Shell & hull fragments
Shell, husk and hull pieces close in color to the kernel, removed on contrast and shape.
Food processing
MAS N Series Nuts Sorter
Nuts are fragile and irregular, so defect visibility depends on orientation and surface exposure — and handling has to limit breakage. This page defines what to inspect; final machine configuration and yields come from a real sample test.
Shell & defect removalGentle handlingMulti-view
Material types
Each nut stream is reviewed on its own.
Kernel size, breakage risk, shell content and surface defects differ by product. Click a type to see what drives the grade.
Almonds
Skin & shapeWhole kernels where shell fragments, doubles, chips and dark or shrivelled pieces are the typical reject classes.
Visible defects
What the sample test should classify.
Fragile product means each class is confirmed on real kernels, with handling tuned to limit breakage.
Broken & split kernels
Halves, chips and splits separated when whole-kernel product is the target grade.
Dark spots & discoloration
Scorched, stained or spotted kernels graded out for appearance and food quality.
Shrivelled & undersize
Immature, shrivelled or undersize kernels outside the accepted size and fill window.
Stones & foreign matter
Stones, glass, metal and plant debris defined as reject during the sample test.
Mould & insect damage
Mouldy, rancid-looking or insect-damaged kernels flagged for removal.
Accept / reject
From mixed kernels to a clean grade.
What stays in the accepted kernel stream and what is gently ejected — confirmed on representative samples.
FeedMixed kernel feed
Incoming product carries good kernels plus shell, broken, dark and foreign pieces.
Illustrative only. Real grading depends on kernel size, breakage, surface exposure, orientation, presentation and the confirmed recipe.
Why it needs testing
Fragile, irregular and orientation-dependent.
Nut grading is built from evidence because handling and presentation strongly affect the result.
Breakage risk
Feeding and ejection must be gentle enough to protect whole kernels while still separating defects.
Orientation-dependent defects
A defect on one face can be hidden on another, so multi-view exposure is often needed to see it.
No numbers without a test
Yield and purity depend on product, sample condition and target defects, so they come from a real machine test.
How it works
From gentle feed to graded reject.
A sample-led loop tuned to fragile product and the accept / reject classes you confirm in testing.
- 01Customer-prepared
Gentle feeding review
Kernels are metered and spread with handling tuned to limit breakage before inspection.
- 02Class definition
Confirm visible defects
Accepted color and size plus shell, broken, dark and foreign reject classes are set from your samples.
- 03Multi-view inspection
Camera exposure planning
Lighting and multi-view cameras are arranged so orientation-dependent defects are actually seen.
- 04Air ejection
Reject stream review
Defects are ejected gently and the reject stream is checked so good kernels are not lost.
- 05Handoff
Operator handoff
Recipe, accepted grade and reject review are handed over once the result is confirmed.
Inspection path
Technologies behind the grade.
Nut sorting combines visible-light inspection with AI-assisted recognition and multi-view inspection for orientation-dependent defects.
Real sample testing
Confirm your nut grade with real kernels.
A nut sorting project should start from your real product, including whole kernels and the shell, broken, dark and foreign pieces that must be removed.
Request a Real Material TestOnline preview
Preview your nuts stream online first.
Upload sample photos for a visual accept / reject preview before sending real material. Real sorting performance still needs a machine test.
Start AI Visual Preview