1. Direct Answer: No, Speed Is Not the Most Important Factor
No. For the vast majority of compact SMT production environments — especially small batch, high-mix low-volume, and frequent-changeover workshops — pick and place speed is not the most important factor when choosing a machine.
Here is why in one sentence: Daily production output depends on how many good boards the entire SMT line can produce steadily, not on how fast one machine places components under ideal conditions.
This does not mean speed is irrelevant. Speed matters, but its importance depends heavily on your production model:
| Production Model | Speed Importance | What Matters More |
|---|---|---|
| Mass production (single product, all day) | High | Speed + line uptime + feeder stability |
| Medium-volume (several products, batch rotation) | Medium | Changeover time + feeder flexibility + speed |
| Small batch / high-mix (many products, short runs) | Low | Changeover speed + software + accuracy + workflow |
| Prototype / NPI (one-off, new designs daily) | Very Low | Ease of programming + component flexibility + speed to first board |
The key insight: A machine with 13,000 CPH and 10-minute changeover can outproduce a 26,000 CPH machine with 45-minute changeover when you are running 5 small batches per day. The math is simple but often overlooked.
2. Decision Formula & Checklist: What Really Determines Output
The Real Daily Output Formula
Daily Output = N × (Placement Time per Board + Overhead per Board)
Where Overhead per Board =
+ (Setup Time + Changeover Time) ÷ Batch Size
+ Feeder Preparation Time per Board
+ Board Loading / Unloading Time
+ Reflow Conveyor Wait Time
+ Inspection Time per Board
+ Rework / Correction Time (quality-dependent)
Feeder Demand Formula
Feeder Demand = Unique 8mm Tape Reels + Unique 12/16mm Tape Reels + Tray IC Slots + Tube Feeder Positions + Backup Feeders (5–10 for common values like 100nF, 10kΩ)
Example calculation:
- 68 BOM line items
- 52 unique 8mm tape values → 52 feeders
- 4 unique 12/16mm tape values → 4 feeders
- 3 tray ICs → 3 tray positions
- 2 tube parts (connectors) → 2 tube feeder positions
- 6 backup feeders for common passives → 6 feeders
- Total minimum feeder requirement: 52 + 4 + 3 + 2 + 6 = 67 positions
Pre-Purchase Decision Checklist
Before comparing CPH numbers, answer these 12 questions:
| # | Checklist Item | Why It Matters |
|---|---|---|
| 1 | How many BOM line items per typical job? | Determines minimum feeder count needed |
| 2 | How many product types per week? | Determines changeover frequency impact |
| 3 | Average batch size per job? | Determines whether setup or speed dominates output |
| 4 | Largest PCB dimensions (L × W)? | Must fit machine's PCB handling range |
| 5 | Smallest component package used (0201? 0402?)? | Determines placement accuracy requirement |
| 6 | Do you use fine-pitch ICs (QFP, BGA, QFN)? | Determines vision system requirement |
| 7 | How many operators available per shift? | Affects feasible changeover complexity |
| 8 | Target daily output (boards per day)? | Back-calculates required line throughput |
| 9 | Available workshop floor space (m²)? | Limits line length and machine footprint |
| 10 | Reflow oven already selected or to be matched? | Oven must match placement output speed |
| 11 | Will jobs repeat or are they mostly one-off? | Repeat jobs benefit from saved programs and feeder maps |
| 12 | Expected growth in 12–24 months? | Choose a machine that can scale, not one at its ceiling |
3. Real Case: Speed vs. Changeover in a Small EMS Workshop
Background
A small EMS (Electronics Manufacturing Service) provider in Southeast Asia was comparing two pick and place machines for their compact SMT line. Their production profile:
- Product types per week: 12–18 different PCB models
- Average batch size: 30–80 boards per job
- Average BOM line items: 45–75 unique components
- PCB size range: 80×60mm to 280×200mm
- Component packages: 0402, 0603, 0805, SOT-23, SOIC, QFP-100, QFN
- Available operators: 2 per shift
- Floor space: 35 m² for the SMT area
The Two Machines Compared
| Parameter | Machine A (High-Speed) | Machine B (Compact Industrial) |
|---|---|---|
| Advertised CPH | 28,000 CPH | 13,000 CPH |
| Placement heads | 8 nozzles | 4 nozzles |
| Feeder positions | 102 | 64 |
| PCB size max | 400×350mm | 330×250mm |
| Typical changeover time | 40–60 minutes | 10–15 minutes |
| Software complexity | High (requires trained engineer) | Moderate (operator-friendly) |
| Machine footprint | 2.4m × 1.6m | 1.6m × 1.2m |
| Price | Higher | Moderate |
Daily Output Calculation (Real Scenario)
Let's calculate the actual daily output for both machines running a typical day with 6 different jobs, average 50 boards each:
| Time Component | Machine A (28K CPH) | Machine B (13K CPH) |
|---|---|---|
| Available time (8-hour shift) | 480 min | 480 min |
| Total changeover (6 jobs × changeover time) | 6 × 45 min = 270 min | 6 × 12 min = 72 min |
| Remaining placement time | 210 min | 408 min |
| Effective CPH (real, ~55% of advertised) | ~15,400 CPH | ~7,150 CPH |
| Components per board (avg 150 comps) | 0.58 min/board | 1.26 min/board |
| Boards producible in remaining time | 210 ÷ 0.58 = 362 boards | 408 ÷ 1.26 = 324 boards |
Result: Despite having less than half the advertised CPH, Machine B produced 89% of Machine A's daily output — while costing less, occupying less floor space, and requiring less operator training. The difference: changeover time dominated the equation.
The customer chose Machine B. After 6 months of operation, their actual average daily output was 280–340 boards across 5–8 jobs, with changeover averaging 11 minutes. They later added a second identical machine when volume grew, keeping changeover consistent across both lines.
4. Key Parameters to Evaluate Beyond Speed
When evaluating a pick and place machine, build your comparison around these 10 parameters — not just CPH:
| Parameter | What to Look For | Compact SMT Range | Industrial Range |
|---|---|---|---|
| PCB Size (max L × W) | Must accommodate your largest board + margin | 300×250mm to 400×350mm | 400×350mm to 500×400mm |
| Feeder Count | ≥ your max BOM line items + 10% buffer | 44–64 positions | 72–102 positions |
| CPH (realistic) | Advertised × 0.5–0.6 for real-world estimate | 6,500–20,000 CPH (adv.) | 20,000–40,000 CPH (adv.) |
| Placement Accuracy | ±0.05mm for 0402; ±0.03mm for 0201 | ±0.05mm to ±0.03mm | ±0.03mm to ±0.02mm |
| Component Range | Smallest passive to largest IC you use | 0402 to 32×32mm IC | 0201 to 55×55mm IC |
| Nozzle Count | More nozzles = fewer tool changes per board | 2–4 nozzles | 4–8 nozzles |
| Vision System | Flying vision + upward-looking camera for ICs | Standard flying vision | Multi-camera with IC alignment |
| Conveyor | Automatic width adjustment, SMEMA compatible | Single-lane, auto-width | Dual-lane optional |
| Feeder Types | 8mm tape, 12/16mm tape, tray, tube, vibration | All types supported | All types + bulk feeder |
| Software | CAD/BOM import, feeder mapping, component library, job recall | Operator-level GUI | Full engineering suite |
Brochure CPH vs. Real CPH: The 40-60% Rule
As a practical rule of thumb, expect your real-world sustained CPH to be approximately 40–60% of the advertised number. The gap comes from:
- Board loading/unloading (conveyor transfer time)
- Nozzle changes (different component sizes require different nozzles)
- Vision recognition time (ICs and odd-shaped parts take longer)
- Fiducial alignment (per-board registration)
- Feeder indexing (advancing tape between placements)
- Non-optimized placement sequence (real BOMs are never perfectly ordered)
5. Configuration Recommendations by Production Scale
Entry-Level Configuration (Small Batch / Startup)
For: Prototyping, in-house R&D assembly, small EMS starting with <10 product types, 10–50 boards per batch.
| Equipment | Recommended Spec | Product Example |
|---|---|---|
| Pick and Place | ~6,500 CPH, 44–50 feeders, 2 heads | HW-T4-44F-50F |
| Stencil Printer | Semi-automatic or manual, 400×300mm | XSE High-Precision Printer |
| Reflow Oven | 4-zone desktop or compact, lead-free capable | HW-R306 |
| Floor Space | ~8–12 m² | |
| Daily Output | 50–150 boards/day (depending on board complexity) | |
Standard Configuration (Growing EMS / In-House Production)
For: 10–30 product types, 50–200 boards per batch, moderate component variety including ICs.
| Equipment | Recommended Spec | Product Example |
|---|---|---|
| Pick and Place | ~13,000 CPH, 64 feeders, 4 heads | HW-T6-64 or HW-DU400 |
| Stencil Printer | Semi-automatic with vision alignment | ASE Automatic Printer |
| Reflow Oven | 6-zone compact, conveyor speed 0–2m/min | HW-R408 |
| Floor Space | ~15–25 m² | |
| Daily Output | 200–500 boards/day | |
Advanced Configuration (Higher Volume / Complex Boards)
For: 20+ product types, larger batch sizes, fine-pitch ICs, BGA, multiple shifts.
| Equipment | Recommended Spec | Product Example |
|---|---|---|
| Pick and Place | ~20,000–28,000 CPH, 72–102 feeders, 6–8 heads | HW-T8-72-80F or HW-M8-102F |
| Stencil Printer | Fully automatic with vision alignment | CP400 Automatic Printer |
| Reflow Oven | 8–10 zone, dual conveyor option, profiling software | HW-R612E or HW-R816 |
| Floor Space | ~25–40 m² | |
| Daily Output | 500–2,000 boards/day | |
6. The Line Balance Principle: Why One Fast Machine Is Not Enough
A compact SMT line is a system, not a collection of individual machines. The output of the entire line is determined by its slowest link:
Line Output = MIN(Printer Throughput, Pick & Place Throughput, Reflow Throughput, Inspection Throughput)
Common Bottleneck Scenarios
| Bottleneck | Symptom | Fix |
|---|---|---|
| Stencil printer too slow | Pick and place machine waits for boards | Upgrade to semi-auto or auto printer |
| Reflow oven too slow | Boards queue before reflow; placement machine stops | Add zones or increase conveyor speed |
| Inspection backlog | Defects found late; rework piles up | Add inline AOI or dedicated inspector |
| Feeder setup bottleneck | Machine idle during changeover | Pre-load feeders offline; use feeder carts |
| Operator bottleneck | One person doing setup, loading, inspection | Split roles; train second operator |
7. When Speed Actually Matters
Speed is not unimportant — it matters in the right context. High placement speed is valuable when:
- Your product design is stable and runs for months without changes
- Production volume is consistently high (500+ boards per job)
- The same PCB runs all day with minimal changeover
- Feeder setup is fixed and rarely reconfigured
- Operators are experienced and the process is standardized
- Upstream and downstream equipment can match the placement speed
- Your quality process is mature — defects are rare
- You have reached the throughput ceiling of your current machine
In these scenarios, upgrading from 13,000 CPH to 20,000+ CPH can directly increase daily output. But for most compact SMT users, the bottleneck is elsewhere.
8. Frequently Asked Questions
Q1: Is pick and place speed the most important factor when choosing an SMT machine?
No. For most compact SMT production, small batch, and high-mix environments, real output depends more on feeder setup, changeover time, placement accuracy, reflow oven matching, and line balance than on placement speed alone. Speed becomes more important only when production is stable, high-volume, and low-mix.
Q2: What is a good CPH range for small batch PCB assembly?
For small batch production (5–50 boards per job, frequent changeovers), 6,000–13,000 CPH is usually sufficient. The HW-T4-44F-50F (~6,500 CPH) handles entry-level needs, while the HW-T6-64 (~13,000 CPH) provides growing room. Machines above 20,000 CPH are better suited for stable, high-volume production where changeover is minimal.
Q3: How do I calculate how many feeders I need?
Use the formula: Feeder Demand = Unique 8mm Tape Reels + Unique 12/16mm Tape Reels + Tray IC Positions + Tube Feeder Positions + 5–10 Backup Feeders. A real customer with 186 components across 68 BOM line items required approximately 54 tape feeders. Always add a buffer of 5–10 positions for common backup values (e.g., 100nF capacitors, 10kΩ resistors). The HW-T6-64 with 64 feeders covers most small-batch scenarios; the HW-M8-102F with 102 positions handles complex BOMs.
Q4: Does higher CPH always mean higher daily output?
No. Daily output = (Available Time − Setup Time − Changeover Time − Downtime) × Effective CPH. A 13,000 CPH machine with 10-minute changeover can outproduce a 26,000 CPH machine with 45-minute changeover when running 5 small batches per day. Changeover time eats into placement time — and in high-mix production, changeover often consumes more time than placement.
Q5: What factors matter more than speed in high-mix low-volume SMT?
Changeover speed, feeder flexibility (support for tape, tray, tube), software ease-of-use (CAD/BOM import, job recall), placement accuracy across varied component packages, and repeatability between job files all matter more than raw CPH in high-mix low-volume SMT. A machine that is easy to reconfigure between jobs will produce more good boards per day than a fast machine that takes an hour to set up.
Q6: Can a compact pick and place machine handle production-level output?
Yes. A well-planned compact SMT line with a 13,000–20,000 CPH pick and place machine (like the HW-T8-72-80F), matched reflow oven, and organized feeder workflow can reliably produce 500–2,000 boards per day depending on board complexity and batch size. The key is line balance — all equipment must be matched in throughput.
Q7: What is the difference between brochure CPH and real CPH?
Brochure CPH is measured under ideal lab conditions — optimized component placement order, zero interruptions, ideal component arrangement. Real CPH is typically 40–60% lower after accounting for board loading, nozzle changes, vision checks, conveyor transfer, and operator interventions. When comparing machines, estimate real output as 0.5–0.6 × advertised CPH, then factor in your specific changeover and setup overhead.
Q8: How does reflow oven speed affect pick and place machine selection?
The reflow oven determines your line's maximum throughput. If the oven can process 120 boards/hour but the pick and place machine produces 150 boards/hour, 30 boards will queue before reflow. Always match oven conveyor speed and zone count to your placement machine's realistic output. For compact lines, the HW-R408 (6-zone) pairs well with 13,000 CPH machines; the HW-R612E (8-zone) or HW-R816 (10-zone) for 20,000+ CPH systems.