THE CABLE FEEDSTOCK BEHIND OUR ASSEMBLIES.

Cable type is the part of the BOM where the assembly's reliability is decided. Choose a stranded conductor class wrong and the harness fails at year three when bend cycles accumulate; choose a shielding stack wrong and the EMI test fails at site acceptance after the customer has already paid the line-set bill. The cable feedstock decision is upstream of the connector decision, upstream of the test plan, upstream of the field-return curve at month nine — and yet most CM proposals treat cable selection as a procurement check-box rather than an engineering choice. We maintain a material library of every cable type listed below, audited by lot at receiving (UL 1581 strand-count verification, IMDS submission auto-generated for IATF customers, RoHS/REACH declarations refreshed per BOM line) and traceable through MES to final test. Specialty cables in particular — TC-ER power, blast-detection VOD, 10 M flex robot cable, FFC ribbon at 0.5 mm pitch, GMSL2 coax for ADAS, EDP for display interconnect, CAN-bus drag-chain rated, drag-chain compliant power+signal hybrid — are held in on-site stock rather than ordered ad-hoc per program. The cost of a cable feedstock substitution mid-program is not the cable itself; it is the requalification cycle (typically 3-6 weeks for a Tier-1 IATF customer), the FAI re-pull on the new lot, and the customer trust write-off. Our policy: every cable type on this hub has 2-3 pre-qualified approved equivalents already on the material card, so substitute swaps within an active program do not trigger a requalification cycle, only a delta PPAP. Pick the cable feedstock your design calls for and you will see, on each cable type page, the UL files that apply, the typical insulation/jacket compound, the temperature and bend-cycle envelope, the EMI shielding stack we run for that cable family, the test plan we run at receiving, and a representative finished assembly that uses it. If the design is not yet locked, send the drawing and we will pre-validate the cable choice against the application's electrical, thermal, mechanical, and chemical-exposure envelope before the connector decision is made — that is a free first-round review with no NDA required, returned inside 24 working hours by the engineering desk.

Insulation system selection (PUR vs PVC vs TPE vs silicone)

Insulation and jacket compound selection is the upstream decision that locks the cable's temperature envelope, oil resistance, flammability rating, and flex-life ceiling for the life of the program. Polyurethane (PUR) is the premium continuous-flex jacket: -40°C to +80°C operating range, 5-10 million flex cycles, intrinsic oil and ozone resistance, and the standard for drag-chain robotic and CNC programs. PVC is the volume default at 60-90°C operating range, low cost, but embrittles below -15°C, leaches plasticizers when warm, and cracks in continuous-flex applications past 1 million cycles. TPE (thermoplastic elastomer) sits between PUR and PVC: -40°C to +90°C, 2-3 million flex cycles, broader chemical resistance than PVC, and recyclable end-of-life — favored on EU programs subject to RoHS and end-of-life vehicle directives. Silicone is the high-temperature specialist: -55°C to +180°C continuous, 200°C peak, but soft jacket abrades fast and offers no oil resistance — a heating-element or industrial oven cable, not a drag-chain cable. FEP and PTFE step up further (-65°C to +200°C, with PTFE at +260°C peak) for aerospace and high-temperature instrumentation programs but at five-to-ten-times the jacket cost of PUR. Flammability ratings layer on top: UL VW-1 vertical flame test for general-purpose, UL FT-2 for industrial, UL FT-4 for tray cable, and IEC 60332-3 Cat A bundled-cable test for control cabinets. Mismatching the flammability rating to the install environment is a UL listing failure that pulls the entire panel off market. We hold UL E-File references for every approved jacket compound on the material card and verify per-lot at receiving against the cable's printed legend; one reel in fifty goes to lab cross-section to confirm jacket thickness and compound integrity against the spec sheet.

Shield architecture: foil + braid vs braid only vs spiral wrap

Cable shielding is the second decision after jacket compound, and the three viable architectures — foil + braid, braid only, spiral wrap — have non-overlapping use cases that the BOM does not always make clear. Foil-only shielding (aluminum-mylar tape with a drain wire) provides 100% optical coverage but only 50-70% effective coverage at the EMI gasket gap and degrades under flex; right for static signal cables in benign environments but fails on robotic or drag-chain programs. Braid only (tinned-copper braid, typically 80-90% optical coverage) is mechanically robust under flex and handles low-frequency EMI well, but leaves coverage gaps at high frequency where the braid weave geometry creates aperture loss; it is the standard for power and motor-drive cables. The combination of foil + braid is the gold standard for ADAS coax (GMSL2, FAKRA), Ethernet, and any cable where high-frequency EMI immunity is the gating spec; the foil closes the high-frequency apertures, the braid carries the bulk shield current at low frequency, and the architecture survives flex. Coverage matters once a number is on the drawing — under 80% braid coverage the cable behaves as if it has no shield above 100 MHz, and the CISPR 25 or FCC Part 15 emissions test will fail at site acceptance. Spiral-wrap shielding (a single layer of helically-wound shield wires) is a legacy approach that lives on in retractable cables and high-flex industrial robot programs where braid would crack at the spiral pitch — 60-75% effective coverage, mechanically softer than braid, but its high inductance under fast switching makes it unsuitable for VFD or PWM motor drives. We hold material-card references for foil + 85% tinned braid as the default on every signal cable and call out spiral wrap explicitly only when e-chain bend radius forces the substitution.

AWG sizing for ampacity vs voltage drop

The conductor-gauge decision is the most over-simplified line on the cable BOM, and field-return data confirms it: about half of 'cable runs hot' complaints trace to undersized conductors picked from an ampacity table without the voltage-drop check. NEC Article 310 publishes ampacity tables for conductors at 60°C, 75°C, and 90°C insulation ratings, and the rated current is the maximum the conductor can carry without exceeding its insulation temperature limit at a 30°C ambient with three current-carrying conductors in raceway. That is the ceiling, not the design point. The design point is voltage drop: NEMA's 3% conductor-voltage-drop rule (or the NEC Section 215.2(A)(1) Informational Note recommending 3% feeder + 5% total) is what keeps the load operating in spec on long cable runs. A 30 A circuit at 12 V running 30 ft round trip on 12 AWG is comfortably under ampacity but loses about 3.6% to conductor voltage drop — out of spec on most VFD or servo loads. The same circuit on 10 AWG drops to 2.3% and meets the rule. Bundle derating compounds the calculation: NEC Table 310.15(C)(1) requires reducing rated ampacity by up to 70% when 7-9 conductors share a raceway. Ambient temperature derating layers in for installations above 30°C, and conduit fill rules cap the usable cross-section at 40% on a three-conductor pull through rigid conduit. Our DFM round on every power cable assembly returns the AWG sizing with both ampacity and voltage-drop arithmetic shown, with bundle derating and ambient correction applied if the application warrants — on roughly 30% of incoming drawings we recommend a one-size-up conductor and document the wattage-loss savings against annual run hours and the energy-cost delta over the program lifecycle, with the cost-pay-back window quantified per the customer's local utility rate.