Warnings explained

When a result is flagged in red or amber, the ⓘ next to that flag links here. Each warning below explains the threshold it tripped, why it matters, and — most importantly — what to do about it. None of these stop the simulation from running; they tell you the result is at the edge of safe operation or at the edge of what the solver can model accurately.

A warning is not a crash. It is the tool pointing at the one thing on this operating point you should look at before you build or fly.

Irreversible magnet warning

Threshold: the hottest motor magnet has exceeded about 120 °C (T_mag > 120 °C).

Why it matters. Above roughly 120 °C the permanent-magnet material in the motor begins to demagnetise permanently — it does not recover when it cools. A motor that has been cooked this way has a lower velocity constant (Kv) and makes less thrust forever, not just for this flight. This is the most serious thermal warning because the damage is irreversible.

What to do.

Reduce the sustained throttle — a brief burst is far safer than a long climb at the same throttle.
Fit a larger or higher-quality motor with more thermal headroom.
Improve cooling: more airflow over the motor (a faster slipstream, a cowling, or forced air) lowers the thermal resistance and drops the magnet temperature.
Use a lower-Cp propeller so the motor draws less current at the same RPM.

Convergence warning

Threshold: the propeller's inner aerodynamic loop did not fully converge — frac_converged dropped below about 0.99 (under ~99% of the blade stations settled to a steady answer).

Why it matters. The solver works the blade in stations from root to tip. When some stations do not converge (often the heavily loaded or stalled tips), the thrust and power reported for that rotor are less reliable — they are the solver's best unsettled estimate rather than a pinned-down answer. The closer frac_converged is to 1, the more you can trust the numbers.

What to do.

Treat that rotor's thrust, power and efficiency as approximate, not exact.
Move to a less extreme operating point — very high disk loading, very high RPM, or a heavily stalled prop are the usual culprits.
Check the per-rotor solver quality table on the Diagnostics tab to see how many stations were affected; a frac_converged of 0.98 is a minor edge case, 0.5 is not.

Steep-descent flag

Threshold: the rotor is descending fast enough that the airflow through the disk stops flowing cleanly through and starts to recirculate — an unsteady regime the steady model is not validated inside.

Why it matters. In a fast, steep descent the air the rotor pushes down can curl back up around the disk and get drawn through again, so the flow becomes unsteady and recirculating. The model bridges through this band to keep the run going, but it cannot resolve the numbers precisely there. This is a flag, not an error: the simulation still completes and every other result is produced — the thrust, power and efficiency at the flagged point are just approximate.

What to do.

Treat the thrust, power and efficiency at that operating point as approximate rather than exact.
Reduce the descent rate — a gentler descent moves the rotor back into the validated range.
Add forward speed — flying forward while descending keeps clean air moving across the disk and avoids the recirculating regime.

Related: Vert force · H-force

Compressibility warning

Threshold: more than about a tenth of the blade stations are running above the drag-divergence Mach number — frac_above_Mdd > 0.10 (more than 10% of the stations). The blade tips are fast enough that air compressibility (the same effect that limits a propeller-driven aircraft) is adding sharp drag.

Why it matters. Near the speed of sound the air no longer behaves as an incompressible fluid; drag rises steeply and the propeller's thrust and efficiency suffer. Just as importantly, the aerodynamic model is operating near the edge of its valid range, so the numbers for that rotor are less trustworthy when this is flagged.

What to do.

Lower the propeller tip speed — the direct levers are a lower RPM or a smaller-diameter propeller.
A smaller, slower-spinning prop or a larger, lower-RPM prop both reduce the Mach tip.
Treat the thrust and efficiency for that rotor as unreliable while this warning is active.

Pack thermal warning

Threshold: the battery pack's coupled thermal model flags that cells are running hot — the pack-thermal threshold is about 60 °C at the cell core.

Why it matters. Hot cells age faster, sag more under load, and in the worst case become unsafe. The pack-thermal warning means the current draw is generating more heat than the pack can shed at this operating point, so the cell cores are climbing into a zone that shortens battery life and erodes the safety margin.

What to do.

Lower the current draw — a less aggressive throttle, or spreading the load across more rotors.
Use a higher-capacity pack (more cells share the heat) or a higher-C-rating pack (lower internal resistance, so less resistive heating).
Improve battery cooling: airflow or slipstream over the pack lowers the cell temperatures.
Check the per-cell core-temperature heatmap to see whether one weak cell is driving the warning.

Discharge cutoff

Threshold: the lowest cell in the pack has reached or fallen below the pack's own minimum cell voltage — 3.0 V for a standard LiPo, and whatever the pack declares for other chemistries. It is not a fixed number the simulator imposes; it comes from the pack you selected.

Why it matters. Under load a cell's voltage sags in proportion to the current drawn and its internal resistance. Past the cutoff you are no longer taking energy out of a healthy pack — you are pulling it below the window it is rated for, which permanently costs capacity and cycle life and, held there, is a safety issue.

The reason this needs a flag at all is that the numbers look ordinary when it trips. A cell sitting right on 3.0 V under load reports a pack voltage that reads unremarkable next to every other figure on the page. Nothing in the thrust, the efficiency or the endurance says the pack is at its limit. Now something does.

What to do.

Draw less current — a lower throttle, or the same thrust spread across more rotors.
Use a pack with a higher C-rating — lower internal resistance means less sag at the same current.
Use a larger pack — more capacity in parallel shares the current, so each cell sags less.
Start from a higher state of charge — the same current takes a half-full pack through cutoff when it would not have taken a full one.

This flag and the pack thermal warning often appear together, because the same excessive current both sags the cells and heats them. They are still different limits with different fixes — one is voltage, one is heat — so treat whichever appears as its own problem.

Photo too blurry

The propeller edge is not sharp enough for the scanner to measure blade width reliably. ThrustLab rejects the image instead of returning a plausible-looking but inaccurate chord curve.

Hold the camera steady and directly overhead, tap the propeller to focus, and make sure the full propeller is visible on a plain contrasting surface. Then take a new photo. More light can help the camera use a faster exposure, but avoid direct glare on reflective blades.

Blade shape at its limit

Threshold: when you add performance data to a propeller you are creating, the tool adjusts the blade's section shape until the predicted numbers line up with the points you entered. This flag means the adjustment ran all the way to the edge of the range it is allowed to use and stopped there, instead of finding a value inside it.

Why it matters. The result is no longer telling you what your data says — it is telling you where the limit is. Your points implied a blade section further from the geometry you scanned or entered than the model will represent, so the answer was capped rather than matched. This is worse than adding no data at all, because the propeller now carries a shape adjustment that does not reproduce your measurements while looking like a propeller that was matched to real data.

Nothing else on the page says so. The result still reports a tier and a residual that look ordinary; the residual can sit comfortably inside its normal range while the shape is pinned at its limit. That is why this needs its own flag.

What to do.

Check your units first — this is the usual cause. Thrust, RPM and air density each have to be in the units the form asks for. A thrust column in grams read as newtons, or the other way round, lands exactly here.
Check the points describe one propeller, not a whole aircraft. Four rotors' worth of thrust against one blade's geometry is unreachable by any blade shape.
Check the geometry you entered — diameter, pitch and blade count. A propeller measured as 18×8 but entered as 18×12 cannot be matched to its own data.
Or save without the performance data. A propeller on its measured geometry alone is honest and usable; one capped at a limit is not.

Related: Section drag at its limit

Section drag at its limit

Threshold: when your performance data includes power (or torque, or battery watts), the tool also adjusts how much profile drag the blade sections carry. This flag means that adjustment ran to the edge of its allowed range — roughly half to two and a half times the baseline drag — and stopped there.

Why it matters. Your points implied more shaft torque than this blade can absorb at any plausible drag level, so the drag was capped rather than matched. The thrust side of the answer may still be reasonable while the power and efficiency are not: they will not reproduce the watts you entered, and a propeller that is right on thrust and wrong on power is exactly the error this data was meant to remove.

This flag and Blade shape at its limit often appear together, because data that is unreachable in one respect is usually unreachable in both. They are still separate limits — one is about the blade's shape, one about its drag — so read whichever appears on its own terms.

What to do.

Check the power figure is for one propeller. Whole-aircraft or whole-pack watts against a single blade is the most common cause of this flag.
Check where the power was measured. Battery watts include the ESC and motor losses; shaft power does not. Entering one where the other is expected shifts the number by a large fraction.
Check RPM and air density match the conditions the measurement was taken in — the same propeller absorbs noticeably different power at a different density altitude.
Or leave the power column out and enter thrust alone. The result then honestly says it constrains thrust only, instead of claiming a power match it did not achieve.

Related: Blade shape at its limit