Mobile: 18688976300 (Mr. Peng) 15222276387 (Mr. Bai)
Email: pengwl@honred.com
QQ: 176496342
Address:Building A1, Runteng Industrial Park, Shiyan Street, Bao'an District, Shenzhen
Bottom line first: to keep screws from dropping, many removal stations magnetise the bit — a magnetic holder, a ring, a quick charger, it all counts. Above M2 this is largely harmless, but in the M0.8-M1.6 band and on stations near Hall sensors, magnetic encoders, MEMS magnetometers or vibration motors it produces three effects at once: iron debris attracted where it does not belong, screws that cling to the bit instead of releasing, and residual-magnetisation offset written into sensitive parts with no traceability. Vacuum is not a natural substitute; each route has a clear boundary.
Magnetic holding relies on residual flux. No air line, no nozzle, nothing to clog, and it is extremely cheap — which is exactly why it gets over-used. But it has a defect vacuum does not: it cannot release actively. Vacuum drops a screw in a millisecond by venting; a magnet can only be stripped mechanically, and when the screw leaves the bit depends on how well it happens to stick.
On a removal process that difference propagates into three downstream metrics: removal completeness, screw count, and foreign-object risk at the next assembly step. A screw that stays on the bit and never reaches the capture box is one missing from the count, and the missing one may still be inside the station, ready to score or crush the next workpiece. In reports this shows up as "occasional low completeness" with no clear cause.
There are four sources. The first is a magnetic bit holder: a permanent ring slipped over the bit, typically 20 to 80 millitesla at the surface, higher with neodymium. The second is a magnetiser, where an operator pushes a whole box of bits through for convenience, leaving an uneven distribution with the strongest field at the tip. The third is the magnetic selection inside a screw feeder, which magnetises screws during feeding and is the most overlooked because it is not equipment at the station. The fourth is the driver itself, where brushless-motor leakage gives a measurable reading close up, though usually far below the first three.
It helps to have an intuition for scale: a dipole field falls off close to the cube of distance, so doubling the distance cuts it to about one eighth. A 50 millitesla field at 1 mm from the tip is under 1 millitesla at 5 mm. Harm therefore concentrates in the "very close" cases — the screw head, the threaded hole, and the moment a part is brought near a device surface.
On the device side, the operate point of many Hall switches sits in the low millitesla range, and MEMS magnetometer offset responds to even less. A normally magnetised bit can therefore produce a measurable offset in some sensors at 2 mm, a distance that nobody would call contact in manual assembly.
Debris is the obvious one. Plating flakes, thread-cutting residue and iron-bearing particles from worn tooling are attracted to the bit tip and the screw head, then dropped into the threaded hole or an optical cavity during removal. On cleanliness-critical stations this is fatal, and much of it is only visible under magnification, so visual inspection misses it.
Cling is the second, already described: the inability to release actively. It hurts automated stations more than manual ones, because automation depends on counting to decide an action completed, while a person can still see it.
The third is residual magnetisation of sensitive parts, and it is the most insidious because the symptom is not immediate failure but zero drift, output offset and lost calibration. By the time final test rejects a sensor reading, the trail leads back weeks to an assembly station whose bit magnetisation was never recorded. Autofocus voice-coil modules in camera assemblies, magnetometers and speakers in phones, and magnetic-encoder joints in robots all belong on this list.
One more point is routinely missed: magnetic holding simply fails on non-magnetic materials. A2 and A4 stainless screws are only weakly magnetic after cold working, titanium screws are non-magnetic, and brass or plastic screws are out of the question. Stations reporting "screws drop occasionally" often trace it to a material batch change while the process card still says magnetic pickup. The fix belongs to the equipment, and vacuum-adsorption models of small screw removal equipment need no hardware change when the material changes, which is one reason the vacuum route is steadier at small sizes.
The boundary for magnetic holding is clear: iron-based screw, no magnetic parts nearby, M1.6 or larger, and acceptable for the screw to travel briefly with the bit. When all four hold, magnetism is the cheapest answer and there is no reason to switch to vacuum for the sake of sophistication.
Vacuum applies when controlled release is required, the screw is small, the material is non-magnetic or unknown, and magnetic parts are nearby. The cost is a stable negative pressure, nozzles that clog, and sensitivity to head geometry — screws with recessed or raised heads are hard to seal against. Negative pressure is typically 40 to 60 kilopascal below ambient, and the nozzle bore must be smaller than the head diameter or the screw will not be held.
Mechanical grippers suit odd head shapes, very large or very small sizes, and cases needing high lateral force. Clamping and release are both controllable, but response is slow, alignment demands are high, and the jaws wear and generate particles. These three routes are not substitutes; most plants run all of them and assign by station.
The real difficulty is mixed material. A line that removes iron-based screws today and titanium tomorrow cannot rely on an operator to identify the material. What an automatic vision screw removal machine contributes is classifying screws by outline and head style first, then selecting the matching pickup parameters, instead of asking people to remember "this batch is titanium". Recognising a head style is not hard; mapping the result directly to pickup method and torque recipe, in a closed loop, is.
First, no magnetic pickup on magnetically sensitive stations. This should be a prohibition, not a recommendation. Any station involving camera modules, magnetic encoders, magnetometers, speakers or vibration motors goes to vacuum or mechanical holding.
Second, a residual-flux threshold and an inspection interval. Measure with a gauss meter at 1 mm from the tip: within 3 millitesla for sensitive stations and 15 millitesla elsewhere; full inspection on incoming bits, weekly checks on bits in use with records kept. Measure the tip, not the shank — the distribution along the shank is very uneven.
Third, a material-to-pickup mapping table. Iron-based goes to magnet or vacuum, austenitic stainless and titanium to vacuum or mechanical, brass and plastic to vacuum. The table must follow the BOM and be updated when material changes, not live only in the machine manual.
Fourth, a release confirmation method. "It looked like it fell" is not a method. Vacuum uses a pressure sensor to watch recovery after venting, magnetic holding uses a through-beam fibre sensor or a drop detector, and the most robust is capture-box weighing plus per-screw counting. At least one of the three, with output into the data record.
Fifth, a rule for demagnetisers. Once a bit is magnetised, the correct treatment is an AC decaying demagnetiser, not "rubbing it backwards on the holder" — which only reverses the direction and keeps the magnitude. Thermal demagnetisation is not an option: the Curie temperature of steel is far above the tempering range of a bit, and by the time it is hot enough to demagnetise, the hardness is gone. An M0.8-M2.0 screw removal solution that ships pickup method, residual-flux threshold and release confirmation together in the process card avoids the familiar sequence of buying the machine, never fixing the parameters, and leaving the line to work by feel.
Related Articles
/ RECOMMENDED NEWS