The challenge
Ready-made frozen hamburger sandwiches are a high volume category, and the packaging line behind them runs fast. Our customer needed to pick six individually wrapped sandwiches at a time and place them into shipping boxes.
The difficulty was not the weight. It was everything a bun does when you handle it badly. A soft bakery product deforms under point pressure, and a deformed bun in a retail box is a rejected unit. The tool had to pick six at once, hold them securely enough to move at line speed, and put them down without compressing a single one.
At the same time the tool needed to stay light. Six pick positions across a wide footprint is a lot of structure, and structure is mass. Every kilogram of tool mass is a kilogram the robot cannot spend on product, and it is also a kilogram the arm has to accelerate and decelerate on every cycle.
Our solution
Our engineering team applied topography optimization to design an organically shaped gripper, then produced it in Nylon PA12 by selective laser sintering on an EOS P396.
The shape is the result, not the starting point
Rather than laying out a frame and thinning it down, we mapped the load paths from the robot flange out to each of the six pick positions and let the structure grow along them. What comes out is the branching geometry visible in the photographs. Every rib carries load. Nothing is there for manufacturing convenience, because with additive manufacturing there is no manufacturing convenience to design around.
Gentle handling is designed in, not tuned in afterwards
Cup selection and the applied vacuum level were matched to the wrapped sandwich rather than to a generic payload figure. The grip is distributed across six positions so no single point carries enough force to mark the product.
Less material means leaner cost and a faster cycle
The material reduction achieved through topography optimization does three things at once. It lowers the printed cost of the tool, it lowers the inertia the arm has to overcome on every move, and it leaves more of the rated payload available for product.
Full specifications
| Material | Nylon PA12 |
|---|---|
| Machine | EOS P396 |
| Layer thickness | 120 microns |
| Build time | 18 hours |
| Post processing | Blue dye vapour polishing |
| Number of components in the assembly | 5 |
| Tool weight | 2.1 kg |
| Units per cycle | 6 individually wrapped sandwiches |
| Gripping method | Vacuum |
Results
Six units move per cycle with no compression or deformation of the bun.
Removing material that carries no load removes printed cost from the part directly.
Lower tool mass means lower inertia, so the arm moves faster within the same duty cycle and uses less energy doing it.
Frequently asked questions
How do you pick a soft bakery product without crushing it?
By distributing the load and matching the vacuum to the product rather than to a generic payload target. Spreading the pick across six positions means each position carries roughly a sixth of the load, which keeps contact pressure below the threshold where a wrapped bun starts to deform. Cup profile matters as much as vacuum level.
What is topography optimization in end of arm tooling?
It is a design method where the load paths through a part are calculated first, and material is then placed only along those paths. The result usually looks organic or skeletal, because that is what an efficient structure looks like when nothing forces it into flat plates and right angles. It is only practical when the part is 3D printed, since a machined version of that geometry would be prohibitively expensive.
Why print a gripper instead of machining it?
Three reasons dominate in bakery and food packaging. Weight, because printed tooling can be 40 to 60 percent lighter for the same footprint. Part consolidation, because a printed assembly of five components replaces a bolted assembly of twenty. And revision cost, because changing the pick pattern for a new product size is a design change rather than a retooling bill.
How many units can one gripper handle per cycle?
This tool handles six. The number is set by the robot payload, the product footprint, and the box pattern it is filling. We have built tools with a single pick position and tools with more than twenty.
Can the same tool handle a different bun size?
Usually yes, with a design revision to the pick pitch and cup selection. Because there is no hard tooling involved, that revision costs design time and a new print, not a new mould or fixture.
Have a similar application?
Send us the part, the payload, and the cycle time. Our engineering team will scope the tool with you and tell you honestly whether additive manufacturing is the right answer for it.