Honeycomb Paperboard Thickness Tolerance: Why Nominal Is Not a Specification
Thickness questions arrive when a panel that measures correctly on the drawing does not fit the frame it was designed for, or when a stack of panels measures differently at the edges than at the centre. Honeycomb board is not a machined product, and its thickness varies with the liner, the core and how the panel was pressed. The answers below cover where the variation comes from and how to bound it.
What determines the thickness of a honeycomb panel?
Three inputs: the core's expanded height, the liner thickness, and the adhesive layer. The core dominates but is the least precise, because expansion height depends on moisture, on the paper's recovery, and on how evenly the core was drawn open. A nominal 20 mm panel can legitimately arrive at 19.2 mm or 21 mm depending on those factors, and the variation is not uniform across a sheet.
Where is the variation largest?
At the edges, and at the corners in particular. The core is less constrained there during expansion and during pressing, so it can expand slightly more or collapse slightly more than the interior. A panel that measures 20 mm in the middle can measure 19 mm within 50 mm of the edge. Since frames and slots are built to the edge dimension, edge variation is what causes fit problems, and a centre measurement is the least useful place to take one.
ISO 534 Paper and board — Determination of thickness, density and specific volume.
How should the measurement be taken?
With a defined contact pressure and a defined dwell, at defined positions. A dead-weight dial gauge with a flat foot is adequate; a caliper with knife edges is not, because it deforms the liner. The practical protocol is nine positions — three along each of two edges plus the centre — reported as a range rather than a mean. The range is the number a frame designer needs, because a frame that fits the mean will bind on the thick edge.
Does humidity change the thickness?
Yes, and it changes it in both directions over time. Dry board is thinner and stiffer, humid board is thicker and softer, and a panel that is pressed to nominal at one moisture content will not measure nominal at another. A specification should state the moisture content at which the thickness is measured, and the acceptance range should be wide enough to cover the service moisture range rather than the measurement room only.
What tolerance is realistic?
For a panel in the 15 to 25 mm class, a range of plus or minus 0.8 mm at the measured positions is achievable for a controlled process, and plus or minus 1.2 mm is realistic for a general-purpose product. Tighter than 0.5 mm requires either a pressing step after expansion or a grading operation, both of which add cost. A buyer asking for plus or minus 0.3 mm is asking for a different product, not a better one.
Standard Test Method for Thickness of Solid Electrical Insulation.
How does thickness affect the other properties?
It scales them. Bending stiffness rises with the cube of thickness, so a 5 percent thickness gain produces roughly 15 percent more stiffness at the same weight per area. That is why a panel at the top of its tolerance range performs noticeably better than one at the bottom, and why a buyer who accepts the full range is accepting a performance spread as well as a dimensional one. If the structural calculation assumed nominal, the calculation should be re-run at the low end of the range.
What should the purchase order say?
State the nominal thickness, the permitted range at defined edge positions, the measurement method with contact pressure, the moisture content at measurement, and the structural assumption used in design. Add a rule for what happens at the low end: either the panel is rejected, or the design must be checked against the low-limit thickness. The second option is usually cheaper, but it has to be a decision rather than an omission.