Make airflow visible before redesigning parts
Generated from
content/lms/race-aerodynamics/06-measure-and-visualize-performance/04-make-airflow-visible-before-redesigning-parts.md; edit the source file, not this page.
Source path: content/lms/race-aerodynamics/06-measure-and-visualize-performance/04-make-airflow-visible-before-redesigning-parts.md
Course: Engineer downforce you can actually use
Module: Measure and visualize performance
Estimated duration: 55 minutes
Principle: do not redesign invisible air
The first job in amateur aerodynamics is not to build a more aggressive part. The first job is to find out what the air is already doing. If you skip that step, you are modifying by feel, intuition, or imitation. That is a weak way to work because two cars that look similar can respond differently, and because aerodynamic parts interact with the bodywork, the ground, the wheels, and the flow arriving from upstream. The better rule is simple: make the airflow visible, learn the pattern, then decide whether the part needs redesign, repositioning, or no change at all.
This lesson is about that first visual step. It is not the same lesson as measuring downforce and drag, and it is not a replacement for normalized one-variable testing. Those lessons tell you how much force changed and whether the test was fair. This lesson teaches you how to see the flow pattern that explains why a change helped, hurt, or did nothing. You use visible airflow to stop guessing where the problem is.
Aero visualization is useful because the important problems often live in places you cannot feel directly from the driver's seat. The driver may feel high-speed understeer, rear instability, or extra drag, but that does not tell you whether the flow is attached over a wing, detached at a diffuser edge, spilling around a spoiler, failing to feed an intake, or being disturbed by an outlet. McBeath's core point is practical: seeing what is happening around wings, spoilers, diffusers, cooling intakes, and outlets gives you understanding and points toward areas to improve. That is exactly the mindset you need before redesigning parts.
The skill is to turn a vague aero complaint into a visible airflow question. Not: the rear is nervous, so add a bigger wing. Ask: is the rear wing seeing attached flow across the span, or is part of the span separated before the wing can work? Not: the diffuser is not working, so make it steeper. Ask: is the diffuser flow leaving cleanly, or is it separated, low velocity, or even moving the wrong way in an outer section? Not: the car looks like it needs an airdam. Ask: how does the streamline pattern ahead of and around the front of the car change with and without the airdam? You are not trying to prove that your favorite idea is right. You are trying to see what the air is doing before you spend fabrication time.
What visible airflow can tell you
Visible airflow is strongest at answering pattern questions. Is the surface flow attached or separated? Is the local flow direction aligned with the part you expect to be working? Does one section of a wing or diffuser behave differently from another section? Does opening a louvre or changing a part disturb the local flow in a way that you can actually see? Does the visual pattern after a change match the direction of the force data or lap-time data?
The cheapest and most accessible technique in the bonded material is tuft testing. Small pieces of yarn or wool are stuck to the car, then the car is driven while the tufts are photographed or videoed. The tuft behavior gives you a readable surface-flow signal. When flow is attached to the surface, the tufts line up in neat rows and flutter only a little. When the flow has separated from the surface, the tufts whirl around and point in random directions. That is the basic language. Neat alignment means attached local flow. Random, swirling, inconsistent direction means separation or strong disturbance.
Oilflow and streamline visualization add a second layer. McBeath's CFD examples distinguish surface information from off-surface information. An oilflow plot shows surface flow direction. Streamlines over the same region reveal more about what is happening away from the surface, including the wing tip and the downstream region. That distinction matters. A surface trace can tell you where the local skin-level flow is going, but the air just above the surface may be curling, spilling, or forming a structure that the surface trace alone does not fully explain. The skilled interpretation links the two views instead of treating one image as the whole truth.
Visible airflow does not directly give you downforce, drag, or lap time. A tuft map that looks neater is not automatically faster. A disturbed tuft is not automatically bad. A local flow change may be useful, useless, or harmful depending on whether it improves the force balance and drag cost of the whole car. That is why this module has separate lessons on instrumenting the two primary forces and normalizing the run before you trust the aero number. This lesson gives you the visual diagnosis. The force and run-quality lessons tell you whether the diagnosis led to performance.
Why the visible pattern matters mechanically
Aero parts work by shaping flow, pressure, and momentum. The bonded corpus gives two useful wing reminders. Downforce comes partly from the reaction of airflow with the upper surface of a wing, but the major part comes from entrainment of air to the lower surface. For your purposes as a driver or club-racing developer, that means the underside and the flow staying attached over the useful working region matter a great deal. If the flow across the span separates too early, the wing cannot keep increasing useful load in the way you expect.
McBeath gives the example that wing twist could be altered so that flow across the whole span remains attached for longer, allowing more downforce before large-scale separation and stall. The important lesson is not that you should immediately twist your wing. The lesson is that a part redesign should be aimed at a visible mechanism. If the visual evidence shows one part of the span losing attachment before the rest, then a spanwise solution may be worth investigating. If the whole wing is clean and the car is still slow, then the problem may not be the wing surface flow at all.
Diffusers show the same principle. One McBeath figure describes flow separation and generally low velocity in a diffuser at 20mm ground clearance. Another notes an Exige diffuser outer section where the flow looked to be going in the wrong direction. Those are not minor cosmetic observations. A diffuser is an underbody device whose performance depends on the flow staying organized through the expansion region. If the visible pattern shows separation, low velocity, or reversed local direction, making a larger or more dramatic diffuser shape without understanding that pattern can simply build a bigger version of the same problem.
Local devices can also change local flow in ways that are visible but not automatically beneficial. The Honda RA107 end-plate louvre example shows the local airflow being modified when the louvres were open, with the upper tuft near vertical. That is a good reminder that a visible change is not the same as a successful change. Opening a path or cutting a slot may clearly alter local flow, but the next question is whether that local alteration supports the rest of the car's aero job. You need the visual evidence first, then the force, drag, or lap evidence.
The basic workflow
Start with one area of the car and one question. The novice error is to cover the car in yarn because it feels thorough. For this skill, thorough means targeted and readable. Choose the part whose behavior you are about to change or the area that plausibly explains the symptom. A rear wing, spoiler, diffuser exit, intake, outlet, or front airdam region is a better starting point than a whole-car decoration exercise.
Write the question in plain language before you run. Good questions sound like this: is the flow attached across the rear wing span at the speeds where the car needs rear load? Is the outer diffuser section flowing rearward cleanly, or is it disturbed? Does opening this outlet change the local flow direction in a way we can see? Does the airdam change the streamline pattern around the front of the car? These questions are narrow enough that the video can answer them.
Prepare the visualization so the answer will be legible. For tufts, you need small yarn or wool pieces placed where their direction can be seen by a camera. The corpus supports the method, not a specific tape brand, exact tuft length, or camera mount, so treat those as practical setup decisions rather than theory. The principle is that the tufts must remain visible while the car is driven, and the driver should not need to look at them while driving. You are gathering evidence for later review, not adding another distraction at speed.
Run the car in the condition that matters. If the complaint is high-speed balance, a slow paddock roll will not answer it. If the problem is a race condition, remember that aerodynamic interactions are a fact of life when cars are racing. Clean-air behavior and traffic behavior can differ. Do not combine all conditions into one conclusion. Treat clean air, following another car, and crosswind or yaw-sensitive conditions as different observations unless your test plan has already controlled for them.
Read the video in slow, boring detail. Look first for attached versus separated behavior. Attached tufts line up in neat rows and flutter a little. Separated tufts whirl and point randomly. Then look for spanwise differences. Does the center behave differently from the end? Does the outer diffuser section look less organized than the middle? Does a louvre or outlet make one tuft stand up or turn across the expected direction? Finally, compare the pattern to the part's job. A wing, diffuser, spoiler, intake, or outlet does not need the same visual answer everywhere, but it does need flow behavior that makes sense for its function.
Only after that should you redesign. The decision sequence is observe, interpret, change, verify. Observe the flow with a simple visualization method. Interpret whether the pattern matches attached flow, separated flow, wrong-direction flow, or a local disturbance. Make the smallest redesign that addresses the visible mechanism. Then verify with the same visual method and, where available, with force, drag, speed, or lap-time data. McBeath's broad warning is important here: it is difficult to generalize in competition-car aerodynamics, and what works on one car may not work on another. Trial and error are part of the development process, but trial and error are only useful when each trial produces evidence.
How to read tufts without fooling yourself
The first reading is direction. On a surface where you expect attached flow, aligned tufts are encouraging. If a row of tufts points steadily rearward with only small flutter, the local flow is behaving in an attached way. If the same row whirls around or points in random directions, you are seeing separated or highly disturbed flow at that surface.
The second reading is consistency. One tuft can lie, especially if it is poorly attached, hidden from the camera, or sitting in a local edge effect. A row is more meaningful than a single piece of yarn. A repeated pattern over multiple runs is more meaningful than one exciting frame. The bonded testing guidance is blunt that aerodynamic testing needs concentration and care, and that mistakes in test execution can make the result worthless. Apply that discipline to visual tests. Do not redesign an expensive part because one tuft looked strange in one moment.
The third reading is location. A tuft near an edge, louvre, outlet, wing tip, or diffuser side may show a strong local structure that is different from the main surface flow. That is not a failure by itself. In the Honda louvre example, the upper tuft angle shows local flow modification when the louvres were open. That tells you the device changed the local flow. It does not by itself tell you that the whole car improved. A local reading must be connected to the question you asked.
The fourth reading is timing. If the tuft pattern changes only at a certain speed, ride height, steering angle, or traffic condition, do not average it mentally into one vague impression. The diffuser example at 20mm ground clearance is a reminder that ride-height condition can matter. The racing-interaction note is a reminder that the flow around cars can change when other cars are nearby. Your visual notes should preserve the condition under which the pattern appeared.
The fifth reading is comparison. If you change a part, compare before and after using the same region, the same camera view, and the same driving condition as far as practical. The lesson on one-variable aero tests handles the full testing discipline, but the visual principle is simple: the comparison is only useful if the viewer can tell whether the tuft pattern changed because the airflow changed, not because the camera view, speed range, or surface region changed.
Using oilflow and CFD-style visualization honestly
Oilflow plots and streamline images are powerful because they show patterns that a driver cannot feel. The oilflow plot shows where surface flow goes. Streamlines can show what is happening off the surface and downstream. McBeath's front-wing example matters because the surface plot and the off-surface streamlines naturally invite comparison. The driver-developer's job is to use that comparison, not to over-read a single image.
When you see surface flow direction, ask what it says about the part's work. Is flow feeding the region you expected? Is it being pulled sideways before it reaches the working area? Is it heading toward an outlet, away from an intake, or across a diffuser section? When you see off-surface streamlines, ask whether they explain the surface trace. A wing tip or downstream vortex structure may make the surface evidence easier to understand. A surface pattern that looks odd may be less mysterious when the off-surface flow is visible.
Do not use pro-level imagery as a reason to skip simple track visualization. The corpus notes that professionals use CFD to model many configurations and wind tunnels to validate solutions, while amateur tools can still assist the search for an aerodynamic solution. It also emphasizes that tools, simple or exotic, have to be used carefully and with common sense. For an intermediate track-day or club-racing driver, that means a tuft video can be a legitimate first tool, not a poor substitute for something you cannot afford. You are not trying to become a CFD department. You are trying to avoid redesigning parts blind.
From visible flow to redesign choice
Once you have a readable pattern, sort the result into one of four decisions.
The first decision is leave it alone. If the flow is attached where you expected attachment, the part is not automatically perfect, but the visible evidence does not justify a redesign aimed at fixing separation. You may still need force measurements or a balance change, but you should not pretend the video showed a surface-flow failure.
The second decision is reposition before reshaping. If a wing or working surface appears to be fed poorly or only partly attached, changing its position, height, or relation to upstream bodywork may be more logical than building a new profile. The ADR rear-wing material in the corpus shows a wing in stock position about to be moved and describes different wing heights being tested. The chunks do not provide the results, so do not invent them. The takeaway is narrower: position is a legitimate test variable, and visual evidence should tell you whether the wing is seeing the flow you think it is seeing.
The third decision is reshape with a specific mechanism in mind. If the visible pattern shows separation over part of the span, then a change aimed at maintaining attachment in that region has a reason. McBeath's wing-twist example gives the logic: a geometry change could keep flow attached across the span for longer and delay large-scale separation and stall. For a diffuser, if the outer section shows wrong-direction or separated flow, the redesign question becomes how to restore organized flow in that region, not how to make the part look more aggressive.
The fourth decision is cross-check with measured performance before committing. A visual improvement should be tested against downforce, drag, speed, or lap-time behavior where possible. The data-logging reference in the corpus is included because logging can help mechanics, engineers, and drivers extract useful information. In this module, that cross-check belongs with the force-instrumentation lessons. Here, the rule is that a beautiful tuft pattern is not the finish line. It is evidence that guides the next measured test.
Calibration cues
You know you are improving at this skill when your notes become more specific. Early notes tend to say the air looked messy. Better notes say the outer third of the diffuser exit showed random tuft direction during the high-speed portion, while the center row stayed mostly aligned. Early notes say the wing looked fine. Better notes say the lower-surface-adjacent tufts stayed aligned across the center span, while the end region showed a repeatable angle change near the tip. Early notes say the outlet worked. Better notes say opening the outlet clearly changed the local tuft angle, but force or speed data still needs to confirm whether that change helped.
You also know you are improving when you stop treating every visible disturbance as bad. The air around a race car is complex. Edges, tips, louvres, outlets, and downstream wakes can create local structures. Your job is not to make every tuft perfectly straight. Your job is to understand whether the flow behavior supports the part's task. On a surface where you need attached flow, random whirling is a warning. Near a device whose job is to redirect or relieve local flow, a visible direction change may be the expected signal.
A third cue is that your redesign ideas get smaller and more testable. Instead of replacing a whole rear aero package, you may decide to move a wing, adjust a local edge, revisit a diffuser outer section, or test an airdam configuration. This is how amateur aero development becomes less wasteful. The corpus repeatedly points away from blind generalization and toward careful tools, common sense, and trial-and-error learning. A good visual test narrows the next trial.
The final cue is that your visual evidence starts to agree with the car's measured behavior. If tuft video suggests improved attachment on a wing and the later force or speed data moves in the expected direction, your confidence rises. If the visual evidence looks better but the car slows down or gains drag without useful load, you have learned something equally important: the visible surface pattern was not the limiting factor, or the change created a cost somewhere else. Either way, you are learning from evidence instead of copying someone else's part.
Safety, legality, and scope
Aero testing still has to be done like car testing. Use appropriate track rules, do not distract the driver, and do not create loose material that can come off the car. The bonded publisher note also flags that some jurisdictions with strict emissions control laws may treat modifications as a legal issue, so compliance is not optional when you start cutting, adding, or redirecting bodywork. This lesson does not give permission to bypass rules, class regulations, or event safety requirements.
Keep the scope narrow. If the goal is airflow visualization, do not simultaneously change springs, wing angle, ride height, tire pressure, and camera placement. That belongs in the one-variable testing lesson, but it matters here because a visual comparison becomes weak when several things changed at once. The clean mental separation is this: visualization tells you what the air appears to be doing, instrumentation tells you what forces changed, and normalized testing tells you whether the run was fair. You need all three skills, but you should not blur them into one vague impression.
The core rule again
Before you redesign an aero part, make the flow visible in the area the part is supposed to control. Use tufts for a cheap, track-usable surface-flow read. Use oilflow or streamline imagery when you have it to connect surface direction with off-surface behavior. Read attached flow, separation, wrong-direction flow, and local device effects as different signals. Then change only what the evidence points at and verify the result. That is how you turn aero development from guesswork into disciplined learning.
Worked example: rear wing attachment before a spanwise redesign
Suppose the car has high-speed rear instability, and the paddock solution being discussed is a larger rear wing. The visible-flow version of the problem starts with a narrower question: is the current wing receiving and maintaining attached flow across the useful span, especially where the car needs rear load?
Place the visualization where the answer will be readable and review the video at the speed range where the symptom matters. If the tufts over the working region line up in neat rows with only small flutter, the local surface flow is attached. That does not prove the wing makes enough load, because visible flow is not a force measurement, but it does argue against guessing that the main failure is simple surface separation. Your next step may be force instrumentation, angle adjustment, or balance work rather than a new profile.
If the video shows repeatable random tuft motion over part of the span, especially if one section loses attachment earlier than another, the redesign question becomes more precise. McBeath's wing-twist discussion gives the mechanism: a geometry change can be used so flow across the span remains attached for longer, delaying large-scale separation and stall and allowing more downforce before that limit. The proper response is not to copy a twist number from another car. It is to use the visible spanwise pattern to decide whether a spanwise change is even relevant.
The wing examples in the corpus also remind you that position can be a test variable. The ADR rear wing is shown in stock position before being moved, and different wing heights are referenced. The chunks do not tell us the result of those height tests, so the lesson should not pretend that higher or lower is automatically correct. The disciplined conclusion is this: before redesigning the profile, find out whether the wing is in the flow you think it is in, whether the flow remains attached where it matters, and whether a position test would answer more than a fabrication job.
Worked example: diffuser outer section before making the diffuser more aggressive
A diffuser problem is a perfect place to practice refusing to redesign blind. The tempting paddock answer is to make the diffuser longer, steeper, wider, or more dramatic. The visual-flow answer starts by looking at whether the existing diffuser flow is organized enough to deserve a more ambitious shape.
The bonded corpus gives two useful warnings. One figure describes flow separation and generally low velocity in a diffuser at 20mm ground clearance. Another notes that the outer section of an Exige diffuser looked as if the flow was going in the wrong direction. Those are visual diagnosis problems. If your diffuser exit or outer section shows random tuft motion, low-energy-looking separated behavior, or a repeatable wrong-direction signal, a more aggressive part may intensify the same failure.
For an intermediate driver, the useful question is local and concrete: does the center, inner section, and outer section behave the same way at the relevant speed and ride-height condition? If the center looks organized and the outer section looks confused, you do not yet have a whole-diffuser answer. You have a section-specific problem. If the problem appears at one ride height but not another, the visual diagnosis has to stay tied to that condition. The McBeath diffuser-at-20mm example is a reminder that the ride-height state can be part of the mechanism.
The redesign should follow the observed failure. Wrong-direction or separated outer flow points toward restoring organized flow in that region, not simply adding area. If the visual pattern improves after a small change, do not stop there. Pair the visual result with the module's downforce and drag instrumentation work. A diffuser that looks cleaner but costs too much drag or upsets balance has not finished the test.
Worked example: airdam streamlines before copying another front-end treatment
Front-end aero is easy to copy badly because airdams, splitters, and spoilers are visually obvious. The corpus includes a streamline comparison 200mm above the ground with and without a 100mm airdam. That is exactly the kind of comparison you want before assuming a shape is good because it looks like a race-car part.
The lesson is not that every car needs a 100mm airdam. The lesson is that the visible streamline pattern can change when the device is present, and the change should be observed before the design is treated as solved. If you are testing an airdam, the visual question is how it changes the air approaching and passing the front of the car. Does the flow stay organized around the region you care about? Does it feed or starve downstream devices? Does it create a local behavior that later shows up as drag or balance change?
This is where sibling lessons matter. The one-variable test lesson keeps the comparison fair. The force-instrumentation lesson tells you whether the front-end change altered downforce or drag. This lesson keeps you from skipping the visible mechanism. You look first, because the same general device can behave differently on different cars and because the corpus is clear that broad generalization in competition-car aerodynamics is risky.
Common mistakes
Mistake one: redesigning from feel alone. The car feels unstable or slow, so you add a part. The corpus argues directly against modifying by feel, intuition, or copying what others have done. Good looks like naming the airflow question first, then using tufts, oilflow, streamlines, or other accessible tools to see the pattern.
Mistake two: treating a single tuft as a verdict. One tuft stands up and the whole part gets blamed. Good looks like reading rows, regions, and repeatable behavior. Attached flow shows tufts lined up in neat rows with only a little flutter. Separated flow shows random whirling and inconsistent direction. A single strange tuft is a clue, not a conviction.
Mistake three: confusing local change with global improvement. The Honda louvre example shows local airflow modification when the louvres were open. Good looks like separating those ideas. Yes, the device changed local flow. No, that does not automatically mean the car gained useful performance. You still need the visual change to connect to the part's job and later to force, drag, or lap evidence.
Mistake four: reading surface flow as the whole airflow story. Oilflow shows surface direction, while streamlines can reveal off-surface and downstream behavior. Good looks like linking surface and off-surface evidence where you have both. If a wing-tip or diffuser-edge pattern is confusing, do not force the surface trace to answer everything by itself.
Mistake five: copying another car's solution. McBeath is explicit that what works on one car may not work on another, even when the cars appear similar. Good looks like using the other car only as an idea source, then testing your own car's airflow. Trial and error are useful only when the trial is observed and recorded.
Mistake six: making the visualization test too busy. Covering several areas and changing several parts at once feels productive, but it can make the video hard to interpret. Good looks like one visual question, one region, one comparison, and enough repetition to trust the pattern.
Drill: three-run visible-flow map
Use this drill at your next test day or practice session when event rules and safety allow exterior visualization. The goal is not to tune the whole car. The goal is to learn how to turn one aero question into a readable visual answer.
Run one is the baseline observation. Choose one region only: rear wing span, diffuser exit, a cooling outlet, or the front airdam area. Apply tufts where the camera can see them, then drive a normal, safe session in the speed range relevant to the question. After the run, review the video and mark three things: where the tufts line up neatly, where they whirl or point randomly, and whether any section behaves differently from the neighboring section.
Run two is the repeat. Do not redesign the part yet. Repeat the same observation as closely as practical. The success criterion is that the main pattern appears again. If the pattern does not repeat, you do not yet have a redesign target. You have a test-quality problem, a condition change, or a pattern that is weaker than you thought.
Run three is the smallest meaningful visual comparison. Make one allowed, reversible change that directly addresses the observed mechanism, or use an existing adjustment if the part has one. The exact change depends on the car and rules, so do not invent complexity. The success criterion is not lap time. The success criterion is whether the visual pattern changed in the expected region. If separated tufts became more aligned in the target area, you have earned a measured force or drag test. If nothing changed, the part you adjusted may not control the flow problem you observed.
Write the result in one sentence after the session. A good sentence is specific: outer diffuser tufts were random on both baseline runs and became more rearward-aligned after the reversible edge change. A weak sentence is vague: aero looked better. The drill is successful when your note names the region, the observed behavior, the condition, and the comparison.
Cross-references inside the module
Use this lesson before the module's instrumentation lessons when you do not yet know where the airflow problem is. Once the visual pattern points to a mechanism, use Instrument downforce and drag before changing aero to find out whether the car gained useful force or paid too much drag. Use Normalize the run before you trust the aero number when you compare before and after results, because visual and measured tests both lose value when the run conditions are sloppy. Use Spot separation before it becomes a setup trap when the visual pattern shows attached flow becoming separated and the temptation is to blame springs, bars, tires, or driver input first.
The separation between lessons matters. This lesson gives you eyes. The instrumentation lessons give you numbers. The normalization lesson protects the comparison. The setup-trap lesson keeps you from chasing chassis changes for an aero-flow problem. Together, they let you change aero parts because the car asked for it, not because another car looked fast with that shape.
Author Review
No quiz questions are attached to this lesson.
Sources
| # | Document | Chunk | Pages | Score | Collection |
|---|---|---|---|---|---|
| 1 | uio julian edgar car aero testing | 06a78fd3-5941-96b8-badf-a5ccb37cf731 | 6 | 1 | uio_books_raw_v1 |
| 2 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 576d96a1-00b7-66dd-f5b1-e33666cc457f | 334 | 1 | uio_books_raw_v1 |
| 3 | Competition Car Aerodynamics 3rd Edition McBeath Simon | dbf70393-f24e-23f2-3f7a-133b7af33c77 | 78 | 1 | uio_books_raw_v1 |
| 4 | Competition Car Aerodynamics 3rd Edition McBeath Simon | c7d0125c-8080-dbcc-df83-3b96d0b84bab | 477 | 1 | uio_books_raw_v1 |
| 5 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 6edca499-2988-7702-ccc8-3d17b516edff | 385 | 1 | uio_books_raw_v1 |
| 6 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 4cd6a5d6-20be-26fa-1ddd-9abe28dae72a | 234 | 1 | uio_books_raw_v1 |
| 7 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 09410e4a-fe94-a192-45b0-1a276a5adb93 | 271 | 1 | uio_books_raw_v1 |
| 8 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 870b343d-3f04-6fe9-e1ba-0a47fd466f68 | 189 | 1 | uio_books_raw_v1 |
| 9 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 9a496275-f006-9cdc-8647-b7acc6459056 | 42 | 1 | uio_books_raw_v1 |
| 10 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 43f9ecd8-7336-a0ec-07a9-5149279141e4 | 43 | 1 | uio_books_raw_v1 |
| 11 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 9e3001fd-e626-5565-9b11-bc3cab151d27 | 281 | 1 | uio_books_raw_v1 |
| 12 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 10acd525-ae45-7603-2847-9b1b9db65585 | 9 | 1 | uio_books_raw_v1 |
| 13 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 61068e74-0999-1e25-03bd-8c545f352d25 | 26 | 1 | uio_books_raw_v1 |
| 14 | Competition Car Aerodynamics 3rd Edition McBeath Simon | ba67dd3a-d141-8206-f5ac-652fb8c13487 | 91 | 1 | uio_books_raw_v1 |
| 15 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 78757d4c-5981-472c-7a07-7b79b28d7bc4 | 218 | 1 | uio_books_raw_v1 |
| 16 | Competition Car Aerodynamics 3rd Edition McBeath Simon | d59eb8fd-22e2-e843-aecb-2b23f2efcbda | 213 | 1 | uio_books_raw_v1 |
| 17 | Competition Car Aerodynamics 3rd Edition McBeath Simon | d788f877-dfdc-2c41-96e0-e6a0de38e907 | 412 | 1 | uio_books_raw_v1 |
| 18 | Competition Car Aerodynamics 3rd Edition McBeath Simon | 4b5e1aa7-14cf-aacf-908a-c47094ea7ba5 | 504 | 1 | uio_books_raw_v1 |
| 19 | Competition Car Aerodynamics 3rd Edition McBeath Simon | b10ba858-0fd5-9b94-7474-322f3c81f9c4 | 8 | 1 | uio_books_raw_v1 |