Pin Tumbler Rekeying Bench Technique

Why this matters

A rekey that fails in the customer's hand almost never fails because the pins were cut wrong. It fails because of what happened between removing the old pin stack and reloading the new one: a driver that ejected off the bench and got swept into the trash with a rag, a security pin seated one chamber off from where the bitting chart says it belongs, a "it turned on the bench" call that skipped the door-mounted test. The visible symptom is a callback. The real cost is worse: a rekey job is a security promise, and a pin stack that looks complete but is one component light, or a spool pin sitting in the wrong chamber, hands the customer a lock that feels rekeyed and is not fully secured against the old key or against manipulation. This is the bench discipline that makes the difference between a rekey and a rekey that holds.

Read the cylinder before a single pin comes out

Before disassembly, run the existing key in with light tension and feel the drag as each pin sets. A clean, even resistance at each position is a standard pin stack. A hard catch that eases slightly past the shear line, on a lock advertised as pick-resistant, is the tell for a spool or serrated security pin at that position, and you want to know that before you have five chambers open on the bench with no reference. Count the pin positions (five is standard residential, six is common on Grade 2 commercial and some import cylinders), confirm the keyway profile against the blank you are about to cut, and confirm whether this is a fixed cylinder or an interchangeable core, because a core comes apart with a control key rather than a retaining clip and disassembling it the wrong way strips the core's own control lugs.

Control the plug so nothing leaves the bench uninvited

Every chamber you open holds a driver riding on a compressed spring. Push the plug follower in unevenly, or let it lag behind the plug as it slides out of the shell, and that spring lets go the instant the chamber clears the shell wall, launching a driver and spring into open air. Work over a parts tray or a folded rag so a launched component has somewhere to land besides the floor, keep the follower flush against the plug face through the entire withdrawal in one continuous motion rather than a series of small pushes, and wear eye protection for this step specifically, not just when a grinder is running. If a component does eject anyway, stop and do not resume the rebuild by feel or by assuming you can eyeball a substitute spring from the kit. Search the bench, the tray, and the floor systematically before reloading that chamber, because a chamber that goes back together short one driver has no spring pressure holding its pin to the plug, which means that pin can sit off the shear line indefinitely and the cylinder will still turn: a rekey with a silently disabled pin position is a worse security failure than the lock it replaced.

Read the new bitting before you pick a single pin

Work from the manufacturer's depth-and-space chart for the cut key, or decode it directly off the new key with calipers against the shoulder stop, never from memory or a verbal number called across the shop. Before reaching for the pin kit, lay the new key flat next to the old one and compare the cut profile visually, position by position: this catches a transposed digit on a hand-written work order before it costs you a full pinning cycle. Pin length in a standard system runs shallow-cut-shortest-pin to deep-cut-tallest-pin (the convention varies by manufacturer, so confirm which direction that specific chart runs before you pick), and the number that matters is the one at each position, not an average across the key. A single digit transposed between two adjacent positions is the most common cause of a rekey that turns under hand pressure but binds the moment the lock carries any load in the door.

Load the stack in order, security pins included

Bottom pin first, to the length the chart calls for at that position, followed by the driver, then the spring, working one chamber at a time so nothing gets crossed between positions. Where the cylinder carries a security pin (spool or serrated), it has to land at the exact chamber the manufacturer's pinning specifies for that keyway family, not at whichever position felt right on the last job. A spool driver seated in the wrong chamber does not fail cleanly: under light rotational tension it can false-set right at the shear line, meaning the plug rotates a fraction and stops as if the lock has opened, when in fact the spool's narrow waist has simply caught at the shear line without the chamber above it actually clearing. That reads as "the lock turned" on the bench and as "the lock will not fully open" in the customer's hand a day later.

Interchangeable cores and a master pin change the stack you are loading

A fixed cylinder comes apart with a retaining clip; a small-format or large-format interchangeable core comes apart with its own control key, and forcing a core apart with a follower the way you would a fixed cylinder shears the core's control lugs, an unrelated repair you have now created on a job that started as a rekey. Where the property runs a master key system, every chamber below the change pin also carries a master wafer sized to let the top master's cut set that chamber's own shear line in addition to the change key's. Pin a master-keyed chamber as if it were a single-key cylinder and you either destroy the master function on that door (the master no longer opens it) or, worse, create a second working combination nobody intended, because a wrong-length master wafer can create its own accidental shear line. If the property is master-keyed, pull the bitting for that position from the system's own pinning record before you touch a chamber, not from the change key alone; the change key only tells you the bottom half of what that chamber needs to contain.

Test in stages, never just "does it turn"

Run the new key dry first, no tension, watching for binding as it seats fully to the shoulder. Then apply light tension and turn slowly, feeling for drag that eases rather than a clean release, which signals a pin still riding above or below the shear line. Only after a clean, repeatable turn on the bench do you reinstall the cylinder in the lock body and repeat the same progressive test with the hardware fully assembled and the latch or bolt engaging real resistance, because a plug that turns freely on an open bench block can still bind once it is carrying the throw of a bolt against a strike. Then, and this is the step a rushed rekey skips, insert the old key and confirm it does not turn: not "turns hard," not "with force it goes," but does not turn. If it turns at all, even reluctantly, one or more pins are still sitting where the old bitting left them and the rekey is not complete. If that verification fails, the fix is not to hand back a lock that "should be fine": go back to the chamber the drag points to, re-check that pin against the chart, and rerun the full progressive test from the start.

Worked example: a five-pin deadbolt with one security position

The new key's bitting, shallow to deep, position 1 nearest the bow: 2-4-1-5-3. The manufacturer's chart for this keyway carries the security pin at position 3, the deepest cut on this key. The technician pins positions 1, 2, 4, and 5 correctly against the chart, but at position 3 reaches for the spool driver out of habit and seats it at position 2 instead, where a standard driver belonged.

Dry test: the new key seats fully, no obvious catch. Light-tension test: the plug rotates about a third of a turn and stops with a soft give, which the technician initially reads as the lock opening. Reinstalled in the door and tested under real bolt load, the deadbolt will not retract: the plug is false-set, not open. The drag signature on a second attempt, a stop that eases rather than a hard mechanical block, is the specific tell for a misplaced spool pin rather than a wrong-depth standard pin, which stops harder and does not ease at all.

The technician pulls the plug back out, removes the spool driver from position 2, restores a standard driver there, and seats the spool driver correctly at position 3 against the chart. Full progressive test repeats from dry insertion: the new key now turns cleanly through a full rotation with even resistance at every position, no false-set catch. The old key is inserted next and stops hard at the first disturbed chamber, unable to turn at all. Both results are recorded on the ticket before the lock goes back on the door.

Verification before the job leaves the bench

Reinstall the cylinder in its housing and the lock in the door, and run the full cycle again with the door closed: bolt or latch extending fully and retracting cleanly on the new key, with no assist from lifting or jiggling the door. Confirm the customer receives the number of keys the ticket calls for, tag any customer-requested "do not duplicate" keys, and record on the work order that the old key was tested and confirmed non-operational. That last line is what closes the job: you opened five chambers to change this lock's security, and the record has to show the security was actually restored, not just that a key was cut.

References

• ANSI/BHMA A156.5, Auxiliary Locks, for the deadbolt bolt-projection figures the door-mounted verification checks against, in the edition the hardware was listed under
• ALOA (Associated Locksmiths of America) pinning and service training materials
• Manufacturer depth-and-space and pinning-chart documentation for the specific keyway being serviced
• See related: Rekey and Lock Change SOP (customer-facing rekey-versus-replace decision and job documentation), Broken Key Extraction Standard SOP