What causes an impulse relay to not latch or to chatter? Four faults account for nearly every non-latching or chattering call on site: coil voltage that drops below the pull-in threshold at the far end of a long push-button run, a mechanical ratchet mechanism that has worn or jammed after years of cycling, push-button wiring that lets two commands interfere with each other, and — on electronic latching types — induced noise on the command line that the relay misreads as a pulse. Each cause leaves a slightly different symptom, and confusing them wastes a site visit swapping a relay that was never faulty. A short, ordered sequence — check the coil terminals during a pulse, check free mechanical movement with power off, check the button wiring for continuity, and check cable routing near noisy loads — isolates the real cause before any part gets replaced.
Chattering vs Not Latching: Two Different Faults
Site reports lump these together as "the relay's not working," but they are different failure modes with different root causes. A relay that chatters is trying: press the button and you hear or feel a buzz, a partial armature movement, sometimes a state change that does not hold. A relay that will not latch at all does nothing — no click, no hum, no change of state regardless of how many times the button is pressed. Diagnosing the wrong one first saves the most time on a troubleshooting call.
Chattering almost always points to a marginal condition: voltage close to but below the pull-in threshold, or noise superimposed on a valid signal. Complete failure to latch points to something binary: no supply reaching the coil, a broken command loop, or a mechanism that is physically jammed. Ask which one you actually have before opening the enclosure — it narrows the search by half immediately.
Key takeaway: Chattering means the coil is seeing a marginal signal; complete failure to latch means it is seeing no signal, or the mechanism cannot move regardless of signal. Treat them as separate diagnostic paths.
Cause One: Coil Voltage Below Pull-In Threshold
An impulse relay coil needs a minimum voltage at its terminals to fully pull the armature over and complete the latch — the pull-in voltage. Below that threshold but above zero, the coil may move partway, buzz, or momentarily change state and bounce back, which is exactly what a chattering complaint sounds like on site. This is the most common cause on long push-button runs, particularly on 24 V or 12 V control circuits — such as the DC-coil variants like the ABB E290-16-10/24 latching relay — where the working voltage is already low and a modest cable resistance eats a proportionally larger share of it.
Vdrop = I_coil × R × 2 × L
Where: Vdrop is the voltage lost along the command loop (V), I_coil is the relay's coil pulse current (A), R is the control cable resistance per metre for the conductor size in use (Ω/m), L is the one-way distance from the supply to the furthest push-button (m), and the factor of 2 accounts for the outgoing and return conductor length together. If the nominal supply minus Vdrop drops below the relay's rated pull-in voltage, the far button pulses the coil but never fully seats the armature.
A corridor circuit on one site had run cleanly for two years on a 24 V DC command loop. A refurbishment added a fourth entry point at the far end of an extended corridor, tapped off the existing bus with a long cable run back to the panel. The new button worked erratically — sometimes latching, sometimes buzzing with no state change — while the original three kept working normally. The relay tested fine on the bench; the fault was voltage drop on the new run, not the device.
Pull-In Voltage
The minimum voltage that must appear across an impulse relay's coil terminals for the armature to complete its full travel and latch. Below this threshold the coil may move without fully seating, which is felt on site as chatter or an unreliable trigger rather than a clean state change.
The fix is a heavier command conductor, a control voltage nearer the supply, or relocating the relay closer to the centre of its push-button loop rather than at one end of the building.
Cause Two: A Worn or Jammed Ratchet Mechanism
Mechanical impulse relays, such as the ABB E290-16-10/230 latching relay, hold their state with a small ratchet pawl that steps the contact carrier one position per pulse, alternating open and closed. After a high number of operating cycles, or after dust and fine metal swarf from nearby drilling or cable-tray cutting work into the mechanism, the pawl can bind partway through its travel. The symptom looks electrical, but the coil is pulling correctly and the fault sits entirely in the mechanism.
The field check does not need power at all. Most modular impulse relays carry a manual override lever or a visible state indicator on the front face. With the panel isolated, work the lever by hand through a full cycle. A mechanism that moves freely but does not respond to an electrical pulse points back to the coil or the command signal; one that resists, sticks partway, or will not complete its travel confirms a mechanical fault, and the relay needs replacing rather than rewiring.
Key takeaway: A manual override check with the panel isolated separates mechanical faults from electrical ones in under a minute, before any meter comes out.
Cause Three: Push-Button Wiring Faults
Parallel push-button wiring is what makes an impulse relay circuit simple, but the same simplicity means a single miswired button affects the whole loop. Two faults show up repeatedly on site: a button wired between the wrong terminals so it holds the command supply against the common return continuously rather than pulsing it, and a break in the shared bus conductor that isolates one section of buttons from the relay while the rest keep working.
A continuously closed contact — a stuck button, a miswired auxiliary contact, a bridged terminal — holds the command input high or low permanently. Some relay designs read a held signal as a stream of pulses and chatter continuously; others latch once, then ignore further pulses until the stuck contact releases. The fix is at the button: isolate each one on the loop and watch whether the fault clears.
| Symptom | Most likely cause | First field check |
|---|---|---|
| Chatters, never fully latches, worse at far end of run | Coil voltage below pull-in threshold | Measure voltage at coil terminals during a pulse from the furthest button |
| No response at all, coil silent | Broken command loop or blown upstream protection | Continuity check of the command bus, terminal by terminal |
| Responds to some buttons, not others | Break or loose terminal partway along the bus | Isolate loop sections and test each in turn |
| Latches, mechanism resists on manual override | Worn or jammed ratchet mechanism | Operate override lever by hand with panel isolated |
| Random state changes with no button pressed | Phantom pulse from induced noise | Inspect cable routing for proximity to power or drive cabling |
Key takeaway: A single miswired or stuck push-button can make an entire parallel loop misbehave — isolate buttons one at a time rather than assuming the relay is at fault.
Cause Four: Phantom Pulses on Electronic Latching Types
Electronic latching relays such as the Schneider iETL iTL16 2P impulse relay read the command input differently from a mechanical ratchet: the pulse is interpreted electronically against a defined threshold rather than mechanically stepping a pawl. That gives cleaner switching on a marginal signal — but it also means noise coupled onto a long command run can be read as a valid pulse, producing an unexplained state change with nobody at any button. See our guide to buzzing coils and phantom switching for more on this.
Long command runs bundled with power cabling, especially near variable frequency drive output leads, are the classic source. Disconnect suspect buttons one at a time and watch whether the triggering stops, then inspect the run for a section routed close to a noisy cable. Separating the command loop from power cabling resolves most of these cases without touching the relay.
Chattering
Repeated, rapid, incomplete attempts by a relay's coil to operate — audible as a buzz, sometimes accompanied by a brief flicker of the switched load — without the armature fully seating into its latched position. It is the signature symptom of a marginal coil signal rather than a complete absence of one.
A Field Troubleshooting Sequence
Work through these checks in order, and stop at the first one that isolates the fault. First, classify the symptom as chattering or complete non-response. Second, with the panel isolated, run the manual override lever through a full cycle by hand to rule out a mechanical jam. Third, restore power and measure voltage at the coil terminals while a button is pressed, comparing the reading against the relay's rated pull-in voltage on the datasheet. Fourth, if voltage is adequate but the fault persists, check continuity and terminal tightness along the push-button bus, isolating sections to find a break or a stuck contact. Fifth, if the loop tests clean electrically and the fault is intermittent state changes with nobody pressing a button, inspect cable routing for proximity to power or drive cabling and consider whether an alternative control voltage or wiring separation resolves it.
This sequence puts the cheapest, fastest checks first. A manual lever test and a coil voltage reading take two minutes and rule out two of the four causes before a single terminal is disturbed. Replacing the relay should be the last step — browse the current range in the relay collection if a swap is needed — but keep the original unit to inspect the mechanism. A worn ratchet on one relay in a batch installed together predicts the same fault on its neighbours.
Key takeaway: Manual override and coil voltage checks together rule out two of the four common causes in under five minutes — do them before assuming the relay itself has failed.
Conclusion
Most impulse relays that "won't latch" or chatter on site are not faulty devices; they are correctly functioning relays responding to a marginal or interrupted signal. Coil voltage drop, a worn ratchet, a miswired push-button, and induced noise on electronic types account for nearly every call. Work the checks in order — mechanical, then coil voltage, then wiring continuity, then cable routing — and the relay you were about to condemn usually turns out to be the one part of the circuit that was working correctly all along.
Frequently Asked Questions
Why does my impulse relay buzz but never fully change state?
This is chattering, and it almost always means the coil is receiving voltage below its rated pull-in threshold. The armature moves partway without fully seating. Check the voltage at the coil terminals during a button press, particularly on long 24 V or 12 V runs, where cable resistance eats a larger share of an already low working voltage.
How do I tell if an impulse relay fault is mechanical or electrical?
With the panel isolated, operate the manual override lever by hand through a full cycle. If it moves freely, the fault is electrical — check coil voltage and button wiring next. If it resists or will not complete its travel, the ratchet has worn or jammed, and the relay needs replacing rather than rewiring.
Can a long push-button cable run stop an impulse relay from latching?
Yes. Voltage drop along the command loop is proportional to cable length and coil current, and on low-voltage control circuits it can push the voltage at the furthest button below the relay's pull-in threshold. The fix is a heavier conductor, a higher control voltage, or relocating the relay nearer the centre of the loop.
Why did my impulse relay start switching on its own after a nearby cable was added?
This is phantom switching, typically caused by capacitive coupling from a power or drive cable routed close to the command loop. It is more common on electronic latching types, which read the command input against a defined threshold rather than a purely mechanical pulse. Separating the command wiring from power cabling usually resolves it.
Should I replace an impulse relay the first time it fails to latch?
Not before checking the cheaper causes. A manual override test and a coil voltage reading take a few minutes and rule out mechanical and voltage-drop faults before any part is disturbed. Replace the relay only after wiring, voltage and mechanism have tested clean, and keep the removed unit to confirm the mechanical fault if that was the cause.