Broken Key Extraction Technique by Lock Type

Why this matters

The field procedure for a broken key, isolate the plug, never push the fragment inward, never introduce glue, prove the cylinder afterward, is the same regardless of what is jammed shut. The tool that actually gets the fragment out is not. A technician who owns one hook-and-pick extraction kit and reaches for it on every call is the technician who converts a disc detainer padlock, a tubular vending lock, or a warded antique into a worse problem than the one they arrived to fix, because the tool that works beautifully on a pin tumbler cylinder relies on a mechanical feature that some lock families simply do not have.

The one question that picks the right tool

Before anything goes into the keyway, ask a single thing: is this fragment being gripped by spring-loaded pins pressing down on it, or is it sitting in a mechanism with nothing bearing on the blade at all? Pin tumbler and most wafer locks are the first case. Disc detainer locks, most warded locks, and tubular locks are the second. The two answers point to opposite techniques for the identical symptom, a snapped key sitting in a keyway: one wants a tool built to exploit spring pressure that is already working in your favor, the other wants a tool that displaces nothing, because there is no pressure to relieve, only geometry and friction to work around.

Read the mechanism before you commit a tool, because the afternoon case below shows how easily a family gets assumed from the outside rather than confirmed. A round, flat face with a small circular bore instead of a slotted keyway is tubular. A wide, simple, uninterrupted keyway you can see light through, with no visible pin chambers along its length, is often warded. A solid-bodied padlock with no exposed keyway rose and a noticeably heavier shackle than its size suggests is worth treating as a possible disc detainer until the extractor confirms otherwise. None of these are certain from the outside alone, which is exactly why the hook extractor's own behavior, engaging cleanly versus finding nothing, is the more reliable confirmation than appearance.

Where the gate says yes: pin tumbler and wafer locks

A hook or saw-tooth extractor works on these mechanisms precisely because the pins are already pressing the fragment against the bottom of the keyway. Sliding a fine hook past the fragment and drawing it back engages the serrations while the pins' own spring tension helps hold the fragment against the hook rather than letting it float free in the channel, which is why this is the fast, reliable case: a clean pull usually clears it in one or two attempts. A spiral or screw-type extractor threaded directly into the fragment is also reasonably safe here, because the chamber walls around a pin tumbler plug are built to tolerate the modest radial pressure of the thread engaging. The full field procedure, tension control, attempt limits, and what to do when this method fails twice in a row, is covered in the shop's extraction standard and is not repeated here; the point of this section is narrower, that the tool choice itself is correct because the mechanism has springs doing half the work for you.

Where the gate says no: disc detainer locks

A disc detainer mechanism, common on higher-grade padlocks and some cabinet and vending locks, holds its security in a stack of rotating discs with light detent springs, not linear pin springs bearing down on the key blade. A hook extractor finds nothing to catch: there is no downward pressure holding the fragment in place, so the hook slides past it without engaging anything. Worse, a screw-type extractor threaded into a disc detainer fragment is the wrong move for a mechanical reason specific to this family, not a general caution: the discs sit close around the blade with very little clearance, and the radial expansion of a screw biting into the fragment can crack or bind a disc against its neighbor rather than free the key. Extraction here works by rotation instead of by pulling, using a thin pick to work the disc pack's true gates back into alignment, or a disc-detainer-specific extraction tool that grips the fragment's flat faces rather than threading into it. If you do not carry that tool, this is a bench or referral job, not a field improvisation with pin tumbler equipment.

Warded locks: the opposite risk from every other family

A warded lock has no pins and no discs, only a shaped keyway the bit has to clear on its way to the bolt. A snapped bit in a warded lock is frequently just loose, because nothing is clamping it the way a pin stack clamps a pin tumbler fragment, and needle-nose pliers or a simple bent-wire hook alone often lift it straight out with no specialized tool at all. The risk here runs the opposite direction from every mechanism above it: over-tooling a warded lock, forcing in an extractor sized and stiffened for a tight pin tumbler keyway, can bend the delicate wards themselves, the thin metal projections the key blade is shaped to clear. Match the tool's size and stiffness to how open and unobstructed this keyway actually is, rather than defaulting to your strongest extractor because it worked on the last three jobs.

Tubular and high-security sidebar cylinders: two families that end differently

A tubular, or circle-key, lock arranges its pins radially around a circular bore rather than along a flat blade, common on vending machines, laundry equipment, and elevator control panels in light-commercial settings. A standard flat hook cannot reach around that geometry, and improvising with one usually just pushes the fragment deeper into the radial bore rather than engaging it. Extraction needs a tool built for the circular format, a tubular-pick-style extractor or a small-diameter tube sized to the specific pin circle, worked around the fragment rather than beside it.

High-security sidebar cylinders (Medeco, Mul-T-Lock, and similar product families where naming the brand is what the reader actually looks up) add a leg that reaches into a slot on the plug in addition to the ordinary pin stack. A fragment can be fully cleared of the visible pin stack and the plug can still refuse to turn, because the sidebar leg is a second, independent point of resistance a standard pin tumbler extraction never has to account for. Set that expectation with the customer before you start, not after two failed field attempts: these are the cylinders most likely to end at cylinder removal for bench disassembly rather than a quick in-place pull, and that is a property of the mechanism, not a sign the technician did anything wrong in the field.

Automotive ignition and steering-column locks: stored energy changes the technique

A steering column held under load by a customer's own attempt to turn a broken key can release with real force the instant tension comes off the fragment, so confirm the wheel is not under load, wheels straight, nobody leaning on it, before working the fragment at all. If the column will not return to rest cleanly, stop rather than forcing it; that resistance means stored torque is still in the mechanism. Beyond that hazard, the tumblers themselves are typically flat and wide rather than the round pin format of a residential deadbolt, which calls for a wider, flatter extractor than a standard pin tumbler hook. Many ignition housings also carry an immobilizer or transponder antenna wound close behind the tumbler pack, so before threading anything past the depth the fragment itself occupies, know how deep that specific column's tumbler pack actually runs; where you do not know it, work in small increments and stop to check position rather than assuming a residential extraction depth is a safe default.

Worked example: two calls, same symptom, opposite tools

Morning call: a residential rear entry deadbolt, pin tumbler, key snapped with a short stub protruding. A hook extractor engages cleanly on the first attempt and the fragment comes free in under two minutes, the outcome the gate predicts for a spring-loaded mechanism.

Afternoon call, same day: a storage-unit padlock, described on the ticket as "just a padlock," snapped key inside. The technician reaches for the same hook extractor that worked that morning. It finds nothing: no drag, no resistance, the hook slides past the fragment without engaging anything, which is itself the diagnostic answer to the gate question, this is not a spring-loaded mechanism. A closer look at the shackle and body confirms a disc detainer padlock, not the standard wafer or pin body it resembled from the outside. The technician stops reaching for the morning's toolkit, switches to a disc-detainer pick, and works the disc pack's gates into alignment with the fragment rather than trying to pull it. The extraction succeeds, but it takes roughly four times as long as the morning's pin tumbler job, and it succeeds because the technique changed with the mechanism rather than because more force was applied with the wrong tool.

Verification, with one addition for detent and sidebar mechanisms

Once a fragment is out, the cylinder-proving cycle from the shop's extraction standard applies regardless of family: repeated smooth cycles with the door open, then with it closed. Add extra cycles specifically on disc detainer and sidebar cylinders, because their detent springs and sidebar leg can mask a partial obstruction, a disc not fully returned to true or a sidebar leg riding rough, that a plain pin tumbler proving cycle would already have caught in fewer attempts. More cycles here is not caution for its own sake; it is compensating for a mechanism that gives you less feedback per cycle than a pin stack does.

References

• See related: Broken Key Extraction Standard SOP (full field procedure, tension control, attempt limits, and documentation, not repeated here)
• 29 CFR 1910.133 for eye and face protection against sprung fragments and snapped extractors
• Manufacturer service literature for disc detainer and high-security sidebar cylinder disassembly
• ALOA (Associated Locksmiths of America) training on mechanism-specific extraction technique