Separate the setup sheet from the car at speed
Generated from
content/lms/race-aerodynamics/04-control-the-aero-platform/04-separate-static-setup-from-running-platform.md; edit the source file, not this page.
Source path: content/lms/race-aerodynamics/04-control-the-aero-platform/04-separate-static-setup-from-running-platform.md
Course: Engineer downforce you can actually use
Module: Control the aero platform
Estimated duration: 60 minutes
Principle: the setup sheet is the starting condition, not the car you are actually driving.
When you write down front wing setting, rear wing angle, front and rear ride height, diffuser hardware, spoiler position, or any other aero configuration, you have recorded what the car looked like in the paddock. That record matters. Without it, you cannot repeat a test. But the skill in this lesson is learning not to confuse that static record with the running aero platform. The useful question is not only what did we set. The useful question is what did the car become at speed, and did that running car make the lap faster, more balanced, and more repeatable.
Aero does its work through pressure distribution. The car has lift, neutral vertical force, or downforce because the pressures over the top surfaces and the pressures under the car create a net vertical result. Instead of trying to reduce the whole car to a large pressure map, the practical track-side method is to measure the outcome at the front and rear axles. If the car rises as speed increases, it is showing lift. If it gets lower as speed increases, it is showing downforce. That simple observation is the first wedge between the setup sheet and the running platform.
For an intermediate driver, the danger is usually not ignorance of the setup sheet. The danger is believing the sheet too much. You can bolt on more rear wing and write a larger number in the logbook, but if the wing is low in disturbed flow, stalled, or just adding drag without enough useful rear download, the static number has fooled you. You can lower the car in the garage, but if it grows in front ride height on the straight or collapses at the rear in high-speed corners, the running platform is not what the setup pad promised. You can trim the car for maximum top speed, but that configuration may be slower over the lap if it gives away too much high-speed cornering and braking confidence.
The working rule is this: treat static setup as the label on the test, and treat running evidence as the result of the test. The label lets you repeat the condition. The result tells you whether the car actually gained useful downforce, lost too much straight-line speed, changed front-to-rear balance, or moved outside the window you thought you were testing.
What belongs on the static side.
The static side is still essential. You need the front aero setting, the rear aero setting, the measured front and rear ride heights at rest, the wing or spoiler position, and any floor, diffuser, strake, undertray, or bodywork configuration that is being compared. You also need notes and times, because a setting without a time and a condition record becomes a story instead of evidence.
Keep this side boring. Do not turn it into a theory essay. It should answer a few repeatability questions: what was installed, where was it set, what was the baseline, what did we change, and when did we run it. If you cannot rebuild the configuration from the notes, the test is weak before the car leaves pit lane.
What belongs on the running side.
The running side is where the lesson lives. You are looking for speed-dependent evidence: front and rear ride height change, suspension deflection or load if available, high-speed corner entry speed, apex speed, exit speed, straight-line speed, sector time, lap time, and driver feedback on aero handling balance. The useful corner-speed evidence is normally from higher-speed corners, around >60 mph or 100 km/h depending on how much downforce the car makes. Low-speed corners still matter for lap time, but they are less useful for isolating aero because mechanical grip dominates more of the result there.
A basic data logger can support this work. You do not need a full professional wind-tunnel program to separate static settings from running behavior. Lap times, sector times, high-speed corner speeds, and straight-line speeds already tell you a lot when you are disciplined about the test. Ride-height sensors make the evidence stronger because they show whether the car is rising or lowering with speed at the front and rear axles. Load cells or strain gauges can directly measure loads through spring, damper, pushrod, or pullrod paths, but those measurements have limits because they do not include vertical aerodynamic forces generated by the wheels. Laser ride-height sensors can be more precise for some cars, but they include tyre deformation, so the useful version is calibrated ride height versus vertical load rather than a raw number treated as truth.
Every measurement has noise. Wind, track irregularities, sensor precision, and the normal scatter of driving all move the result. That does not mean the test is useless. It means you need enough repeatability to know whether a change is larger than the noise. If your downforce measurement repeatability is only plus or minus a few percent, do not pretend a tiny apparent change proves anything. You are looking for changes big enough to survive the measurement method.
Why one-change testing matters.
The cleanest aero comparison starts from a mechanically sorted car. If the car has a major mechanical balance problem, the aero test becomes muddy because you cannot tell whether the car improved from a pressure change, a tyre state change, a driving adaptation, or a suspension behavior. Once the mechanical baseline is stable enough, change only the aerodynamic configuration you are testing. That is the discipline behind the five-lap wing comparison method in the corpus: run one configuration, record average lap performance, discard abnormal high or low times, change only the wing configuration, and compare the result.
This is not about worshiping five laps as a magic number. It is about protecting the test from wishful thinking. A single good lap can be a better tow, cleaner traffic, a braver driver, or a tyre that happened to be in a better moment. A single bad lap can be traffic, a missed brake point, a gust, or a tyre peak falling away. A small average, with obvious outliers removed, is crude but useful. It makes you less likely to chase the setup sheet because one lap flattered it.
You also return to baseline during the session. Conditions drift. Tyres deteriorate. Weather changes. Wind changes. Track grip changes. If you run baseline at 9:00, test part A at 9:30, test part B at 10:00, and never go back, you may be measuring the morning instead of the part. A periodic baseline run anchors the session. If baseline itself has slowed, heated the tyres differently, or changed balance, your comparison needs that context.
Separating mechanical balance from aero balance.
You cannot always completely separate mechanical balance tuning from aero balance tuning. Real test days are short, traffic exists, weather moves, and the car wears. But you can design your observations so the two influences are less tangled.
Use low-speed and higher-speed corners differently. In a low-speed corner, the car is telling you mostly about tyres, weight transfer, mechanical geometry, dampers, roll stiffness, braking release, and throttle application. In a higher-speed corner, the same mechanical system is still there, but the pressure field has more authority. If the car understeers in slow corners and high-speed corners in the same way, do not automatically blame aero. If the car is reasonably balanced in slow corners but pushes at high-speed entry after adding rear wing, you may have moved the aero balance rearward. If the car is nervous only in the fast section after a front aero change, the setup sheet may have created more front authority than the rear platform can match.
The point is not to diagnose everything from driver feel. Driver feedback is one data channel, not the whole test. Combine it with the speed trace, the sector time, straight-line speed, and ride-height behavior. If the driver reports high-speed understeer and the data shows rear ride height dropping more than front ride height after a rear-wing change, the story is coherent. If the driver reports more stability but the high-speed sector does not improve and straight speed falls, the setup may feel nicer without being faster. If the car gains high-speed corner speed but loses enough straight speed that the sector or lap is worse, you have learned the drag cost.
Top speed is a trap.
Many drivers get attracted to the biggest number on the straight. It is clean, simple, and easy to compare in the paddock. But the fastest top-speed setup is rarely the fastest lap-time setup. Downforce exists because cornering, braking confidence, and exit speed can be worth more than the last few miles per hour at the end of the straight. The real comparison is segment and lap time, not ego speed.
This does not mean drag is irrelevant. Drag is one of the costs you must account for. Straight-line speed, coastdown testing, throttle-stop testing, or other drag checks help you see whether a device is earning its cost. But the right conclusion is not maximum downforce at any drag cost or minimum drag at any cornering cost. The right conclusion is the balanced setting that produces the best performance for that venue and condition.
Building the balance table.
A single good setup is useful. A balance table is more useful. The table is built by starting at a low-downforce balance, increasing rear downforce, running the car until you can sense and measure the new balance, then adjusting the front until the car is balanced again. Record the front setting, rear setting, notes, and times. Then repeat until you reach the maximum rear downforce setting you can practically use.
Now you have a set of matched front and rear settings from low to high downforce. That table is not a perfect universal law. It was produced at a test venue, in certain conditions, with a specific car, tyres, driver, and test method. But it is far better than guessing. It lets you go to another event with a known menu instead of starting from scratch. If you return to the same track in the rain and want maximum downforce, you can choose the maximum rear setting and the front setting that previously balanced it, then spend practice learning the wet track rather than hunting blindly for balance.
Notice the phrase practically use. The setup sheet may accept a larger number. The car may not. The wing may stall. The strakes may scrape. A deep plywood strake can improve diffuser-side flow in a test but be impractical on the road or track. A shorter rubber replacement may be more usable but less effective. Practical aero setup lives in that gap between what the drawing promises and what the moving car can survive and repeat.
Using ride height without duplicating the ride-height lesson.
The sibling lessons in this module teach direct running ride-height measurement and sensitivity mapping in more detail. Here you use those skills as evidence. You are not just asking how low the car ran. You are asking whether the static configuration produced the intended running platform.
A car that rises with speed has lift. A car that lowers with speed has downforce. Front and rear ride-height traces let you see whether the aero load is arriving at the axle you expected. If a rear wing change produces a meaningful rear ride-height drop but very little front change, expect the balance to move rearward unless the front is adjusted. If front ride height drops with a front device change but high-speed rear stability falls away, the running platform is telling you the front and rear are no longer matched.
Be careful about what the sensor actually measures. Suspension deflection or suspension load does not include all wheel-generated vertical aerodynamic force. Laser ride height includes tyre deformation. Track bumps and wind add noise. The skill is not pretending the measurement is perfect. The skill is using the measurement honestly, alongside sector time and driver feedback, so you stop treating the static sheet as the result.
Using drag evidence honestly.
Drag can be measured more directly than downforce in some practical settings, but drag measurements also need discipline. A coastdown test gives total drag force, and total drag includes mechanical resistance as well as aerodynamic drag. Maximum-speed calculations need reliable power, frontal area, gearing, and enough space, which are not always available. Throttle-stop testing can be useful for comparing drag changes when the test is controlled. Straight-line speed from the data logger is often enough for the driver-level question: did this configuration give away speed on the straight, and did it earn that loss somewhere else.
Do not let drag testing pull you away from the lap-time question. A rear wing that is in stalled flow may add drag like a large spoiler without making the rear platform better enough. A wing in cleaner freestream can produce a different downforce and drag result at the same apparent angle. Pitot tube testing can help locate cleaner airflow above the car. Tufts can show whether the wing is attached or behaving badly. Rear ride height can help show whether rear download is actually increasing. None of those facts live on the original setup sheet. They are running-platform evidence.
Decision method after a run.
After each controlled run, answer the questions in order. First, did the running platform change in the expected direction. Look at front and rear ride height, speed dependence, and any load evidence. Second, did the balance change in a way the driver could feel in aero-speed corners. Third, did the performance change where aero should matter: high-speed corner entry, apex, exit, and sector time. Fourth, what did the change cost in straight-line speed. Fifth, is the result bigger than the noise of the test.
Only after that do you touch the setup sheet again. If the rear setting improved high-speed stability but created understeer, you may need to add front aero to rebalance rather than remove the rear change immediately. If the car gained straight speed but lost high-speed cornering and the sector slowed, the lower-drag sheet is not better. If the data is inconsistent and baseline has drifted, run the baseline again before deciding. If the measurement is too small to trust, record it as inconclusive instead of inventing certainty.
The advanced habit is to make the sheet serve the car. The poor habit is to make the car serve the sheet. If the sheet says more wing and the car at speed says stalled, draggy, or unbalanced, believe the car. If the sheet says lower ride height and the car at speed says the floor or diffuser no longer works in its running window, believe the car. If the sheet says top speed and the lap says slower, believe the lap.
Cross-references inside the module.
Use Measure the car's running ride height when you need the sensor installation, calibration, and data-capture method. Use Track front and rear aero sensitivity and Map front and rear aero sensitivity at speed when you need a fuller map of axle response across configuration changes. Use Keep the floor honest at running ride height when the evidence suggests the underbody, diffuser, or floor has moved outside its useful window. Use Define the window before you chase more downforce when you need to set the acceptable ride-height and balance range before adding load.
This lesson is the connector between them. It teaches the discipline that keeps all those measurements from becoming decoration. The setup sheet says what you meant to test. The running platform says what actually happened. Your job is to keep those two columns separate until the evidence justifies changing the sheet.
Worked example: five-lap wing comparison without believing the label
You arrive with a baseline rear wing setting that the setup sheet says is stable but modest. The first run is not a hero run. You do five representative laps, record lap and sector times, note high-speed corner entry, apex, and exit speeds, record straight-line speed, and save the driver's balance notes. If you have ride-height sensors, you also look at front and rear ride height versus speed.
Now you add rear wing angle and change nothing else. The static sheet now looks more aggressive, but you do not conclude anything yet. You run another five representative laps. You discard obvious abnormal laps rather than letting traffic or a mistake dominate the average. Then you compare. If rear ride height is lower at speed, high-speed rear stability is better, and the fast sector improves more than the straight speed falls, the running car has probably earned the extra wing. If the driver says it feels planted but high-speed sector time does not improve and straight speed falls, the sheet may have bought comfort instead of speed. That can still matter in a race, but it is a different conclusion.
If the car now understeers in faster corners, do not immediately say the new rear setting failed. The corpus-supported balance-table method is to increase rear downforce, run until you can sense the new balance, and then adjust the front until the car is balanced again. The lesson is that the rear-wing entry on the sheet is only half the story. The running car may be telling you that rear load improved, but front load did not keep up.
Before you leave the session, return to baseline. If the baseline is now slower or feels different, tyre deterioration, weather, or track condition has moved the reference. That does not erase the test, but it changes how much confidence you should put in the comparison.
Worked example: rain return using a balanced setup table
Earlier in the season, you used a slow but disciplined test day to build a table from minimum to maximum downforce. Each rear setting had a front setting that restored balance, and each pair had notes and times. That table cost time and tyre wear to build, but it removed a large amount of future guesswork.
Now you return to the same venue and it is raining. Practice time is precious. Without the table, the team might spend the first wet session trying rear wing, then front trim, then another rear change, all while the driver is also learning where the standing water and grip changes are. With the table, you can choose the maximum practical rear downforce setting and the matching front setting that previously balanced it. That does not mean the car is automatically perfect in the wet. It means your first wet setup is anchored to a known front-to-rear relationship instead of a guess.
The running-platform discipline still applies. After the first wet laps, you look for whether the car is balanced in the corners where speed is high enough for aero to matter, whether straight-line speed loss is acceptable, and whether the driver can use the added stability. The table saves the search time. It does not excuse you from checking what the car actually became at speed.
Worked example: a wing that looks aggressive but does not work cleanly
A large rear wing mounted low behind the car can look convincing on the setup sheet. The angle is bigger. The hardware is obvious. The driver expects more rear grip. But the running platform may say something else.
The first check is whether the wing is in useful airflow. Pitot tube testing can help identify how high above the car the wing needs to sit to reach freestream air, meaning air moving about as fast as the car. If the wing is buried in slower or disturbed flow, the same angle on the sheet is not the same aerodynamic condition. Then check whether the wing is attached or stalled. Tufts can show poor flow behavior, and throttle-stop drag testing can help reveal whether the device is adding drag out of proportion to useful downforce. Rear ride height is another practical clue: if the wing angle change does not produce the expected rear ride-height change at speed, do not let the static number convince you the rear platform improved.
The right action may be to raise the wing, reduce angle, change the mounting, or accept that the device is acting more like a draggy spoiler than a clean wing in that configuration. The mistake would be to keep adding angle because the setup sheet still has unused adjustment range. The car at speed has already answered the better question.
Common mistakes
Mistake 1: treating paddock ride height as running ride height. Static front and rear ride height tell you where the car starts. They do not tell you whether the car rises with lift, lowers with downforce, pitches into a different window, or changes axle balance at speed. Good looks like comparing the static line with front and rear running evidence.
Mistake 2: changing aero and mechanical balance together. If you change wing angle, anti-roll bar, spring, damper, and tyre pressure in the same run, the data may improve but the lesson is blurred. Good looks like a mechanically stable baseline, one aero change at a time, and a written note when a mechanical change was unavoidable.
Mistake 3: chasing top speed as the main score. The lowest-drag setup may win the straight and lose the lap. Good looks like comparing straight-line speed against high-speed corner speeds, sector time, lap time, and the driver's ability to use the car.
Mistake 4: trusting one lap. A single lap can be traffic, wind, a mistake, a better tow, or a tyre moment. Good looks like repeated laps, average comparison, discarded abnormal high or low times, and a baseline return when conditions are changing.
Mistake 5: forgetting baseline drift. Tyres deteriorate, weather changes, and track conditions move. If every later run is slower, the part you tested may not be the reason. Good looks like returning periodically to the baseline setup so you can separate configuration effect from session drift.
Mistake 6: measuring a force path and assuming it is total aero. Suspension deflection and pushrod or spring loads can be useful, but they miss vertical aero forces generated by the wheels. Laser ride height includes tyre deformation. Good looks like using each measurement for what it can honestly show and cross-checking against speed and balance evidence.
Mistake 7: installing a wing by angle alone. Angle on the setup sheet does not prove clean flow. Many wings can be run stalled or in poor air. Good looks like checking freestream height, tuft behavior, drag cost, and rear ride-height response before trusting the setting.
Drill: two-column platform audit
Run this drill at the next test day or open-track event where you can make controlled changes safely. The goal is not to find the perfect aero setup in one day. The goal is to teach yourself to keep the static setup sheet and the running platform separate.
Preparation: before session one, create two columns in your notes. The left column is Static setup. The right column is Running evidence. In the static column, record front aero setting, rear aero setting, front and rear static ride height if available, wing or spoiler position, and the exact change you plan to test. In the running column, leave space for high-speed corner comments, front and rear ride-height change if logged, straight-line speed, sector time, lap time, and driver balance.
Session 1: baseline. Run five representative laps with no aero changes. Do not hunt for a lap record. Drive consistently enough that the data is usable. Record the average lap and sectors, high-speed corner speeds, straight speed, and balance notes. Mark obvious abnormal laps so they do not drive the conclusion.
Session 2: one aero change. Change only one aero variable, ideally rear wing or rear spoiler setting if that is the available adjustment. Run the same five-lap structure. Record the same evidence. Do not tune around the result in the same run. You are trying to see what the running car did.
Session 3: baseline return and rebalance decision. Return to baseline first. If baseline has moved, record that drift before judging the change. If the previous change clearly added rear authority and created high-speed understeer, make one front aero adjustment and run again. If the change only cost straight speed without a high-speed sector gain, return it or mark it as not useful for this venue.
Success criterion: by the end of the drill, you should be able to write three plain sentences. First, what static change was made. Second, what the car at speed actually did. Third, whether the lap or sector result justified updating the setup table. If you cannot write all three without guessing, the drill result is inconclusive, which is still better than a false conclusion.
Calibration cues and recovery rules
A good test has a coherent signature. More useful rear downforce should tend to show lower rear ride height at speed, more rear security in aero-speed corners, and a sector or lap effect that can justify any straight-line speed loss. More useful front aero should tend to show front running-platform change and sharper high-speed front response, but it must not create a rear stability problem that costs the driver confidence or exit speed. Lower-drag trimming should show better straight-line speed, but it must not slow the aero-speed sectors enough to lose the lap.
A bad or unclear test usually has a broken signature. The setup sheet changed, but ride height did not move in a meaningful way. The driver felt a difference, but the high-speed sector did not agree. Straight speed improved, but every corner that needs downforce got worse. The measured change is smaller than the expected noise. The baseline run after the test is no longer comparable. In those cases, recover by returning to baseline, repeating the run, simplifying the change, or calling the result inconclusive.
The strongest recovery rule is to avoid rescuing a weak test with a confident story. If wind, traffic, tyre deterioration, or measurement scatter defeated the comparison, write that down. A clean inconclusive note protects the next test. A fake conclusion sends the car down the wrong development path.
Author Review
No quiz questions are attached to this lesson.
Sources
| # | Document | Chunk | Pages | Score | Collection |
|---|---|---|---|---|---|
| 1 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 4adf8cb4-89c7-1b45-bd4d-9bb03634ecf3 | 345 | 1 | uio_books_raw_v1 |
| 2 | Competition Car Aerodynamics 3rd Edition McBeath Simon | c0cd0f54-6d9c-7f08-e9af-37c31b3421d3 | 345 | 1 | uio_books_raw_v1 |
| 3 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 80bde176-e318-b515-e3d5-5de74a7cd507 | 476 | 1 | uio_books_raw_v1 |
| 4 | Competition Car Aerodynamics 3rd Edition McBeath Simon | e69e50b8-72e1-795d-d8ff-b80dec2cc10c | 352 | 1 | uio_books_raw_v1 |
| 5 | uio julian edgar car aero testing | b886348a-1227-a258-aa63-68d2e768be77 | 65 | 1 | uio_books_raw_v1 |
| 6 | uio julian edgar car aero testing | 237ab1cc-b3cd-b239-83e4-b9ffcef75fdf | 91 | 1 | uio_books_raw_v1 |
| 7 | uio julian edgar car aero testing | 561b0517-420a-cee0-a0d0-cbb161f44d01 | 92 | 1 | uio_books_raw_v1 |
| 8 | Competition Car Aerodynamics 3rd Edition McBeath Simon | c87c89fe-58c4-8968-6248-4a307e39f9e2 | 346 | 1 | uio_books_raw_v1 |