Publish Time: 2026-11-02 Origin: Site
Hanwha SMT mounter evaluation · Published by I.C.T · For production, process and manufacturing engineers
When engineers compare Hanwha SMT placement equipment, the conversation lands on speed, accuracy and price. Those describe the machine on the day it ships. What decides whether it is still earning its keep three years later is different: can you get parts when something fails, is the machine configured to your real product rather than a brochure, and will the supplier install it, program it and train your operators so it runs? Evaluating Hanwha mounter availability, configuration and service support answers those questions before purchase. The hardware context is the Hanwha pick and place machine range, and this guide turns the three pillars into a checklist for any candidate.
The article is three pillars. The first two chapters cover availability — spare parts, lead time, local stock, uptime and mean-time-to-repair. The middle two cover configuration — feeders, nozzles and heads, then programming and line integration. The last two cover service support — installation, commissioning and acceptance, then training and maintenance. It closes with a checklist and FAQs. Every example figure is explicitly hypothetical, not a quoted performance or customer result.
The headline specification is a snapshot: a placement rate under ideal conditions, an accuracy figure, a component list. It is real, but only one input to a decision. The other inputs decide total cost of ownership and uptime over the five to seven years you will own the machine. A mounter slightly slower on paper but always available, correctly configured and well supported will out-produce a faster one idle waiting for a part, fighting a wrong nozzle set, or blocked behind an unskilled changeover. The three pillars — availability, configuration and service support — are the parts of that picture the spec sheet leaves out.
Price is paid once; availability, configuration and support are paid for — or lost — every week the machine is on your floor. A cheaper machine with a two-week lead time on a line-stopping part can cost more in one stopped day than the price difference to a better-supported alternative. A machine configured only generically, with nozzles and feeders chosen for a demo board, rarely hits its nominal rate on your product. A machine installed without commissioning and training underperforms for months while everyone learns by error. None of this appears in the quotation, yet all of it shapes the number that matters: good boards per month at the quality the customer accepts.
By the end you should walk into a supplier meeting with a structured evaluation: a way to classify spare parts and set a lead-time target, a formula to turn vendor reliability claims into an availability budget, a configuration checklist tying feeders, nozzles and heads to your BOM, a programming and integration checklist, and an installation, training and maintenance checklist that becomes your acceptance document. Each chapter closes with questions to put to the supplier and items to write into the order, and every worked example is marked hypothetical so it illustrates the arithmetic without pretending to be a quoted performance. This is the discipline the Hanwha mounter selection guide for PCB lines also applies, here narrowed to availability, configuration and support.
Availability begins with parts. A mounter is a system of mechanical, pneumatic, electronic and software elements, any of which can fail. What separates a brief stop from a multi-day outage is whether the failed part can be sourced quickly — from your shelf, a local distributor, or only the OEM overseas. Evaluating spare-parts availability means asking exactly which parts are stocked, where, and how long a critical failure really takes. Even a value-oriented model such as the SM481 Plus low-cost SMT mounter stands or falls on this question in daily production.
Not all parts are equal. A failed head module stops the whole machine; a worn nozzle reduces yield gradually; a spare cable is an inconvenience. Classifying by failure impact tells you what must be on your shelf and what can wait. The categories are line-stopping critical parts, high-use consumables, scheduled wear parts and optional accessories — each with a different stocking rule, because over-stocking convenience items while under-stocking critical ones is a common, expensive mistake. Drive the classification from your actual configuration, not a generic list, because the critical parts depend on what the machine contains.
"Spares are available" without a number is not a commitment. For each critical and high-use part, extract the stocking location, the typical lead time, and whether the quoted time is the part alone or a technician-assisted repair. Ask for a distribution — typical and worst case — because a "48-hour" promise that means "after the overseas depot confirms stock" is a different risk from one backed by a local warehouse. Hypothetically, a critical module with a 5-day regional lead and a 1-in-10 stockout whose worst case is 20 days implies about 1.5 extra downtime days per event; a local hub cutting the worst case to 8 days drops that to 0.3. The figure is illustrative, but the structure turns "available" into a comparable, contractible item.
Hold on-site the critical parts with the longest lead time and the consumables with the highest burn rate; rely on the supplier for the rest — but fire a standing alert when stock hits a minimum sized to the part's lead time and your criticality. Document the stock list and re-order rules as part of the purchase, and review them at each new product introduction.
Spare tier | Typical examples | Failure impact | Suggested stocking rule |
|---|---|---|---|
Critical (line-stopping) | Head module, control board, conveyor drive | Whole machine stops | 1 on hand + supplier SLA |
High-use consumable | Nozzles, feeder gears, belts, pads | Gradual yield loss | 1–3 months of consumption |
Scheduled wear part | Sliders, bearings, filters | Predictable degradation | Per PM plan |
Optional / accessory | Spare PC, jigs, cables | Convenience only | As budget allows |
Used as a stocking matrix, the table forces a decision for every part class before purchase rather than after the first failure. The critical row matters most: a line-stopping failure with no local part is the difference between a coffee break and a lost shift.
A warranty is only as useful as the parts and response it actually covers, so read the terms against the failure modes that stop your line rather than the headline years. Many warranties exclude wear items and consumables — precisely the nozzles, belts and sliders that fail most often in daily production — and cap labour or travel, leaving you to pay for the most likely events yourself. What you want written into the order is a spare-parts service-level agreement: a committed delivery time for each critical part, measured from your request to the part on your floor, with the worst case named and, ideally, penalised for missing it. Hypothetically, a commitment of next-business-day regional dispatch backed by a local hub returns a failed module to service in hours, while an overseas-only commitment can stretch the same fault to days and turn a minor stop into a lost shift. Tie the SLA to the stocking matrix above: hold the long-lead critical parts yourself, and contract the supplier for the rest with a number you can actually enforce. Confirm whether loaner units, advance-replacement or consignment stock are included, because a loaner head while yours is repaired is availability you can plan around, not a surprise outage. Together the warranty and the SLA convert "the machine is covered" from a reassuring sentence in the quotation into a measurable, contractible commitment — the real difference between a stop that gets resolved fast and one that lingers while everyone argues about blame.
Parts availability is necessary but not sufficient: the machine must also stay up. The language here is reliability — mean time between failures (MTBF) and mean time to repair (MTTR) — and the derived figure, availability, is the fraction of time the machine is capable of producing. A high MTBF with a long MTTR can yield worse real availability than a lower MTBF with a fast repair, so the two must be read together. A model such as the DECAN F2 budget SMT mounter is judged not only on nominal reliability but on how quickly a fault is resolved where you operate.
The formula is availability = MTBF / (MTBF + MTTR). If a module fails every 500 hours and takes 10 to repair, availability is 500 / 510 ≈ 98.0 percent. The inputs are where care is needed: both are distributions, and the figure you care about includes the failures and repairs you will actually experience — especially the wait for a part or technician, where availability is usually lost. The mistake is to quote only MTBF and ignore MTTR and the waiting time inside it. Ask for MTTR as actually experienced, including parts and travel, and combine it with MTBF; that combined availability, not the headline MTBF, is what to compare and accept against.
Vendor MTBF figures are measured under controlled conditions and often exclude the wear items that fail most in production, so applying them directly overstates availability. Derate the vendor MTBF — many teams use 30 to 50 percent as a planning assumption — then confirm the real figure after the first months. As a hypothetical budget, 6,000 panels a year at 30 minutes each is about 3,000 machine-hours of demand; at a 95 percent single-shift target, allowable downtime is roughly 150 hours a year, and a 12-hour repair across a few expected events sits inside that — until an overseas part turns days into hundreds and the budget breaks.
Local support is itself availability. A local presence that diagnoses remotely, ships same-day and dispatches an engineer quickly beats a slightly better MTBF from three time zones away. Capture response time as a committed number with conditions — remote diagnosis within hours, on-site within days, what is contracted versus billable — because for many failures the wait, not the repair, dominates downtime. A service contract that caps the worst case is a more valuable availability feature than a marketing claim of high reliability.
Configuration is how a general machine becomes your machine. Two Hanwha mounters of the same model can behave differently depending on feeder slots fitted, nozzle set chosen and head type installed. The nominal CPH is achieved only when the configuration matches the product, so evaluating configuration means reading your BOM and placement profile and checking them against what the machine will actually be built with. For a high-speed architecture such as the DECAN S1 chip-shooter mounter, the question is whether its head and feeder design fits a high-count, repeated-placement product or a mixed one — the answer decides whether you see its speed in practice.
Feeder slots are the most common bottleneck. The distinct components on your panel, times boards per panel and reels per component, sets a minimum that is often higher than it looks once multi-up panels and buffer reels are counted. Too few slots forces reel changes mid-run or splits a product across two machines, destroying the promised throughput. Count real feeder demand from the BOM, including a buffer reel of high-consumption parts to avoid a stop at reel end, and confirm the candidate can be fitted with that many slots physically and in software. A supplier who sizes feeders from your actual BOM, not a maximum, demonstrates the configuration discipline this pillar requires.
Nozzles are the interface with every component, and the wrong set quietly caps throughput and yield. Different package sizes, heights and fragilities need different nozzle geometries and vacuum behaviours; a set chosen for a demo board rarely matches a real mixed BOM. List your component types — chips, ICs, connectors, odd-forms, tall or fragile parts — and confirm the machine can present and switch the nozzles they require, including spares, because a worn nozzle without a spare on hand is a small failure with a line-stopping consequence. Head type has the largest effect on throughput: a high-speed head favours a product dominated by the same component placed thousands of times, while a flexible head suits a high-mix product. Derive the head from the BOM's placement distribution, not a default, and record the reasoning.
Product profile | Feeder strategy | Nozzle set | Head choice |
|---|---|---|---|
High-mix, low-volume | Many tape/tray slots, flexible | Broad multi-size set | Flexible head |
High-volume, few parts | Dense identical reels | Repetitive set | High-speed head |
Tall / odd-form parts | Tray + tube feeders | Dedicated nozzles | Flexible + shuttle |
Long panels | Wider conveyor, support | Standard + lens | Per panel length |
The matrix is a starting point; the real choice comes from your BOM's placement distribution and panel geometry. Used as a checklist, it forces a configuration decision for feeders, nozzles and heads before order — when those decisions are cheap. A configuration that passes the matrix against your actual product has a realistic chance of reaching its nominal rate.
Configuration does not end with hardware. A mounter is also a programmed system, and how it is programmed and how it talks to the rest of the line determine whether the nominal rate is ever realised and whether the machine runs unattended between changeovers. For a multi-machine cell, the balancing of multiple Hanwha mounters in a line depends precisely on this integration working, because an unbalanced or non-communicating line throttles every machine on it.
Programming turns a CAD dataset and BOM into a running program, and the time it takes is configuration availability you either recover or lose on every new product. A good workflow imports CAD and centroid data, maps components to nozzles and feeders, and validates the program before the floor. Offline preparation — building and checking the program away from the machine — means the line is not stopped while an engineer types coordinates, directly protecting throughput. Ask what the toolchain supports: clean CAD import, a library matching your parts, offline simulation, an easy changeover path. A supplier whose software makes offline prep routine shortens every setup; one requiring on-machine manual entry lengthens every one, and because setup recurs with every new product, that workflow is a per-product availability tax the configuration choice can lower.
A mounter is one station, and it only contributes when it hands boards to the next cleanly. The interface most SMT lines use is SMEMA, carrying board-ready and board-complete signals so the line runs without manual intervention; a mounter that implements it unreliably creates misloads and stops that erase its speed advantage. Confirm conveyor length, support and interface before purchase, and verify the mounter's output — panel size, component height, accuracy — feeds a reflow profile and inspection step that can handle it. Fix these checkpoints in writing before order: feeder slots sized from the real BOM including buffer; nozzle set covering every component type with spares; head type from the placement distribution; programming toolchain supporting offline prep; line interface matching the existing line. Each maps to a later acceptance test, so a vague answer becomes a failed test at installation rather than a surprise in production.
A mounter rarely works alone; it is one station in a line whose throughput is set by the slowest step, so integration and balancing decide whether the nominal rate is ever actually reached. The interface most SMT lines use is SMEMA — board-ready and board-complete signals passed between stations so conveyors start and stop without manual intervention and boards are never double-fed or starved. A mounter that implements SMEMA unreliably creates misloads, jams and stops that erase its speed advantage, so confirm the handshake is tested against your real upstream printer and downstream reflow, not merely claimed in the brochure. Beyond the signal itself, line balancing is the discipline of allocating placements so no machine waits on another: if your mounter can place twenty thousand components an hour but the printer only feeds fourteen thousand reels, the line runs at the lower number regardless of the mounter's rating. The balancing of multiple Hanwha mounters in a line is covered in our dedicated guide, and the same principle applies here — size each station, including this one, to the real panel mix rather than a generic maximum. Verify conveyor height, direction and width match the existing line, and that the mounter's output, panel size, component height and accuracy, feed a reflow profile and inspection step that can absorb it without a bottleneck. Fixing these checkpoints in writing before order turns "integrates with our line" from a hope into a tested acceptance criterion, and it is where configuration and service support meet the reality of your factory floor.
Service support begins the moment the machine is delivered, and the quality of that beginning shapes everything after. Installation, commissioning and acceptance are where configuration is verified, the line is made to run, and the baseline is set that you maintain for years. Done well, they turn a pallet of equipment into a producing asset; done poorly, they seed chronic problems no later visit fully removes. The installation and training checklist itself is documented in the dedicated Hanwha installation and operator training checklist, which this chapter frames as your acceptance backbone.
Installation starts before the truck arrives. The site must be ready — power, air and vacuum at specification; floor flatness and load; environmental control; clearance for service and material flow. A machine delivered to an unready site waits, and the delay counts against your project though no fault exists. Sign off a site-readiness checklist by both sides, and confirm the interface to the existing line — conveyor height and direction, the SMEMA handshake, the reflow and inspection the mounter must feed. Commissioning then runs a Factory Acceptance Test at the supplier and a Site Acceptance Test at yours, each with pass criteria tied to your product: envelope on your longest panel, accuracy on your finest pitch, real cycle time on your actual mix, line handshake with your equipment. A documented FAT/SAT converts the supplier's claims into evidence you can rely on.
Acceptance is where vague promises become specific commitments. Insist on measurable, product-specific criteria: placement accuracy verified on your panel, not a coupon; real cycle time on your mix, not the datasheet rate; spare-parts lead time committed in writing; a line-integration check that the board moves cleanly to reflow. Each criterion should name the test, the target and the condition, so "performs to specification" is never the only record. Installation quality is the seed of availability: a machine levelled, aligned and integrated correctly wears evenly and faults rarely, while one rushed into service develops chronic stops no later service fully cures — which is why service support ranks with availability and configuration rather than below them.
After installation, the machine's fate is mostly in your hands — those of the operators who run it and the maintenance routine that keeps it. Operator training and preventive maintenance are the daily activities that determine whether availability and configuration deliver or decay. A well-trained operator catches drift before it becomes scrap; a PM plan replaces a wear part on schedule instead of after it fails. The maintenance backbone is the Hanwha preventive maintenance checklist, which should be adopted; the training side is the human half no maintenance plan replaces.
A curriculum is only useful if specific, so define it as a checklist tied to roles: safe operation and emergency stop; loading and verifying programs; feeder and nozzle changeover with the correct procedure; first-article and in-process checks; recognising and clearing common faults; basic daily upkeep. Each item should be hands-on, assessed and recorded, because a trainee who has read about a changeover is not one who has performed it under supervision. Training that builds internal capability is itself availability — a line that can change over and recover without waiting for a visit keeps running — and the record becomes part of the machine's documentation. Name the deliverables — a schedule, a hands-on assessment, a signed record — as acceptance items, because trained operators are the cheapest, highest-leverage availability investment in the whole evaluation.
Preventive maintenance (PM) is the planned replacement and inspection of wear items — nozzles, belts, sliders, filters, pads — before they fail, converting random stops into short, scheduled interventions. A PM plan sized to your actual run hours, not a generic calendar, keeps the machine inside its availability budget and turns the earlier MTTR discussion into a managed reality. The plan must be resourced: assign an owner and stock the wear parts it consumes, tying back to the spare-parts pillar, because a PM plan without parts or an owner is a document, not a practice. Confirming the plan is resourced and recorded turns maintenance from a hope into a control, and it is where escalation, remote support and documentation close the loop: a fault logged with cause and fix trains the next response, and a known escalation path removes the panic that wastes the most time.
When a fault occurs, the recovery clock starts immediately, and the fastest recoveries often begin before a technician boards a train. Remote diagnostics — a secure login the supplier uses to read alarms, logs and live machine state — can frequently pinpoint the failed module and even stage the fix without a site visit, because most faults declare themselves in the error log if someone qualified is watching. Build remote access and a named escalation path into the service agreement: who answers, how fast, and what they can do remotely versus what requires a physical visit. Pair that with spare-parts logistics, which is where availability is genuinely won or lost in daily practice. A regional or on-site hub that holds your critical parts means a failed module is swapped in hours, while a depot three time zones away turns the same fault into days of lost production. Consignment or loaner stock, where the supplier parks a spare head or control board at your site and bills only when it is used, is the strongest form of logistics because it removes the wait entirely. Tie the logistics back to the stocking matrix and the warranty SLA from the availability chapters: the parts you hold, the parts the supplier holds locally, and the committed dispatch time together define your real mean-time-to-repair. A fault logged with its cause and fix then trains the next response, and a known escalation path removes the panic that wastes the most time — closing the loop from the first failure to a faster recovery the next time, and completing the three-pillar picture for the specific machine you operate.
Support element | What to verify | Why it affects availability |
|---|---|---|
Installation & commissioning | FAT/SAT against your product | Wrong setup seeds chronic downtime |
Operator training | Hands-on, assessed, recorded | Skilled ops prevent errors and stops |
Preventive maintenance | Scheduled plan with stocked parts | Avoids unplanned wear-part failures |
Escalation & remote support | Response time, remote login | Faster recovery from faults |
The table is the service-support summary to lift into the acceptance document. Each row is a commitment that, written into the order and verified, converts the service pillar from a promise into a managed capability — completing the three-pillar picture for your specific machine.
Evaluate the machine on the three things the spec sheet hides
Speed and price decide the shortlist, but availability, configuration and service support decide whether the machine earns its place on your floor for the years you own it. Run this three-pillar framework against your real product: classify the spare parts and set a lead-time target, size the feeders, nozzles and head from your BOM, confirm the programming and line-integration path, and write the installation, training and maintenance commitments into the order as acceptance criteria. A candidate that clears all three pillars on your actual panel — with answers recorded, not assumed — is a machine you can commit capital to with confidence, while one fast on paper but vague on parts, config and support is a risk wearing a good brochure. Start from the Hanwha pick and place machine range and evaluate each model against the product you actually build.
Because CPH is a snapshot under ideal conditions, while the three pillars describe how the machine behaves across the years you own it. A faster machine with a two-week lead time on a line-stopping part, a nozzle set that does not fit your BOM, or no local training can produce fewer good boards per month than a slightly slower, better-supported one. The pillars turn the question from "which is fastest on paper?" into "which will keep placing my product at the rate I need?" — the only version that affects your bottom line.
Classify parts by failure impact, then ask the supplier for each critical and high-use part its stocking location, typical lead time and worst case, stated as a distribution rather than a best case. A simple exposure estimate multiplies the stockout probability by the extra days versus the typical lead time. Treat any such number as hypothetical until replaced with your supplier's actual figures, but use the structure to compare candidates on equal terms and to decide what to hold on site.
There is no single target; it follows from your volume, shift pattern and risk tolerance, usually 90 to 98 percent for a single mounter in a balanced line. Build an availability budget from MTBF and MTTR — including the wait for parts and a technician — and derate the vendor MTBF to reflect your environment. The budget tells you whether a machine is reliable enough for your committed output, which is more useful than comparing raw MTBF numbers, and you should revisit it after the first months with actual data.
Derive the configuration from your BOM and placement profile, not a default. Count feeder slots from the real components including buffer reels; choose a nozzle set covering every package type with spares; pick the head type from the placement distribution; confirm the programming toolchain supports offline prep; verify the line interface matches your existing standard. Each item becomes a yes-or-no acceptance test, so a vague configuration promise becomes a failed test at installation rather than a surprise in production.
Yes, because trained operators are the cheapest, highest-leverage availability investment in the whole evaluation. A curriculum tied to roles — safe operation, program loading, feeder and nozzle changeover, first-article checks, fault recovery, basic upkeep — performed hands-on, assessed and recorded, keeps the line running without waiting for external visits. Training that builds internal capability is itself availability, and the record becomes part of the machine's documentation you maintain as staff change.
A signed site-readiness check before delivery; a FAT at the supplier and a SAT at your site, each with product-specific pass criteria; acceptance on your real panel for accuracy, real cycle time and line integration; a committed spare-parts lead time in writing; and a recorded training and maintenance plan. Each criterion should name the test, the target and the condition, so the record is specific and reusable when the product changes. The dedicated installation and operator-training checklist provides the item-level backbone to lift into the order.
Preventive maintenance is planned replacement and inspection of wear items before they fail, converting random stops into short, scheduled interventions. A PM plan sized to your actual run hours, with an owner and the consuming parts stocked, keeps the machine inside its availability budget year after year. Without the parts or an owner, the plan is only a document. PM is where the availability pillar and the service-support pillar reinforce each other, turning the earlier MTTR discussion into a managed reality.