Safety light curtain wiring diagram: why the E-stop does not go on the same input

I get the same question on every cell retrofit where one safety relay has to cover the whole thing. There is an E-stop station, there is a light curtain, there is one relay, and the obvious move is to string both devices together and land them on a single pair of input terminals. Both sit closed when things are fine. Both open when they are not. Why complicate it.

That is where a safety light curtain wiring diagram goes wrong, and it goes wrong quietly, because the circuit works. Break the beam, the machine stops. Hit the button, the machine stops. It just is not doing what you think it is doing, and the only thing that would have told you is a validation test nobody was planning to run.

An E-stop contact and an OSSD are not the same kind of signal

Get this part straight and the rest of the wiring falls out of it on its own.

An E-stop is dead metal. Two normally closed contacts, no supply of its own, no way of telling anyone it has failed. So the relay has to go looking for faults itself. It generates test pulses on S11 and S21, pushes them out through the two E-stop contacts, and reads them back on S12 and S22. Each channel carries its own pulse, and that is what lets a dual-input relay catch a loose wire, a short to 24V, a short to ground, a contact welded closed, and a cross fault between the two channels.

An OSSD is the opposite animal. Two PNP transistor outputs that sit live at 24V whenever the field is clear, already driving the line and already testing themselves. The curtain pulls each output low for a few tens of microseconds on a repeating cycle and checks that the line really did go low. On an IDEC SE4D the spec sheet gives 14 ms max to drop the output when the beam breaks, and 80 to 90 ms to bring it back. If the off pulse does not come back looking right, the receiver calls it an output or wiring error and holds both OSSDs off.

One device is waiting to be tested. The other tests itself and needs sole ownership of its own conductor. Series them onto one input and the relay's test pulses now have to travel through a transistor that is busy driving that same wire.

Left, the safety light curtain wiring diagram that gets drawn first. Right, the one that survives a validation test.

 

The safety light curtain wiring diagram for one dual-input relay

A 440R-D22R2 carries two dual-channel universal inputs, and that is the entire reason to reach for it here. One device per input.

IN1 takes the E-stop. S11 goes out through the first normally closed contact and returns on S12, S21 goes out through the second and returns on S22.

IN2 takes the curtain. OSSD1 lands on S32, OSSD2 on S42, straight in, with nothing in series with either of them.

Reset sits on S34, and it is more than a button. The normally closed auxiliary contacts of both contactors sit in that circuit ahead of the pushbutton, which is worth a section of its own further down. The two safety outputs, 13/14 and 23/24, drive the coils of a pair of 100S-C contactors, and those contactors are what actually take power off the motor. Stop category 0. The motor coasts.

Then the bit people skip, because it is a rotary switch rather than a wire. The logic selector goes to position 2, which is L12 OR IN 1 AND IN 2. Leave it wherever it arrived set from the last job and you can hand over a beautifully wired panel in which one of the two devices does nothing at all.

Terminal	What lands there
A1, A2	24V DC supply to the relay
S11, S21	Pulse test outputs, one per channel, out to the two E-stop contacts
S12, S22	Input 1 return, one terminal per E-stop contact
S32, S42	Input 2, OSSD1 and OSSD2 from the light curtain receiver
S34	Reset input, through both contactor N.C. auxiliary contacts and the reset button
13/14, 23/24	Safety outputs to the two contactor coils
Y32	Solid-state auxiliary output, status to the standard PLC
L11, L12	Logic terminals for the input logic selected on the rotary switch

Terminal assignment for a safety light curtain wiring diagram built on a 440R-D22R2.

Worth being clear about what this arrangement actually buys. The risk assessment behind it only called for Category 3 and PLd. Wired this way, with a Type 4 curtain, a dual-input relay and two monitored contactors, both safety functions come out at Category 4 and PLe. That margin is not bought by picking a fancier relay. It comes from not sharing an input.

Small one that generates calls: the reset has to be held for longer than a quarter of a second and released inside three. A stab at it does nothing. Neither does leaning on it.

Two dual-channel inputs on one unit. That is what makes a single-relay safety light curtain wiring diagram possible at all.

 

The feedback loop that only bites you at reset

Those two contactor auxiliary contacts sitting in the reset circuit are external device monitoring, and on a panel somebody else built they are the first thing I go looking for, because they are the part most likely to be missing.

Picture a welded contactor. Main contacts stay closed, the motor stays powered, and the normally closed auxiliary contact on that same contactor stays open. The relay can no longer see a complete reset circuit, so it refuses to reset. Take those contacts out of the loop and you get a reset that appears to work, on a machine that never actually lost power.

The test that surprises people is worth running on a machine that is up, because the result is not what anyone expects. Break the link between the K1 and K2 feedback contacts while it is running. Nothing happens. The machine keeps going. Now press the E-stop, watch everything drop, release it, press reset, and the relay will not come back until you restore that link. Contactor monitoring is a reset-time check, not a run-time one.

One thing not to do twice. On this pairing the curtain's own EDM wire is simply left open at the receiver, so its monitoring stays inactive. The relay is doing the contactor monitoring through the reset circuit. Two devices watching the same pair of contacts is not twice as safe. It is two things that can disagree.

External device monitoring in a safety light curtain wiring diagram lives in the reset circuit, which is why a broken feedback link does nothing at all until the next stop.

 

Terminal names change with the brand, and so does where the monitoring lives

Swap brands and the functions survive, but the labels do not. Rockwell identifies everything by terminal number on the relay. IDEC identifies everything by wire color coming out of the curtain, and pushes several functions that Rockwell keeps in the relay down into the curtain itself. Reading a safety light curtain wiring diagram from one brand with the other brand's habits is how a correct-looking panel ends up with automatic reset that nobody asked for.

Function	Rockwell 440R-D22R2 with a 440L GuardShield	IDEC SE4D
Power to the curtain	Brown +V, blue 0V	Brown +V, blue 0V
Safety output from the curtain	OSSD pair into relay terminals S32 and S42	Black wire is control output 1 (OSSD1), white wire is control output 2 (OSSD2)
Manual against automatic reset	Deliberate press on the relay reset input S34, held between a quarter second and three seconds	Pink emission-halt/reset wire: connected (to 0V or +V) for manual reset, left open for automatic reset
Contactor monitoring	Contactor N.C. auxiliary contacts in the relay reset circuit at S34, with the curtain's own EDM monitoring switched off at the receiver	Yellow-green external device monitoring wire, 300 ms window as shipped, 100 to 600 ms available through the optional controller
Turning contactor monitoring off	Not a thing you do. The relay monitors through the reset circuit either way	Leave the yellow-green external device monitor wire open
Non-safety status signal	Y32 on the relay	Yellow-green and black auxiliary output

The same safety light curtain wiring diagram functions, named two completely different ways.

That reset wire is the one I look at twice on an unfamiliar panel. Pink tied to +V is manual reset. Left open, it is automatic reset, which means the outputs come back on by themselves the moment somebody steps out of the beam. One wire, and it is the difference between a machine that waits for a person and a machine that restarts around them.

The tests that prove it

Wiring that looks right proves nothing. A safety light curtain wiring diagram is validated by injecting faults into it, and the list below is the short version of what to run before signing anything off. Every one of these should stop the machine, and none of them should allow a reset while the fault is still in place.

What you do	What has to happen
Pull the OSSD wire off S32 while running	Outputs drop immediately. Reconnecting is not enough: the beam has to be broken and restored before a reset is accepted
Short S32 to 24V	Outputs drop after a moment and the red OSSD off indicator on the receiver flashes. Reset does nothing until both the receiver and the relay have been power cycled
Short S32 to 0V	Same behavior as the short to 24V, including the power cycle
Short S32 straight to S42	Same again
Repeat all four using S42 instead of S32	Identical results
Pull the E-stop wire off S12	Outputs drop. After reconnecting, the E-stop has to be pressed and released before a reset goes through
Briefly jump 24V onto S12	Outputs drop and the PWR/Fault indicator goes steady red. A power cycle is needed before it will reset
Briefly jump 0V onto S12	Outputs drop and the PWR/Fault indicator goes steady red
Briefly jump S12 straight to S22	Outputs drop and the PWR/Fault indicator blinks red
Break the K1 to K2 feedback link while running	Nothing. The machine keeps running. The next stop is the one that will not reset

Fault injection tests for a safety light curtain wiring diagram with an E-stop sharing the relay.

The power cycle line catches people out more than anything else in that table. You clear the short, you press reset, nothing happens, and the natural conclusion is that something else broke while you were in there. It did not. Some faults latch until the receiver and the relay both see a fresh power-up.

Common questions

Can one safety relay handle both an E-stop and a light curtain?

Yes, as long as it has two independent dual-channel inputs and a logic selector to AND them together. A dual-input unit gives you IN1 and IN2 for exactly this job. What it does not give you is permission to put both devices on the same input.

Does a safety light curtain wiring diagram need external device monitoring?

If contactors are what remove the power, then something has to watch those contactors. Where that monitoring lives is a brand decision. Rockwell puts the contactor auxiliary contacts into the relay's reset circuit. IDEC gives the curtain its own external device monitoring input on the yellow-green wire, with a 300 ms window as shipped and 100 to 600 ms available through the optional controller. Run one or the other, not both.

Why will the relay not reset after a fault on a safety light curtain wiring diagram?

Usually because clearing the fault is only half the job. On the curtain side the beam has to be broken and restored before a reset is accepted. On a short to 24V, to 0V, or between the two OSSD lines, the receiver and the relay both want a power cycle first. And if the contactor feedback loop is open for any reason, the relay refuses regardless of what else you have fixed.

Can the OSSD outputs go into a standard PLC input?

For status, yes, and that is what the auxiliary outputs exist for. For the safety function, no. A standard input card does not compare the two channels against each other and does not react to the test pulses, so it cannot tell a healthy curtain from two wires shorted together.

Should a safety light curtain wiring diagram use automatic or manual reset?

Manual, on anything a person can stand inside. Automatic reset restores the outputs the instant the beam clears, with no deliberate action from anybody, and on a cell with a blind spot behind the curtain that is a machine starting up with somebody in it. The Rockwell arrangement does not leave it to a wire at all: the reset is a deliberate press on S34 that will not go through while the beam is still broken.

Sourcing the parts

None of this hardware is exotic, which is exactly why the surplus market is worth a look once a retrofit budget has already gone into the mechanical side. IQElectro stocks Guardmaster 440R safety relays, 100S-C safety contactors and Type 4 safety light curtains, which covers every box in the safety light curtain wiring diagram above. If you are matching a relay to a curtain that is already on the machine, or working out which 440R variant a panel actually needs, the catalog number is the place to start and we can check it against what is on the shelf.