Separate drag from downforce before you tune aero
Generated from
content/lms/race-aerodynamics/01-model-aero-as-speed-dependent-load/01-separate-drag-from-downforce.md; edit the source file, not this page.
Source path: content/lms/race-aerodynamics/01-model-aero-as-speed-dependent-load/01-separate-drag-from-downforce.md
Course: Engineer downforce you can actually use
Module: Model aero as speed-dependent load
Estimated duration: 55 minutes
Principle
You separate drag from downforce because an aero change can help the car in one part of the lap and hurt it in another. Downforce helps by increasing the normal reaction at the tires, which raises the frictional force available for cornering, braking, or traction. Drag hurts by absorbing power and reducing straight-line speed or acceleration. The useful aero setup is not automatically the one with the most load or the least drag. It is the setup where the cornering gain, balance gain, and sector-time gain are worth the straight-line cost.
The practical rule is simple: read drag where the car is mostly going straight, and read downforce where the car is trying to carry speed through faster corners. For an intermediate driver, that usually means you look at straight-line terminal speed, acceleration along the straight, and coastdown behavior for drag. You look at higher-speed corner entry, apex, and exit speeds, plus driver feedback on balance, for downforce. McBeath points to higher-speed corners above roughly 60 mph or 100 km/h as the sort of places where aero effects begin to become visible, while also warning that the threshold depends on downforce level. Slow corners can still matter to lap time, but they are a poor place to diagnose aero because mechanical grip, power delivery, line, and driver timing can dominate the result.
This lesson is not about choosing the final aero target, locating the center of pressure, or using dynamic pressure as the speed dial. Those are sibling skills. Your job here is narrower and more useful at the track: after you change a wing, splitter, spoiler, diffuser, vent, or trim setting, you must be able to say which part of the result was drag, which part was downforce, and whether the trade was actually faster.
The two signals you are trying to separate
Think of every aero change as producing two linked but different signals. The first signal is the load signal. If the change adds useful downforce, the car should be able to carry more speed through relevant faster corners, or it should hold the same speed with more margin, depending on how hard you are driving. The balance may also move. More front load, more rear load, or a shifted center of pressure can change whether the car tends toward understeer or oversteer in the parts of the lap where aero load is meaningful. In basic track testing, you normally infer this load signal through sector times, higher-speed corner entry speed, apex speed, exit speed, and driver feedback on aerodynamic handling balance.
The second signal is the resistance signal. If the change adds drag, it can reduce top speed, flatten acceleration, or increase deceleration in a coastdown test. Drag is easier to measure directly than downforce with modest equipment, but even then you must be precise about what you have measured. A coastdown test measures total drag force, and that total includes a mechanical resistance component as well as an aerodynamic one. If you call the whole deceleration trace aero drag, you have already lost the separation you came to learn.
The reason this matters is that the stopwatch can hide the cause. Suppose a wing angle change makes the lap time better. That does not automatically mean the wing is aerodynamically efficient. It may have added a lot of drag but even more useful cornering speed. Suppose another change makes the top speed worse. That does not automatically mean the change is bad. Top speed is not usually the primary determinant of lap or elapsed time in most forms of motorsport, with high-speed oval racing and possibly Le Mans called out as important exceptions. You separate the signals so you can avoid both errors: rejecting a good downforce trade because the straight looks slower, or accepting a bad drag penalty because one lap happened to be quick.
What clean separation requires
The first requirement is a stable baseline. Aero testing is unusually easy to contaminate because you are looking for small changes that show up through lap time, sector time, speed traces, and subjective balance. Weather, track condition, and tire deterioration can move the baseline during a session. That is why the test plan must include returning to the baseline setup periodically, especially if conditions are changing. If you run baseline in the morning, make two changes, and then compare the final run to the first run after the track and tires have changed, you may be measuring the day more than the car.
The second requirement is one aero change at a time. The Carroll Smith method described by McBeath compared two different wings over five-lap runs, with only wing configuration changes made. That detail is not administrative housekeeping; it is the whole experiment. If you change wing, tire pressure, ride height, and driver target in the same outing, you can still have a fun session, but you cannot honestly separate drag from downforce. The car may be faster or slower, but the reason will be blurred.
The third requirement is repeatable samples, not heroic laps. Five laps per configuration is a useful model because it gives you enough data to average without asking for an entire test day. Discarding abnormally high or low times is crude, but it is a practical way to keep a missed apex, traffic hold-up, or unusually clean lap from deciding the test. A single best lap can be emotionally convincing and technically useless. You are looking for a repeatable change in the data and a repeatable change in what the driver feels.
The fourth requirement is sector-level thinking. A lap time is a summary, not a diagnosis. McBeath lists lap times, sector times, higher-speed corner entry, apex and exit speeds, and straight-line speeds as the sort of traditional test outputs you can use. Those channels let you divide the lap into places where drag is likely to show and places where downforce is likely to show. If the car loses speed only on the longest straight but gains speed through the fast corner complex, you are looking at a trade. If it loses straight speed and does not improve the fast corners or balance, you are likely looking at drag without enough useful downforce. If it gains faster-corner speed with little straight-line penalty, you may have found a more efficient configuration.
How to read drag without confusing yourself
Start with the straight. Drag shows up when the car is trying to accelerate against air resistance or when the car is allowed to slow without power. On track, the basic drag indicators are straight-line speed, the rate at which speed builds along the straight, and terminal speed before braking. These are indirect measurements, but McBeath is clear that indirect measurements of configuration effects on sector or lap times and speeds are often all you really need to know. If a configuration change consistently reduces end-of-straight speed, that is a drag warning. It is not yet a verdict, because the setup may also be paying you back in faster corners.
If you want a more direct drag test on a modest budget, the coastdown method is the practical starting point. You do not need a racetrack, but you do need a long, straight, flat, smooth road. You accelerate to a defined speed, remove drive in a controlled way, and record how the car decelerates. A configuration that creates more total resistance should slow faster. This is why drag is the easier of the two major aero forces to measure with simple tools.
The limitation is built into the measurement. Coastdown gives total drag force, not pure aero drag. Mechanical resistance is inside the result. Wheel bearings, drivetrain losses, tire rolling resistance, and any other mechanical resistance are part of what the car is fighting as it slows. The bonded corpus does not give a complete correction procedure, so the honest lesson is to treat coastdown as a comparison method under controlled conditions, not as a magic CD machine. If you keep the car mechanically unchanged and repeat the same procedure, a before-and-after change can still be useful. What you must not do is claim you have isolated the aerodynamic component perfectly unless you have the instrumentation and assumptions to support that.
There are more sophisticated ways to measure drag. McBeath mentions horizontal suspension load measurement and driveshaft strain measurement. Both can provide total vehicle drag, but the sensors and data acquisition required may exceed a club racer's budget. Another route is to measure maximum speed and calculate a drag coefficient if you have enough space, gearing, reliable at-the-wheels horsepower, and frontal area. McBeath is blunt about the practical problem: rarely are all the required elements available. For most Tracky drivers, the right conclusion is not that drag measurement is impossible. The right conclusion is that you should use straight-speed data, coastdown comparisons, and common sense with a clear understanding of their limits.
How to read downforce without pretending you measured it directly
Downforce is useful because it increases tire loading and therefore available frictional force. The mechanism is simple enough, but the track diagnosis is not. You rarely see downforce directly with the simple tools available to most drivers. You infer it from what the car can do in speed ranges and sections where aero load matters. That is why the corpus points you toward higher-speed corner entry, apex, and exit speeds, plus driver feedback on aerodynamic handling balance.
The key is to choose corners where the aero change has a fair chance to matter. A very slow hairpin can be decisive for lap time, but it is often a poor aero diagnostic. A faster bend, a quick entry where the car is loaded, or a high-speed transition is more useful. If the car can enter faster with the same confidence, hold a higher apex speed, or exit a fast corner with less balance trouble, the configuration may have added useful load or improved the load distribution. If those gains appear only in sections where the car is above the approximate aero-relevant speed range, the diagnosis is stronger.
Driver feedback belongs in the test, but it must be disciplined. The useful feedback is not vague approval. Useful feedback describes where the balance changed and whether the change occurred in the aero-relevant part of the corner. For example, did the car gain support in the fast entry but still behave the same in the slow corner? Did the balance shift in a way that matches the wing or trim change? Did the car feel better, but the speed trace and sector time refuse to confirm it? The corpus supports driver feedback on aerodynamic handling balance as a supplement to logged data, not as a replacement for it.
You should also separate downforce from confidence. A new aero part can make a driver feel more committed because the car looks more serious or because the session target changed. That is why you average laps, return to baseline, and compare sectors. If the driver claims more grip but the high-speed corner speeds, sector times, and balance notes do not repeat, the claim is weak. If the same change appears in the data and the feedback after a baseline return, it becomes much more credible.
Deciding whether the trade is worth it
Once you have separated the signals, the next step is to decide whether the drag cost bought enough downforce benefit. The corpus describes this as the downforce-to-drag compromise, and it matters especially in high-downforce formulas and in categories where aerodynamic efficiency is key, such as Le Mans prototype sports cars. For club racing and HPDE, you may not have a professional simulation department, but the logic is the same. The car must get around corners, accelerate down straights, brake for the next corner, and repeat the process for the whole lap.
A useful mental model follows the performance simulation described by McBeath. The model looks at a corner, estimates the maximum speed around it from radius and grip, then uses that corner exit speed as the starting point for acceleration along the next straight. It then accounts for acceleration, changing aerodynamic forces, braking distance to the next corner, and the next corner after that. The whole lap is built from sectors. You do not need to run a professional model to use the idea. It teaches you where to look: a downforce gain is not isolated to the apex, because higher corner speed can change exit speed and the following straight. A drag penalty is not isolated to top speed, because slower acceleration may affect the time spent on the whole straight.
This is why sector times matter more than isolated bragging points. If the added wing angle costs a small amount of terminal speed but gives a larger gain through and after a fast corner, the lap may improve. If it costs straight-line speed and gives no repeatable high-speed corner gain, it is probably only slowing the car. If it improves one fast sector but ruins another part of the lap through balance, you may have found a center-of-pressure problem rather than a simple drag problem, which belongs in the related balance and center-of-pressure lessons.
Also remember that top speed is context-dependent. McBeath notes that top speed is not usually of primary significance in most forms of motorsport, while high-speed oval racing and possibly Le Mans are different. That means the same drag penalty can be acceptable at one venue and unacceptable at another. On a tighter circuit with important fast bends, a downforce-biased setup may win. On a venue dominated by long full-throttle time, the same setup may be too costly. The separation skill lets you avoid treating either answer as universal.
Worked example: five-lap wing comparison
Imagine you are comparing two rear wing configurations at a test day. The correct structure comes straight from the corpus: run each configuration over five laps, change only the wing configuration, average the lap times, and throw out abnormal highs or lows. Add a baseline return if conditions are moving or if the tires are aging enough to change the reference.
For each run, you record lap time, sector time, straight-line speed, and higher-speed corner entry, apex, and exit speeds. After the session, do not start with the best lap. Start with the map of where the car changed. If Wing B is slower at the end of the straight but faster at the apex and exit of the fast corner, you have separated the two effects. Wing B likely added drag and added useful downforce. The decision is whether the sector gains paid for the straight loss.
If Wing B is slower down the straight and the fast-corner speeds are unchanged, the answer is much less flattering. You have drag without a visible load benefit. It may still have changed balance in a way the driver likes, but without repeatable sector or speed evidence you should be cautious. If Wing B improves the fast corner but the lap is still worse because the following straight is long and acceleration suffers, that is not a failed test. It is exactly the information you came for. The wing may be useful at a different circuit, a different trim level, or a different balance target, but this configuration did not win this lap.
If Wing B is faster through the fast section and the straight speed loss is small, you may have a good compromise. You still need the baseline return because tire deterioration and track conditions can fake a trend. If the baseline return no longer matches the original baseline, adjust your confidence. The data may still point in the right direction, but you should not pretend the session was cleaner than it was.
Worked example: coastdown after adding aero trim
Now imagine you added an aero trim change and want to know whether it increased drag. You find a long, straight, flat, smooth road and run a coastdown comparison before and after the change. The result shows the trimmed car decelerates more quickly. That is a useful warning, but it is not yet a complete aero verdict.
The first discipline is language. The coastdown result shows increased total resistance under your test conditions. Because coastdown total drag includes mechanical resistance as well as aerodynamic resistance, you do not claim a pure aero number unless your method accounts for the mechanical component. If the car was mechanically unchanged and the procedure was repeatable, it is still reasonable to treat the difference as evidence that the aero trim added resistance.
The second discipline is linking the drag result to the track result. If the same configuration also gives repeatable gains in higher-speed corner entry, apex, or exit speed, then you have the classic trade: more resistance, more useful load. If the track data shows no faster-corner benefit and no better balance, the coastdown result becomes a strong reason to remove or revise the change. If the track data improves but the coastdown penalty is large, you do not decide from either test alone. You decide from the lap and sector effect at the venue you care about.
This example also shows why drag is easier to measure than downforce but not easier to interpret. A bigger deceleration number can be seductive because it feels objective. The lesson is to keep it in its lane. It tells you about resistance. It does not tell you whether the car became faster through the corner that matters.
Worked example: when top speed matters more
The corpus singles out high-speed oval racing and possibly Le Mans as exceptions to the usual warning that top speed is not the primary performance question. It also names Le Mans prototype sports cars as a category where aerodynamic efficiency is key. Use that as a calibration example. If your venue or class spends a large part of the lap at high speed, a drag increase carries more weight. You still separate drag from downforce, but the acceptance threshold changes.
On a circuit shaped by long straights, the same wing setting that helped a fast bend on a tighter course may lose too much time in acceleration and terminal speed. On a course where a high-speed corner controls the following straight, the downforce gain can still matter because a better exit speed can feed the next straight. That is why the simulation logic is useful: corner speed, exit speed, acceleration, drag, and braking distance are linked. The lesson is not to worship top speed. The lesson is to judge the drag cost in the actual lap structure.
Calibration cues
A good separation process leaves you with several matching cues. The straight-line cue is whether the car reaches the same speed at the same point, accelerates similarly through the straight, and arrives at the braking point with the same terminal speed. A consistent drop here is a drag clue. A consistent improvement here may mean reduced drag, although power, wind, and exit speed must also be considered.
The faster-corner cue is whether the car carries more entry, apex, or exit speed in the corners where aero load is likely to matter. This is where you look for downforce. The balance cue is whether the driver reports a repeatable change in aerodynamic handling balance in those same sections. The sector cue is whether the part of the lap containing the fast corner improves enough to explain the lap change. The baseline cue is whether returning to the original setup returns the car toward the original data. Without that final cue, your confidence should be lower.
The instructor version of the same calibration is direct. If you tell me the car is faster because of aero, I want to know where. If the answer is only that the lap felt better, we keep testing. If the answer is that the car gave up straight speed but gained repeatable speed in the fast corner and improved that sector average over five-lap runs, now we have a useful statement. If the answer is that a simulation predicted the change, I still want to know what assumptions went into the model and whether the track data agrees.
Common mistakes
The first mistake is treating lap time as a cause. Lap time is the outcome. It does not tell you whether the improvement came from downforce, drag reduction, driver execution, traffic, tires, weather, or a balance shift. Good looks like sector and speed evidence that explains the lap.
The second mistake is changing too many things. If you change wing configuration and another setup item at the same time, you cannot honestly assign the result to drag or downforce. Good looks like one aero variable, repeatable laps, and notes that make the test boring enough to trust.
The third mistake is chasing top speed at every venue. Top speed is visible and easy to compare, but McBeath warns that it is not usually the primary significance in most motorsport forms. Good looks like asking whether the straight speed loss was repaid by faster corner and sector performance, while treating high-speed oval and Le Mans-type situations as different cases.
The fourth mistake is calling coastdown pure aero drag. Coastdown measures total drag force and includes mechanical resistance. Good looks like describing it as a controlled total-resistance comparison unless your instrumentation supports a more exact separation.
The fifth mistake is testing aero in the wrong corners. If you judge an aero change mainly from slow-corner feel, you may be measuring mechanical grip or driver timing. Good looks like using higher-speed corner entry, apex, and exit speeds, then cross-checking those with sector times and balance feedback.
The sixth mistake is trusting a model without respecting its assumptions. Performance simulations can compare downforce and drag values versus lap time, and they can be very useful, but McBeath stresses that they are based on assumptions with variable validity. Good looks like using simulation to choose a starting point and track testing to fine-tune under the conditions you actually have.
The seventh mistake is ignoring baseline drift. Tires deteriorate, weather changes, and track condition changes. Good looks like returning to the baseline setup periodically and being honest when the baseline no longer matches itself.
Drill: one-variable aero separation test
At your next suitable event, run a three-part aero separation drill. The count is three runs: baseline, one aero change, and baseline return. Each run should be five clean laps if the event format allows it. The duration is one session block if you can run continuously, or two to three shorter sessions if traffic and schedule require it. The success criterion is not a faster lap. The success criterion is that you can write one defensible sentence after the test: this change altered straight-line speed by this pattern, altered faster-corner speed or balance by this pattern, and the sector or lap average did or did not repay the trade.
Before the first run, choose the straight where drag should show and the faster corner or sector where downforce should show. Do not use every corner on the track. Pick the evidence before you see the result. Run the baseline and mark the normal speed range for the straight and the chosen faster corner. Run the aero change with no other setup change. Record the same channels. Then return to the baseline and record again. If the baseline return has moved because tires or conditions changed, write that down instead of forcing a conclusion.
Afterward, sort the evidence into three boxes. The drag box gets straight-line speed, acceleration pattern, terminal speed, and any coastdown comparison if you did one. The downforce box gets higher-speed corner entry, apex, exit, and balance feedback. The trade box gets sector averages and lap averages after removing obvious abnormal laps. If all three boxes point in the same direction, you have a strong result. If the boxes disagree, you learned where the next test needs to be cleaner.
How to use simple tools without overclaiming
A basic data logger is enough to begin. McBeath notes that low-cost loggers make this kind of test easy to execute, but only with a disciplined approach. That distinction matters. The equipment does not make the test honest. Your procedure makes it honest. If all you have is lap time, sector time, and speed, you can still separate a great deal. If you have suspension loads, driveshaft strain, wind tunnel data, or a full simulation, you can ask more precise questions, but the same discipline applies.
Flow visualization also has a place, but in this lesson it is supporting evidence rather than the main separation method. McBeath notes that seeing what the air is doing around wings, spoilers, diffusers, cooling intakes, and outlets can help you understand what is happening and point toward development areas. That can explain why a change might add load or drag. It does not replace the lap, sector, straight, and corner evidence that tells you whether the change worked on track.
Cross-references
Use the dynamic-pressure lesson when you need to understand why aero effects grow with speed and why the same part can feel irrelevant in one corner and decisive in another. Use the center-of-pressure lessons when the car gains load but the balance moves in a way that costs confidence or speed. Use the aero-behavior-through-handling lesson when your main evidence is balance feedback rather than speed traces. Use the aero-target lesson before you start trimming parts, because separation tells you what happened, while the target tells you what you wanted to happen.
Final standard
You have separated drag from downforce when you can explain the car in lap structure, not adjectives. You can point to the straight-line evidence for resistance, the faster-corner and balance evidence for load, the sector evidence for the trade, and the baseline evidence that the comparison is fair. If you cannot do that, do not call the setup good or bad yet. Call the test incomplete, clean up the method, and run it again.
Worked example: five-lap wing comparison
Run each wing configuration over five laps, change only the wing configuration, average the laps, and remove obvious abnormal highs or lows. Compare straight-line speed against higher-speed corner entry, apex, and exit speeds. If the second wing gives up speed on the straight but gains speed through the fast corner and improves the sector, you have found a downforce gain with a drag cost. If it gives up straight speed without a repeatable faster-corner or balance gain, the change is mostly penalty. A baseline return keeps tire deterioration, weather, and track change from masquerading as aero.
Worked example: coastdown separates drag cost
Use a long, straight, flat, smooth road for a before-and-after coastdown comparison. If the car slows faster after the aero trim change, treat that as evidence of increased total resistance, not automatically pure aerodynamic drag. The method includes mechanical resistance, so the value is most useful as a controlled comparison. Then connect it to the track: more total resistance plus faster high-speed corners is a trade, while more total resistance with no corner or balance gain is a strong reason to revise the change.
Common mistakes
The common failures are treating lap time as the cause, changing multiple setup items at once, chasing top speed at every venue, calling coastdown pure aero drag, judging aero from slow corners, trusting simulation without respecting assumptions, and ignoring baseline drift. Good testing uses one aero variable, repeatable five-lap samples, straight-line speed for drag clues, faster-corner speeds and balance feedback for downforce clues, sector averages for the trade, and periodic baseline returns when conditions or tires are changing.
Drill: one-variable aero separation test
Run three five-lap samples if your event format allows it: baseline, one aero change, and baseline return. Before driving, choose one straight for drag evidence and one faster corner or sector for downforce evidence. Success is not setting a faster lap. Success is being able to write a defensible sentence explaining how the change affected straight-line speed, how it affected faster-corner speed or balance, and whether the sector or lap average repaid the trade.
When the principle changes emphasis
The principle does not disappear on high-speed tracks, but the weighting changes. McBeath identifies high-speed oval racing and possibly Le Mans as cases where top speed can matter more than usual, and he notes that aerodynamic efficiency is key in Le Mans prototype sports cars. At those venues, a drag increase needs a stronger downforce payoff. On tighter circuits with important fast bends, a straight-line penalty may be acceptable if it buys repeatable speed through and after the corner that controls the lap.
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 | c87c89fe-58c4-8968-6248-4a307e39f9e2 | 346 | 1 | uio_books_raw_v1 |
| 4 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 2cdbaed7-3768-54ee-edaf-38283586198c | 59 | 1 | uio_books_raw_v1 |
| 5 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 893cce66-5e94-8af0-6d98-00acc7cbd324 | 383 | 1 | uio_books_raw_v1 |
| 6 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 90d1fa19-5889-6e05-3847-4f1454f3babb | 384 | 1 | uio_books_raw_v1 |
| 7 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 6edca499-2988-7702-ccc8-3d17b516edff | 385 | 1 | uio_books_raw_v1 |
| 8 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 2abb3a1a-1abc-3549-8f79-9fce704061d6 | 334 | 1 | uio_books_raw_v1 |