Flag Lean Carburetor Settings on a Dyno Test (2026)
Catching a lean carburetor setting after the pull is over means the engine already ran hot for however long that pull lasted. Instead of eyeballing an RPM trace and guessing where the needle went wrong, set up a dyno workflow that flags the lean condition the moment RPM and temperature data cross a threshold you defined ahead of time — so the flag fires during the pull, not after the piston already saw it.
TL;DR
Flag lean carburetor settings during a dyno test by pairing RPM climb with a temperature threshold, not RPM alone.
A reliable lean flag needs a sustained signal over about two seconds, not a single spike.
Run one calibration baseline pull per engine before trusting any automated flag in 2026.
Rich-side flags use the same threshold logic in reverse — falling RPM and temperature under load.
Recalibrate thresholds after any needle, glow plug, or fuel change — old thresholds produce false flags.
Why this matters
A lean needle setting shows up as a fast RPM climb and a temperature spike under sustained load, and by the time a driver or tuner hears the engine "go sharp," the piston-to-sleeve mating has already been under thermal stress for several seconds. On a dyno stand that data is already being captured — connect a digital tachometer to the tuning app and the RPM feed is live the whole pull. The only piece missing is a rule that watches that feed and raises a flag the instant the numbers cross a line, instead of a human trying to spot it in real time on a moving trace.
This workflow builds that rule using two signals — RPM and temperature — because either one alone produces false positives. RPM alone catches transient throttle blips that look identical to a lean spike for a second or two. Temperature alone lags the actual carburetor behavior by several seconds because heat takes time to build in the head. Together, they confirm each other.
Before you start
A digital tachometer wired into the dyno stand and paired with the tuning app, logging RPM against time for the full pull.
One completed baseline pull on the engine at its current needle setting, saved and not overwritten.
An EGT probe or head temperature sensor if you want the temperature confirmation layer — RPM alone still works as a soft flag without it.
The gotcha that catches most first attempts: thresholds are engine- and needle-specific. A flag rule built around a .21-size engine's RPM band will fire on nearly every pull of a larger .28, or never fire at all on a smaller engine. Build a new baseline for every engine you flag, not one rule for the fleet.
Run the baseline pull
Run the engine at its as-received needle setting for one full controlled load pull with no adjustments mid-run.
Log RPM against time and note the RPM ceiling — the highest sustained value under load, not a momentary peak.
If a probe is fitted, record head or exhaust temperature at the same load points.
Save this pull as the reference set. Nothing gets flagged yet — this pull is the yardstick everything else measures against.
Expected result: one clean data set showing how this specific engine, on this specific needle setting, behaves under controlled load — with no flag rules configured against it yet.
The tachometer feed captured here is the same signal the lean-flag rule watches.
Configure the RPM-based lean threshold
Take the RPM ceiling recorded in the baseline pull — that number is your working reference, no adjustment needed.
Set the flag rule to trigger when RPM climbs roughly 3-5% above that baseline ceiling under sustained load.
Require the RPM rise to hold for at least two seconds of the pull before firing. A momentary blip at a throttle transition should not trip the flag.
Save the rule.
Expected result: a lean-flag rule tied specifically to this engine's baseline RPM behavior, not a generic number pulled from another engine's data.
Add the temperature confirmation layer
If a temperature probe is fitted, set a second rule: flag when head or exhaust temperature exceeds the baseline reading at the same load point by roughly 15-20°F, sustained for the same two-second window.
Require both the RPM condition and the temperature condition to trigger together for a "confirmed lean" flag. Let either signal alone log as a lower-priority "watch" flag instead.
Save.
Expected result: two-tier flagging — a soft warning when one signal moves, a hard confirmed flag when RPM and temperature agree.
A short paragraph before the next block, because the four steps above are the whole configuration in sequence — RPM baseline, temperature baseline, lean threshold, confirmation layer — and it's worth seeing the order laid out once before running the actual test pull.
Each step depends on the one before it — skip the baseline and the thresholds have nothing to measure against.
Run the test pull
Adjust the high-speed needle a quarter turn lean from the baseline setting — the standard increment used in nitro carburetor tuning.
Run the same load profile used for the baseline pull.
Watch for the flag to fire at the point in the pull where RPM and temperature separate from the baseline curve.
Expected result: the flag fires while the pull is happening, at the exact point the current needle setting crosses the threshold — not as a note added after reviewing the printout later. If you need help reading what the printout says the needle should do next, translate the pull into needle settings using the same baseline data.
Flag rich settings with the same workflow
A rich carburetor setting shows the opposite signature — RPM and temperature falling below baseline under sustained load instead of climbing above it. Mirror the same two rules below the baseline line rather than above it: flag when RPM drops 3-5% under the baseline ceiling for two seconds, and flag when temperature falls a comparable margin below the baseline reading at the same load point.
Running both the lean-flag and rich-flag rules against one baseline pull gives a full read on where the current needle setting sits, in both directions, without a second calibration run. This is the same principle behind flagging powerband characteristics at any point on the curve, not just at the top end.
Get a dyno-verified needle setting
Controlled load testing and clear findings on your engine's actual carburetor behavior.
Troubleshooting
Flag fires on nearly every pull, regardless of needle setting. The baseline pull itself had mechanical inconsistency — a glow plug variance or an inconsistent load profile. Rerun the baseline before trusting the rule.
Flag never fires even on an engine that sounds obviously lean. The percentage margin is set too wide for this engine's displacement. Tighten the RPM and temperature margins and retest.
RPM data is noisy or jumps erratically mid-pull. This is almost always a tachometer connection issue, not a carburetor issue — loose wiring or sensor placement drifting off the flywheel. Recheck the pairing between the tachometer and the app.
Temperature flag fires but the RPM flag doesn't, or vice versa. The probe is likely positioned differently than it was during the baseline pull. Reposition it to match exactly, or trust the confirmed flag over either signal alone.
Flags disagree pull to pull on what should be an identical needle setting. Fuel batch, glow plug, or ambient temperature changed since the baseline. Recalibrate the baseline after any consumable swap, not just after a needle adjustment.
Customize the workflow
Once the two-tier lean flag runs reliably on one engine, extend it rather than stopping at "it works." Log peak RPM automatically on a dyno pull alongside the lean flag so every pull in 2026 records both the ceiling and the moment it was crossed. Race teams running several engines through the same session should build this per engine, not per class — see dyno testing for race teams for how that scales across a fleet without cross-contaminating baselines. If the current tachometer or logging hardware is the bottleneck rather than the workflow, a comparison of nitro RC engine tuning tools is worth reading before adding more rules on top of unreliable data.
FAQ
What's the best way to flag lean carburetor settings during a dyno test?
Pair an RPM threshold with a temperature threshold rather than relying on either signal alone. RPM alone catches throttle blips as false positives; temperature alone lags the actual lean condition by several seconds.
Is RPM data alone enough to catch a lean carburetor setting?
RPM alone works as a soft warning but produces false positives from momentary throttle transients. Requiring the rise to hold for at least two seconds, and confirming with temperature, cuts false flags significantly.
How much margin should a lean flag threshold use above baseline?
A margin of roughly 3-5% above the baseline RPM ceiling, sustained for two seconds, is a workable starting point for most nitro engines. Tighten or widen it based on how often the flag misfires on your specific engine.
Can the same dyno workflow flag rich carburetor settings, not just lean ones?
Yes. Mirror the same RPM and temperature threshold logic below the baseline instead of above it, using falling RPM and falling temperature under sustained load as the rich-side signal.
Do I need a temperature probe to flag lean settings, or does RPM alone work?
RPM alone works as a lower-confidence watch flag without a probe. Adding head or exhaust temperature turns that watch flag into a confirmed flag and cuts false positives.
How often should the baseline pull be recalibrated?
Recalibrate after any needle adjustment, glow plug change, or fuel batch change. A threshold built on an old baseline produces flags that don't match the engine's current behavior.
What's the standard needle adjustment increment used in nitro carburetor tuning?
A quarter turn is the standard increment for high-speed needle adjustments in nitro tuning, and it's a useful step size when running a test pull against a lean or rich flag rule.
One last thing
RPM is almost always the leading indicator and temperature the confirming one — on most pulls the RPM signal crosses its threshold a couple of seconds before the temperature signal catches up, which is exactly why a two-second hold window on the RPM rule, combined with a temperature confirmation, catches real lean conditions without chasing every throttle blip on the trace.
