The Lap Analysis Workflow tells you the process. This page covers the channels themselves - what each trace shape means at every phase of a lap, what good numbers look like in a club racing context, how AiM EVO4S and VBOX HD2 record each channel differently, and - critically - how to tell whether an anomaly in the data is a sensor issue or a genuine driving problem. These are the channels used every weekend in Porsche Club GB. The ranges and observations come from real data, not textbook theory.
Speed, G-Force and Rotation
The channels that do the diagnostic work in a post-session debrief. Every other channel is secondary to these six.
What the trace shape tells you
The speed trace is the primary lap analysis channel. Its shape across a corner defines whether you are looking at an entry problem, an apex problem, or an exit problem. The three key markers are: where deceleration begins (braking point), the lowest point of the trace (apex speed), and where the trace begins climbing consistently (throttle application point).
A smooth, continuous deceleration curve from braking point to apex indicates controlled trail braking. An abrupt flat section partway through deceleration - where speed briefly stops falling - can indicate a brake release and reapplication, a hesitation at turn-in, or a change in confidence or line. Confirm with Ax, video, and corner context before drawing the conclusion. This pattern is one of the most common findings in club coaching but the cause needs establishing rather than assumed.
On corner exit, the speed trace should climb without hesitation from the throttle point. A trace that flattens or dips slightly before resuming climb indicates exit hesitation. That may be wheelspin correction, a confidence lift, or the driver managing a balance issue. Confirm with RPM, Ax, and video before calling it wheelspin.
AiM EVO4S vs VBOX HD2
The EVO4S and VBOX HD2 both provide GPS-derived speed, but they do not sample, filter, or present it in exactly the same way. The EVO4S uses the GPS09 external module supplied with the kit. The VBOX HD2 records GPS data at a rate that depends on the specific variant - the standard RLVBVDHD2 logs at 10 Hz, while the RLVBVDHD2-H (V10) logs GNSS data at 25 Hz. Check your unit's part number if you are unsure which variant you have. For coaching work, compare broad speed-trace shape, apex speed, and sector loss first. Be more cautious with very short transient comparisons -braking onset and release timing in particular -when working across different hardware families, and be aware that a 10 Hz GPS system and a 25 Hz GNSS system will not give identical braking-point resolution even when used on the same car.
The two systems will not give identical speed numbers at the same moment due to differences in update rate, positional averaging, and filtering. When comparing a VBOX reference lap against an EVO4S driver lap, expect some offset in peak speed readings that is instrument variance, not a real difference. Use the shape of the trace for coaching and the delta for timing, not absolute speed numbers across different devices.
What the trace shape tells you
Lateral G is the channel that reveals how hard the car is being worked in cornering and - more usefully - how the grip is being used through each phase of the corner. The shape matters more than the peak.
A clean arc: Ay builds smoothly from corner entry, peaks at or just past the geometric apex, then decays progressively as the car unwinds toward the exit. This is a driver using grip continuously, not in bursts. The decay rate on exit reflects how aggressively they are applying throttle - a faster decay usually means earlier, more committed throttle.
A flat plateau mid-corner: Ay reaches a level and stays there without continuing to build. The car has stopped rotating. This is understeer - the front has run out of grip and the driver is not generating any more yaw. The plateau level tells you the grip ceiling the front is hitting.
A mid-corner spike: A sharp upward feature in the Ay trace that does not correspond to a tighter part of the corner is nearly always a driver correction - a steering input to catch a slide, or a correction after running wide. Cross-reference with video immediately before concluding this is a handling problem.
Typical ranges for club racing
These figures are for dry conditions on a representative club circuit. Porsche Boxster Cup cars on Pirelli P Zero Trofeo R control tyres with single adjustable dampers and free springs, no aero. Sprint/hillclimb cars on semi-slicks will sit at the higher end or above.
Road tyres, standard suspension: use medium-speed corners as the most reliable Ay benchmark. Peak values in dry conditions commonly sit around the low-to-mid 1 g range, depending on circuit, corner radius, tyre state, and logger installation. Fast corners with long sustained arcs give the cleanest comparisons - hairpins are contaminated by combined braking, rotation, and traction loads and are not reliable Ay benchmarks.
Boxster Cup (Pirelli Trofeo R, adjustable dampers): use medium-speed corners with sustained lateral loading as the primary comparison points. Well-driven cars on this control semi-slick will show materially higher peak Ay than a road-tyred club car. The most reliable benchmark is always the same corner, same day, same tyre state, against a reference lap in the same car - an absolute ceiling figure is less useful than the delta to the reference.
Wet conditions (Pirelli WH Wet): Peak lateral G drops significantly on the control wet tyre. All dry benchmarks are invalid in wet conditions. Absolute Ay figures in the wet are less consistent than in the dry because they vary more with track surface water depth, tyre temperature, and line choice. The more useful wet analysis metric is consistency of Ay shape lap-to-lap -a driver generating smooth, sustained lateral G in the wet is managing the tyre better than one with identical peak numbers but a spiky, erratic trace. Treat any wet Ay benchmarks as session-specific references rather than fixed targets.
For how these G benchmarks translate into setup diagnosis -and what each channel signature means for specific permitted adjustments -see Part 3: Setup Correlation.
Semi-slick club saloon: peak Ay can extend well beyond a road-tyred club car, depending on tyre compound, circuit, fuel load, and logger installation. Slick tyre race car with aero: 2.0 g and above is achievable in the right conditions, but again the reference lap in the same car on the same day is more useful than any absolute figure.
What the trace shape tells you
Longitudinal G covers braking (negative, typically plotted below zero) and acceleration (positive). It is the channel that most directly reveals braking technique - both the efficiency of the stop and how the brakes are being released into the corner.
Under braking: The ideal trace builds rapidly to peak braking force, holds that level briefly, then releases progressively as the driver begins trail braking toward the apex. The release slope is where most club racers lose time - an abrupt vertical drop to zero means the driver has released the brakes completely before the apex rather than trailing them in. Trail braking holds the front tyres loaded, maintains rotation, and allows a higher apex speed.
Under acceleration: Ax should build smoothly from the throttle application point. Regular oscillation under acceleration - a repeating wave pattern - is wheelspin being corrected either by the driver lifting or by traction control intervention. A single large spike followed by a drop is usually a gear change.
In the corner: Any significant positive Ax event mid-corner (throttle burst while still turning) that is not matched by a corresponding reduction in Ay usually indicates the driver is using a throttle input to manage understeer or to recover balance - not to accelerate.
Typical ranges for club racing
Peak braking deceleration: A well-driven club car on road tyres with standard brakes achieves 1.0-1.2 g of peak longitudinal deceleration. A car on semi-slicks with uprated brakes can reach 1.4-1.6 g. If a driver is consistently below 0.8 g at their peak braking point, they are either braking very early (no need for hard braking) or not maximising braking force.
Peak acceleration: Highly dependent on power-to-weight. A Boxster Cup car (230-240 BHP, ~1350 kg) will show 0.3-0.45 g sustained acceleration out of medium corners. A sprint car with slicks and lighter weight will sit higher. The absolute number matters less than consistency lap-to-lap and the shape of the build.
The most useful braking metric is not peak Ax on its own but the area under the braking curve - the integral of deceleration over time. In practical terms: how quickly braking builds to effective peak deceleration, how long the driver sustains it, and how cleanly it is released toward the apex. A driver who builds to 1.0 g quickly and holds it stops faster than one who peaks at 1.1 g but takes twice as long to build - the area under the first trace is larger even though the peak is lower.
What the trace shape tells you
Yaw rate is the rate of rotation about the car's vertical axis - how fast the car is turning its nose. It is the channel that most directly reveals balance: whether the car is rotating as expected for the corner being driven, rotating too much (oversteer), or not enough (understeer).
At a given speed and corner radius, there is a predictable yaw rate for a neutral car. If the measured yaw rate is lower than expected, the front is not generating enough rotation - understeer. If it is higher, the rear is contributing more rotation than the front can control - oversteer building.
The yaw rate spike: A sudden sharp increase in yaw rate mid-corner, particularly on corner exit when the driver is applying throttle, is the rear stepping out. The driver's corrective steering input will appear almost simultaneously - a brief reduction in yaw rate as the front counters the slide. This sequence is a clear oversteer signature, and it tells you whether the snap is predictable and managed or sharp and reactive.
Yaw rate lag at turn-in: If yaw rate builds slowly after the steering input, the car is reluctant to rotate. That can come from entry technique, brake release timing, line choice, or front-limited balance. Check it against the speed trace, brake release shape, and video before attributing it to a setup issue - technique and setup produce the same signature here.
AiM EVO4S IMU notes
The EVO4S measures yaw rate from its internal 6-axis MEMS IMU. The IMU must be mounted securely and level - any flex in the mounting bracket can introduce cross-axis contamination and noise into yaw rate, lateral G, and longitudinal G, making genuine vehicle events harder to separate from mounting artefact.
Mounting checks: The arrow on the EVO4S body must point toward the front of the car. Any rotation about the vertical axis (yawing the unit on the mount) mixes lateral and longitudinal G channels and produces results that look like genuine handling events but are not. Even 5 degrees of mounting rotation creates a measurable cross-axis error at high G levels.
Typical yaw rate values: A Porsche Boxster at medium speed through a 90-degree corner will show 20-35 deg/s peak yaw rate. A hairpin taken slowly will show higher values - 40-60 deg/s is normal for a tight slow corner. Values above 80 deg/s in a car without significant aero usually indicate a slide event rather than cornering.
What the trace shape tells you
TPS is the channel that reveals driver commitment and throttle discipline. The trace should show three clean phases on most corners: fully or near-fully closed through braking and corner entry, a disciplined transition around rotation, and then a clean progressive build to 100% on corner exit with no unnecessary reversals.
The partial throttle exit: A TPS trace that builds to 40-60% and pauses before continuing to 100% is a driver who has applied throttle, felt the car move, and backed off. This is almost always a confidence issue with corner exit balance rather than a genuine car problem - the car is not actually on the limit at that point. Cross-reference with Ax and Ay at the same moment: if neither channel shows a strong corresponding G event, the lift is more likely to be confidence- or balance-related than a genuine grip limit event.
Throttle in the braking zone: Any TPS value above zero during the braking phase can mean the driver is left-foot braking (normal and usually fine), has not fully released the throttle before applying the brake (overlap that can cause confusion on some drive-by-wire systems), or is applying throttle to manage understeer mid-corner (worth investigating).
CAN vs analogue TPS input
On the Boxster Cup cars, TPS is pulled via CAN from the Porsche DME. This gives calibrated percentage values at ECU resolution with no sensor noise. The reading is what the ECU sees, which is the correct reference for understanding engine and traction control behaviour.
On a car without CAN access, TPS is wired as an analogue 0-5V input to the EVO4S. The channel must be calibrated: close the throttle fully and zero the channel, then open to 100% and set the full-scale value. Uncalibrated TPS that shows 5% at closed throttle and 95% at full throttle will make every analysis appear as though the driver never fully closes or opens the throttle - a common source of confusion in first-time data setups.
The VBOX HD2 does not log TPS natively without an external input -it requires a vehicle CAN connection or serial data feed. If TPS is not available in VBOX data, longitudinal G under acceleration is the practical proxy: smooth Ax build indicates progressive throttle application; oscillation indicates on/off inputs or traction control intervention. For full TPS access on VBOX-equipped cars, verify whether the installation includes a CAN interface to the vehicle ECU.
What the trace shape tells you
RPM cross-referenced with speed and TPS reveals gear selection, engine response, and whether the driver is consistently using the same gear in the same place each lap.
Gear selection consistency: Overlay 3 laps and check whether the driver is changing gear at the same RPM and same track position. Variation of more than 200 RPM at the same upshift point across laps indicates inconsistency in shift timing - this may be driver habit or it may indicate the car has a gear engagement issue that is forcing early shifts.
Downshift into corners: The RPM trace at the corner entry should show a clean rise as the driver blips and downshifts through gears. A missed blip shows as an RPM drop followed by a large spike as the gearbox drags the engine up - this will also appear as a brief longitudinal G event that looks like additional braking but is actually drivetrain shock.
RPM at peak corner: The RPM at apex tells you whether the driver is in the correct gear. If apex RPM is consistently below the torque peak for the engine, the driver is a gear too high and not in the strongest acceleration part of the powerband for the exit phase.
Wheelspin identification
The classic wheelspin signature in data is: RPM rising faster than speed. If GPS speed shows the car accelerating at 0.3 g while RPM is climbing sharply, the wheels are spinning faster than the car is actually moving. This is confirmed if you have wheel speed logged separately - GPS speed and wheel speed diverging under acceleration is unambiguous wheelspin evidence.
Without wheel speed, the combination of fast-rising RPM, flat or slowly-rising GPS speed, and an oscillating Ax trace is the wheelspin diagnostic. On a traction-control equipped car, TC intervention will appear as a step-down in RPM (cut) followed by recovery - a sawtooth pattern under acceleration rather than a smooth build.
Heading, Sector Delta and Lap Time
Heading is the compass direction of travel derived from GPS position changes. It is not directly useful for lap-by-lap coaching but is the channel that generates the track map in AiM Race Studio and Circuit Tools - the overhead view of the circuit that everything else is spatially referenced against.
Track map accuracy: If the track map looks distorted or corners appear in the wrong place, the heading channel is drifting due to GPS positional error - usually caused by low satellite count or multipath interference at that part of the circuit. This does not necessarily corrupt speed or G data, but it does mean the spatial alignment of channel events to circuit position will be inaccurate. Re-check at circuits with grandstands, trees, or large structures close to the track.
Line comparison: Overlaying the GPS track maps of two drivers on the same circuit gives a useful visual reference for where the lines diverge. The heading channel rate of change through a corner also reveals corner radius - a faster heading rate means a tighter radius, which cross-referenced with speed tells you whether a driver is taking a geometric line or a late-apex line. This is a useful setup discussion aid but should always be backed by the speed and G traces rather than used in isolation.
Sector delta is the cumulative time gained or lost versus the reference lap within each defined circuit sector. It is the primary diagnostic channel in the lap analysis workflow - the first thing to look at, every session, before opening any other trace.
How it accumulates: The sector delta is cumulative within a sector, not per-corner. A delta of +0.4 s at the end of sector 2 means the driver is 0.4 s behind the reference across the whole of that sector. It does not tell you which corner in the sector caused the loss - that requires the speed trace to diagnose. But it tells you with certainty that the problem is in sector 2, and that spending analysis time on sectors 1 and 3 is a lower priority.
Sector placement matters: Sector boundaries placed at braking zones (before a corner) include the corner exit in the current sector and the braking phase of the next corner. This can attribute a corner exit loss to the wrong sector. Ideally, place sector boundaries at the end of the main straight - at the braking point marker - so each sector captures a full braking zone, corner, and exit. If the sector splits were placed at circuit setup and cannot be moved, understand the boundary positions before interpreting the delta.
GPS lap time vs sector delta: Always verify that the GPS start/finish line placement gives a lap time consistent with the timing transponder within 0.1-0.2 s. If the GPS lap time is significantly different from the transponder, the GPS gate is placed incorrectly and the sector splits will be offset - every split will be wrong by the same amount, which can make a fast sector look slow and vice versa.
Which Channels to Look at Together
No channel in isolation tells the full story. These are the pairings used in every post-session debrief, and what each combination is designed to reveal.
| Channel combination | What you are diagnosing | What to look for |
|---|---|---|
| Ay + Yaw Rate | Understeer vs oversteer balance | Ay without corresponding Yaw Rate increase = front not generating rotation (understeer). Yaw Rate rising faster than Ay would suggest = rear leading (oversteer building). The ratio between the two channels at the same moment is the balance indicator. |
| Ax + TPS | Throttle discipline and braking technique | Any TPS above zero while Ax is still negative (braking) needs explaining -left-foot technique, overlap, or mid-corner throttle used to manage balance. TPS building progressively while Ax transitions from negative to zero on corner exit = trail throttle, generally fine. TPS applying and holding a small non-zero value through the mid-corner phase while Ax is near zero -without a corresponding clean exit build -can indicate the driver is using maintenance throttle to prevent lift-off oversteer on a trailing-throttle-sensitive car such as the 986 Boxster. Cross-reference with Ay and yaw rate at the same moment. TPS reversing mid-exit while Ax drops = driver has lifted to manage a problem. |
| GPS Speed + Ay | Grip usage and cornering efficiency | The G-v diagram (lateral G plotted against speed) shows the cornering envelope the driver is reaching at different speeds. A driver who reaches high lateral G at low speed but not at high speed is underdriving fast corners. A driver who hits the same Ay ceiling regardless of corner type is tyre-limited, not technique-limited. |
| GPS Speed + RPM | Wheelspin and gear selection | RPM rising faster than GPS speed = wheelspin. RPM lower than expected for the GPS speed in a given gear = driver is in the wrong gear. RPM at apex consistently below torque peak = one gear too high into the corner. |
| Ax + Ay (combined G) | Combined acceleration demand and friction circle usage | The combined G vector (sqrt of Ax squared plus Ay squared) represents the total longitudinal-plus-lateral acceleration demand in the plane of the road. Plotting this over time shows whether the driver is using the available grip continuously or leaving gaps between braking, cornering, and exit drive where the demand drops unnecessarily. |
| Sector Delta + GPS Speed | Where in the sector the time is being lost | Delta growing rapidly in the middle of a sector, then stabilising toward the sector end, points to a mid-sector corner. Delta growing only near the sector end points to the last corner before the boundary. Match the delta rate of change to the track map to identify the specific corner. |
| Yaw Rate + GPS Heading | Line validation and IMU accuracy check | Heading rate of change and yaw rate should be consistent with each other - both describe rotation of the car. If they diverge significantly, either the IMU mounting is incorrect (yaw rate error) or the GPS has a positional artifact (heading error). Use this pair as a sanity check on the quality of the data before drawing coaching conclusions. |
Configuration Pitfalls
Bad configuration is worse than no data. If the inputs are wrong at the system level, every channel is subtly or obviously wrong - and the errors are invisible unless you know what to look for. These are the mistakes that get discovered mid-season rather than at setup.
What goes wrong
Wheel speed is calculated by counting pulses from the ABS ring or dedicated sensor and multiplying by the rolling circumference you enter in the logger configuration. If that number is wrong - even by 2-3% - every wheel speed reading is wrong by the same proportion, every lap.
The most common source of error is entering the static tyre circumference from the tyre manufacturer's data sheet rather than the loaded rolling circumference under race conditions. A tyre that measures 1,910 mm circumference on a static tyre gauge will typically have a dynamic rolling circumference of 1,870-1,885 mm under load at operating temperature and pressure. Using the static figure overstates wheel speed by 1.5-2%.
On a Boxster Cup car at 180 km/h, a 2% error is 3.6 km/h. That is larger than most of the speed deltas you are trying to coach.
How to find and fix it
GPS cross-check: On a straight section at constant speed, GPS speed and wheel speed should agree within 1-2 km/h. A consistent offset - wheel speed always reading higher or lower than GPS across the whole session - is a circumference constant error. The correction factor is: correct circumference = entered circumference x (GPS speed / wheel speed).
Practical method: Roll the car exactly one wheel revolution on a flat surface, mark the contact patches, measure the distance. Do this at operating tyre pressure, not cold. Repeat three times and average. Enter this as your rolling circumference.
Warning sign: GPS and wheel speed that agree at constant speed but diverge under acceleration indicates genuine wheelspin - that is real data. A constant offset everywhere including constant speed is the circumference constant. Do not confuse the two.
What goes wrong
The AiM EVO4S lets you configure individual channel logging rates. The internal IMU can sample at up to 200 Hz but the default logging rate for many channels is 10 Hz or 25 Hz. If a critical channel has been inadvertently set to 5 Hz or 10 Hz, the data appears valid but important events are being smoothed out or missed entirely.
At 5 Hz (200 ms between samples), a braking zone that lasts 1.5 seconds gives you 7-8 data points. The shape of the longitudinal G buildup and trail braking release is completely unresolvable at that rate. You can see that braking happened. You cannot see how.
At 10 Hz (100 ms), you get 15 points across the same braking zone. Better, but still inadequate for braking technique coaching. The trail braking release - which happens over 0.3-0.5 seconds - is represented by 3-5 samples. Any detail in that phase is lost.
Recommended rates and how to check
GPS speed and position: log at the highest rate your hardware supports. The EVO4S uses the GPS09 external module -the supported GPS rate depends on the GPS09 firmware and configuration; check your AiM logger settings before the season. For the VBOX HD2, the GPS logging rate depends on the variant: the standard RLVBVDHD2 records at 10 Hz; the RLVBVDHD2-H (V10) records GNSS data at 25 Hz. Do not configure the logging rate lower than your hardware provides -you cannot recover resolution that was never sampled.
Ax, Ay (longitudinal and lateral G): 25 Hz minimum, 50 Hz preferred. At 50 Hz you have 20 ms resolution - sufficient to see braking initiation events, steering inputs, and throttle application detail.
Yaw rate: 25-50 Hz. At lower rates the shape of oversteer events is compressed and the onset speed cannot be determined.
Engine channels (RPM, throttle, gear): 10-25 Hz is adequate. These change slowly relative to dynamics channels.
How to check: In AiM Race Studio, open Channel Properties for any channel and check the recorded frequency. The timestamp delta between consecutive samples confirms the actual logged rate. If it does not match your configuration, check the logger firmware version - some older EVO4S firmware has a known issue where certain channels revert to default rates after a firmware update.
What goes wrong
In AiM Race Studio and Circuit Tools, GPS lap timing is triggered when the car's GPS position crosses a software-defined gate on the track map. If the gate is not precisely positioned at the physical start/finish loop location, GPS lap times will differ from transponder times - and more importantly, the sector splits will be measuring different parts of the circuit than intended.
A gate that is 20 m past the actual start/finish line adds a constant offset to every lap time but the error is consistent - all laps are wrong by the same amount, so relative comparison is still valid. A gate that is placed at a point where GPS lock is unreliable (near a bridge, grandstand, or overhead structure) produces variable errors - some laps trigger early, some late - and the sector times become noise.
Sector gates are more damaging than the start/finish gate when misplaced. A sector 2 gate that is 50 m away from where you believe the sector boundary is means every sector 2 time includes 50 m of sector 1 or sector 3 road. The sector delta analysis - the entry point of the whole workflow - is measuring the wrong segment of the circuit.
How to verify and correct
AiM Race Studio: Open the track map view. The gate is shown as a line across the circuit. Zoom in and check it sits on the start/finish line, not a car length before or after it. The GNSS track map is typically accurate to 2-5 m - sufficient to place the gate correctly at most circuits.
Circuit Tools: Open the run, go to the Event Setup view, and check the trigger line position on the satellite map overlay. Move it if needed.
Validation: Compare GPS lap time against transponder time for 3-4 laps. A consistent offset of less than 0.2 s is acceptable - this is GPS positional accuracy. A consistent offset larger than 0.3 s means the gate is in the wrong place. A variable offset that changes lap to lap means GPS signal quality is poor at the gate location - move the gate to a section of the circuit where GPS lock is reliable.
At circuits with overhead structures: Thruxton pit straight has no issue. Silverstone National pit lane entrance can cause dropout. If you have problems at a specific venue, shift the gate to the first or last sector boundary instead and offset your sector split comparison accordingly.
What goes wrong
The AiM EVO4S has a defined axis system printed on the unit: the arrow on the body must point toward the front of the car, and the unit must be mounted level (not angled up on a sloped surface). If either condition is not met, the G channels are measuring a mixture of the axes you want and the adjacent axes - the output appears plausible but is systematically wrong.
Rotation about vertical axis (yaw mounting error): If the unit is rotated 10 degrees clockwise when viewed from above, braking events appear partly as lateral G and cornering events appear partly as longitudinal G. The channels both still respond - they just respond to the wrong inputs. This error is very difficult to spot without the IMU check procedure because the traces look reasonable in isolation.
Tilt about longitudinal axis (roll mounting error): Gravity introduces a constant offset into the lateral G channel. A 5-degree tilt introduces approximately 0.087 g of constant lateral bias. In a Boxster Cup context where you are looking at lateral G differences of 0.1-0.2 g between drivers, a 0.087 g bias is significant.
The three-check procedure
Static check: Car on level ground, ignition on, EVO4S powered. Open the live channel view in AiM Race Studio or on the EVO4S display. Ax should read 0.00 g, Ay should read 0.00 g, Az (vertical) should read 1.00 g. Any offset on Ax or Ay greater than 0.02 g indicates a mounting tilt problem. Use AiM's built-in tilt correction after physically correcting the mount angle.
Direction check: Drive out of the pit lane and brake firmly. Ax should go negative. Turn left out of the pit lane - Ay should go positive. If either is reversed, the unit is mounted back-to-front or mirrored left-right. This check takes 30 seconds and should be done at every new circuit and after any logger removal or reinstallation.
Dynamic validation: After the session, compare yaw rate with the broad shape of GPS heading-rate change as a gross sanity check. If they diverge badly, re-check the physical mounting against the car axes and repeat the static and direction checks. Do not attempt to correct a mounting error by rotating the logger to make the traces visually match - that masks the error rather than fixing it.
What goes wrong
The Motorsport VBOX HD2 records GPS-derived speed and can also log wheel-speed-related data via vehicle CAN or connected Racelogic input modules, depending on the installed hardware and configuration. At most UK club circuits, GPS-derived speed is the correct starting point. Where wheel-speed data is available, validate it against GPS before relying on it for derived calculations such as braking-point distance, sector splits, or distance-based overlays.
The more common issue is the opposite: a wheel speed input is configured as primary but the wheel circumference constant has not been verified (see above). In this scenario, GPS speed is the more accurate channel but the software is using wheel speed for all derived calculations - lap time, sector splits, distance reference, braking point distance. Every downstream calculation inherits the wheel speed error.
Recommended configuration
GPS as primary speed source is the correct starting point for most UK club circuits where satellite coverage is good. The GPS logging rate of your VBOX HD2 determines the spatial resolution available for braking-point and transient analysis: the standard RLVBVDHD2 logs at 10 Hz, which is sufficient for broad speed shape, apex speed comparison, and sector timing. The RLVBVDHD2-H (V10) logs GNSS data at 25 Hz, which gives improved resolution for finer braking-point comparison work. Both are usable for club-level coaching; the difference becomes relevant when comparing very short transient events or closely spaced braking markers.
When wheel speed matters: If the circuit has known GPS dropout areas, a verified wheel speed input provides continuity of speed data through those sections. Verify the wheel circumference constant before relying on wheel speed for any derived calculation - an incorrect constant introduces a systematic error into every downstream figure. The circumference check procedure is in the pitfall above.
Check your specific configuration: The VBOX HD2 supports various input and configuration options. Refer to the Racelogic documentation for your specific firmware version for the correct speed source setup procedure rather than relying on generic guidance - the available options and their behaviour depend on the installed firmware and connected hardware.
Channel Questions from the Paddock
The questions that come up when a driver looks at their data for the first time.
It depends on the frequency of the noise. High-frequency noise on the Ay channel from an AiM EVO4S is almost always one of three things: IMU mounting flex, electrical interference, or the sample rate being displayed at full resolution without smoothing.
IMU mount flex: The most common cause. If the EVO4S is mounted on a bracket that resonates with engine or road vibration, the MEMS accelerometers pick up that vibration as lateral G signal. The noise frequency will be consistent with the vibration source - engine RPM harmonics or road surface frequency. Fix by stiffening or damping the mount. The unit should be on a solid surface with no flex in the bracket.
Display smoothing: AiM Race Studio applies a smoothing filter to channel display by default. If you have turned this off, the raw IMU data at 200 Hz looks much noisier than the filtered view. For lap analysis purposes, the default 10 Hz or 25 Hz display rate is correct - you do not need raw IMU data for coaching work, only for detailed dynamics analysis.
Road surface: A Ay trace that is noisier on certain sections of the circuit but smooth elsewhere is the road surface transmitting through the suspension and chassis. This is real data, not instrument noise - the car is moving laterally due to surface irregularity. There is nothing to fix.
For braking-point work, higher GPS rates give better spatial resolution. At 10 Hz and 150 km/h you get a position update roughly every 4 m -sufficient to identify where a braking zone begins and to compare apex speeds. At 25 Hz the update interval drops to approximately 1.7 m at the same speed, which meaningfully improves resolution for finer braking-point and transient comparison work. The standard VBOX HD2 (RLVBVDHD2) logs GPS at 10 Hz. The VBOX HD2 V10 variant (RLVBVDHD2-H) logs GNSS at 25 Hz. The AiM EVO4S GPS rate depends on the GPS09 module configuration -check your logger settings before the season, not just before the session.
For G-force channels (Ax, Ay, Yaw), the EVO4S IMU runs at up to 200 Hz internally. The useful analysis rate for coaching is 25-50 Hz -sufficient to see the shape of braking build and release, corner entry events, and exit throttle progression. Higher rates add file size without adding actionable information for a post-session debrief. For longitudinal G analysis of very short braking zones -under 50 m -or for detailed ABS analysis, 100 Hz is useful. For everything else at club level, 25 Hz is the practical minimum and 50 Hz the comfortable ceiling.
For G-force channels (Ax, Ay, Yaw), the EVO4S IMU runs at up to 200 Hz internally. The useful analysis rate for coaching is 25-50 Hz - sufficient to see the shape of braking build and release, corner entry events, and exit throttle progression. Higher rates add file size without adding actionable information for a post-session debrief.
For longitudinal G analysis of very short braking zones - under 50 m - or for ABS and traction control detailed analysis, 100 Hz IMU data is useful. For everything else at club level, 25 Hz is the practical minimum and 50 Hz is the comfortable ceiling.
Three checks, in order.
Static check: With the car stationary on level ground, all three G channels (Ax, Ay, Az vertical) should read approximately 0g, 0g, and 1g respectively. If Ax or Ay shows a static offset, the unit is mounted on an angled surface. A small tilt introduces a gravitational component into the measurement that appears as a constant G bias on every reading. The AiM software has a tilt correction function - use it, but also correct the physical mount if the tilt is more than 5 degrees.
Direction check: Drive a straight section of road and brake hard. Longitudinal G under braking should be negative (deceleration). If it is positive, the unit is mounted back-to-front. The arrow on the EVO4S body must point toward the front of the car. Similarly, turn left and check that lateral G is positive. If not, the unit is mounted facing the wrong direction laterally.
Dynamic check: After a session, compare yaw rate with the broad shape of GPS heading-rate change as a gross sanity check. If they diverge badly, re-check the physical mounting against the car axes and repeat the static and direction checks. This is a gross check only - it is not a precision alignment method and should not be used to tune the mount position.
The transponder is right for the official lap time. The GPS lap time is a derived measurement based on when the GPS position crosses a software-defined gate. If the gate is positioned correctly at the start/finish line and the GPS has good lock at that point on the circuit, the two times should agree within 0.1-0.2 s.
A difference of more than 0.3 s usually means the GPS gate is placed at the wrong position. Open the track map and check where the gate line falls - if it is offset from the actual start/finish loop, move it. In AiM Race Studio, you drag the gate markers on the track map view.
A difference that varies lap-to-lap (sometimes close, sometimes 0.5 s different) usually indicates GPS dropout at the start/finish area - often caused by a bridge, grandstand, or overhead structure. In this case the GPS lap time is unreliable and sector splits derived from it will be incorrect. The transponder is the valid time; use GPS channels for shape and delta analysis only, not for absolute timing at that circuit.
Yes, with caveats. The channels that compare reasonably well across different hardware are GPS-derived speed shape, apex speed, lap time, sector structure, and broad delta trends. What does not compare as cleanly is very fine transient detail - exact braking onset, brake-release timing, and short-lived events - because different systems use different update rates, filtering, and positional averaging. Treat broad comparisons as valid; treat precise transient comparisons with more caution when the hardware differs.
G-force channels do not compare as cleanly across different hardware or sensor stacks as GPS-derived speed and timing channels do. An EVO4S internal IMU and a different sensor configuration on another logger can give different absolute values at the same event because the sensors, mounting positions, update rates, and filtering may all differ. Use G channels for shape comparison across hardware - the arc, the onset rate, the decay - not for absolute value or precise timing comparison. Within the same hardware platform in identical mounting positions, G channel comparison is valid.
Within the same hardware platform -two EVO4S loggers, or two VBOX units of the same variant in identical mounting positions -G channel comparison is valid and is used regularly in multi-car team analysis. If two VBOX HD2 units in the same programme are different variants (one RLVBVDHD2 at 10 Hz and one RLVBVDHD2-H V10 at 25 Hz), the GPS sampling rate differs between them. Speed shape and apex speed comparisons remain valid; very fine transient detail should be treated with the same caution as any cross-hardware comparison.
A sudden drop to zero in the middle of a corner where you would expect sustained lateral G is almost never a genuine event - cars do not unload all lateral grip instantaneously in the middle of a corner without a very dramatic reason. Check for a data dropout first: if all channels go to zero or flat simultaneously, the logger has paused or the file has a corruption artefact. This is the most common cause.
If only Ay drops to zero while other channels continue normally, look at the video. A brief zero in Ay can be caused by the car going over a significant kerb or crest that momentarily unloads the suspension - the car is briefly airborne or at minimal wheel load and lateral G genuinely falls. This is real data.
If Ay drops to near-zero but not exactly zero, and the car was not on a crest or kerb, the driver may have straightened the steering briefly mid-corner - either to correct an entry error or because they ran out of road and had to abort the corner. The GPS track overlay will show this clearly - the car line will go straight across the corner rather than following the arc.