -1.9%

OKFilters are fine (delay ~1.4 ms)Filters & noise

Noise after filtering: 0.53 deg/s.

OKyaw: response is finePID

Overshoot 6% (90%: 5%–11%), 50% reached in 5 ms.

OKPropwash: 1.3 deg/s after throttle chopsPID

46 throttle chops, oscillation at 15–80 Hz within 0.1–0.6 s after the chop.

OK4S, 3.84 V/cell at startBattery

Minimum under load (sustained, 0.5 s) 3.23 V/cell, at the end 3.66 V/cell, peak current 72 A, ~248 mAh used.

OKInternal resistance ≈ 7 mΩ/cell (estimate)Battery

Based on 182 segments with sharp current changes. Includes the wires and connector; voltage and current are measured by different slow tasks, so this is an estimate.

Charts