Squareness Problems in Case Erecting: Root Causes and How Servo Drives Correct Them
Squareness is the property that decides whether an erected case runs cleanly through the rest of the line or causes trouble at every station it touches. It is measured as the difference between the two diagonals of the erected case, and the common acceptance threshold in general packaging is under 2 mm.
The trouble is that squareness failures are usually blamed on the machine when the cause is upstream — in the board, in the storage, or in the blank itself. This article separates the causes that a servo drive can correct from the causes it cannot, because buying a servo machine to fix a board problem is an expensive way to remain unsquare.
The Short Answer
A servo-driven erector holds position repeatability of plus or minus 0.3 to 0.5 mm against plus or minus 2 to 3 mm for a pneumatic machine. That difference corrects the squareness problems that originate in positioning — fold rail position, mandrel centring, compression plate alignment — and it does nothing for problems that originate in the blank.
The diagnostic that separates the two categories is simple. Erect ten cases from the same pallet and measure all ten. If the deviation is consistent in direction and magnitude, the cause is positioning and a servo machine will correct it. If the deviation varies case to case, the cause is board variation and no amount of servo precision will help.
What Squareness Actually Measures
Two diagonals of the erected case are measured, from the bottom outside corner to the top outside corner on each side, and the difference is the squareness error. On a 400 mm case, a 2 mm difference represents an angle error of about 0.3 degrees.
That small angle has disproportionate consequences. A case that is 2 mm out of square is 2 mm out of position at the taper, 2 mm out at the labeller, and 2 mm out at the palletiser. Each station then applies its own tolerance, and the errors accumulate rather than cancel.
The measurement should be taken with the case empty and standing on a flat surface, using a steel tape and a fixed reference. Measuring on a conveyor introduces the conveyor’s own variation.
Cause 1: Fold Rail Position
The most common positioning cause. Fold rails determine the angle at which the flaps bend, and if a rail is set 1 mm out over a 400 mm length, the resulting case is visibly out of square.
How to identify: measure the rail gap at both ends and the middle against the drawing. A rail that is parallel but displaced produces a consistent error in one direction.
Can a servo drive fix it? Only if the rail position is a servo axis with a stored position and the stored position is wrong. Otherwise it is a mechanical adjustment, and the fix is a setting sheet with the correct value rather than a machine change.
Cause 2: Mandrel Centring and Wear
The mandrel is the component that squares the case, and it is also the component that wears. Mandrel corner radii round off with use, and a rounded corner allows the case to rotate slightly as the flaps close.
How to identify: inspect the mandrel corners for visible radius. Compare a new mandrel against a used one if available. A mandrel with more than about 0.5 mm of corner radius should be reconditioned or replaced.
Can a servo drive fix it? No. Mandrel wear is geometry, not position. Reconditioning or replacing the mandrel is the fix.
Cause 3: Uneven Compression
If the compression plates or belts are not parallel, the case is pressed asymmetrically and the glue bond sets in a skewed position. This is most often seen as a case that measures square before compression and unsquare after.
How to identify: measure plate parallelism with a feeler gauge at four points. A compression section that is more than 0.3 mm out of parallel across its width will produce a measurable squareness error on a 400 mm case.
Can a servo drive fix it? Yes, on a machine where the compression height is a servo axis, because the axis stores an absolute position and can be recalibrated. On a pneumatic machine the parallelism is set mechanically and drifts with wear.
Cause 4: Suction Cup Position and Pick Square
If the blank is picked slightly off-square, every subsequent operation preserves the error. A pick that is 3 mm off centre on a 400 mm blank translates to a squareness error of similar magnitude.
How to identify: watch the blank as it is picked. If it arrives at the mandrel visibly rotated, the pick is the cause.
Can a servo drive fix it? Partly. A servo pick axis can be repositioned precisely and repeatably, which removes variation. It cannot correct a cup layout that grips the blank asymmetrically — that is a mechanical design question.
Cause 5: Board Warp
Warp is the curvature of a blank, and a warped blank that is forced flat by the fold mechanism retains internal stress that shows up as a non-square case when the case is released.
How to identify: measure the blank’s deviation from flat over its width. Beyond 3 mm, warp is a contributor to squareness error, not just to pick failures.
Can a servo drive fix it? No, and this is the cause most often misdiagnosed as a machine fault. The fix is storage — pallets off the floor, stack heights within the supplier’s recommendation, and protection from one-sided moisture gain. Storage criteria for board are set out in our support section.
Cause 6: Board Moisture Outside the Band
Board below 6 per cent moisture is brittle and cracks at the score, and the score line then becomes the fold line, which may not be where the case geometry intended. Board above 9 per cent is soft, folds where the stress takes it rather than where the score directs it, and the case is unsquare as a result.
How to identify: a moisture meter on a sample from each pallet. This is a two-minute check that answers a question which otherwise takes a shift to diagnose.
Can a servo drive fix it? No. Humidity control and storage discipline are the fixes.
Cause 7: Score Quality from the Blank Supplier
Score depth and width determine where the board folds. A blank with inconsistent scores folds inconsistently, and the variation appears as squareness variation between cases.
How to identify: erect ten cases and measure. Variation between cases from the same pallet points to score quality or calliper variation rather than machine positioning.
Can a servo drive fix it? No. The answer is an incoming inspection criterion and a conversation with the blank supplier. This is the point at which many plants discover that the machine they were about to replace was not the problem.
What Servo Drives Actually Change
The servo advantage is not “better accuracy” in the abstract. It is three specific capabilities.
Absolute positioning. A servo axis moves to a stored position rather than to a relative offset, so accuracy does not drift with accumulated adjustments over a shift. A pneumatic machine set by hand drifts as seals wear and fittings loosen.
Repeatability under load. A servo holds position against varying load; a pneumatic cylinder’s position depends on the resistance it meets. On thick board, a pneumatic fold axis arrives at a slightly different position than on thin board, and the difference shows up as squareness variation between grades. This is the mechanism behind the mandrel design used across the Single-Piece Carton Erector range.
Electronic correction between formats. When a new format is selected, the servo returns to that format’s validated geometry rather than to a mechanical approximation. This is what makes sub-five-minute changeover compatible with consistent squareness.
Reference Values
| Parameter | Pneumatic machine | Servo-driven machine |
|---|---|---|
| Position repeatability | Plus or minus 2 – 3 mm | Plus or minus 0.3 – 0.5 mm |
| Squareness on a 400 mm case | 1.5 – 3 mm | Under 1 mm |
| Drift over a shift | Present, accumulates | Negligible |
| Dependence on board grade | Position varies with resistance | Position independent of resistance |
| Correction between formats | Manual re-adjustment | Stored validated geometry |
| Mandrel wear effect | Not compensated | Not compensated |
The last row matters. Servo drives do not compensate for mechanical wear. A servo machine with a worn mandrel is still an unsquare machine, and it will hold its error with great precision.
Alarms Related to Squareness
| Alarm | Meaning | First three checks |
|---|---|---|
E-33 Servo Follower Error |
Axis deviation beyond tolerance | Belt tension, encoder coupling, mechanical obstruction |
E-41 Flap Detect Missing |
Bottom flap not detected in position | Sensor alignment, fold rail gap, board moisture |
E-12 Vacuum Low |
Pick-up vacuum below setpoint | Cup condition, filter, generator |
E-27 Glue Pressure Low |
Glue supply pressure below the working band | Tank level, filter screen, pump |
E-33 Servo Follower Error is worth interpreting carefully. On a servo machine it can indicate a mechanical problem — a belt losing tension or a coupling developing backlash — that is producing a squareness error the controller is trying to correct. When the alarm appears intermittently before any squareness complaint has been raised, it is the machine reporting the cause of a problem the operator has not yet noticed.
Maintenance Rhythm
Every shift — measure squareness on one case at the start of the shift, check that the current format’s settings match the setting sheet, and verify glue line position on a sample.
Every week — inspect fold rails for alignment and for glue or dust build-up, check the mandrel surface for wear or contamination, and verify compression plate parallelism.
Every month — measure squareness on five cases across the case size range, check servo belt tensions, and review whether any format has drifted outside its validated values.
Every quarter — inspect mandrel corners for radius, check cam followers and bearings for play, and re-verify the pick squareness by watching blank placement at the mandrel.
Annually — recondition or replace the mandrel, re-grease and re-tension all drive elements, re-baseline squareness across the format range, and re-verify the setting sheets against measured results.
FAQ
Will a servo machine fix my squareness problem?
Only if the cause is positioning. Erect ten cases from one pallet and measure them: consistent error means yes, varying error means the cause is board or blank quality.
What squareness should I specify?
Under 2 mm on a 400 mm case is a common acceptance threshold for general packaging. For applications where the case must interface with automated handling, specify under 1.5 mm.
How often should the mandrel be replaced?
Inspect quarterly and recondition when corner radius exceeds about 0.5 mm. On a machine running three shifts, a mandrel may need reconditioning on a two to three year cycle.
Can squareness be corrected downstream?
Not reliably. A taper or a labeller cannot pull a case square; it can only work with the case it is given. Squareness is set at the erector and preserved or lost thereafter. Where the format is a multi-piece construction, the forming sequence introduces additional geometry to control — see the Three-Piece Carton Erector for how that is handled, and the FAQ for the common measurement questions.
Dealing with squareness variation? Send the case dimensions, board grade and moisture readings, plus ten measured case diagonals from the same pallet, to [email protected] or WhatsApp +86 13681839278. Our engineers will identify whether the cause is positioning or board, and specify the correction, within 24 hours.
Want To Know More About Our Upgraded Machines?
Send your carton specifications and output demand, our engineer will provide free technical scheme & quotation
Subscribe To Get Latest News & Technical Guides
Leave your business email to receive free industry articles and new product updates