Cable Harness Crimp Force Monitoring: Baselines, Reject Limits and Traceability Evidence

Direct answer: specify crimp force monitoring as a controlled production-detection system, not as a one-line equipment claim. A defensible RFQ identifies the monitored terminal-wire-applicator combination, how the accepted force signature is established, which variables and limits are controlled, and how every alarm is contained and reviewed. It also states which complementary crimp-height or destructive checks remain required and which traceability fields connect each result to the delivered harness. Monitoring can add 100% non-destructive visibility at the crimping station, but it does not by itself validate terminal design, prove every crimp characteristic or replace the released application specification.

The purchasing decision is therefore not simply “Does the supplier have a crimp force monitor?” It is:

That question turns a machine feature into an auditable quality control.

Why a monitor name is not an acceptance plan

Crimp force monitoring, often abbreviated CFM, observes force behavior during the press cycle. Depending on the equipment and configuration, the system can compare a force signature or derived features with a learned reference and flag a cycle that falls outside configured limits. The Komax QpLite 2 product page, checked on 2026-10-08, describes model-specific features including selectable tolerance profiles, real-time machine-status information and data logging. Those are capabilities of the named equipment family, not proof that a supplier has installed, configured or validated them for a particular crimp.

The TE Connectivity crimp-quality white paper, checked on 2026-10-08, makes a second boundary clear: force monitoring is useful as a 100% non-destructive process check, while crimp height remains an important quality metric tied to the intended compression. TE also explains that force alone does not validate the original terminal design intent and that different methods detect different variables. The procurement implication is simple: use complementary evidence rather than asking one signal to prove everything.

A complete sourcing package separates at least six decisions:

On small screens, swipe or scroll sideways to read every column. Keyboard users can focus the table and use the arrow keys.

If a quotation answers only the second or fourth row, the buyer still has a major evidence gap.

Start with the connection design authority

The monitor cannot decide whether a wire, terminal and seal combination is correct. Begin with the controlled source that defines the connection.

Ask for:

• terminal manufacturer and exact terminal identifier;
• wire conductor material, construction and size as released for the design;
• insulation and seal information where relevant to the termination;
• approved applicator, die, locator and press requirements;
• target crimp dimensions and tolerances from the authorized source;
• strip length and conductor preparation instructions;
• any lubrication, splice, double-wire or special application restrictions;
• applicable workmanship document and customer-specific additions;
• the role authorized to approve an alternative.

The phrase “terminal manufacturer specification” is not sufficient if the document number and revision are missing. A supplier should be able to identify the controlled application instruction without disclosing unrelated proprietary data.

Where the design authority has not released the terminal-wire combination, force monitoring should not be used to create an appearance of validation. The unresolved combination belongs in engineering approval, not in a production-monitor learning cycle.

Keep three baselines separate

Procurement discussions often use the word baseline for different things. State which one is meant.

1. Product baseline: the released wire, terminal, seal, splice and drawing configuration.
2. Process baseline: the approved press, applicator, tooling, setup, maintenance condition and operating method.
3. Monitoring baseline: the reference force signatures, calculation settings, tolerance profile and software or program revision used to judge production cycles.

A valid monitoring baseline depends on the first two. If the product or process baseline changes, the monitoring baseline may need review, revalidation or replacement.

Understand what the signal can and cannot tell you

The press force develops as material is positioned, bent, compressed and released. The recorded curve is influenced by the terminal, conductor, insulation support, applicator, press, speed, mechanical condition and signal-processing choices. A deviation may indicate a missing strand, wrong terminal, missing wire, double feed, tool damage, mispositioning or another process change. Whether a specific fault is detectable depends on its effect on the signal relative to normal variation and on the configured analysis.

Do not write a universal list promising detection of every possible defect. Instead, require a validation matrix for the error modes that matter to the purchased connection.

On small screens, swipe or scroll sideways to read every column. Keyboard users can focus the table and use the arrow keys.

The right column is the key. A monitor result should be supported by another method when the target condition is not reliably observable in the force signal.

Separate detection from acceptance

A monitoring alarm means the cycle departed from the configured rule. It does not automatically identify the root cause. Likewise, a cycle within the monitoring limits means the signal matched the approved rule; it does not prove every product attribute.

Use precise language:

• monitoring pass: the analyzed signal stayed within the approved monitoring rule;
• monitoring alarm: the analyzed signal crossed an alarm rule or the system could not make a valid evaluation;
• product acceptance: all contractual workmanship, dimensional, electrical and other required evidence is conforming;
• process release: the setup and controls are approved for production;
• shipment release: all product, alarm, nonconformance and documentation conditions are closed.

This vocabulary prevents an operator from overriding an alarm merely because the crimp looks acceptable, and prevents the buyer from treating a monitoring pass as a complete certificate of conformity.

Establish a baseline only from known acceptable production

A learned reference is only as strong as the samples and conditions used to create it. The baseline should be created after the terminal application and process setup are approved, using representative conforming crimps rather than whatever cycles happen to be available.

A documented baseline workflow can follow these steps.

Step 1: freeze the configuration

Record the product number, drawing revision, terminal and wire identifiers, seal or secondary component where applicable, applicator and tooling identity, press identity, software/program revision and applicable terminal manufacturer instruction.

Step 2: release the mechanical setup

Complete the specified setup checks. These may include crimp height and other measurements required by the released application specification. Use calibrated equipment and record the actual result, not only “pass.”

Step 3: collect representative conforming cycles

Collect cycles under stable production conditions across the normal operating variation that the approved process is expected to see. Do not deliberately widen the process to hide instability. The equipment maker's instructions and the supplier's validated procedure should govern collection quantity and conditions; this guide does not prescribe a universal sample count.

Step 4: confirm the reference samples

Verify the baseline samples by the product's required methods. Depending on the application, that may include visual inspection, crimp height, destructive pull, cross-section or another authorized test. A monitoring system should not learn from unverified output.

Step 5: challenge relevant error modes

Use controlled, authorized challenge samples to demonstrate sensitivity to selected errors. Protect tooling and people, and do not create unsafe or unapproved conditions simply to satisfy a checklist. The validation plan should state which conditions were challenged, how they were produced, and whether detection was repeatable.

Step 6: set and approve the monitoring rule

Document the selected evaluation mode, tolerance profile, analysis regions, alarm limit or other relevant parameters supported by the system. Record who approved the rule and why. Do not publish proprietary algorithms when a controlled program identifier and validation record are sufficient.

Step 7: lock and release

Apply access control so routine operators cannot silently change the baseline or limits. Release the monitor program together with the production traveler or electronic work order.

Step 8: retain a recoverable version

Back up the approved program and preserve its version, checksum or another unambiguous identity. If a parameter changes, the record should show the old and new versions, reason, validation and approval.

This sequence prevents the learning phase from becoming an undocumented operator action.

Define reject limits through validation, not a copied percentage

A tolerance value that works on one terminal-wire-tool combination may be ineffective on another. Signal magnitude, curve shape, normal variability and defect sensitivity change with the application and equipment. Therefore, do not insert a universal monitoring percentage into a multi-part RFQ.

The approved limit needs four forms of evidence:

1. Known-good distribution: the normal signal behavior for verified conforming production.
2. Challenge separation: evidence showing whether relevant error modes move the evaluated signal enough to be detected.
3. Repeatability over expected conditions: results after normal restarts, material lot changes and maintenance states included in the approved process.
4. Operational performance: a review of false alarms, confirmed defects and any escapes after release.

A tight limit can create repeated false rejects and encourage operators to bypass the system. A loose limit can pass meaningful process changes. The correct setting balances detection and stability for the specific application while preserving the product's separate acceptance criteria.

Use warning and reject actions only when their meanings are controlled

Some systems and procedures support more than one level of response. A warning band may trigger a setup check or review while a reject band stops production and segregates output. If the supplier uses both, the RFQ should ask:

• What parameter crosses each band?
• Does a warning unit remain eligible for shipment?
• Who reviews the warning?
• How many warnings trigger line stop or escalation?
• Can an operator clear the indication?
• Is the action and authorization logged?
• Does a power cycle or job change reset an accumulated warning condition?

A colored screen is not a reaction plan. Connect every displayed state to an explicit product and process disposition.

Do not use limit tuning to make alarms disappear

When a stable program starts producing alarms, the first action should be to investigate the process and monitoring system. Widening the limits without evidence can convert a useful signal into a decorative feature.

Require change control for:

• adjustment of alarm thresholds;
• change of analysis regions or features;
• replacement of the learned reference;
• software or firmware changes that affect evaluation;
• sensor, press or applicator replacement;
• program copying to another machine;
• disabling or bypassing the monitor;
• changes to result-retention settings.

The record should state the reason, before/after identity, validation performed, approver and effective date.

Verify the measurement chain

The monitoring result depends on more than the algorithm. The sensor, mounting, press mechanics, signal acquisition, software and application setup form a measurement chain. Procurement should request evidence that this chain is fit for the intended use.

Questions for the supplier include:

• What sensor or signal source is used, and how is it identified?
• How does the equipment detect a sensor or communication failure?
• What startup or reference check is required?
• What preventive maintenance applies to the monitor and press?
• How are repaired or replaced components revalidated?
• Is the program bound to a specific machine, applicator or job?
• How are time, user identity and record integrity controlled?
• What happens if data storage is full or unavailable?
• How are monitoring results backed up and retrieved?
• Which software or firmware revision generated the result?

Not every system requires the same calibration model as a dimensional gauge. Use the equipment manufacturer's instructions and the supplier's validated quality procedure. The RFQ should ask for traceable maintenance and verification evidence without inventing a calibration interval that does not apply to the system.

Relate monitoring to crimp height

Crimp height is a direct dimensional control for the compressed conductor area when it is defined by the terminal manufacturer or another design authority. TE's public paper describes crimp height as an effective quality metric and explains that crimp force monitoring and crimp-height measurement address different aspects of quality.

A strong production plan can use:

• crimp height to release and periodically verify the mechanical setup;
• force monitoring to observe each configured production cycle;
• destructive pull to assess mechanical behavior at an approved frequency;
• cross-section analysis when the authorized specification requires hidden-geometry evidence;
• visual inspection for features the force signal does not reliably resolve;
• electrical testing for circuit-level release.

Do not let a force-monitor pass erase a failed crimp-height result. Likewise, a conforming sampled crimp height does not justify ignoring a force-monitor alarm on another unit.

Relate monitoring to pull-force testing

Pull-force testing is destructive and normally sampled. It can reveal insufficient mechanical strength, but it does not necessarily distinguish every over-compression or geometry problem. TE's public comparison explains why pull force, crimp height and force monitoring should be understood as different metrics. The Molex Quality Crimping Handbook provides manufacturer-specific training guidance on crimp-height, pull and process capability checks.

When using the Molex handbook or another terminal manufacturer's manual, apply it only within its scope. Training examples and recommended frequencies are not universal requirements for unrelated terminals. The RFQ should name the actual terminal authority and approved values.

Relate monitoring to cross-section evidence

A cross-section can show internal geometry on a prepared sample. It is destructive, requires controlled preparation and evaluation, and cannot be performed on every delivered crimp. Use it for validation, first article, investigation or periodic verification where the approved plan requires it.

The related SINAWATTS guides on crimp tooling, pull and cross-section evidence and crimp cross-section geometry and void evidence help define those records. Force monitoring should link to them rather than replace them.

Build an alarm reaction plan before production

Every alarm creates a product-control decision. If the reaction is improvised, the monitor can generate data without protecting the shipment.

Use a documented sequence:

1. Stop or interlock as specified. Prevent continued production when the approved rule requires a stop.
2. Identify the alarmed unit. Preserve its serial, circuit, terminal position, cycle or container relationship.
3. Contain the exposure window. Segregate the alarmed unit and any additional units defined by the control plan, including units since the last accepted verification when applicable.
4. Preserve the original data. Do not overwrite the curve, result or program identity.
5. Inspect the product by the approved methods. Use visual, dimensional, destructive or other checks stated in the reaction plan.
6. Inspect the process. Check material identity, wire preparation, terminal feed, tooling, press, sensor and setup.
7. Classify the cause. Distinguish a confirmed product nonconformance, process change, monitoring-system problem and unexplained alarm.
8. Discipline the disposition. Route use-as-is, rework, repair or scrap decisions to the authorized role. A monitor operator should not approve an engineering deviation unless the contract grants that authority.
9. Reverify before restart. Complete the defined setup and challenge checks.
10. Bound affected output. Use traceability and evidence to determine what can be released.
11. Record the event. Link alarm, investigation, disposition, restart evidence and affected product identity.
12. Escalate recurring events. Define when repeated alarms require corrective action or customer notification.

The plan should also cover a monitor that is unavailable. If production may continue under a temporary alternative control, state who approves it, which increased checks apply, how output is segregated, and whether customer notification is required. “Monitor bypass” must never be an undocumented convenience.

Prevent alarmed parts from re-entering production

Physical and digital controls should agree. Consider a locked reject chute, cut-off action, barcode hold, container segregation or another method suitable for the equipment. The RFQ need not dictate a specific mechanism, but it should require the supplier to explain how the alarmed terminal or harness is prevented from being mixed with accepted output.

If the crimping machine automatically cuts or marks a rejected piece, request evidence that the function is active for the quoted configuration. If removal is manual, require a second verification or an equivalent control where product risk justifies it.

Make traceability useful for containment

A database with millions of curves is not useful if the buyer cannot connect a result to a delivered harness. Define the traceability unit and relationship before quotation.

A monitoring record may need:

On small screens, swipe or scroll sideways to read every column. Keyboard users can focus the table and use the arrow keys.

The required granularity depends on risk and commercial practicality. A low-volume critical assembly may justify connection-level traceability. A high-volume commodity assembly may use approved lot-level retention plus exception records. The buyer should state the intended containment unit rather than demanding “full traceability” without definition.

Decide what data travels and what remains retrievable

Separate the evidence into four buckets:

1. Quote evidence: proposed equipment, monitoring concept, known limitations and exceptions.
2. First-article evidence: baseline validation, challenge results, setup identity and sample verification.
3. Lot evidence: production summary, alarm count/status, required sampled results and release sign-off.
4. Retained evidence: detailed curves, cycle data, user logs, configuration history and investigation records available for audit.

A buyer rarely needs every raw curve attached to every shipment. Requiring unnecessary files can add cost and complicate review. Instead, specify retention duration, retrieval time, format, time zone, access control and how a record is protected from silent editing.

If raw data are unavailable because the proposed system does not support retention, the supplier should state that limitation in the quotation. The buyer can then decide whether lot summaries and exception evidence are sufficient.

Qualify the system with detection challenges and production evidence

The validation plan should demonstrate both detection and stability.

For detection, select realistic error modes through risk analysis. For stability, run the accepted process across the conditions included in routine production. Document:

• challenge sample definition;
• responsible engineer or quality authority;
• safe preparation method;
• reference sample verification;
• number and sequence of trials as approved by the plan;
• actual monitor response;
• independent confirmation of the introduced condition;
• limits or conditions of detection;
• conclusion and approval.

Do not present a single dramatic screen image as proof of robust detection. The evidence should show that the response is repeatable enough for the intended control.

For production stability, review false alarms and confirmed process findings over a defined observation period. The plan should explain how a new material lot, shift, restart or approved maintenance event is evaluated. A high false-alarm rate can indicate an unstable process, an unsuitable baseline, poor sensitivity selection or a measurement-chain problem. The answer should be investigation, not automatic limit expansion.

Compare inspection methods by the question they answer

Do not rank all methods on one “best” scale. Each method answers a different question.

On small screens, swipe or scroll sideways to read every column. Keyboard users can focus the table and use the arrow keys.

The control plan should map the product risks to this table. A method should be included because it closes a defined risk, not because its name sounds more advanced.

Ask suppliers questions that expose the real control

A structured bid form makes responses comparable.

Equipment and scope

1. What press, applicator and monitoring equipment are proposed?
2. Which terminal-wire combinations will be monitored?
3. Are all crimping stations covered, or only named stations?
4. What system or configuration limitation should the buyer know?
5. Can the system link results to unit, circuit or lot identity at the required level?

Baseline and limits

1. Who creates and approves the baseline?
2. Which verified samples are used?
3. What independent checks confirm the baseline samples?
4. How are relevant error modes challenged?
5. Which parameters are locked, and who may change them?
6. How is the monitoring program version identified?

Production operation

1. Does every configured cycle receive a valid result?
2. What happens to a cycle with missing or invalid data?
3. What interlock or removal control applies to rejected output?
4. Can production continue while monitoring is unavailable?
5. What alternative control and authorization apply if it can?
6. How are repeated warnings or rejects escalated?

Evidence and change control

1. Which fields appear in lot reports?
2. Are raw curves or detailed records retained?
3. How long are they retained, and how quickly can they be retrieved?
4. What events trigger baseline revalidation?
5. How are software, sensor, press, applicator and site changes controlled?
6. Which alarm and override events are visible to the buyer?
7. Which exceptions affect price, timing or feasibility?

Request a direct response or a written exception for every question. Marketing literature can support equipment identification, but it cannot replace the supplier's product-specific answer.

Apply a two-gate approval process

Gate 1: technical quotation

Before award, close:

• connection design authority;
• exact terminal-wire-seal configurations;
• proposed equipment and station coverage;
• required traceability granularity;
• complementary measurement and test methods;
• data retention and delivery expectations;
• proposed alarm reaction and bypass rules;
• exceptions and commercial assumptions.

At this gate, the supplier may not yet have an approved production baseline. It should describe the validation method and identify limitations honestly.

Gate 2: first-article or production readiness

Before routine production, approve:

• released product and process configuration;
• mechanical setup results;
• baseline sample evidence;
• error-mode challenge results;
• approved monitoring limits and program identity;
• access control;
• alarm reaction plan;
• record linkage and retrieval trial;
• revalidation triggers;
• final first-article disposition.

The battery cable first-article and change-control guide provides a broader configuration-control framework. Use the monitoring record as one part of that package.

A bounded worked example

Consider a buyer sourcing a harness with two crimped circuits. The released design identifies the terminal manufacturer, terminal part numbers, approved wires and application tooling. One terminal-wire combination has sufficient force signal for a validated production monitor; the other has not yet demonstrated separation between a relevant small-process change and normal variation.

A defensible plan could be:

• apply force monitoring to every production cycle of the validated combination;
• identify the monitor program, press and applicator on the lot record;
• use the authorized crimp-height target and frequency for setup and routine verification;
• retain cycle results and all alarm investigations;
• segregate an alarmed unit and the exposure window defined by the reaction plan;
• continue visual, crimp-height and sampled destructive verification required by the control plan;
• treat the second combination as not yet qualified for CFM and use its approved alternate controls;
• run a separate validation study before claiming CFM coverage for the second combination.

The buyer should not force a nominal “100% CFM” statement by applying a monitor where validation does not support useful detection. Honest partial coverage with strong alternative controls is more defensible than a site-wide equipment claim.

Example lot summary

A concise lot summary might include:

On small screens, swipe or scroll sideways to read every column. Keyboard users can focus the table and use the arrow keys.

This example does not supply numerical crimp dimensions, pull forces, sample intervals or alarm limits. Those values must come from the released design and validated control plan for the exact connection.

Define false rejects and escapes without hiding uncertainty

A false reject is a monitoring alarm where the authorized investigation finds the product and process conforming and identifies no product nonconformance. An escape is a product nonconformance that the released monitoring rule did not flag. Both require careful definitions because an unexplained alarm is not automatically false, and a defect found later may not have been within the monitor's validated detection scope.

Ask the supplier to trend:

• alarm count by program and station;
• confirmed product or process causes;
• monitoring-system faults;
• unexplained alarms;
• authorized false-reject dispositions;
• known escapes within the validated error-mode scope;
• baseline or limit changes;
• bypass time and output disposition;
• corrective actions.

The trend should drive review. It should not become a target that pressures personnel to reclassify alarms. A very low reject rate is not proof of a capable process if the challenge evidence is weak.

Set revalidation triggers

A monitoring setup is not valid forever. Define which events require review, partial challenge or full baseline replacement.

Candidate triggers include:

• terminal, wire, seal or splice change;
• drawing or application-specification revision;
• supplier or material construction change;
• applicator, die, locator or press change;
• major tooling repair, rebuild or adjustment;
• sensor replacement or relocation;
• software or firmware change affecting evaluation;
• monitor program or limit change;
• transfer to another machine or manufacturing site;
• significant maintenance;
• repeated unexplained alarms;
• confirmed escape within the claimed detection scope;
• long shutdown or restart condition defined by the control plan.

Not every event requires the same response. Build a decision matrix that states the required review and approval for each trigger. The important point is that the decision is recorded before affected product is released.

Preserve portability limits

A program created on one press should not be copied to another machine merely because the part number matches. Machine dynamics, sensor installation and process variation may differ. Require validation at the destination equipment and identify whether a baseline is machine-specific, applicator-specific or otherwise constrained.

Likewise, a program validated on one wire construction should not be applied to another conductor or insulation design without an authorized assessment.

Write the RFQ in testable clauses

The following structure can be adapted to a controlled request.

Clause 1: application identity

Clause 2: monitoring scope

Clause 3: baseline approval

Clause 4: detection validation

Clause 5: production action

Clause 6: complementary evidence

Clause 7: traceability

Clause 8: change control

Clause 9: evidence boundaries

These clauses establish auditable obligations while leaving application-specific numerical values in their proper controlled documents.

Connect monitoring evidence to the full harness release

A crimp is one connection within a harness. The final release may also need:

• component and wire traceability;
• splice and branch controls;
• contact insertion and retention verification;
• connector lock and seal checks;
• dimensional and routing inspection;
• continuity and miswire testing;
• insulation-resistance or dielectric testing where authorized;
• labeling and serialization;
• final visual inspection;
• packaging and preservation.

Use the cable harness continuity, insulation-resistance and hipot guide to separate electrical limits and records. The end-of-line fixture guide explains fixture and false-failure controls. Monitoring evidence should share configuration and traceability keys with those records.

Procurement checklist

Before awarding a monitored-crimp program, confirm:

• [ ] Each terminal-wire combination has an identified design authority.
• [ ] The exact application specification and revision are controlled.
• [ ] Proposed press, applicator, tool and monitor identities are known.
• [ ] Covered and uncovered crimp operations are listed.
• [ ] The baseline workflow uses independently verified conforming samples.
• [ ] Risk-based error-mode challenges are planned.
• [ ] Monitoring limits are application-specific and approved.
• [ ] Operators cannot silently alter or bypass the released program.
• [ ] Warning, reject and invalid states have defined actions.
• [ ] Alarms create traceable containment and disposition records.
• [ ] Crimp height and other complementary checks remain in the plan.
• [ ] Unit or lot linkage matches the buyer's containment need.
• [ ] Raw and summary data retention are defined.
• [ ] Equipment and program changes trigger documented review.
• [ ] Site transfer and program portability are controlled.
• [ ] The lot release closes every alarm or cites an approved deviation.
• [ ] Price, MOQ and lead-time assumptions are stated separately from quality evidence.

When these points are ready, identify the component family in the SINAWATTS catalog. Use the Send a crimp-force-monitoring evidence RFQ to send the released terminal and wire identifiers, drawing revision, expected quantities, monitoring scope, traceability level, required evidence and delivery terms. SINAWATTS should confirm only capabilities, equipment, certifications, materials and commercial terms supported by the specific quotation.

Evidence boundaries

This guide provides a procurement framework based on publicly accessible manufacturer information checked on 2026-10-08. It does not prescribe a terminal design, alarm threshold, sample interval, crimp dimension, pull value, cross-section criterion or universal monitor setting.

The TE material describes general relationships among crimp height, force monitoring and pull testing. The Komax pages describe capabilities offered by named product families. The Molex handbook provides manufacturer training and process guidance. None of those sources proves that a specific supplier uses that equipment, follows that procedure or can detect a particular defect on the buyer's connection.

A supplier's capability must be demonstrated for the quoted manufacturing site, equipment and terminal-wire configuration. Certifications, equipment ownership, installed software, production capacity, materials, inspection capability, stock, price, MOQ and lead time must be confirmed in the quotation and released order. Challenge samples must be prepared safely under an approved validation plan. Customer drawings, terminal manufacturer specifications, applicable workmanship standards and legal requirements remain controlling.

Buyer FAQ

Is crimp force monitoring the same as measuring crimp height?

No. Crimp force monitoring evaluates a production-cycle signal against a validated rule. Crimp height measures a physical dimension at a defined location using an authorized target and tolerance. The two methods can complement each other, but neither result should overwrite a failure from the other.

Can force monitoring replace destructive pull testing?

Not automatically. Pull testing answers a different mechanical question on a destructive sample. Whether it is required, at what frequency, and to what limit must come from the terminal manufacturer, customer specification or approved control plan. A force-monitor pass is not a substitute unless the design authority has explicitly changed the requirement based on valid evidence.

Does a monitor detect every missing strand?

No universal claim is defensible. Detectability depends on the terminal-wire combination, process variation, sensor and algorithm configuration, baseline, threshold and number or location of missing strands. Require controlled challenge evidence for the error mode and configuration that matter to the product.

Should the RFQ specify one reject-limit percentage for all terminals?

No. A single percentage copied across different terminal-wire-tool combinations can cause false rejects on one application and weak detection on another. Require application-specific limit validation, approval and change control.

What is the difference between a warning and a reject?

The supplier's approved procedure defines the states. A warning may call for review while a reject normally creates containment and disposition. The RFQ should state whether a warned unit may be released, who reviews it, when repeated warnings stop production, and how every action is recorded.

What happens if the monitoring system is offline?

Follow the approved contingency plan. It may require production stop, customer notification, temporary increased inspection or segregated production under authorized alternative controls. The supplier should not bypass the system and release output without documented authorization and evidence.

Does storing every force curve provide full traceability?

Not by itself. The record must connect the curve or evaluated result to the correct part, circuit or terminal position, material, equipment, program revision, work order and shipment at the traceability level agreed in the contract. Unlinked data cannot support reliable containment, disposition or shipment release.

How long should monitoring data be retained?

There is no universal period in this guide. Set the duration from customer, regulatory, warranty and product-risk needs, then state it in the RFQ. Also define retrieval time, file format, access control, backup and protection from silent editing.

When should a monitoring baseline be rebuilt?

Use an approved revalidation matrix. Product, terminal, wire, tooling, press, sensor, software, program, site or major maintenance changes are common review triggers, as are unexplained alarm trends and confirmed escapes. Some events may need a documented review; others may require full new baseline and challenge evidence.

Can one program be copied to another press?

Do not assume portability. Press dynamics, sensor installation, applicator condition and signal behavior can differ. Validate the program on the destination equipment and record its approved scope before releasing product.

What should be included in a first-article monitoring package?

Include configuration identity, released application source, press/applicator/tool/monitor identities, setup measurements, verified baseline sample evidence, challenge matrix and results, approved limits/program revision, reaction plan, traceability demonstration, complementary test results, exceptions and approval. Link the package to the actual first-article units.

How should a buyer review a high false-alarm rate?

Treat it as a signal requiring investigation. Review process stability, material and setup variation, baseline quality, sensor condition, tooling, analysis settings and data integrity. Do not simply widen the limits. Record the cause, validation and authorization for any change.

Does “100% monitored” mean every delivered crimp is acceptable?

No. It means each configured production cycle is intended to receive a monitoring evaluation, subject to system validity and record integrity. Shipment acceptance still depends on the full workmanship, dimensional, electrical, material, alarm-disposition and release requirements.

Is a crimping machine brochure enough evidence for an RFQ?

No. A brochure can confirm publicly stated equipment capabilities. The buyer still needs the quoted model and configuration, station coverage, validation method, monitoring program identity, alarm reaction, data linkage, retention and product-specific evidence from the proposed manufacturing site.

How should reworked or recrimped connections appear in the data?

The original alarm or nonconformance should remain visible. Record the authorized instruction, affected connection, person, new cycle or result, repeat inspections/tests and final disposition. A successful later cycle should not silently replace the original event.

Which source should define crimp dimensions and pull requirements?

Use the released terminal manufacturer application specification or another authority explicitly approved by the design owner. Do not copy values from a handbook example, unrelated terminal family or monitoring software screen.

What should a buyer send to receive a comparable quotation?

Send the controlled drawing and BOM, terminal and wire identifiers, required standards, expected production volume, proposed traceability level, required monitoring scope, complementary inspections/tests, data-retention needs, first-article package, change-notification rules and delivery terms. Ask bidders to list exceptions and assumptions line by line.

Official and manufacturer sources checked on 2026-10-08

1. TE Connectivity, Crimp Height: Employing an Effective Crimp Quality Metric — public overview of crimp height and quality-control relationships.
2. TE Connectivity, Crimp Height: Meeting Contemporary Quality Standards — manufacturer white paper comparing crimp height, crimp force monitoring and pull-force evidence, including method boundaries.
3. Komax QpLite 2 — official product page describing individual-crimp monitoring, configurable tolerance profiles and data-logging options.
4. Komax Cable and Harness Quality Tools — official overview of complementary quality-tool categories.
5. Molex Quality Crimping Handbook — terminal manufacturer's public training reference for crimp process, measurement and capability concepts; apply within the relevant product and document scope.
6. IPC Meet Your Standards: IPC/WHMA-A-620 — official public overview of cable and wire-harness workmanship scope and stated limits.

Public pages, document revisions and product configurations can change. Recheck the applicable manufacturer document and equipment configuration when the RFQ and production package are released.