English
Views: 0 Author: Vinci Zhang Publish Time: 2026-11-10 Origin: Site
A PCB handling machine rarely works alone. It sits at the very entrance and the very exit of an SMT line, and the moment it has to hand a board to a printer or receive one back from an unloader, the conversation between machines is no longer mechanical, it is electrical and logical. That conversation is governed by SMEMA, and when two machines do not speak the same dialect of it, you do not get a crash you can see, you get a line that stops for no obvious reason every few hundred boards. This guide is a working checklist you can take to a vendor or a site survey so the loader and unloader you buy actually shake hands with the equipment on either side instead of fighting it.
The loader and unloader are the two points on an SMT line where a board leaves human control and re-enters it, and both points are where the upstream and downstream machines must agree on whether a board is ready to move. SMEMA is the de facto language for that agreement, and ignoring it is the fastest way to turn a new cell into a troubleshooting project.
SMEMA, the interface specification originally published by the Surface Mount Equipment Manufacturers Association and now maintained under the IPC umbrella, defines a small set of dry-contact signals carried on a standard connector so that any two machines from any two vendors can agree on when to send a board and when to accept one. It is deliberately minimal: it does not describe recipes, it does not describe widths, and it does not describe speeds, it only describes the yes-or-no signals that keep a board from being pushed into a machine that is not ready. That minimalism is exactly why it has survived for decades while every proprietary bus around it came and went, because a loader only needs to know two things about its neighbor, and SMEMA gives it those two things reliably. The connector pinout, the signal names, and the basic handshake timing are what you verify first, and everything else in this checklist builds on the assumption that those basics are already correct.
When the SMEMA handshake is slightly wrong, the line does not fail loudly, it fails intermittently, which is worse, because an operator learns to nudge a board or toggle a mode and the stop disappears for an hour, so the root cause is never found. A missing board-request signal means the loader keeps boards until the upstream machine times out, a stuck stop signal means the downstream machine never asks, and a polarity mismatch means the signal is there but inverted so the logic reads the opposite of reality. These are all SMEMA problems, and they all present as the same symptom: the line pauses, someone walks over, the board moves, and the next one stalls again. The checklist in this guide exists to catch those mismatches on paper, before commissioning, so that the first boards through the line are not also the first field test of whether two vendors interpreted the standard the same way.
The loader is the first transmitter and the printer is the first receiver, the unloader is the last receiver and the upstream inspection or buffer is the last transmitter, and the SMEMA chain runs continuously through every machine in between, so a single broken signal anywhere breaks the whole line. When you verify compatibility you are not verifying two pairs of machines, you are verifying one continuous daisy chain from the loader all the way to the unloader, because a board that starts at the loader must be allowed to flow, stop, and resume at every station without any station contradicting its neighbor. This is why a loader that works perfectly on a bench can still stall a real line: the bench test only checks the loader, while the line checks the loader plus every signal it must agree with, and the unloader at the far end is the final judge of whether the chain ever let a board through cleanly.
Before any signal is checked, the two machines have to physically agree on where the board travels, because no amount of correct SMEMA signaling helps if the board would fall off the edge or jam at a height step. Height and width are the mechanical foundation of compatibility, and they are the first items on any serious pre-installation survey.
Most SMT lines adopt a pass-line height in the 900 to 1000 millimeter band measured from the floor to the top of the conveyor rail, and a loader or unloader that lands even twenty millimeters off that band will force a board to tip or a frame to need shims that should never have been necessary. You confirm the number by reading the line specification, not by eyeballing the machine, because the printed value on the neighbor is the one your new equipment must match within a tight tolerance, and a site that mixes brands often discovers the hard way that two machines claiming the same nominal height actually differ by a shim or a foot. The checklist item is simple to write but easy to skip: record the exact pass-line height of every machine the loader or unloader touches, and require the new unit to match it, because a height mismatch is the one compatibility error that no setting inside the PLC can fix after delivery.
A loader or unloader must cover the full board-width range your line actually runs, plus the boards you expect to add later, and it must return to the same width every time the auto-change fires, because a width that drifts by a few millimeters will eventually misregister a board at the neighbor even if the SMEMA signals are perfect. You verify range by listing your minimum and maximum board widths and checking them against the equipment specification, and you verify repeatability by asking the vendor how the width is sensed and locked, since a lead-screw with a hard stop behaves very differently over ten thousand cycles than a friction clamp does. The compatibility question is not only whether the machine can reach your width, it is whether it can reach it the same way every time the line changes over, because the downstream printer does not care that the loader was close, it only cares that the board arrived centered.
Every SMT line has a fixed edge, the side the board is referenced against as it travels, and a loader or unloader that references the opposite edge will deliver boards that are shifted by the full offset between the two references, which then has to be absorbed by every machine downstream. The checklist captures which edge is fixed on the neighboring equipment and requires the new unit to use the same one, and it also captures the direction of travel, because a board that enters the printer from the correct edge but the wrong end is still a board the printer has to reject or misplace. Products such as the I-C.T magazine loader and the vacuum and magazine combination loader are built to align to the line fixed edge, but the alignment still has to be specified at order time, because the hardware supports it and the default may not be your line default.
Once height and width agree, the next thing two machines must agree on is which way the board moves and which way is up, and these seem obvious until you put a machine from one regional convention next to a machine from another and discover they disagree on left-to-right.
The dominant convention in most SMT lines is left-to-right travel when viewed from the operator side, but many lines, especially those that wrap around a room or feed from a different department, run right-to-left or even reverse at one cell, and a loader or unloader shipped with the wrong assumption will deliver boards against the flow. The checklist asks you to draw the line as actually built, not as drawn in the brochure, and to mark the travel direction at the exact point where the new machine connects, because the direction at the connection point is the only direction that matters for the handshake. A wrong direction cannot be fixed with a setting if the mechanical infeed and outfeed are built for the opposite flow, so this is a confirm-at-order item, and it is one of the most common reasons a perfectly good loader arrives and simply cannot be placed where it was intended.
Closely related to direction is the fixed-edge side, front or back relative to the operator, and mixing these is the silent cause of boards that arrive rotated or offset by the rail gap at the neighbor. You record the fixed edge on the adjacent machine and require the new loader or unloader to match it, and you do this explicitly rather than assuming, because two machines from the same vendor can still ship with opposite fixed edges if the order forms were filled differently. The orientation error shows up later as intermittent misregistration at the printer or as boards that the unloader indexes one pitch off, and by then the SMEMA signals have been blamed for a problem that was mechanical from the start, which is exactly the confusion this checklist is meant to prevent.
Boards with tall components on the bottom, or fixtures and carriers, need clearance under the rail and over the top as they pass, and a loader or unloader whose transport height eats into that clearance will scrape or stall boards that every other machine on the line handles fine. The checklist records the maximum component height above and below the board for the products you run and requires the handling equipment to clear it with margin, because the neighbor may transport the board flat while the loader lifts it on a different mechanism that reduces clearance. This is a compatibility item that sits between mechanical and process, and it is worth writing down alongside the height and width, since a board that is compatible on paper but scraped in practice is not compatible at all, and the damage shows up as unexplained rejects at the first inspection station after the handoff.
With the board physically able to travel, the electrical handshake is what actually moves it, and this is the section most often assumed rather than verified. The signals are few, but each one has a polarity and a timing that must match on both ends.
The two core SMEMA signals are the downstream machine telling the upstream one it is ready to accept a board, and the upstream machine telling the downstream one a board is available, and when these two are cross-wired or both asserted at the wrong time the line either never starts or never stops. The checklist verifies the pin assignment and the logic level of both signals against the neighbor, because two vendors can label the same function with different names and still be compatible only if the physical connection and polarity are mapped correctly. You want this written down as a small table, signal by signal, so that during commissioning a single glance resolves any contradiction instead of a half-day of probing, and the table is also what you hand to the technician who wires the interlock so the field work starts from a confirmed plan rather than a guess.
Beyond the basic request and ready pair, the line needs a stop signal that propagates instantly when any station faults, so a jam at the unloader tells the loader to halt before it stacks boards into a stopped machine, and this stop must travel through the SMEMA chain without being swallowed by a mode or a timeout. The checklist confirms that the stop and alarm signals are connected end to end, including through the loader and unloader, because a handling machine that listens only to its own local sensor and ignores the line stop will keep feeding a stalled neighbor. For cells that separate good and rejected boards, the NG/OK unloader is one place where the stop and divert logic must agree with the upstream inspection verdict, and the SMEMA stop is what keeps a fault from becoming a pile of boards on the floor.
SMEMA runs on low-voltage DC signals, typically around twenty-four volts with optical isolation so that a ground loop between two machines does not corrupt the handshake, and a polarity or isolation mismatch is the classic cause of signals that read backwards or that work on the bench and fail on the line. The checklist records the signal voltage and the isolation method on the neighbor and requires the new equipment to match, and it points to the IPC resources on the SMEMA interface as the place to confirm the current published specification rather than relying on memory. This is the one item where a vendor who says it is standard is not enough, because standard and identical are different things, and the voltage that is one volt off or one polarity reversed is the voltage that stops your line at three in the morning.
Even with signals and height correct, two machines can be incompatible if one can move boards faster than the other can accept them, because the slow machine becomes a permanent bottleneck and the fast one becomes a source of board collisions at the boundary.
The loader, the line, and the unloader each have a minimum and maximum conveyor speed, and compatibility requires a real overlap where all three can run, not just a point where two of them happen to match. The checklist lists the speed range of every machine at the boundary and confirms the new unit overlaps the others with margin, because a loader that tops out below the line tact time will starve the printer no matter how good its signals are. You also confirm the acceleration and deceleration behavior, since a loader that snaps boards into the line can shock a neighbor that eases them in, and the mismatch shows up as intermittent jams exactly at the handoff the SMEMA signals were supposed to make smooth.
Pitch is the center-to-center distance between boards as they travel, and if the loader releases boards with a different pitch than the neighbor expects, the downstream machine sees boards arriving too close or too far and either collides them or gaps the line. The checklist records the pitch and the minimum gap the neighbor requires and confirms the loader or unloader can hold that pitch reliably, because pitch is a function of both speed and the index logic, and a machine that is perfect at one speed can drift pitch at another. This is where how to place loaders and unloaders in the line matters, because placement that shortens the handoff distance reduces the room for pitch error to accumulate before the board reaches the neighbor.
Where speed mismatch is unavoidable, a small buffer or an accumulation section at the boundary absorbs the difference so neither machine stalls the other, and the loader or unloader must be compatible with that buffer rather than fighting it. The checklist notes whether the cell uses a magazine buffer or a vertical buffer at either end and confirms the handshake treats the buffer as a normal downstream or upstream station. A buffer that the loader treats as ready when it is actually full is a buffer that overflows onto the floor, so the SMEMA logic at the boundary has to respect the buffer state exactly the way it would respect a printer or an unloader, and writing that requirement down is what keeps the accumulation from becoming an accident.
The frame the machine sits in has to live on the same floor as the line, connect to the same power, and join the same safety chain, and these unglamorous items are where many installations slip a week behind schedule.
Beyond the conveyor height, the overall footprint has to fit the gap you actually have, and the machine has to reach the pass-line height without a custom stand, because a loader that is the right height on paper but needs a fabricated pedestal becomes a custom job the day it arrives. The checklist records the available floor space and the exact height of the line at the connection point and confirms the new unit fits both without modification, and it also records the service clearance behind the machine, since a unit that cannot be opened for maintenance because it is jammed against a wall is a unit that will not be maintained. This is mechanical compatibility in the plainest sense, and it is worth a measured drawing rather than a verbal yes, because the floor is rarely as square as the brochure assumes.
A loader or unloader on casters that are not locked will drift and break the height and fixed-edge alignment you so carefully specified, so the checklist confirms the machine is leveled and anchored at install and that any casters are locked or removed once placed. You also confirm the leveling feet can reach the floor given your line height, because a machine that floats on partially extended feet will settle under the weight of a full magazine and change the pass-line height by exactly enough to cause the intermittent jams you were trying to avoid. Anchoring is also a safety item, and it ties into the E-stop chain in the next section, so the mechanical and the electrical compatibility checklists meet here at the frame.
The SMEMA signals and the safety signals both have to reach the neighbor, and a site that forgets to plan the cable route ends up with a cable draped across a walkway or a safety chain that does not actually include the new machine. The checklist confirms the E-stop daisy chain passes through the loader and unloader so a stop at any point halts the whole cell, and it confirms the signal cable is routed in a tray that will not be walked on or pinched when the machine is moved for service. This is the part of compatibility that is easiest to defer and most likely to bite during the safety audit, so writing it into the plan before delivery turns a likely finding into a non-event, and it also protects the SMEMA wiring from damage that would otherwise surface as the random stops described earlier.
The signals are physical, but the behavior is decided in the PLC, and two machines with identical wiring can still behave incompatibly if their mode logic disagrees about who is in control.
When an operator puts the printer into manual, the loader must know to stop sending boards, and when the line goes auto, the loader must resume, and a mode interlock that is one signal short will let the loader feed a machine a human is inside. The checklist confirms the mode state is exchanged over SMEMA or a hardwire interlock and that the loader and unloader honor it before they move a board, because mode is the difference between a safe cell and a dangerous one, and it is a compatibility item that sits between the electrical handshake and the safety chain. You verify it by asking the vendor exactly which signal carries mode and what each state means, and you write the answer down, because the mode signal is the one most often assumed to be present and most often found missing during commissioning.
Every machine has its own alarm vocabulary, and the loader or unloader has to map its alarms onto the line stop in a way the neighbor understands, so a magazine-empty alarm at the loader halts the line rather than just blinking locally. The checklist records which alarms are mapped to the SMEMA stop and which are local only, and it confirms the acknowledgment behavior, because an alarm that stops the line but cannot be cleared from the neighbor forces an operator to walk the cell for every minor fault. For an NG/OK unloader after inspection this mapping is especially important, since a divert fault must stop the line before a rejected board is misrouted, and the alarm compatibility is what enforces that.
Modern lines change over by sending a recipe that includes the board width, and a loader or unloader that cannot receive or act on that recipe will sit at the old width while the rest of the line changes, stalling the very first board of the new product. The checklist confirms the new unit participates in the auto-width-change handshake, or at minimum that it signals ready only after it has reached the commanded width, so the line does not release a board into a mismatched rail. This is where choosing the right loader and unloader for a high-mix line pays off, because a unit that understands recipe-driven change is compatible with the line you are building, while a manual-only unit is compatible only with the line you used to have.
The sections above are the theory, and this section turns them into a short list you can actually send to a vendor and take to a site survey, so the compatibility question is answered before the machine is built rather than after it is delivered.
Start with a one-page questionnaire asking the vendor for the exact pass-line height, the fixed edge, the travel direction, the signal voltage and polarity, the width range and repeatability, the speed range, and the mode and alarm mapping, because a vendor who answers all of these in writing is a vendor whose machine you can place with confidence. You compare those answers line by line against the neighbor specification, and any mismatch becomes a question before order rather than a finding after install, which is the entire point of the checklist. The questionnaire also surfaces the small items, like whether the unit ships with the connector pre-wired or expects field wiring, because a machine that arrives unwired is a machine whose compatibility you now have to guarantee yourself on site.
Before you trust the line drawing, you measure the actual pass-line height at the connection point, you confirm the actual fixed edge and travel direction as built, and you record the available footprint and service clearance, because drawings drift from reality and the machine must fit the reality. You also photograph the neighbor connector and note the cable route you will use, so the install day has no surprises about what is actually behind the panel, and you verify the floor is level where the machine will sit so the leveling feet have room to work. These measurements are the input to the final compatibility decision, and they are what turn the vendor answers from plausible into verified, which is the difference between a smooth commissioning and a week of field fixes.
The checklist you build here is also the seed of the factory acceptance and site acceptance tests described in the companion guide on SMT loader and unloader FAT and SAT, because every compatibility item above becomes a test point you can tick during acceptance. You note, for each item, how you will prove it on the floor, whether by a voltage reading, a width measurement, or a mode-interlock trial, so acceptance is a confirmation of the compatibility plan rather than a fresh investigation. Treating the checklist and the acceptance test as one continuous document is what keeps the line from passing acceptance on paper while failing it in production, and it is the habit that separates a cell that works from a cell that almost works.
SMEMA covers only the handoff signals, the yes-or-no exchange that lets one machine know when its neighbor is ready, and it deliberately leaves width, speed, height, and recipe to the equipment and the line design, which is why a compatibility checklist has to cover those mechanical and logical items separately instead of assuming the standard handles them.
Yes, if both follow the published signal convention for voltage, polarity, and pin assignment, and if the mechanical items like height, fixed edge, and direction also match, because the standard exists precisely so mixed-vendor lines can interoperate, though the mixed part is exactly why the verification in this guide matters more, not less.
The most common mismatch is the fixed edge or travel direction being opposite between the new handling machine and the neighbor, followed closely by a polarity error on the stop signal, both of which produce intermittent stops that look like random faults until someone checks the interface against a written checklist.
The conventional SMEMA signaling uses a low-voltage DC level in that range with optical isolation, but the exact value and the isolation method should always be confirmed against the neighbor and the current specification rather than assumed, because a one-volt or one-polarity error is enough to invert or weaken the handshake.
It should, because the loader or unloader must respect the buffer full or empty state exactly as it would respect a printer or an unloader, and a buffer the handling machine treats as always ready is a buffer that will overflow, so the handshake logic has to include it rather than bypass it.
Use the vendor questionnaire and the site survey from this guide as a single document, confirm every item in writing against the neighbor specification, and carry those same items into the acceptance test as tickable points, so compatibility is verified on paper first and confirmed on the floor second rather than discovered during production.
