The tooling ran clean for eight months. Then the cell got sped up to pull two seconds out of the cycle, and now parts land on the floor instead of the conveyor. Nothing on the tooling changed. That’s the tell.
A vacuum end effector, meaning the cups, mounts, and vacuum circuit that lift a part off the core, holds that part with a force that depends directly on how fast you move it. Speed up the robot and you raise the force requirement on tooling that was sized correctly at the old speed. Most molders have lived this. Almost none have seen it explained.
So here’s how to tell whether vacuum will hold your part.
Will vacuum hold this part at all?
Answer that before anyone starts running numbers. Vacuum is the cheapest gripping method available, it runs a faster cycle than a mechanical grab, and on most molded plastic it puts less blemish and scratch risk on the part than steel jaws do. On a smooth non-porous face it’s hard to beat. It’s the default on molding cells, and it deserves to be.
It also fails in ways predictable enough to rule out early. Wet or oily faces drop the friction coefficient to somewhere between 0.1 and 0.3 in Schmalz’s published vacuum engineering figures, which guts everything downstream of it. Parts without enough flat contact area for a cup to seal against are out, and so are parts heavy enough that the cup count needed stops fitting on the face you have to work with. One rule specific to molding is worth more than the rest combined: per Plastics Technology, a part that sticks in the mold, even occasionally, normally calls for a pneumatic gripper instead. Occasional sticking is the whole problem, because vacuum tooling sized for a clean release has nothing left when the part fights back. That’s where mechanical grippers stop being the fallback and start being the right answer.
Now the interesting half. One condition that looks disqualifying isn’t.
Porous parts still run on vacuum. They need high flow rather than deeper vacuum, which means a high-flow circuit with the generator mounted on the tooling, or a foam-faced area gripper that seals wherever material happens to be. Specifying a stronger pump is the common mistake. It doesn’t fix a leak-rate problem.
Here, less is more. Every additional cup you hang on the same vacuum circuit adds to the total leak, and the holding force actually gets weaker. Adding cups to a leaky circuit is the instinct, and it’s the wrong move.
Sizing the circuit, not just the cups
Then there’s air. Work out the flow each cup needs on your part’s surface, multiply by cup count, and check that against what your supply actually delivers.
Undersized lines and fittings are a quiet failure mode, where the cups are right, the generator is right, and the circuit still can’t evacuate fast enough to grab at cycle speed. A rule of thumb worth holding onto: no more than six to eight cups on a single vacuum circuit, depending on cup size. Past that, split the circuit. A vacuum check valve such as the STV shuts off supply to a cup that never made contact, so one unseated cup can’t bleed the rest of the plate.
One more thing, plainly. Theoretical holding force is not achieved holding force. Vacuum level, cup wear, seal quality, and leakage all take a bite out of it. Size the plate with margin for that gap instead of to the number on paper.
Marking, heat, and cup material
Published sources disagree about whether vacuum marks parts, and both sides are describing different parts.
Vacuum cups can absolutely mark a high-gloss lens cover or a piano-black trim bezel. On those, a faint cup ring shows up under inspection lighting and the part goes in the regrind bin. Try the cheap fix first: wipe the cup face with rubbing alcohol. A cup carries residue left over from molding the cup itself, and cleaning that off sometimes gets rid of the ring on the part. When it doesn’t, HE compound is the next move, because >ASS< HE vacuum cups are a high-temperature elastomer with good non-marking characteristics.
On a textured interior trim panel or a structural bracket, vacuum is the gentler option, because the load spreads across the whole cup face instead of concentrating at two jaw contact points. The substrate decides.
Cup material carries the rest, and heat is usually what forces the choice. Nitrile runs to 250 °F (120 °C), silicone to 392 °F (200 °C), and Viton to 428 °F (220 °C). Parts coming off the core while they’re still hot are what push you up that ladder, and silicone covers most of it while tolerating rougher surfaces than harder compounds do.
Silicone carries one disqualifier that catches people late, though. It’s not suitable for parts that get painted or coated afterward, and plenty of plants ban silicone cups from the building outright for exactly that reason. Check your plant’s rules before you spec it, not after the cups show up.
Get a cup count for your part
Send us the part and we’ll tell you whether vacuum holds it. Mass, orientation on the core, surface finish, and target cycle time are enough to get a real answer and a cup count, backed by 40-plus years of application experience from the parent company in Germany. Over 2,000 standardized components cover most of what a plate needs, and the pieces that don’t exist yet get printed in-house. EOAT design help starts in Plymouth, Michigan, where a person actually answers the phone.