Lumipad
Register

Before things break.

repair.html covers what to do after something goes wrong; this page covers what to do before. Scheduled inspections, cycle-based replacement, calendar checks, and the maintenance logbook discipline that turns "fly until it breaks" into "drones that keep flying." The cohort experience: graduates who maintain on schedule have ~60% fewer mid-mission failures than graduates who only fix when something breaks. The discipline takes ~2 hours per month per drone; saves dozens of hours of reactive repair work.

Version 1.0 · Updated 05·2026 Author: Lumipad Engineering License: CC-BY-SA-4.0 Languages: EN · TL · CEB

Triggers, schedules, and the discipline that connects them.

Maintenance happens on three different rhythms: cycle-based (every X flight hours), calendar-based (every X weeks regardless of use), and event-based (after specific events). Sections 2-4 cover each rhythm; Section 5 covers the logbook discipline that makes them work; Section 6 covers fleet maintenance for partner orgs.

The schedules in this page are calibrated against cohort default 5"/7"/10" builds operating in tropical Mindanao conditions — heat, humidity, monsoon-season storage, dust during dry-season operations. Builds in different climates may need different intervals; but cohort defaults are conservative enough that following them anywhere is rarely wrong. The downloadable pack provides printable schedules sized for the field bag and workshop wall.

Section 01 The framework before the schedules

The maintenance mindset.

The shift from reactive to preventive maintenance is psychological more than technical. "Fly until it breaks" feels efficient until something breaks at the wrong time — mid-mission, far from spares, with a cooperative manager waiting. Cohort default operations make the shift early because the cost of a single mid-mission failure exceeds many months of preventive maintenance work.

The four maintenance principles:

Principle What it means Why it matters Common counter-argument
1
Schedule beats memory Maintenance happens at fixed intervals you don't need to remember. The schedule is on the workshop wall; the logbook tracks what's due.
Memory degrades; "I should check that" becomes "I forgot to check that" becomes "the thing I should have checked failed."
"I'll remember to do it." You won't. program graduates surveyed at month 6 reported 80% recollection of "things they should be checking" — much lower than they thought.
2
Replace before failure, not after Components have lifespans. Replace at ~80% of typical lifespan, not when they fail.
A prop that breaks mid-flight costs you imagery, possibly the drone, and a return trip. A prop replaced on schedule costs ₱200 and 10 minutes.
"It's still working." Yes — and tomorrow it might not be. Cost asymmetry is what makes preventive replacement economic.
3
Inspect under load, not at rest Many issues only manifest at full motor RPM, full battery current, full flight envelope. Bench inspection misses these.
Vibration, intermittent connections, motor bearing degradation — all show up at flying conditions, not in the workshop.
"It looks fine on the bench." Bench inspection is necessary but not sufficient. Periodic test flights are part of maintenance.
4
Document what you find Every maintenance event produces data: what was inspected, what was found, what was replaced. The pattern across drones and time is what matters.
Without documentation, every drone is its own special case. With it, the cohort learns which components fail when, why, and how.
"It's just a quick check." Quick checks accumulate into a maintenance pattern. Document them, even briefly.

The reactive-vs-preventive cohort data

program graduates surveyed at the 12-month mark, comparing those who maintained on schedule vs those who flew until something broke:

  • Mid-mission failures: scheduled-maintenance group had ~60% fewer than reactive-only group.
  • Mission completion rate: scheduled group completed planned missions at 94%; reactive group at 78%.
  • Total drone-care hours per month: scheduled group spent ~6-8 hours; reactive group spent ~10-15 hours (more total work because reactive repair takes longer than preventive replacement).
  • Per-cohort cost over 12 months: scheduled group spent ~₱8,000 in maintenance parts; reactive group spent ~₱18,000 in repair parts plus ~₱5,000 in lost-mission costs.

The data is consistent: preventive maintenance is faster, cheaper, and produces more reliable operations than reactive repair. The argument for reactive maintenance ("I'm saving time by not doing maintenance") is empirically wrong. The argument is for habit formation, not cost.

The "trust but verify" principle. Cohort default operations trust components within their service life — props, motors, FCs are reliable when fresh. But "trust" doesn't mean "ignore." Verify regularly that components are within their expected behavior — that's what scheduled inspection is.

The verification is brief: 5 minutes of pre-flight checks (covered in fc-setup.html Section 10), 30 minutes of weekly inspection (this page Section 2-3), 2 hours of monthly thorough check (this page Section 3). The total time is small; the failure mode without it is large.

For new graduates: the maintenance discipline feels like overkill until your first failure that maintenance would have prevented. After that experience, the value becomes obvious. Build the discipline before that experience, not after.

Section 02 Counted by flight hours and battery cycles

Cycle-based maintenance.

Most maintenance is cycle-based — components wear with use, and use is measured in flight hours and battery cycles. This is the largest maintenance category; missing these intervals is what produces most preventable failures. Cohort default schedule below is conservative; aggressive operations may need shorter intervals; light use may stretch them.

The cohort default cycle-based schedule:

Component Inspection interval Replacement interval Cohort cost (PHP)
PROP
Propellers Visual inspection for cracks, chips, imbalance.
Every flight (pre-flight check)
~25 flight hours OR after any visible damage. Replace as a set, never individually.
~₱200/set
MOT
Motors Spin by hand for smooth feel; check screw torque; visual inspection of leads and mounting.
Every 10 flight hours
~150-200 flight hours OR at first sign of bearing wear (gritty feel). Cohort default: replace as set if multiple show wear.
~₱1,200/motor
BAT
Battery packs Inspect for swelling, soft spots, discolouration; verify voltage; check connector for wear.
Every charge cycle (visual)
~70-100 cycles OR any concerning signs. Retire from mission work; use for training only after retirement.
~₱1,500-2,500/pack
XT60
XT60 connectors Visual inspection for charring, looseness; reseat connection feel.
Every 25 flight hours
~200 flight hours OR any charring/looseness. Common wear point especially with frequent battery swaps.
~₱100-200/connector pair
FC
Flight controller Configurator-based health check; sensor calibration verification; mounting standoff inspection.
Every 25 flight hours
No fixed replacement interval; replace when failed. Reflash firmware annually for security/bugfix updates.
~₱2,500-4,500 if failed
FRAME
Frame Crack inspection (focus on arm-to-body joints); fastener torque check; standoff integrity.
Every 25 flight hours
No fixed interval; replace any cracked component immediately. Frame typically outlasts other components by far.
~₱500-1,500 per arm; ~₱2,500-4,000 full frame
RX/GPS
Receiver and GPS module Antenna integrity; mounting; connector inspection.
Every 25 flight hours
No fixed interval; replace if antenna damaged or module shows performance degradation.
~₱600-1,200 each

Tracking flight hours practically

Cohort default flight hour tracking uses three sources:

  • FC firmware: most modern FCs track total armed-time. Visible in configurator → Status. This is the authoritative source.
  • Battery cycle counter: per-pack cycle count tracked in the maintenance logbook. Approximation: 1 cycle ≈ 0.15 flight hours for cohort default 5"/7" build.
  • Field log entries: noted per mission. ~7-9 minutes per battery × number of batteries used. Sums into a per-mission flight-hour estimate.

The FC firmware number is most accurate; field log estimates are usually within ±15%. Use whichever is convenient; consistency matters more than precision.

For partner orgs running multiple drones, the per-drone flight hour tracking is essential — drones diverge in usage rapidly, and assumptions like "all drones flew 50 hours this quarter" are wrong within months. Per-drone tracking from day one prevents this.

The "replace as set" rule for props. When any prop shows damage or reaches replacement interval, replace all four. Reasons:

  • Imbalance: mixing fresh and worn props creates rotor imbalance even if individual props look fine. The drone vibrates; FC compensates; performance suffers.
  • Wear correlation: if one prop reached end-of-life, the others are typically close. Replacing one only delays the next replacement by a few hours.
  • Cost economics: ~₱200/set is trivial. The savings of replacing one prop instead of four are dwarfed by the cost of vibration-induced issues.

The same principle doesn't apply to motors — motor replacement is more expensive (~₱1,200 each) and motors usually fail individually. Replace as a set only when multiple show concurrent wear, which is rare.

Battery cycle tracking deserves its own discipline. Each pack has a unique identifier (cohort default: small label with #1, #2, #3 etc.); the maintenance logbook tracks cycle count per pack; voltage at each charge confirms expected behavior. Patterns emerge: pack #3 sags faster than the others by month 4; pack #5 stays strong past 80 cycles. The cycle data informs both retirement decisions and future battery purchases.

Section 03 Time-based regardless of use

Calendar-based maintenance.

Some maintenance happens because time has passed, regardless of how much the drone has been used. Tropical Mindanao conditions accelerate calendar-based wear — heat softens plastics, humidity corrodes connectors, UV degrades wire insulation. Calendar-based intervals are typically shorter than they'd be in temperate climates.

The cohort default calendar-based schedule:

Interval What you do Why this rhythm Time required
WEEK
Weekly check Brief visual inspection: drone external, battery storage, field bag inventory.
Every 7 days regardless of flight activity
Catches developing issues — connector corrosion, battery storage problems, missing tools — before they affect a mission.
~10 minutes
MONTH
Monthly thorough check Full inspection per the cycle-based schedule (Section 2), plus calendar-specific items: connector reseat-test, FC firmware verification, GPS lock test.
Every 30 days regardless of flight activity
Catches issues that develop slowly: gradual connector loosening, configuration drift, slow GPS performance degradation.
~2 hours per drone
QUARTER
Quarterly deep maintenance Includes monthly items plus: full disassembly inspection, complete reseat of all wiring, ESC firmware verification, battery rotation review, frame fastener torque check with thread locker refresh.
Every ~90 days
Aligns with seasonal patterns; pre-monsoon and post-monsoon are natural breakpoints in Mindanao operations.
~4 hours per drone
ANNUAL
Annual rebuild consideration Decide whether to continue current build, rebuild with same parts, or transition to next-generation cohort default.
Every 12 months
Cohort default builds evolve over time; new versions improve on prior issues. Annual review is when to consider migration.
Decision time, not work time

What weekly check covers (~10 minutes):

  • External visual: drone in storage, no visible damage, no insects nesting (yes, this happens), props undamaged.
  • Battery storage: packs at storage voltage (3.85V/cell ±0.1V), no swelling, fireproof bag in place.
  • Field bag inventory: tools accounted for, spare props present, soldering iron and small spares replenished if used.
  • Charger health: brief power-on check; verify display and outputs working.

What monthly thorough check adds (~2 hours):

  • Full cycle-based items (Section 2): props, motors, batteries, frame, etc.
  • Connector reseat-test: disconnect and reconnect every connector. Tropical humidity causes contact oxidation; reseating cleans the surface.
  • FC firmware verification: connect to configurator; confirm firmware version matches cohort default; check for any unintentional config drift.
  • GPS performance test: powered hover with GPS lock; verify satellite count and HDOP match expected baseline.
  • Calibration verification: accelerometer, magnetometer if present. Re-calibrate if drift detected.

What quarterly deep maintenance adds (~4 hours):

  • All monthly items.
  • Partial disassembly: separate top plate from frame; inspect internal wiring; verify no abrasion.
  • All-wiring reseat: every connector, every soldered joint visual inspection.
  • ESC firmware verification: ESC firmware version, response check.
  • Frame fastener torque check: every screw checked with proper torque driver if available; thread locker refreshed where needed.
  • Battery rotation review: which packs are aging fastest? Retirement decisions for any nearing end-of-life.
  • Test flight: complete the maintenance with a full-envelope test flight, not just a hover.

Tropical climate considerations

Mindanao operations face calendar-based wear faster than temperate climates. Specific factors and cohort defaults that address them:

  • Humidity → connector oxidation. Monthly reseat-test is the cohort default; without it, connectors develop intermittent failures within 3-6 months.
  • Heat → plastic softening. Storage in air-conditioned spaces is ideal but rare. Second-best: shaded, ventilated storage; never direct sun. Some cohort graduates keep drones in foam-lined hard cases that buffer temperature.
  • UV → wire insulation degradation. Wires routed through the airframe are mostly protected; wires that route externally degrade faster. Inspect external wiring in monthly checks.
  • Monsoon humidity. June-October has highest humidity; equipment stored in this period needs more frequent inspection (weekly rather than monthly is common cohort default during these months).
  • Salt air in coastal cooperatives. Coastal operations accelerate corrosion further; consider monthly disassembly-and-clean rather than quarterly for builds operated near the sea.

The cohort default schedule above assumes generic Mindanao conditions; specific operating contexts may need shorter intervals. Adjust based on what you actually find during inspections — if monthly checks reveal accelerated wear, shift to bi-weekly.

The annual rebuild consideration is decision time more than work time. Each year, an alumna evaluates each drone:

  • Cumulative flight hours: compare against cohort default 100-200 hour lifespan.
  • Cumulative repair history: how many times has this build been worked on?
  • Performance trend: is the drone still meeting cohort default specs, or has it degraded noticeably?
  • Cohort BOM evolution: have new components become available that would justify rebuild rather than maintain?
  • Operator confidence: do you still trust this drone for high-stakes missions?

Three outcomes from annual review: continue maintaining (typical for builds under 80 hours and no major issues), major refurb (replace several aging components in one session, refresh the build), or retire and rebuild (per repair.html Section 6 retirement criteria). The annual decision is when to stop reactive maintenance and make a strategic call.

Section 04 Triggered by specific events

Event-based maintenance.

Some maintenance is triggered by specific events rather than schedules — a hard landing, water exposure, extended storage, post-firmware-update verification. Event-based maintenance is what catches the issues that calendar and cycle schedules miss: the rare but real situations that cause concentrated damage. The protocols below are cohort defaults for the most common triggering events.

Event Trigger criteria Inspection items Notes
HARD
Hard landing Any landing harder than normal — the drone bounced, tipped over, or the operator winced.
Pilot judgment; "harder than I'd want to repeat" is the threshold.
Full post-crash inspection per repair.html Section 4 (10-step). Even if no visible damage, accelerometer recalibration and frame fastener check.
~30-45 minutes
CRASH
Crash Drone landed not on the launch zone; impact was significant. Even minor crashes count.
Any unintended ground contact at speed.
Full post-crash inspection plus replace the prop set as a precaution. Test flight in controlled conditions before any mission.
~60 minutes minimum; longer if damage found
WATER
Water exposure Rain, splashed, briefly dropped in puddle. Excludes full submersion (covered as crash).
Any water contact, even if drone seemed fine afterwards.
Power off; remove battery; air-dry for 24+ hours; rice or silica gel storage if available. Inspect for corrosion before next flight; especially connector pins.
24+ hour drying minimum; inspection before re-use
STORE
Storage 30+ days Drone hasn't been flown for over a month. Common during monsoon season.
Any extended period without flight activity.
Full pre-flight check including all 22 items from fc-setup.html Section 10. Battery condition verification. Configuration verification. Test flight before mission use.
~1 hour rebuild discipline
FW
After firmware update FC, ESC, or receiver firmware was updated.
Any firmware change to any component.
Full configuration verification; sensor recalibration; bench arming test; tethered hover; free hover; controlled flight before mission use. Per firmware.html Section 6.
~1-2 hours including test flight progression
PART
Component replacement Any non-prop component was replaced (motor, ESC, FC, receiver, GPS).
Any work that touched soldered connections.
Complete progressive test sequence per repair.html Section 4: bench → tethered hover → free hover → controlled flight → mission rehearsal.
~30-45 minutes for verification
TEMP
Extreme weather exposure Operated in unusually hot conditions (>38°C ambient), extreme humidity, or dust storm.
Conditions outside normal cohort default operations envelope.
Cool-down before storage; thorough cleaning; connector reseat-test; battery temperature check before charging.
~30 minutes

The post-storage rebuild discipline

Drones returning from extended storage (especially monsoon-season hibernation) need full pre-mission-rebuild treatment, not just a quick check. program graduates report higher failure rates in the first mission after storage than in any other context — something always seems to need attention.

Why storage triggers issues:

  • Battery self-discharge. Even at storage voltage, packs drift over time. Some may have dropped below safe minimums.
  • Capacitor formation. Electrolytic capacitors in ESCs and FC need brief reformation after extended storage. Usually self-resolves within first power-on, but worth verifying.
  • Configuration drift. Some FC parameters can be subtly different after long unpowered periods (rare but documented).
  • Mechanical settling. Screws, mounts, and standoffs can loosen during temperature cycling without flight vibration to keep them seated.
  • Operator skill atrophy. The pilot is also "stored" — flight skills degrade in a few weeks. Simulator practice (simulator.html Section 6) before first post-storage mission is cohort default.

The cohort default rebuild discipline: treat the post-storage drone as essentially a new build needing pre-flight verification. Skip nothing on the first flight after extended storage.

The "minor crash that doesn't count" trap. Pilot lands harder than ideal; drone tips over but seems unharmed. The temptation: "no damage, continue flying." The cohort default: treat it as a hard landing; full inspection per repair.html Section 4; possibly recalibrate accelerometer; verify no developing issues.

The current cohort had two cases where graduates continued flying after "minor" hard landings, only to have a more serious failure 1-3 missions later. The post-mortem in both cases revealed damage from the original hard landing that wasn't obvious initially. Hard landings always count; the inspection time is small relative to the cost of compounded failures.

The water-exposure paranoia principle: when in doubt, treat any water contact as significant. Tropical humidity is forgiving of brief contact; monsoon-grade water is not. The 24-hour drying-and-inspection cycle costs you a day of flying; flying a corroded FC costs you everything from this point forward. The asymmetry justifies caution.

Section 05 The discipline that connects everything

The maintenance logbook.

The maintenance schedule, by itself, is just a checklist. The logbook is what connects the schedule to your specific drone's history — what was done when, what was found, what patterns emerged. Without the logbook, every maintenance event is independent; with it, the cumulative record reveals which drones, which components, and which conditions are reliable, and which aren't.

The cohort default logbook structure: per-drone, per-component, with 5 fields per entry.

Field What goes here Why Format
1
Date When the maintenance event occurred.
Calendar tracking; correlates with environmental conditions.
YYYY-MM-DD
2
Trigger Why this maintenance happened: scheduled (cycle/calendar), event-based, repair, etc.
Pattern recognition: are you doing more reactive than scheduled?
Short label: "scheduled-monthly" / "post-hard-landing" / "repair-yaw-drift"
3
Items inspected/replaced Specific components touched: visual list.
Per-component history; which specific parts have how many cycles.
Bullet list: "props replaced (set), motor #2 inspected ok, XT60 reseated"
4
Findings What you observed beyond expected baseline.
The pattern data: what's wearing faster than expected, what's lasting longer, what's anomalous.
Free-form notes; 1-3 sentences typical
5
Time and parts cost How long the maintenance took; what it cost in parts.
Operating cost tracking; informs pricing decisions and partner-org budgets.
"45 min, ₱200 props"

What the logbook reveals over time:

  • Per-drone reliability patterns: drone #1 needs less maintenance than drone #2 → maybe issues with drone #2's assembly or build conditions.
  • Per-component lifespan distributions: cohort default motors lasted 165 hours average vs 200 hour expectation → maybe operating conditions are harder than baseline assumes.
  • Seasonal patterns: more issues during/after monsoon; specific failure modes correlated with humidity.
  • Operating cost trends: ₱X per flight hour in maintenance; how this trends informs pricing and capacity planning.
  • Build comparisons: cohort default 5" vs 7" maintenance costs; which justifies the build choice for given operations.

Cohort engineering aggregates anonymised logbook data across graduates for cohort-default refinement. If 60% of graduates report a specific motor failing earlier than expected, the cohort default BOM gets updated. Your logbook entries contribute to better defaults for future cohorts. Participating graduates Slack discussions sometimes reveal patterns nobody saw individually.

Logbook format options

Different formats work for different operators; pick whichever you'll actually use:

  • Paper notebook: cohort default for many graduates. Always available, no charging needed, dirt-resistant. Doesn't aggregate well across drones.
  • Spreadsheet (Google Sheets, Excel): cohort default for partner-org fleet operations. Columns per drone; rows per maintenance event. Pivots well for pattern analysis.
  • Notes app on phone: convenient if you always have your phone. Less structured but adequate for individual graduates.
  • Custom log software: a few graduates have built personal logging tools. Useful if you have the inclination, not necessary.

The cohort default download pack (top of this page) includes a printable paper notebook template and a Google Sheets template that maps to it. Pick one; stick with it; the value compounds.

The single biggest mistake is starting a logbook and abandoning it after a few weeks. Sparse but consistent beats detailed but abandoned. Five fields per entry, 2 minutes per maintenance event — sustainable indefinitely.

Reviewing the logbook periodically. Quarterly review of the per-drone logbook reveals patterns that aren't obvious in individual entries:

  • Which drones are trending toward retirement decisions?
  • Which components are wearing faster than expected — is the cause environmental, operational, or build-quality?
  • Are scheduled maintenance intervals appropriate, or should they be adjusted based on what you're actually finding?
  • What's the per-flight-hour operating cost trend?

The quarterly review also identifies maintenance habits that work and habits that don't. The discipline calibrates itself if you let it — what gets measured improves; what doesn't get measured drifts.

Section 06 For partner orgs running multiple drones

Fleet maintenance.

Individual alumna with one drone can maintain it ad hoc with the schedules above. Partner orgs with 5+ drones need structured fleet maintenance — staggered schedules, shared spares, dedicated maintenance time, fleet-wide pattern recognition. This section covers what changes at fleet scale and the cohort recommendations for partner-org operations.

What changes at fleet scale:

Aspect Individual alumna Partner-org fleet (5+ drones) Why the difference
SCH
Schedule structure How maintenance time is allocated.
Ad hoc; alumna fits maintenance around mission schedule.
Staggered fleet schedule: maintenance for drone #1 on week 1, #2 on week 2, etc. Distributes workload evenly.
Fleet-scale work would overwhelm a single dedicated day; staggering keeps capacity up.
PERS
Personnel Who does maintenance.
The alumna who flies it.
Designated maintenance lead (~10-20% of someone's time) supporting per-operator inspection. Specialised for complex repairs.
Specialisation produces better outcomes; fleet scale justifies dedicated capacity.
SPAR
Spares inventory What's kept on hand.
Field bag (~₱4,000-6,000) + workshop minimums.
Centralised fleet inventory ~2-3x per-drone equivalent. Bulk-order pricing. Faster restocking. Standardised across drones.
Fleet pool is more cost-effective; bulk orders save 10-15% vs individual purchases.
LOG
Logbook Maintenance record-keeping.
Personal paper notebook or single spreadsheet.
Shared spreadsheet with per-drone tabs; dashboard view of fleet health; automatic alerts when maintenance is due.
Pattern recognition needs structured data; ad hoc notes don't aggregate.
RTM
Replacement timing When components get replaced fleet-wide.
Per-drone basis as components reach interval.
Batched replacements: when one drone needs new motors, check the others — those approaching interval get replaced too.
Batching reduces total downtime and shipping cost; takes advantage of bulk pricing.

The cohort default fleet maintenance schedule for partner orgs:

  • Daily: per-drone pre-flight and post-flight checks (responsibility of operating pilot).
  • Weekly: maintenance lead does fleet visual review; schedules upcoming maintenance.
  • Monthly: rotating monthly thorough check — one or two drones per week, all drones covered each month.
  • Quarterly: deep maintenance batched for the fleet, typically over 2-3 maintenance days.
  • Annual: fleet-wide rebuild/retirement review; cohort default BOM update consideration.

Staggering matters: a partner org with 8 drones doesn't do all 8 monthly thorough checks on the same day. Two drones per week, four weeks per month, all drones covered in rotation. The maintenance lead's time is steady rather than spike-loaded.

Fleet maintenance cost economics

Cohort experience with partner-org operations:

  • Maintenance time cost: ~10-15% of total operator-hours go to maintenance. Less than that suggests under-maintenance; more suggests inefficiency.
  • Maintenance parts cost: ~5-8% of fleet capital cost annually. ~₱25,000 per ₱400,000 fleet, roughly.
  • Fleet downtime: well-maintained fleets have ~5-8% drone-days unavailable per month. Reactive-only fleets average 15-20% — fewer drones available than the books suggest.
  • Maintenance lead salary cost: 10-20% of one role; ~₱8,000-15,000/month at typical Davao rates. Pays for itself in fleet uptime improvement.

The economics favour structured fleet maintenance once you're at 5+ drones. Below that scale, the maintenance lead role is too much overhead; individual operator-maintenance is appropriate. Above 10 drones, the maintenance lead becomes essential rather than optional.

Cross-fleet pattern recognition is the major fleet-scale advantage. With 5+ drones operating in similar conditions, patterns become statistically meaningful: motor #X is failing 30% earlier than expected → maybe a bad batch, or maybe operating conditions are harder than estimated. Individual graduates see "my drone has motor issues"; partner-org fleets see "motor model Y has shorter life in our operations" — the difference shapes future BOM decisions.

Cohort engineering and partner-org maintenance leads share aggregated data via the graduates Slack workspace. Anonymised reports become the basis for cohort default updates. If you're running a partner-org fleet, contributing your aggregated data makes everyone's maintenance better.

The fleet retirement calendar. Partner orgs typically replace ~20-30% of their fleet annually — drones reaching retirement criteria (per repair.html Section 6) get rebuilt or replaced; that becomes the new addition to the fleet. Planning for this rotation is part of fleet financial management. The cohort default fleet capital cost is amortised over ~3-4 years; planning for rebuild costs over that period prevents budget surprises.

Six numbers across maintenance operations.

Reference values for cohort default schedules and the cost economics of preventive vs reactive maintenance. Useful for budgeting time and capital.

~25 hr
Prop replacement interval
Replace as set, not individually
~70-100
Battery cycles before retirement
From mission work; training-only after
~150-200 hr
Motor lifespan
Cohort default 5"/7"; tropical conditions
~2 hr/mo
Per-drone monthly time
Cohort default thorough check
~60%
Failure reduction
Scheduled vs reactive maintenance
~5-8%
Annual fleet maintenance cost
As fraction of fleet capital cost

Four cases from cohort and partner-org operations.

Specific situations from cohort 02 and 03 records, plus partner-org reports, illustrating where maintenance discipline produced different outcomes than reactive operations would have.

"The motor caught on schedule."

Later-cohort alumna, year 1 post-graduation

During monthly thorough check at 145 flight hours, motor #3 showed slight gritty feel when spun by hand — early bearing wear. Cohort default replacement interval is 150-200 hours; the alumna replaced motor #3 on schedule plus motor #1 (which showed similar early wear). Total time: 90 minutes; cost: ₱2,400. Without the scheduled check, motor #3 would likely have failed mid-mission within the next 20-30 hours. Lesson: scheduled inspection caught the wear at "replace conveniently" stage, not "fail unexpectedly" stage.

"Post-monsoon rebuild paid off."

Current-cohort graduate, after 4-month wet season storage

Drone stored from June to October during heavy monsoon. Pre-storage was at storage voltage; bag was sealed; environment was relatively dry. Pre-flight rebuild discipline (Section 4) revealed: 1 battery slightly below safe minimum (retired), 3 connectors with mild oxidation (cleaned and reseated), accelerometer drift on configurator (recalibrated). None individually critical, all together would have caused subtle issues. The 1-hour rebuild prevented what would have been a marginal first-mission performance. Lesson: storage rebuilds catch the accumulated minor issues that storage produces.

"The XT60 that nobody noticed."

Later-cohort alumna, mid-mission failure

Drone lost power mid-mission; emergency landing in cleared field, no damage. Diagnosis: XT60 connector had developed internal contact issue after ~250 charge cycles. Visible only after disassembly. This is the connector pattern that cohort default monthly XT60 inspection now exists to catch. Pre-incident XT60 wasn't inspected since build; ~₱100 connector replacement would have prevented the mid-mission emergency. The maintenance schedule was updated cohort-wide as a result of this incident.

"Fleet logbook revealed the bad batch."

Partner-org with 8-drone fleet, late 2024

Fleet maintenance lead noticed 3 of 8 drones had shown unusual ESC issues within 60 days of each other. Logbook review: all three drones had ESCs from the same batch, ordered together. Confirmed via ESC firmware version and serial-number ranges. The batch had a manufacturing defect. Action: replaced ESCs on the affected 3 drones proactively (before failures) and avoided that batch in future orders. Lesson: fleet logbook revealed a pattern invisible to individual drones. Without aggregated data, this would have been "three random ESC failures" instead of a manageable systemic issue.

Questions worth answering carefully.

The schedules feel like a lot of work. Can I get away with less? +

You can — but cohort experience is that "less" usually means "reactive operations" which costs more total time and produces worse outcomes. The trade-off depends on your operations:

  • Casual hobby flying: yes, the cohort default schedule is overkill. Annual visual inspection and obvious-damage replacement is enough.
  • Occasional paid surveys: at minimum, the cycle-based schedule (Section 2) and post-event triggers (Section 4). Calendar-based can be lighter.
  • Regular paid surveys / commercial work: the full cohort default. The discipline is the difference between reliable-livelihood and intermittent-failures.
  • Partner-org fleet operations: full schedule plus fleet-specific structures (Section 6). Anything less doesn't scale.

Honest assessment: the cohort default is calibrated for working graduates who depend on their drones. If you fly weekly for clients, follow it. If you fly monthly for fun, scale back. The schedule should match the stakes of your operations.

What if I don't track flight hours? Can I do this by calendar alone? +

Partially. Calendar-based maintenance (Section 3) works without flight hour tracking. Cycle-based maintenance (Section 2) needs hours; without them, you're estimating.

Estimation approach if you don't track:

  • Estimate flying frequency: typical alumna flies ~4-8 hours per month during active season, 0-2 hours per month during monsoon.
  • Multiply by months since last maintenance event.
  • Apply cycle-based intervals against the estimate.

Estimation is ±30-50% accurate. That's acceptable for non-critical components (frame fastener checks); insufficient for critical ones (motor lifespan, battery cycles).

Cohort recommendation: start tracking. The FC firmware tracks armed-time automatically; reading it monthly costs zero effort. Within 6 months you'll have accurate data; until then estimate conservatively. Time investment: ~30 seconds per check.

Can I skip maintenance during the rainy season when I'm not flying? +

No — calendar-based maintenance still applies. In fact, monsoon storage often requires more attention, not less.

Monsoon-specific issues that develop without flight activity:

  • Battery self-discharge below safe levels: packs at storage voltage drift over months; some drop below 3.5V/cell which damages cells permanently.
  • Connector oxidation: high humidity causes contact corrosion even on idle equipment.
  • Mould and fungal growth: tropical humidity in unventilated storage creates conditions for biological growth on components.
  • Insect intrusion: insects nest in inactive equipment. Common Mindanao issue with stored drones.

Cohort default monsoon-season maintenance: weekly visual check (~10 min); monthly battery voltage verification and brief power-on; thorough rebuild before resuming season operations (Section 4 STORE event protocol). The total time is similar to active-season maintenance; it just shifts toward storage care rather than flight prep.

What about the maintenance schedule if I'm flying rarely (monthly or less)? +

Cycle-based intervals stretch out (you accumulate hours slowly), but calendar-based intervals don't. The maintenance work doesn't reduce proportionally — it just shifts toward calendar-based items.

Practical approach for low-frequency flyers:

  • Weekly visual check: still applies, especially during storage.
  • Monthly thorough check: still applies, even if no flight has happened.
  • Cycle-based replacements: stretch to whatever the actual hour-count justifies. Props at 25 hours; if you fly 1 hour per month, that's 2+ years.
  • Pre-flight rebuild: any flight after 30+ day gap gets full rebuild treatment per Section 4 STORE event.

Counter-intuitively, low-frequency flyers often do more maintenance per flight hour than high-frequency flyers. The calendar-based items dominate; the cycle-based items barely accumulate. This is one reason occasional drone owners often have higher per-mission failure rates than active graduates — paradoxically, frequent use is better for drones than rare use.

What's the cohort recommendation on third-party maintenance services? +

Cohort default is self-maintenance for individual graduates; some partner orgs use third-party services for specific work. Trade-offs:

  • Self-maintenance: builds skill, lower marginal cost per event, fast turnaround. Requires time investment and tooling.
  • Third-party services: faster for complex work, no skill investment needed, predictable cost. Higher marginal cost per event; longer turnaround for complex repairs.

Practical division for cohort default operations:

  • Routine maintenance (Sections 2-3): self-maintained. The schedules don't require specialised tooling; the time investment is small per drone.
  • Event-based work (Section 4): usually self-maintained, except for complex water damage or significant crashes that exceed alumna skill.
  • Partner-org fleet maintenance (Section 6): dedicated maintenance lead role (10-20% of someone's time); third-party services for specific complex work.

Davao has 2-3 drone-capable repair shops with cohort experience. Use them for complex cases or time-pressured situations; don't rely on them for routine work that you should be doing yourself.

What changes for 10" research-grade builds? +

10" cohort default builds (ArduPilot, larger motors and props, multispectral camera) follow similar principles with adjusted intervals:

  • Props: 10" props last longer per flight hour than 5" props (lower RPM, less flex). Cohort default replacement at ~40 hours rather than 25.
  • Motors: larger motors typically last longer; cohort default ~250 hours rather than 150-200.
  • Batteries: 6S 4000mAh packs cycle slower than 4S 1500mAh; ~80-110 cycles before retirement.
  • Frame: 10" frames are mechanically more conservative; less flex stress per flight; longer typical lifespan.
  • Camera and sensors: multispectral cameras need more careful calibration verification; quarterly rather than annual.
  • Mission Planner / ArduPilot side: parameter verification and firmware updates more important than INAV equivalents; quarterly review recommended.

Otherwise, the discipline is the same: cycle, calendar, event-based; logbook discipline; fleet considerations apply. The cohort default schedule download includes 10"-specific intervals.

What about preventive replacement of components that haven't failed? +

The core preventive-maintenance question. Cohort defaults reflect specific decisions:

  • Replace preventively: components with predictable failure modes and high consequences if they fail (props, motors, batteries). The economics favour scheduled replacement.
  • Replace reactively: components with unpredictable failure modes or low consequences (FC, frame). Failure is rare; replacement before failure is mostly waste.
  • Inspect frequently, replace as needed: components in between (ESCs, GPS, receivers). Visual condition often signals impending failure.

The cohort default schedule (Section 2) reflects these distinctions. Don't replace an FC that's working fine; do replace props after 25 hours regardless of appearance.

Specific gotcha: don't replace too aggressively. Some graduates in their first year over-maintain — replacing components that still have life because the schedule said so. Cohort default schedules are calibrated against typical operations; if your operations are gentler, components last longer; respect the actual data over the schedule. The logbook is what reveals when intervals can be stretched.

How do I know my maintenance is working? +

Quantitative indicators that scheduled maintenance is doing its job:

  • Mid-mission failure rate: should trend toward zero. Aim for <1 failure per 50 flight hours after 6 months of structured maintenance.
  • Mission completion rate: should exceed 90%. Below this suggests maintenance gaps or operational issues.
  • Per-mission setup time: should be consistent (~60-90 minutes per cohort default). Increasing time suggests developing equipment issues that maintenance should be catching.
  • Repair time and cost: should be predictable. Spikes in repair work suggest accumulated maintenance debt.
  • Component lifespan tracking: actual vs expected. Trending below expected suggests harder operating conditions or maintenance gaps.

Qualitative indicators:

  • Drones feel reliable; you trust them for high-stakes missions.
  • Surprises are rare; what happens during missions is mostly what you expected.
  • Pre-flight checks reveal nothing concerning most of the time.
  • You're not constantly fixing things between missions.

The combination — quantitative data plus the sense of reliability — confirms the discipline is working. If either is off, review the maintenance approach: are you actually doing the schedules, or just recording that you did?