Gap and flush are the two basic measurements of how well a car’s body panels fit together. Gap is the width of the seam between two neighbouring panels, and flush is the height difference between their surfaces across that seam. Below: definitions, why it matters, where it is checked and three ways to measure it.
Gap, flush and parallelism: the definitions
Gap
Gap is the opening between two panel edges, such as a front door and a fender. An early Ford patent on gap and flush gauging describes it as “the distance between the edges of the panels measured in the planes of the panels”. Body-design research from Chalmers describes the same idea as the distance perpendicular to the surface normal between two parts.
Car panel edges are rounded, so the number depends on where the edge is picked and in which direction you measure. LMI’s Gocator measurement manual, for example, lets the user choose whether gap is measured along the fitted reference surface, perpendicular to the reference edge, or as the straight-line distance between the two edge points. Two tools set up differently can report different gaps on the same seam.
Flush
Flush (also called step or mismatch) is how level the two surfaces sit. Hexagon’s PC-DMIS documentation defines it as “the height difference between two mating sheet metal parts”. PC-DMIS reads flush at a set distance back from the gap’s edge, which it calls the indent, so the indent is part of how a flush value is defined.
Sign conventions
Gap is normally reported as a width. Flush can be positive or negative, depending on which panel sits higher. That means every flush value needs a reference side. In PC-DMIS, the sign depends on whether the second panel is higher or lower than the “Master” side, and the Master side is always to the left of the scan direction. LMI’s tool returns a signed flush by default, with an option to return an absolute value instead.
Conventions vary between OEMs and tools. The drawing or check sheet should state which panel is the reference (say, the fender on a door-to-fender seam) and whether “positive” means proud or sunk. If the spec, gauge and audit sheet disagree, the same car can pass on one and fail on another.
Parallelism
Parallelism describes whether the gap stays even along the length of a seam. For example, a seam could be 4 mm at the top and 5.5 mm at the bottom and still be “in tolerance” at each point, yet look tapered to a customer. Some OEMs specify it directly. A Navistar service bulletin on hood fit, for example, lists a “Parallelism (A to B)” limit and a left-versus-right gap difference limit alongside the gap and flush ranges. A Chalmers thesis written with Volvo Cars notes that the visual quality impression is lowered by a wide gap and non-parallelism of the split line.
Why gap and flush matter
- Perceived quality. Nikon calls panel fit “one of the first things that a customer notices even before getting in the car”. A 2018 study on SUV body gaps found gap behaves as a “the smaller the better” quality characteristic.
- Sealing and water. Keyence points out that water entering through a door or hood can damage electrical and interior components.
- Wind noise and leaks. Cognex links accurate panel alignment to avoiding wind noise and leaks. Nikon adds squeaks and excessive wear from poor door alignment.
- Function. A measuring-tunnel patent notes that a door has to have a specific gap within a very narrow margin so that it opens and closes properly.
- Customer audits and surveys. Fit problems surface in your own PDI and customer product audit, and later in owner surveys. J.D. Power’s India Initial Quality Study counts problems in the first two to six months of ownership as PP100 (problems per 100 vehicles), and “vehicle exterior” is one of its eight problem categories.
Where on a car gap and flush are measured
Checks cover the visible split lines between closures and the body, plus lamps. Hexagon lists doors, bonnet, boot and fenders. A Volkswagen Autoeuropa project describes checks such as the alignment of the tailgate and the correct assembly of the rear lights at several stages along the line.
| Zone | Typical seams | What makes it tricky |
|---|---|---|
| Hood (bonnet) | Hood to fender (left and right), hood to headlamp, hood to bumper | Long seam; left-right balance is very visible from the front |
| Front lamps | Headlamp to fender, hood and bumper | Transparent lens; gaps around headlights and taillights are harder to measure |
| Side doors | Front door to fender, front door to rear door, rear door to quarter panel, door to roof | Flush changes along the seam; door sag shows up as poor parallelism |
| Rear | Tailgate or boot lid to quarter panels and roof, tail lamps | Many mating parts meet in a small area |
| Small closures | Fuel lid (or charging port lid) to quarter panel | Short, curved seam; small panel easily set proud or sunk |
Three ways to measure gap and flush
1. Feeler gauges, taper gauges and manual audit
The manual method uses hand tools. A 1981 Candid Logic patent describes inspectors using a linear scale, shim gauges or cylindrical plug gauges, taking the largest gauge that fits as the gap. A taper gauge works by sliding the tapered blade into the gap and reading the graduation where it stops. Keyence notes that feeler gauges are used for gaps and calipers for flush.
Manual tools are simple and familiar. Their limits:
- They can only be applied at certain selected points along the gap.
- Contact gauges often scratch vehicles, and checks tend to be done by batch rather than on every car.
- Results depend on the operator. Hyundai’s gap-measurement patent notes that inspection “through a gauge or only with naked eyes” relies on the operator’s method.
- Mechanical tools struggle on complex surfaces and transparent parts.
2. Handheld laser gap-and-flush gauges
Handheld laser gauges project a laser line across the seam and read its profile by triangulation. The operator holds the device on the seam and gets gap and flush in one shot, with the reading stored against the point. Examples include:
- Third Dimension GapGun, which uses laser triangulation, measures gap and flush at the same time and logs the data automatically.
- Hexagon CALIPRI C15, a wireless handheld system using laser light section technology for inline and near-line flush and gap inspection, where deviations from tolerance are highlighted in colour.
- LMI Corporation LaserGauge, whose datasheet quotes repeatability in the 0.20 mm range, a reading in under three seconds and a day or two of operator training.
Handheld gauges suit audit stations, rework checks and new-model launches, and need no fixed installation. They still depend on someone walking the car, so they are usually a sample check rather than 100% coverage.
3. Inline and automated 3D systems
Automated systems measure every car without an operator. Two layouts are common.
Robot-mounted sensors. A 3D laser sensor such as an LMI Gocator is mounted on an industrial robot and programmed to move to each measurement point. At Stellantis’s Brampton plant, a Cognex system with four laser profilers was placed directly after door assembly and before painting. Robots give flexibility, but Nikon lists the trade-offs at final assembly: short sensor stand-off near a painted car, reach limits at the front, rear and roof line, conveyor tracking and safety.
Drive-through tunnels and arches. Here sensors sit on a fixed or rail-guided frame and the car moves through on the conveyor. Nikon’s APDIS Gap & Flush uses three Laser Radars on linear rails, up to 170 measurement locations, conveyor speeds up to 140 mm/s and a stand-off greater than 0.5 m. Camera-based tunnels are another route; one patented design uses video cameras, LED lights and a position encoder on the conveyor to rebuild seam edges in 3D.
Comparing the three approaches
| Manual gauges | Handheld laser gauge | Inline / automated 3D | |
|---|---|---|---|
| Coverage | Sample cars, selected points | Sample cars, many points per car | Every car, fixed set of points |
| Contact with paint | Yes | Device rests on or near the panel | No contact |
| Data | Written on a check sheet | Stored digitally per point | Stored per vehicle, linked to line data |
| Operator dependence | High | Lower, but still hand-placed | Low once programmed |
| Flexibility for new models | High | High | Needs programming and set-up per model |
| Best fit | Quick checks, low volume | Audit, launch, rework, supplier checks | 100% inspection and trend tracking |
Where DeepVision Tunnel fits
DeepVision Tunnel is Indus Vision’s drive-through option for final assembly. Structured-light 3D cameras on a bolted aluminium arch read each panel seam as the vehicle moves on the conveyor, with no stop and no fixture. DeepMeasure software computes gap, flush and parallelism along the seam and files the results against the vehicle ID. Seams covered include hood, headlamp, door to fender, door to door, rear door, fuel lid and tailgate. It runs on-premise, so data stays in the plant, and exports to quality and MES systems. The first measuring station was trialled in August 2026 on a door seam with the car moving at 32 mm/s; trial values are not a guaranteed specification. For how limits are set seam by seam, see our guide to gap and flush tolerances, or read more about our automotive inspection work.
Frequently asked questions
What is the difference between gap and flush?
Gap is the width of the opening across the seam. Flush is the height difference between the two surfaces. A seam can have a perfect gap and still fail on flush.
Is flush positive or negative?
It can be either. The sign depends on which panel is taken as the reference and how “proud” or “sunk” is defined. Tools such as PC-DMIS tie the sign to a master side. Always confirm the convention on the drawing and the gauge before comparing results.
What is parallelism in gap and flush?
Parallelism is how much the gap changes along one seam, for example between the top and bottom of a door. Some OEM specifications give it a separate limit, along with a limit on left-to-right difference.
Can a feeler gauge measure flush?
Not directly. Feeler and taper gauges measure gap width. Flush needs a height reference across the seam, such as calipers or a laser profile.
Do we need inline measurement if we already audit with handheld gauges?
It depends on what you need. Handheld audits suit sampling, launch and rework. Inline systems measure every car and build a trend per seam, which helps catch drift before it reaches the customer product audit. The two can work side by side.