Dual-head laser thickness measurement for moving sheets

Close the geometry before choosing a resolution.

T = C − (A + B)

T is total thickness. C is the calibrated separation between the two distance reference planes. A and B are positive distances measured from each head toward its facing surface along a common axis. Relative or sign-inverted outputs must first be transformed to these coordinates. Mounting dimensions alone do not establish C.

In the ideal aligned arrangement, rigid sheet translation increases one distance and decreases the other by the same amount. Actual performance also depends on frame expansion, sheet tilt, lateral spot mismatch, sensor linearity and optical response. A digital output increment is not the uncertainty of the thickness result.

For parallel faces tilted by θ to the perpendicular of the measurement axis, the intercepted thickness is Tₙ / cos θ, where Tₙ is thickness normal to the sheet. Even aligned spots need an agreed tilt limit. Triangulation cannot see through opaque electrode coatings; separating individual layers needs process-reference measurements or another sensing method, with supply scope confirmed separately.

Match acquisition timing to web motion.

Ask how each head is triggered and how readings are paired. A PLC polling two values in sequence does not establish simultaneous acquisition. Record timestamps, invalid-value handling, averaging windows and the behavior when one head loses the surface.

For locally constant normal sheet velocity, timing mismatch produces an approximate error magnitude |vₙ × Δt|. An illustrative 0.1 m/s normal motion and 100 μs mismatch correspond to 10 μm. This is geometry arithmetic, not a measured LDS result; web travel speed along the sheet is a different quantity.

Choose a usable rate after exposure and filtering are fixed. For a moving web, along-track sample spacing is line speed divided by effective paired sample rate. A single track is not a full-width thickness map: traversing or multiple tracks require a separate position and synchronization plan.

Qualify the frame, surfaces and delivered data together.

Begin with LDS-PRO for demanding dynamic measurements and review LDS-TRI where the required tolerance and speed permit. Neither family name qualifies the complete gauge. Keep the expected sheet excursion inside both heads’ usable ranges, including thickness extremes and mounting tolerances.

Electrode texture, wet coatings and metallic substrate exposure can produce different return signals. Request a trial on representative material at the intended line speed. Agree how dropouts, coating edges, seams and transitions are flagged instead of silently substituted by held values.