Shenzhen Jiatel Hardware Electronics Co.,Ltd - Custom Pogo Pin and Wire Harness Manufacturer
Request Quote

Terminal Extraction Force: Why the Housing Breaks Before the Terminal Comes Out

Engineering Note #3 — Jiatel Hardware Electronics

When a terminal comes out of a connector, the terminal is almost never the thing that failed. The retention feature — the lance that locks it into the cavity — has either been defeated by a tool, or it has plastically deformed the plastic shoulder it locks against.

The second case is the expensive one, because the housing is now scrap. The terminal can be replaced. A cavity whose locking shoulder has been planed flat will not hold a new terminal either, and nothing about the part looks damaged.

This note covers what actually breaks, what the standards measure, and the operating rules that keep the housing intact.

1. Two forces that get used interchangeably

Force What it is What it acts on Measured by
Retention force The force the locking feature resists before the terminal moves in its cavity The lance and its shoulder EIA-364-35 (insert retention, axial)
Extraction force The force an operator or a tool actually applies to remove the terminal Lance, shoulder, and whatever the tool touches Not a standard test — it is an assembly action
Mating / unmating force Force to fully mate or unmate the connector Housing, latch, all contacts together EIA-364-13

These are three different numbers, and only one of them is a specification.

Retention force is a designed value, measured on a sample, and it is what the housing can survive. Extraction force is whatever you do to the part. Mating force is a connector-level number that says nothing about a single cavity.

The engineering rule that follows: your extraction method must keep applied force below retention force. If it does not, you are not removing a terminal — you are testing the housing to destruction, one cavity at a time.

2. What the housing actually does when it fails

Four failure modes, in order of how often they appear:

1. Shoulder planing. The lance sits against a moulded plastic step. Pull hard enough and the lance slides up over the step, shaving it. The plastic does not crack — it yields and flows. Retention drops to near zero and the cavity looks fine. This is the most common and the least visible.

2. Lance deformation. The lance is a cantilever spring. Overload it and it bends past its elastic limit. It no longer springs back to full engagement depth, so retention is permanently reduced even if the shoulder survived.

3. Wall cracking. Thinner housings crack at the cavity wall rather than planing the shoulder. Visible, and usually caused by side-loading rather than pure axial pull.

4. Cavity distortion. Repeated insertion of an oversized probe, or levering a tool against the cavity wall, opens the cavity. A new terminal then sits loose in a cavity that has lost its interference.

Diagnostic that catches all four early: measure retention force on a sample from the batch, not just continuity. A cavity that has been planed will still pass a continuity check with the terminal held in by friction.

3. The TPA trap: the most common cause of destroyed housings

Many modern connector systems have a Terminal Position Assurance element — a secondary lock, usually a comb or a slider, that sits behind the terminals and prevents them backing out.

TPA exists for a good reason: primary lances alone can be defeated by vibration or by a poorly seated terminal. The secondary lock is what makes the connector reliable.

It is also the single most common cause of a wrecked housing, because of one sequence:

The technician pulls the terminal without releasing the TPA first.

With TPA engaged, the terminal cannot come out. The force goes into the TPA feature instead of the lance, and the TPA seat is not designed to be a load path. The result is a housing that will never lock again — and the failure is invisible until the harness is in the field and a terminal backs out under vibration.

Rule: if the connector has a TPA, releasing it is step one, not an optional step. Any work instruction that does not start with “release TPA” is a work instruction that produces scrap.

4. Decision matrix: lance type × dominant failure

Retention type How it holds Dominant failure when abused Operating rule
Single cantilever lance One plastic shoulder catches the lance tip Shoulder planing Axial force only; never lever sideways. Use the correct extraction tool, not a pin or a blade.
Double lance (both sides) Two shoulders, symmetric One side fails first under off-axis pull, then the second Pull must be straight. Any visible tilt means one lance has already taken the whole load.
Box / dual-beam contact with lance Lance on the box body, shoulder in cavity Box distortion before lance failure Do not squeeze the contact body to “help” it out; that reduces the interference that holds it.
TPA / secondary lock Mechanical block behind the terminal TPA seat destroyed if pulled while engaged Release TPA first, every time, before any extraction force is applied.
Front-finger / sealed connector Retention finger plus interfacial seal Seal torn, finger bent Seal is not a grip point. Tool must not pass through the seal area.

5. The wire-pull mistake

Pulling the wire instead of the terminal body is not a small technique error. It changes the mechanics of the failure.

When you pull the wire, the force is applied at the crimp, offset from the terminal axis by the bend of the wire and the depth of the cavity. That offset creates a moment, not just a tension load. The terminal rotates slightly in the cavity, which:

  1. Loads the lance on one edge instead of across its width — concentrating stress at the point most likely to plane the shoulder
  2. Loads the cavity wall laterally, which is the failure path for thin housings
  3. Can pull the crimp itself, which is a separate and worse failure — a partially withdrawn conductor that still looks terminated

A wire that comes out with the insulation stretched is evidence that the crimp was loaded, not the lance. That terminal must be cut off and re-terminated, not reinserted.

6. The 0.8 rule, and what it is not

A practical working limit used in harness assembly:

maximum extraction force = rated retention force × 0.8

The intent: keep a 20 % margin so that the extraction operation never approaches the retention limit, which leaves the shoulder and lance inside their elastic range.

What this is: an industry practice heuristic. It is a sensible margin, widely used, and it keeps housings alive.

What this is not: a standard requirement. EIA-364-35 defines how to measure retention; it does not tell you what fraction of it you may apply during service or rework. Anyone presenting 0.8 as “the EIA limit” is misquoting.

The number you should actually use is on the connector’s datasheet as a minimum retention force. If it is not there, ask for it — a supplier who cannot state retention force for a removable-contact connector has not characterised the part.

7. Re-termination, and the limits of rework

Two separate questions get conflated here.

Can the housing be reused? Only if retention force still meets the minimum after extraction and reinsertion. Measure it. A housing that has been planed will accept a new terminal and feel normal in the hand.

Can the terminal be reused? For crimped terminals, no. Crimping is a cold-weld deformation of the conductor strands and the crimp barrel. Re-crimping the same terminal on a shortened wire reduces strand count and changes the crimp geometry. Every re-termination should use a new terminal and cut back to fresh conductor.

There is also a re-termination count for the contact-to-housing interface itself. Each insertion cycle wears the lance tip and the shoulder. Ask for the rated number of cycles; it is usually well below the connector’s mating cycle rating.

8. How Jiatel specifies retention and extraction

For harness assemblies we state, per connector series:

  1. Minimum retention force per cavity, measured per EIA-364-35
  2. The TPA type and release method, written into the work instruction rather than left to the operator
  3. The extraction tool — part number, not “a suitable tool”
  4. Rated re-termination cycles for the contact-to-housing interface
  5. The retention check used in QC, and the sample rate

If a harness arrives without items 2 and 3, the first rework will damage housings, and the damage will not show up until the field.

9. What to do next

If you are writing an assembly or rework instruction now:

  1. Find out whether the connector has a TPA, and put “release TPA” as the first step
  2. Get the minimum retention force from the datasheet
  3. Set your extraction force limit below it, with margin
  4. Specify the extraction tool by part number
  5. Check retention on a sample after rework, not just continuity

→ Cable Assemblies and Wire Harnesses — assemblies built with documented retention values and tooling.

→ Request a Custom Harness Quote — send the connector series and we will return retention force, TPA release method and extraction tooling with the quotation.

Related notes: How to Calculate Pogo Pin Spring Force covers the contact-force side of the same reliability question; SMT Battery Holder Reflow Profile covers what heat does to contact force after assembly.

References

  • EIA-364-35 — Insert Retention Test Procedure. Applies an axial load to a contact to verify it does not pull out of its housing; defines the measurement, not a permissible service load.
  • EIA-364-13 — Mating and Unmating Force Test Procedure. Connector-level forces, measured at a rate not exceeding 25.4 mm/min (Method A).
  • EIA-364-08 — Crimp Tensile Strength Test Procedure. Relevant to the wire-pull failure mode described in section 5.

The 0.8 factor in section 6 is an assembly practice heuristic and is explicitly not presented as a standard requirement.