Incoming Quality Control for LED Therapy Devices — What OEMs Must Test Before Production Starts
The most expensive place to find a defective component is after it has been soldered onto a PCB. Incoming Quality Control is not a cost center — it is the first line of defense in a manufacturer’s quality system.
In a typical LED therapy device, there are between 200 and 600 individual component types across the BOM — LEDs, drivers, PCBs, sensors, cables, connectors, enclosures, silicone pads, adhesives, packaging materials, and labels. Each of these components has specifications. Each specification has a tolerance. And each tolerance deviation — if it reaches the production line undetected — can result in a defective finished device.
Incoming Quality Control (IQC) is the process of verifying that incoming components meet their specifications before they are released to production. For LED therapy devices, where component defects directly affect optical performance, electrical safety, biocompatibility, and regulatory compliance, IQC is not optional — it is the foundation of everything that follows.
The question is not whether IQC should be done. The question is what to test, how to test it, how many to test, and how to handle the results.
This article is written from the perspective of an LED therapy OEM manufacturer — sharing how we have designed our IQC program, what we test, and what we have learned about where IQC pays for itself in reduced production costs and improved outgoing quality.
Why IQC Is Different for LED Therapy Devices Than General Electronics
Most general electronics OEM factories have an IQC process. But LED therapy devices create specific IQC demands that generic electronics IQC programs often miss:
Optical components require specialized measurement equipment — wavelength, radiant flux, and spectral bandwidth of LEDs cannot be verified with a multimeter or visual inspection. A production IQC without an integrating sphere, spectroradiometer, or calibrated photodiode will not catch LED bin drift, wavelength mislabeling, or radiant flux degradation.
Biocompatibility requirements create substance-level verification demands — skin-contact materials (silicone pads, mask seals, head straps) must meet ISO 10993. This means verifying not just the material type, but the specific grade, filler content, and additive composition — information that must come from the material supplier’s certificate of composition.
Regulatory traceability creates documentation demands — components used in a medical device must be traceable to the supplier, lot, and manufacturing date. If a field issue occurs, the manufacturer must be able to identify which component lots were affected. A factory that does not lot-trace incoming components cannot meet this regulatory requirement — and cannot conduct an effective recall if needed.
Thermal and electrical co-dependence — LED output, driver efficiency, and thermal management are interdependent. A driver that meets electrical specs at room temperature may fail thermal performance at operating temperature. The IQC for LED therapy devices must account for these co-dependencies — which requires understanding the system-level design, not just the component-level specification.
The Five Component Categories That Require IQC in LED Therapy Devices
Category 1 — LED Emitters (Highest Risk)
LEDs are the highest-risk incoming component for LED therapy devices — because their performance directly determines whether the device delivers the clinical effect that the brand is marketing.
What to test:
Wavelength (λp): Measure the peak wavelength of a sampled number of LEDs from each incoming reel using a spectroradiometer or spectrometer. Compare against the purchase specification (e.g., 630 nm ±3 nm). Reject the lot if any measured wavelength falls outside the specified tolerance. Note: a lot where 5% of LEDs are at the bin edge but within tolerance is still a borderline lot — flag it for additional sampling before releasing to production.
Forward voltage (Vf): Measure Vf at nominal drive current on the same sampled LEDs. LEDs from the same bin should have Vf within ±0.2 V of the bin center. Unexpectedly high or low Vf indicates possible die attach issues, wire bond problems, or a mis-binned lot.
Radiant flux (Φe): Measure radiant flux per LED at nominal current using an integrating sphere. Compare against the bin minimum (typically expressed in mW or W/sr). LEDs below the bin minimum should be rejected. Also check for abnormally high radiant flux — which may indicate the LED is being driven above its rated current at the test condition, masking thermal performance problems.
Spectral bandwidth (FWHM): For narrowband applications (e.g., 660 nm for a specific CCO absorption peak), verify the full width at half maximum (FWHM) is within specification. Excessively wide bandwidth reduces spectral purity and may indicate the bin is mixed.
Sampling plan: AQL General Inspection Level II, with tightened inspection (reduced AQL from 1.0 to 0.65) for any new supplier or new lot from an existing supplier. Minimum sample size: 20 LEDs per reel, tested across the reel (beginning, middle, end positions).
Equipment required: Integrating sphere with spectrometer or spectroradiometer (initial test), calibrated photodiode with narrowband filter set (production screening), LED test fixture with constant current source and thermal control stage.
Category 2 — LED Drivers and Power Supplies
The LED driver determines electrical safety, EMC performance, and LED operating conditions. A defective driver can cause hipot failure, excessive EMI emissions, or premature LED degradation.
What to test:
Output voltage and current accuracy: Verify output voltage and current under nominal load at 120 VAC and 230 VAC input. The driver should deliver the rated current within ±5% across the specified input voltage range. Reject drivers with output variation greater than ±5% — they will cause LED output inconsistency across units.
No-load voltage: Measure the driver output voltage with no LED load connected. This voltage must be below the maximum LED reverse voltage rating. A driver that outputs excessive no-load voltage will destroy LEDs when first powered without a load protection circuit — and may not have that protection.
Input power and efficiency: Measure input power at nominal load. Efficiency should be within the datasheet specification (typically 85–92% for a well-designed LED driver). Drivers with efficiency significantly below spec will generate excess heat — degrading LED performance and potentially causing safety issues.
Isolation (hipot pre-screen): Perform a scaled-down hipot test on the driver input-to-output isolation at the incoming stage — not a full IEC 60601-1 hipot, but a 500 VDC dielectric withstand check to screen for transformer winding insulation failures. Reject any driver that fails this pre-screen.
EMC pre-screening: For new driver designs or new suppliers, perform conducted emissions pre-screening using a spectrum analyzer and LISN (Line Impedance Stabilization Network). Drivers that fail pre-screening at incoming IQC are not ready for production — and will cause expensive failures at the EMC laboratory.
Sampling plan: 100% incoming inspection for new driver designs or new suppliers. AQL General Inspection Level II (AQL 1.0) for production-repeated orders from qualified suppliers with a clean history.
Equipment required: AC power source (with 120 VAC and 230 VAC capability), electronic load, power analyzer or wattmeter, hipot tester (scaled), spectrum analyzer with LISN (for EMC pre-screening).
Category 3 — PCBs (Printed Circuit Boards)
PCBs are the structural foundation of the device’s electrical system. Failure modes include incorrect trace widths (causing safety hazards), insufficient creepage/clearance distances (causing hipot failure), and solder mask defects (causing short circuits).
What to test:
Visual inspection: Check for scratches, voids in solder mask, unwanted copper residues, and incomplete via plating. Use a microscope or magnifying lamp at minimum 10× magnification. A single surface defect on a high-voltage trace can cause a hipot failure.
Dimensional verification: Measure critical dimensions — particularly creepage and clearance distances between primary (line-voltage) traces and secondary (low-voltage or accessible) traces. Verify against the PCB design file. For IEC 60601-1 compliance, primary-to-secondary clearance must meet minimum requirements (typically 4 mm for 250 VAC working voltage, pollution degree 2). A PCB that meets consumer electronics spacing requirements but not medical device requirements will fail at the testing laboratory.
Solderability: Perform a solderability test on PCB pads (dip a representative pad in flux and solder — the solder should wet the pad uniformly within 3 seconds). This is especially important for PCBs that have been in storage for more than 6 months — oxidation can make pads unsolderable without rework.
Electrical continuity: Verify that all traces that should be connected are connected, and traces that should be isolated are isolated. Use a flying probe tester or a simple continuity tester for simple boards.
Sampling plan: First article inspection (FAI) on every new PCB revision — 100% dimensional verification. AQL General Inspection Level II for production-repeated PCB lots from the same revision.
Equipment required: Microscope or magnifying lamp (10–40×), digital caliper and micrometer, dimensional reference drawings, solderability test kit, flying probe tester or multimeter continuity mode.
Category 4 — Skin-Contact Materials (Silicone, Polymers)
Silicone pads, mask seals, and any material that contacts skin require the most rigorous IQC of any component category — because biocompatibility is non-negotiable, and material substitution is the most common fraud vector in this category.
What to test:
Material identification by FTIR: Perform Fourier-transform infrared spectroscopy on a sample from each incoming lot of silicone or polymer material. Compare the spectrum against the approved reference spectrum for the approved material grade. Any deviation indicates a formulation change — reject the lot and contact the supplier.
Hardness verification: Measure Shore hardness (for silicone, typically Shore A 30–60 for soft pads) using a durometer. Verify against the approved specification ±5 A. Lot-to-lot hardness variation can indicate filler content changes that affect biocompatibility and mechanical durability.
Color verification: For colored silicone (skin-toned, white, or tinted for NIR transmission), measure color using a colorimeter. Express as CIE Lab* values and calculate ΔE (color difference) against the approved reference. ΔE > 2.0 indicates a formulation change — reject or require supplier explanation.
Thickness and dimensions: Measure pad thickness and dimensional tolerances against the specification. Silicone pads that are 0.2 mm thinner than specified may have insufficient structural integrity for repeated use — and may fail adhesion or tear in production assembly.
Material certificate review: Every incoming lot must be accompanied by a Certificate of Conformance (C of C) listing the material grade, lot number, manufacturing date, and a statement of compliance with the specified standard (e.g., ISO 10993 grade). Review this certificate before releasing the material — not after.
Sampling plan: AQL General Inspection Level II for dimensional and hardness testing. FTIR testing on every new lot — regardless of supplier history. The cost of FTIR testing (approximately USD 10–30 per sample, 30 minutes) is negligible compared to the cost of a batch of devices failing ISO 10993 biocompatibility testing.
Equipment required: FTIR spectrometer, Shore A durometer, colorimeter (with CIE Lab* software), digital caliper, material certificate review protocol.
Category 5 — Cables, Connectors, and Wiring Harnesses
Cables and connectors are frequently treated as low-risk components — and therefore receive insufficient IQC. This is a mistake. Cable failures (broken conductors, intermittent connections, insulation breaches) are among the most common causes of field failures and safety incidents in powered skin-contact devices.
What to test:
Visual inspection: Check for jacket damage, kinking, insufficient strain relief at terminations, and connector housing integrity. Any visible defect on the outer jacket of a cable that will be flexed during use should be rejected.
Continuity and resistance: Measure the resistance of each conductor in the cable assembly. Reject any conductor with resistance more than 10% above the specification — excessive resistance causes heat buildup at the connector terminations and can cause safety issues.
Hipot pre-screen: For power cables (AC input cables), perform a scaled hipot test on the cable assembly — checking that insulation resistance between the live conductor and the outer jacket meets the specification (typically >10 MΩ for 500 VDC test).
Connector mating cycle test: For connectors specified for a minimum number of mating cycles (e.g., 500 cycles for a USB charging port), verify the connector’s mechanical specification. The incoming IQC does not need to perform a full cycling test — but the production process must track mating cycles and replace connectors that exceed the rated limit.
Sampling plan: AQL General Inspection Level II with AQL 1.0 for major defects (continuity failure, insulation failure) and AQL 2.5 for minor defects (minor jacket cosmetic defects).
Equipment required: Multimeter (continuity and resistance), hipot tester (scaled), magnifying lamp.
Building the IQC Sampling Plan — How Many to Test
The number of units to test from each incoming lot is determined by the lot size and the acceptable risk level.
ANSI/ASQ Z1.4 (formerly MIL-STD-105E) is the standard reference for attribute sampling plans in manufacturing. For most LED therapy device components:
- Critical components (LEDs, drivers, skin-contact materials): Use Tightened inspection (reduced AQL of 0.65) or 100% inspection for new suppliers or new component lots
- Major components (PCBs, cables, connectors): Use Normal inspection (AQL 1.0) for qualified suppliers with established quality history
- Minor components (packaging materials, labels): Use Normal inspection (AQL 2.5) — the cost of a packaging defect is low; excessive inspection of low-risk items wastes resources
Lot Tolerance Percent Defective (LTPD) is an alternative approach — instead of accepting a small percentage of defects (AQL), you define the maximum defective rate you are willing to accept in any single lot, and design the sampling plan around detecting that lot with a specified probability (typically 90%).
For LED emitters from new suppliers, we recommend an LTPD of 1.0% — meaning the sampling plan must have a 90% probability of detecting a lot with 1.0% defective LEDs. This typically requires a sample size of 200+ LEDs per lot.
What to Do With IQC Failures
Hold the lot: Any lot that fails IQC must be placed on incoming hold — physically segregated from production material — until the failure is dispositioned. Do not release a failed lot to production “while you investigate.”
Supplier notification: Notify the supplier within 24 hours of the failure. Provide the IQC test report — not just “the lot failed.” Include specific measurements, lot number, and the specification that was not met.
Supplier response requirements: Require the supplier to respond with: (1) identification of the root cause, (2) containment actions for the affected lot and any shipped lots, (3) corrective actions to prevent recurrence. A supplier who responds only with “we will retest and send new samples” has not identified the root cause — accept only a response that includes systemic corrective action.
Disposition of failed lot: Options are return to supplier, downgrade to a less-critical application (if the defect is minor and does not affect safety), or rework and re-inspect. For LED therapy devices, do not use defective LEDs in production — even at a lower irradiance grade — without explicit written authorization from the brand and a re-verification of optical performance.
The IQC Documentation System — Why Records Matter
Every IQC inspection must generate a record — not just a pass/fail result on a worksheet that disappears after the lot is released.
Required IQC records:
- Lot number and supplier identification
- Date of inspection and inspector ID
- Sampling plan used (reference to ANSI/ASQ Z1.4 table)
- Actual measurements taken (not just “passed”)
- Number of defects found (if any)
- Disposition (released, held, returned)
- For held lots: root cause and corrective action from supplier
Retention: Retain IQC records for a minimum of 3 years for non-regulated components, and for the device’s service life plus 2 years for components used in medical devices. This is not just good practice — for ISO 13485-certified manufacturers, it is a regulatory requirement.
IQC data as a supplier management tool: The records from IQC — defect rates by supplier, by component category, and over time — are the raw data for the supplier scorecard. A supplier’s IQC performance trend over 4+ quarters is the most reliable predictor of their future quality performance.
How We Run IQC at RainbowDO
RainbowDO’s IQC program covers all five component categories described in this article. Here is how it operates in practice:
LED emitters: 100% incoming inspection using an integrating sphere with spectrometer — wavelength, Vf, and radiant flux measured on a sample of 30 LEDs per reel, taken from beginning/middle/end positions. New suppliers require FAI (100% inspection) for the first 5 lots. All lots lot-traced to supplier manufacturing date.
LED drivers: 100% incoming inspection for all new driver designs and new suppliers. For established suppliers, AQL General Inspection Level II (AQL 1.0). Each driver tested at 120 VAC and 230 VAC input for output voltage/current accuracy and efficiency. Scaled hipot pre-screen on 100% of drivers.
PCBs: First Article Inspection on every new PCB revision — dimensional verification, creepage/clearance measurement, solderability test. AQL inspection for repeated production lots.
Skin-contact materials: FTIR verification on every new lot — regardless of supplier quality history. Hardness and color verification on AQL sampling. Full material certificate review with lot traceability.
Cables and connectors: AQL inspection with hipot pre-screen on power cables.
Documentation: All IQC records maintained in our quality management system (Q-Pulse), accessible by lot number, supplier, and date. IQC data feeds the quarterly supplier scorecard. CAPA opened within 4 hours for any critical component failure.
Certifications: ISO 13485, MDSAP, ISO 9001.
📧 layla@rainbowdo.com | WhatsApp: +86 135 9032 9742
Incoming Quality Control — Common Questions
Q1: We are a small brand with a first OEM order. Can we require our factory to run IQC on components, or is this something we have to do ourselves?
The short answer: the factory’s quality system must include IQC for components used in your product. If the factory does not have an IQC program, it is a significant quality system gap — and you should either require them to implement one before your first order, or factor the expected higher defect rate into your quality agreement. As the brand, you own the product compliance. You cannot fully delegate quality system design to a factory without a verified quality system.
Q2: What is the difference between IQC and incoming inspection?
In practice, these terms are often used interchangeably. Technically, incoming inspection can refer to any verification of incoming materials — including a simple count and visual check. IQC implies a systematic, measurement-based verification against specifications — with defined sampling plans, acceptance criteria, and records. For LED therapy devices, “incoming inspection” should mean the full IQC program described in this article — not a visual count.
Q3: How do we know if the factory’s IQC is actually working?
The most reliable indicator is the trend in your defect rate reported at incoming inspection by your own inspection agency (if you use one) or at your customer’s incoming inspection. If defect rates are decreasing over 3+ consecutive orders, the IQC program is working. If defect rates are flat or increasing despite the factory’s claims of IQC implementation, request their IQC records — specifically the lot-by-lot defect rates by component category. A functioning IQC program has data. A non-functioning IQC program has a binder that says “IQC” on the cover.
This article is written from the perspective of an LED therapy OEM manufacturer with an ISO 13485-certified quality system — sharing how our IQC program is designed and why each element is included. The standards referenced (ANSI/ASQ Z1.4, ISO 10993 series) are publicly available. The specific IQC requirements for any product should be defined in the brand’s quality agreement with the OEM, tailored to the device’s risk profile and target market regulatory requirements.