DEV Community: Asher Hu

Snap-Fit Tabs That Survive Two Teardowns: Designing Serviceability Into the Enclosure

Asher Hu — Tue, 29 Sep 2026 11:50:55 +0000

Many products are effectively scrapped the first time a user opens them for repair. The enclosure tabs snap off, the screw bosses strip, the wiring harness tears. The position we work from is simple: serviceability is not a post-sales concern. It has to be written into the structural design specification on day one as a hard functional requirement. That decision determines whether a product can be repaired and resold, or becomes e-waste.

Why designers should care: repairability is decided by geometry drawn months before anyone opens the product. It is cheap to design in and expensive to retrofit, and it shows up directly in after-sales cost.

The direct answer, in three parts: define a snap-fit cycle life before drawing the tab, plan a teardown route through the assembly, and police the structural-electronic interface. On a blood pressure monitor project the client required the enclosure to survive at least five non-destructive teardowns without losing ingress protection or structural strength. Snap fits are the mainstream way to get fast assembly, but "it clips shut" and "it still holds after five teardowns" are separated by the whole structural team's judgement.

Part 1: Snap-Fit Survival — Set the Standard Before You Draw

The core answer: the original snap-fit design had a life of only two to three cycles. After two teardowns, material fatigue and deformation opened a permanent gap at the mating feature, the sealing gasket was no longer compressed, and the unit failed its airtightness test.

We started with a benchmark: five mainstream products were collected and measured for the force required to open them the first time, plus the condition of the tabs afterwards. The data was direct. The deflection of a plastic snap on first disassembly depends mainly on the cantilever length of the tab arm and the thickness at its root. A longer, thinner arm feels softer on the first open and loses retention force faster on re-mating.

Our design parameters were set as follows. For PC+ABS, a common material for this job, the draft angle on the snap was standardised at 0.5 to 1 degree. Root thickness was taken as 0.6 times the adjacent wall thickness, and the mating portion as 0.8 times, which keeps rigidity adequate. We then simulated ten assembly cycles focusing on the stress concentration at the root of the arm. The simulation showed that with a cantilever longer than 8 mm, root stress approaches the material's yield limit by the third cycle. Every over-length cantilever was therefore given a reinforcing rib or converted to an interlocking latch.

Testing does not use a vague "tear it apart until it breaks" approach. We build a dedicated teardown fixture that reproduces the motion of a service technician prying with a fingernail or a flat-blade screwdriver, with force and angle tightly controlled, and run cycles. The final blood pressure monitor design, after optimisation, retained roughly 95 percent of its snap force after five non-destructive teardowns, fully meeting assembly and ingress protection requirements.

Part 2: Designing the Teardown Route

The core answer: letting after-sales open the product is not only a snap-fit survival question, it is a system plan. The teardown path has to be laid out the way a city plans roads.

First, layer and group. Parts that are replaced often, such as batteries and sensors, go in the top layer where the fewest steps reach them. Parts with low failure rates and complex connections, such as the mainboard, go below. In a handheld device project we made the battery compartment cover a separate module that opens by removing one captive screw, deliberately a cross-head pan head so no special tool is needed.

Second, reduce fastener types and count. A device should ideally use only two screw specifications, for example 2 mm and 2.5 mm cross-head. Every additional screw type raises the chance that after-sales fits the wrong one or strips the head. During layout we mark screw positions and matching lengths with colour-coded dots.

Third, design poka-yoke and guidance in. Where parts must be assembled in a specific order, we use trapezoidal or dovetail features so the wrong orientation simply will not go in. Cables are given enough slack near their connectors and secured with ties or channels, so a teardown does not pull on solder joints. We once took apart a competitor's unit and found an antenna routed around a module that has to be rotated out. The user could not possibly service it. That lesson went straight into our design checklist.

Part 3: Where Structure Meets Electronics

The core answer: serviceability design requires structural designers to work closely with electronics and industrial design. Most problems appear at the interfaces.

We hold one rule firm: any two parts connected by a flex cable must be either removable together or disconnectable independently. Structurally that means reserving working space for the FPC connector latch. We place connectors at the board edge and make sure that once the enclosure is open, a finger or a tool can reach the latch directly rather than being blocked by other components.

Locating features are the other key. We strongly recommend locating posts on the PCB and battery, with matching holes in the enclosure. A service technician then only has to align and press, seating the board in one motion and avoiding crushed edge components. On a Bluetooth speaker project we calculated that adding two locating posts costs very little in tooling but reduces "mainboard not seated correctly" repair cases by roughly 30 percent.

Serviceability design is, in essence, prepaying a measure of friendliness into the after-sales stage of the product's whole-life cost. It slightly increases initial design and tooling complexity while substantially reducing the brand's after-sales cost and reputational risk. The next time you pick up a new device, try asking: if this breaks, can I fix it, and where would I start? That may be the most honest test of a product's structural design.

FAQ: Serviceability Design Questions

Q1: What snap-fit material is best? Is harder always better?

A: For snap fits that must survive repeated teardowns, toughness matters more than hardness. We usually specify modified PP or PC+ABS, which balance toughness, rigidity and cost well. Purely hard materials, such as some filled nylons, give a high initial retention force but tend to fracture brittlely after repeated cycles. A useful reference is notched impact strength: where five or more non-destructive teardowns are required, we lean toward materials rated above 5 kJ/m², confirmed against the supplier's datasheet.

Q2: What standard do you test teardown cycles against?

A: There is no universal national standard, so we write an internal standard from the client's requirement and the product's expected use. For consumer electronics we take the maximum number of teardowns likely within one after-sales repair cycle and multiply by a safety factor. Methodologically we reproduce the real repair scenario with controlled tool, angle and force, then record permanent deformation, retention force decay and whether overall ingress protection is affected after every cycle. All of it is logged and used as design review evidence.

Q3: How do you justify the cost of serviceability design to a client?

A: From two angles. The first is direct after-sales cost: a repair that is easy to perform reduces engineer time on site and lowers the risk of secondary damage. The second is brand reputation cost: a product that is easy to repair improves user loyalty and word of mouth. Our proposals list the initial tooling and structural cost that serviceability adds, alongside the corresponding after-sales scenario analysis. That investment is usually far below the cost of a large-scale recall or brand damage caused by poor design.

Hertz Industrial Design (Dongguan, China) works across product appearance design, structural design, CMF and manufacturing follow-through, and more than 320 of its designs have reached mass production. Website: www.hezidesign.com.

Originally published on the Hertz Industrial Design website (hezidesign.com), Design Experience column.

Trial Production Tooling: Soft Molds, Aluminum Molds and Rapid Steel Molds

Asher Hu — Tue, 29 Sep 2026 04:31:07 +0000

A 200-unit trial run does not need a 120,000 RMB production tool. Choosing the tooling tier by the wrong criterion is one of the most expensive mistakes in small-batch production, and the money is only part of the cost.

Why designers should care: tooling is quoted per mold, but it is selected per purpose. Knowing what the trial run is actually for tells you which tier to buy, and what you will not be able to test if you buy the cheapest one.

The direct answer, in three tiers: soft molds (silicone, resin or low-strength alloy) cost 5,000 to 15,000 RMB and deliver in about a week, but last only tens to a couple of hundred shots and cannot run ABS or PC. Aluminum molds cost 20,000 to 40,000 RMB, hold roughly ±0.1 mm and run several hundred to one or two thousand shots. Rapid steel molds, simplified single-cavity tools in softer steel such as P20, cost 30,000 to 60,000 RMB and run a few thousand shots. For 100 to 500 units in a real engineering plastic, an aluminum mold is usually the right balance.

Tier 1: Soft Molds - The Cheapest Ticket In

The core answer: soft molds are the fastest and cheapest way to get a physical part, and the wrong way to promise delivery. A set runs 5,000 to 15,000 RMB with a lead time that can be compressed to under a week, which makes it the standard choice for very early testing below 100 units.

The limits show up quickly. Silicone tooling typically survives only tens to a couple of hundred shots. Dimensional accuracy lands around ±0.3 mm, and the part may carry a visible parting line. Most importantly the material is restricted: soft tooling suits free-flowing, lower-strength plastics such as silicone, TPU and PP. Testing the strength of ABS or PC is simply not possible.

We have seen this fail in the field. A client produced a functional prototype from a soft mold, and early user feedback was excellent. They then took that "perfect prototype" to a channel partner to sell stock, and the soft mold could not keep up. The following 100-unit order could not be delivered on time, and the customer relationship nearly ended. A soft mold validates a design; it does not honor a sales commitment.

Tier 2: Aluminum Molds and Rapid Steel Molds

The core answer: this is where cost meets tool life. When you need 300 to 500 units and the material must be an engineering plastic such as ABS or PC, the choice is between an aluminum mold and a rapid steel mold.

Aluminum molds cost 20,000 to 40,000 RMB. Aluminum is softer than steel and machines faster, which is why it costs less than a conventional steel tool. Accuracy reaches ±0.1 mm with a life of several hundred to one or two thousand shots. For the speaker project mentioned above, a 200-unit order made an aluminum mold at roughly 30,000 RMB the more prudent choice: the parts are almost indistinguishable from final production parts, so the launch quality can be simulated honestly.

Rapid steel molds are simplified steel tools: often single cavity, in a softer grade such as P20, without complex hot-runner systems. Cost lands around 30,000 to 60,000 RMB with a life of a few thousand shots. They outlast aluminum, but remain a simplification compared with a full production tool.

The decision line: with a budget of 20,000 to 50,000 RMB, a requirement of 100 to 500 units, and a need to validate the real material and the real molding process, choose the aluminum mold first. Rapid steel molds suit a requirement closer to small-scale sales, above roughly 500 units, where the tool is expected to bridge into early production.

Tier 3: Four Questions Before Choosing a Mold

The core answer: do not start by asking which mold is cheapest. Ask these in order.

1. What is the core job of this batch? Showing functionality to investors, letting early users try it, or selling it to recover cost? If it is the last one, quality requirements are high and a soft mold is effectively excluded.

2. Is the quantity fixed? Fifty units and five hundred units follow completely different tooling logic. Quantity is the hard constraint.

3. Can material and surface finish be compromised? If only the structure needs validating, the difference between ABS and PC matters less and a soft mold can still work. If the plan includes painting or IMD, an aluminum or steel mold is required.

4. Are the schedule and the budget locked? Tight money and tight time make a soft mold the only option. With a little more room, an aluminum mold does the job properly in one pass.

Cost is not the only price of a mold. Waiting and failure are the other two. Delays from the wrong tool, quality that misses the target, and sales commitments that cannot be met all cost far more than the price difference between tiers. At the trial production stage, tooling spend is better understood as validation investment.

Tier 4: What Designers Can Take Away

Match the tool to the purpose, not to the unit price. A 120,000 RMB production tool for a 200-unit channel test spends most of its value on shots that will never be taken.

A soft mold validates, an aluminum mold demonstrates, a steel mold supplies. Each tier answers a different question, and the questions are not interchangeable.

Neither aluminum nor rapid steel molds can be upgraded in place. What carries over is the design data and the validation conclusions, which is exactly the most expensive part of the trial-and-error phase. A production tool is designed and built fresh, with full cooling, multiple cavities and long life.

Ask for an itemized mold quote. Material grade, machining processes such as CNC, EDM and wire cutting, and whether trial shots and mold repair are included all matter more than the total. Real quotes list the steel grade, for example 718H or NAK80, estimated machining hours, standard component brands and the number of trial shots before delivery.

FAQ: Small-Batch Tooling Questions

Q1: Can soft-molded parts actually be sold?

A: They can be sold, but the customer must be clearly told the parts come from a soft mold, a non-production process, with material strength, surface precision and batch consistency that cannot match production parts, and with no guarantee of continued supply. Soft-molded parts suit a limited edition or prototype release to a very small fan base or a crowdfunding campaign, not regular commercial channels. Marking the product as a prototype in the sales contract matters, because it prevents later quality disputes.

Q2: Can an aluminum or rapid steel mold be upgraded into a production mold later?

A: Not directly, but the mold design data and the product validation conclusions carry over completely, which means the most expensive trial-and-error phase is already done. The value of an aluminum or rapid steel mold is confirming design feasibility, manufacturability and market acceptance. Once that passes and mass production is decided, a new production steel tool is designed and built from the validated data, with full cooling and heating, multiple cavities and long life.

Q3: How do you judge whether a mold factory's quote is reasonable?

A: Ask for an itemized quote and focus on material grade, machining processes such as CNC, EDM and wire cutting, and whether trial shots and mold repair are included. Comparing totals alone tells you very little. A reasonable quote lists the steel grade, for example 718H or NAK80, estimated machining hours, standard component brands and the number of trial shots before delivery. Take that list to two or three suppliers; the differences usually appear in how tightly they control machining detail and material cost.

Hertz Industrial Design (Dongguan, China) works across product appearance design, structural design, packaging design and CMF, and more than 320 of its designs have reached mass production. Website: www.hezidesign.com.

Originally published on the Hertz Industrial Design website (hezidesign.com), Design & Craft column.

From Design to Mass Production: A 10 to 20 Week Timeline, Its Bottlenecks, and What Can Be Compressed

Asher Hu — Mon, 28 Sep 2026 15:15:14 +0000

Starting from zero and reaching a trial-molded part takes 10 to 20 weeks for a typical consumer electronics product, roughly two and a half to five months. It breaks into four stages: appearance design 2 to 4 weeks, structural design 3 to 6 weeks, prototype validation 1 to 2 weeks, and tooling plus trial production 4 to 8 weeks. Products with a lithium battery, wireless charging, a motor, or 3C, CE or FCC certification tend to sit at the 20-week end.

Why designers should care: the schedule is not negotiated at the end, it is decided by when the requirements are frozen. Most overruns trace back to one meeting that did not happen.

Cost context: appearance design alone runs 5,000 to 12,000 RMB, appearance plus structure 15,000 to 40,000 RMB, and a full package covering appearance, structure, CMF and tooling follow-up 30,000 to 80,000 RMB.

Stage 1: How the Timeline Splits, and What Each Part Costs in Days

The core answer: the four stages add up to the total. Appearance design takes 2 to 4 weeks. Concept sketches settle the direction in 3 to 5 days, and the rest goes into 3D modelling and CMF options; a client feedback round is commonly 3 to 5 days, so the pace of client feedback matters as much as the design work.

Structural design, 3 to 6 weeks, is the largest block. A complete structure means assembly drawings, part drawings, draft angles, snap fits and screw bosses, plus confirming standard-part lead times with suppliers. For products with 30 to 60 parts, four to five weeks is the normal rhythm.

Prototype validation takes 1 to 2 weeks. A CNC set with painting, silk screening and assembly is typically about a week. SLA or silicone moulding is faster, but the surface is far from a production part, so it validates structure rather than feel or appearance.

Tooling and trial production takes 4 to 8 weeks. From receiving the 3D to the T1 trial shot is usually 3 to 5 weeks, and one revision after the trial adds another 1 to 2 weeks. That is normal for mould shops in Dongguan and Shenzhen; rush orders can be squeezed in, depending on machine availability.

Stage 2: Where the Timeline Actually Gets Stuck

The core answer: two places. Structural review rounds and mould revisions after the trial shot. A structural revision costs 3 to 7 days, a mould revision costs 1 to 3 weeks, and together they typically add 3 to 6 weeks beyond the plan.

Structural delays usually are not the designer's fault, they come from information that was never locked down. On a blood pressure monitor we built, the battery compartment was revised three times, and two of those were because the client had not yet decided between two AAA cells and a lithium battery: compartment dimensions, button positions and the display flex all moved with that decision. Confirming that class of information at the first review saves about two weeks.

Mould delays come from trial parts that do not match design intent. Sink marks, warpage and weld lines are invisible in 3D and only appear at T1. The mould shop also has to slot the revision into its machine schedule, and two to three weeks in peak season is normal.

Certification is the bottleneck people forget. 3C, CE and FCC take 3 to 6 weeks and should run in parallel with tooling. Waiting until the mould is finished before submitting pushes the whole schedule back by a month.

Stage 3: What Can Be Compressed and What Cannot

The core answer: appearance and structure can be compressed. Appearance from four weeks to two is realistic, and structure from six weeks to three or four is feasible, provided the requirements are stated once and the review does not exceed two rounds. Tooling at 4 to 8 weeks cannot be compressed, because mould machining and trial scheduling are physical time, and the 1 to 2 week prototype stage is hard to shorten either.

The single most effective action is a requirements lock-in meeting before kick-off, settling the battery solution, button count, interface types, target cost and certification list. Half a day there saves one to two weeks later.

The second is bringing the structural engineer into the appearance review early. Handing the design over only after the appearance is finalised means a week lost whenever an interference is found and the appearance has to change.

The third is running tooling and certification in parallel: use the T1 parts for a baseline certification screening test while the mould shop keeps revising. What cannot be saved should be stated plainly, including mould machining days, trial scheduling, injection cooling and secondary operations. When negotiating lead time with a mould shop, writing the revision rounds into the contract works better than chasing by phone.

Stage 4: How Design Fees Relate to the Timeline

The core answer: our quotes come in three tiers. Appearance design alone 5,000 to 12,000 RMB; appearance plus structural design 15,000 to 40,000 RMB; the full package covering appearance, structure, CMF and tooling follow-up 30,000 to 80,000 RMB. More revision rounds and a longer follow-through to trial production both raise the fee.

Mould cost is separate and is paid directly to the mould shop, not included in the design fee. For small appliances we have seen anything from tens of thousands to over a hundred thousand RMB, depending on part count and the number of cavities.

Sixty-five percent of our clients are repeat customers, and we have worked on more than 320 products. The pattern is consistent: the few hours spent up front defining requirements, certification and cost targets are worth more than chasing the mould shop ten times later.

What Designers Can Take Away

Four stages, 10 to 20 weeks. Appearance 2 to 4, structure 3 to 6, prototype 1 to 2, tooling 4 to 8.

Two bottlenecks dominate. Structural review rounds at 3 to 7 days each and mould revisions at 1 to 3 weeks each.

Run certification in parallel with tooling. Three to six weeks submitted after the mould is finished is a month lost.

Compress the front, not the physical time. Lock the requirements, bring structure into the review early, and write revision rounds into the mould contract.

FAQ: Design to Production Questions

Q1: How long does product design to mass production take?

A: 10 to 20 weeks, about two and a half to five months. Appearance 2 to 4 weeks, structure 3 to 6 weeks, prototype 1 to 2 weeks, tooling 4 to 8 weeks. Products with batteries or certification requirements sit at the 20-week end.

Q2: What does appearance design alone cost and take?

A: 5,000 to 12,000 RMB over 2 to 4 weeks. With structural design it is 15,000 to 40,000 RMB, and a full package with tooling follow-up is 30,000 to 80,000 RMB. Quotes move with product complexity and how far the follow-up goes.

Q3: Why does structural design take longer than appearance design?

A: More parts, plus coordination with moulding processes and standard-part purchasing, which is normally 3 to 6 weeks. For a 30 to 60 part product, assembly drawings, part drawings, snap fits and screw bosses plus supplier confirmation make four to five weeks routine.

Q4: How long does tooling take, and can it be rushed?

A: 3 to 5 weeks from frozen 3D to the T1 trial shot, 4 to 8 weeks including revisions. Rushing depends on the mould shop's machine schedule, and peak season is hard to squeeze. Writing revision rounds into the contract is more effective than chasing.

Q5: Can a product be finished in three months?

A: Yes, with everything in place: requirements locked in one pass, no more than two review rounds, no queue at the mould shop, and certification running in parallel with tooling. Miss one and three months will not hold.

Hertz Industrial Design (Dongguan, China) works across product appearance design, structural design, CMF and tooling follow-up, and more than 320 of its designs have reached mass production. Website: www.hezidesign.com.

Originally published on the Hertz Industrial Design website (hezidesign.com), Company News column.

Product Packaging Design Costs: Three Tiers, What Drives Them, and When to Start

Asher Hu — Mon, 28 Sep 2026 10:01:58 +0000

Small and mid-sized companies typically see packaging design quotes between 3,000 and 20,000 RMB. The spread is more than six-fold, and almost all of it comes down to one question: does the work involve structure? A printed carton and a rigid gift box are not the same job with a different price tag.

Why designers should care: packaging is quoted per project, but it is scoped per decision. Knowing which decisions drive the tier tells you what to prepare before the first meeting.

The direct answer, in three tiers: flat packaging (cartons, card boxes, labels) runs 3,000 to 6,000 RMB over 5 to 7 working days. Structural packaging (inserts, irregular boxes, transport packaging) runs 6,000 to 12,000 RMB over 7 to 15 working days. Rigid gift boxes run 12,000 to 20,000 RMB over 10 to 20 working days. Start once the product form is signed off and its dimensions are frozen, which normally saves a round of rework compared with commissioning packaging after product development ends.

Tier 1: What the Three Quotes Cover

The core answer: flat packaging handles the visual layer only. The work is layout, typography, colour and print-ready artwork; a carton goes from brief to deliverable files in 5 to 7 working days for 3,000 to 6,000 RMB, which suits simple, fast-moving products sold online or on supermarket shelves. When a series is extended across SKUs, the second SKU onward is normally priced at 30 to 50 percent.

Structural packaging adds how the product is held, shipped and packed. On top of the design fee, a structural engineer produces die-cut drawings, insert cavities and a stacking test plan. We often review this alongside the product structure review, because packaging dimensions depend on product dimensions: change the product and the packaging follows. Quotes run 6,000 to 12,000 RMB over 7 to 15 working days.

Rigid gift boxes add materials and the opening experience. Magnets, telescoping lids, EVA or paper inserts and multi-material combinations are typical, and the designer has to work with the printer on process feasibility. Quotes run 12,000 to 20,000 RMB over 10 to 20 working days, and for a seasonal launch the work should start two to three months ahead.

Tier 2: Where the Money Actually Goes

The core answer: three blocks. Design hours account for roughly half the quote; structural validation, meaning die-cut drawings, inserts and stacking or drop-test planning, accounts for 20 to 30 percent; the remainder covers sampling follow-up and pre-press file preparation.

Hour differences are the most concrete way to see the tier. Flat packaging normally delivers two or three directions, with two or three revision rounds on the chosen one. Structural packaging requires assembly relationships and developed dimensions, and changing an insert usually forces a change to the outer box as well, which makes the hours two to three times those of a flat project. A rigid box is higher still: the insert alone needs a physical sample to confirm how it feels.

Die costs are a one-off charged by the printer and are not included in the design fee. An irregular box or a box with an insert typically carries a die cost from several hundred to two or three thousand RMB, higher as the shape gets more complex. Sampling a carton is usually in the hundreds of RMB, while a rigid gift box, with several rounds of material sampling, commonly reaches the thousand-RMB range. Ask about both at the design stage, or the budget gap will surface later.

Tier 3: What Makes a Quote Move Up or Down

The core answer: four variables dominate. Number of directions, revision rounds, material and finishing complexity, and whether a series is extended.

More directions multiply hours directly. Asking for five directions usually costs more than 50 percent above two, and the extra directions are not necessarily useful in the decision. Revision rounds work the same way: beyond the rounds agreed in the contract they are billed per round, at a level comparable to a tenth or more of the initial design fee.

Material and finishing is the hidden cost. Foil stamping, spot UV, embossing and spot colours each require an extra finishing specification file from the designer and an extra press trial at the printer. When a series exceeds five SKUs, we recommend settling one master visual first and extending it in batches, which is more economical than quoting each SKU separately.

Tier 4: When to Bring in a Packaging Design Company

The core answer: once the product form is signed off and the internal structure is no longer changing. Starting too early means rework; starting too late misses the channel window.

Packaging dimensions are tied to the product's outer dimensions, the accessory count and the thickness of the manual. If packaging is finished before the product dimensions are frozen, then a change on the structure side forces the box dimensions, insert cavities and artwork to be redone, and the die and the samples have to be paid for a second time. We have seen this more than once, and it always comes down to sequencing.

Three things worth preparing yourself: confirm the final product dimensions and the pack quantity; decide the channel, because e-commerce shipping, supermarket shelves and gift channels have different strength requirements; and assemble the selling points and required markings, including barcode, ingredients, warnings and certification marks. With those three ready, the communication cycle is roughly halved.

What Designers Can Take Away

The tier is decided by structure, not by visual complexity. Flat packaging is artwork; structural packaging is engineering plus artwork.

Die and sampling costs sit outside the design fee. Ask early, because they are charged by the printer and can surprise a budget.

Four variables move the quote. Directions, revision rounds, finishing complexity and series extension.

Freeze the dimensions first. Packaging designed before the product is frozen will be redone, dies included.

FAQ: Packaging Design Questions

Q1: What does packaging design usually cost?

A: Flat packaging 3,000 to 6,000 RMB, structural packaging 6,000 to 12,000 RMB, rigid gift boxes 12,000 to 20,000 RMB, mostly including two or three directions and two revision rounds. Quotes normally exclude die costs and sampling, which the printer charges separately.

Q2: How long does packaging design take?

A: Flat packaging 5 to 7 working days, structural packaging 7 to 15 working days, rigid gift boxes 10 to 20 working days, with sample confirmation adding 3 to 7 days. For seasonal gift boxes, start two to three months ahead to leave room for material sampling and a trial run.

Q3: Can packaging and product appearance design be done together?

A: Yes. One team keeps the visual language consistent and removes one round of dimension hand-over, which usually shortens the overall schedule. Among our repeat clients, 65 percent commission appearance, structure and packaging together.

Q4: How do you choose a packaging design company in Dongguan?

A: Three checks: does it have a structural engineer, does it produce die-cut drawings, and does it follow the sampling? A visual-only team cannot do inserts or irregular boxes. The Pearl River Delta has dense printing capacity, so a local team can handle sampling and rework faster.

Originally published on the Hertz Industrial Design website (hezidesign.com), Company News column.

Structural Noise in Injection-Molded Assemblies: A Troubleshooting Order

Asher Hu — Sun, 27 Sep 2026 13:39:24 +0000

Ten machines from the same batch: some run quietly, some hum, some click as they turn. When that happens, my troubleshooting order is fixed at three steps: classify the fault by sound signature first, then check assembly stress, then work through moving parts, cabling and resonant cavities. Eight cases out of ten are located in the first two steps. Guessing at a fully assembled machine is the slowest route, because the same noise can come from an unbalanced motor, from a screw pulling the housing out of shape, or from a cable rubbing against the edge of a hole.

Why designers should care: noise is a return and a complaint, and almost every cause traces back to a tolerance, a preload or a cable route that was never fixed on the drawing.

The order: sound classification, assembly stress, moving parts and cabling, resonant cavities. Continuous friction points to interference or flash, periodic clicking points to rotating parts and hard stops, and humming at one particular speed points to housing resonance. The three field tests are: loosen the torque and listen for a change, press the housing and listen for the noise to stop, and remove one layer at a time and listen for it to disappear.

Step 1: Classify by Sound Before Taking Anything Apart

The core answer: spend ten minutes classifying the noise and save half a day. There are three families.

Continuous friction, a rustle or squeak that persists whenever the machine is moving, points to interference, flash or an over-pressed fit. Two surfaces that should have 0.1 to 0.2 millimetres of clearance have closed up after assembly, and they rub continuously.

Periodic clicking, once per revolution or once per open-and-close, follows the motion, and points to gear meshing, snap fits, hard-stop bumps or an eccentric bearing seat. On one of our small appliances the click tracked the motor and turned out to be the gearbox output shaft running with more than 0.5 millimetres of one-sided clearance in the housing bore.

Humming under specific conditions, only at a certain speed, load or attitude and gone as soon as the speed changes, is almost always about housing stiffness, cavity volume and support point positions.

After classifying, do two things: record a clip with a phone pressed against the machine and slow it down to hear the rhythm, then press the palm of your hand against different areas of the housing in turn. Wherever the noise drops noticeably is the main radiating surface, and that is where you look at supports and ribs. In the field this beats any instrument.

Step 2: Assembly Stress - Loosen the Torque Half a Turn First

The core answer: after classifying, do not touch the motor, touch the screws. Assembly stress is the most easily overlooked and most easily verified family.

The typical symptom is a machine that is quiet when assembled and then, overnight or after a few degrees of temperature change, clicks once by itself or develops a light plastic rubbing sound. A plastic part has been forced into a deformed position and the stress is releasing slowly. The cause is usually insufficient housing flatness: tightening four screws twists the part.

Snap fits behave the same way. On ABS parts we hold single-sided snap interference inside an empirical band, because too much of it creates preload that speaks up as soon as the machine warms. Where materials with different shrinkage meet, such as a PC/ABS part snapped onto a pure ABS part, the male and female halves of the same rib shrink differently, and the temperature rise pushes them against each other.

The test: loosen each of the four suspect screws half a turn and power up again. If the noise drops noticeably or disappears, the root cause is assembly stress, and the direction is flatness, boss height tolerance and snap interference. If nothing changes, go straight to the next step and stop polishing the housing.

Boss height tolerance deserves its own look. Four bosses on one housing panel differing by 0.2 millimetres mean only the two tall ones carry load once tightened, and the panel is bent into a slight arc. The deformation is invisible, and the symptom only appears as a click when you press the housing by hand.

Step 3: Moving Parts, Cabling and Resonant Cavities - Rebuild One Layer at a Time

The core answer: our internal method is layered rebuild. Strip the machine down to the core module, listen once, then listen again after each layer goes back on: bracket, harness, top cover, bottom cover. Whichever layer makes the noise worse contains the root cause.

Cabling is where we trip most often. A cable that is not secured knocks against the housing as soon as the machine vibrates, a rhythmic slap that follows the vibration frequency. Routing channels, cable clips and grommets together solve most of it. Clips spaced too far apart still let the middle section swing, so we place fixing points near the grommet and near the connectors at each end.

Motor and fan isolation pads are the next item. A pad that is too hard transmits vibration straight into the housing; too soft and the module moves further, which needs swing clearance so it does not strike anything inside the cavity. There is no universal value: try several durometer grades against module weight and rotational frequency and judge by ear and hand.

Resonant cavities are the most direct diagnosis of all. Nudge the speed slightly and the noise disappears: the original speed was hitting the natural frequency of a panel or cavity. The usual fix is to raise the stiffness of the flat area. Ribs should be 0.5 to 0.6 times the main wall thickness, or a large flat panel can be changed to a curved surface. Ribbing is more economical than thickening the whole wall, which brings sink marks and a new cosmetic problem.

Root-Cause Criteria

Noise that changes with speed or load points to the excitation side: motor, fan balance, gear meshing. Noise that changes when screws are loosened points to assembly stress. Noise that stops when the housing is pressed points to housing resonance or insufficient stiffness in the radiating surface. Noise that disappears when a particular layer is removed points to that layer's fits or cable fixing. A single click after standing or a temperature change points to stress release or a shrinkage mismatch between materials.

What Designers Can Take Away

Classify before you dismantle. Ten minutes of listening and pressing saves half a day of guessing.

Loosen the torque first. Half a turn on four screws tells you whether the root cause is assembly stress.

Rebuild layer by layer. The layer that makes the noise worse contains the fault, and cabling is the usual suspect.

Fix four items on the drawing. Snap interference, housing flatness, boss height tolerance and cable routing. A noise complaint costs far more than the two extra hours on the drawing.

FAQ: Structural Noise Questions

Q1: The machine was quiet when assembled, then clicked once overnight. Why?

A: Most likely stress release: a plastic part forced into a deformed position relaxes back. Check housing flatness and boss height tolerance first, then whether snap interference is excessive. Where two materials with clearly different shrinkage are mated, a temperature change triggers it too. The verification is to loosen the screws and leave it the same amount of time.

Q2: How do you tell motor noise from housing resonance?

A: Shift the speed slightly. If the noise changes or disappears, it is resonance; if it follows along, it is the excitation source. Pressing different areas of the housing with the palm is the other method: wherever the noise drops is the main radiating surface.

Q3: Do you need professional equipment to troubleshoot noise?

A: Structural noise is located in the field with a phone recording, layered rebuild and hand probing. Frequency analysis is for excitation-side faults such as motor electromagnetic noise or bearings, and those go to the supplier. On the structural side the point is isolating variables.

Q4: How do you avoid noise problems at the design stage?

A: Fix four items on the drawing and in the 3D: interference, flatness, boss height tolerance and cable routing. Route cables during modelling with fixing points defined rather than pulling them at the assembly bench, and rib large flat panels at 0.5 to 0.6 times the main wall thickness.

Hertz Industrial Design (Dongguan, China) works across product appearance design, structural design and CMF, and more than 320 of its designs have reached mass production. Website: www.hezidesign.com.

Originally published on the Hertz Industrial Design website (hezidesign.com), Design Experience column.

Injection Molding Color Variation: A Shop-Floor Troubleshooting Order

Asher Hu — Sat, 26 Sep 2026 12:17:30 +0000

When a batch of parts comes out with colors that do not match, the reflex is to start adjusting the machine. In our experience the cause sits in one of three places: whether the variation between parts follows a pattern, whether the masterbatch ratio or batch changed, and how long the material sat in the barrel. Work through them in that order and the fault is usually bracketed within half an hour, which is far cheaper than repeated trial shots.

Why designers should care: color is a specification, not a machine setting, and the tolerance has to be agreed in writing before mass production rather than argued about after shipment.

The order: distribution pattern, then masterbatch (ratio, carrier, batch), then base resin grade and regrind ratio, then thermal history (melt temperature, back pressure, screw speed). The distribution pattern alone points the direction in nine cases out of ten.

Step 1: Classify the Variation - Where It Differs Matters More Than How Much

The core answer: the first move is not to measure with a spectrophotometer but to ask about distribution. Is one of four cavities consistently darker? Does the same cavity run progressively yellower from start-up to the afternoon? Is the whole batch a step away from the previous batch? Is there a gradient across a single part from gate to end of fill? These four patterns point to four completely different places.

Cavity-to-cavity difference points to the tool. The usual cause is a difference in runner length or gate size between cavities, which changes shear rate and therefore the actual melt temperature; the cavity with the smaller gate sees more shear heating, disperses masterbatch better and often comes out darker. Mold temperature plays a part too: if the cooling circuits are unbalanced and cavities differ by more than 5 degrees Celsius, crystalline resins such as PP, POM and PA develop different degrees of crystallinity and the color follows.

Drift along the time axis points to thermal history. Parts running light for the first few dozen shots and settling afterwards usually mean residual material from the previous batch has not been purged, or that the residence time from hopper to screw is too long. A gradient within one part points to fill and shear: near the gate the shear is strong and the temperature high, at the far end both are low, so masterbatch dispersion differs, which is most visible on light-colored parts.

Step 2: Masterbatch - Check Ratio, Carrier and Batch Separately

The core answer: masterbatch takes the blame most often, and it is also the easiest link to verify. Typical addition rates run 1 to 4 percent, and white or light-colored parts sometimes go to 4 to 6 percent. A few tenths of a percent matter: on a light part a 0.3 percent difference is visible to the eye. That is why we require gravimetric dosing units in the shop rather than hand blending, where uniformity depends on the operator and one drum differs from the next.

Carrier resin has to match. A PE-carrier masterbatch used with ABS or PC disperses poorly, producing color streaks, specks or an overall mottled look. When selecting masterbatch, ask what the carrier is and keep it in the same family as the base resin. Damp masterbatch also causes variation; it is normally dried with the base resin under the same conditions, which means roughly 2 to 4 hours at 80 degrees Celsius for ABS and about 4 hours at 120 degrees Celsius for PC.

Batch-to-batch variation is the most concealed trap. For the same color code, a difference of a few tenths to one ΔE between production batches is normal. Our practice is to require a color plaque and a retained sample with every batch, to check color on arrival, and to reject a batch that fails. When switching masterbatch batches on the same machine, a purge shot goes in between so two batches never mix inside one shipment.

Step 3: Thermal History - Three Extra Minutes in the Barrel Yellows a Light Part

The core answer: we had a run of PP housings that started normally and drifted yellow through the afternoon, getting worse as the shift went on. The masterbatch checked out and the base resin had not changed. The cause was regrind: the ratio of sprue and runner material had been raised on that shift. Every thermal pass raises the yellowness index of PP a little, so the more regrind, the more visible the shift.

We normally hold regrind below 15 to 20 percent, and more conservatively still on light-colored and cosmetic parts, where the preference is to avoid it altogether.

Residence time is the most overlooked item. Shot size should sit between 20 and 75 percent of the machine's maximum shot capacity. A small tool on a large machine means the material sits in the screw for a long time and degrades into yellow; we have seen more than once a color problem that disappeared when the same material was run on a smaller machine.

Melt temperature, back pressure and screw speed all change shear heat, and with it both masterbatch dispersion and thermal degradation of the resin. When adjusting, change one parameter at a time and run at least ten shots before judging. After a stoppage, discard the first dozen shots: their thermal history is not the same as the steady state. Color changes are their own discipline, and black to white is the hardest combination; it takes a purging compound, and plain resin often will not clear it.

Step 4: Root-Cause Criteria - Decided in Minutes

The core answer: the criteria collapse into a short table that can be read on the shop floor. If the cavity-to-cavity difference clearly exceeds the time-to-time difference within one cavity, look at the tool: runners, gates, mold temperature distribution. If the color is off at start-up and recovers, look at barrel residue and residence time. If it jumps when a material batch is changed and recovers when it is changed back, look at the masterbatch batch, the base resin batch and the regrind ratio. If the gradient runs along the flow direction within a single part, look at fill and shear. If none of the four applies, look at the measurement itself.

Numerical criteria belong in the contract before production rather than in an argument after shipment. The industry normally works in ΔE: up to 1.0 is accepted by most customers, 1.0 to 2.0 is a negotiation zone, and above 2.0 the difference is obvious to the eye. Measurement conditions must be fixed: a D65 light source with a 10 degree observer, consistent plaque thickness, and no direct comparison between a textured surface and a polished one, which produces wildly different readings.

We normally seal three plaques at each limit, upper and lower, together with a visual criterion: no perceptible difference under a D65 standard light booth. On a blood pressure monitor with off-white ABS cosmetic parts, that was the specification we agreed with the customer, ΔE at or below 1.0 plus no visual difference under D65, with a part pulled every two hours during production and the reading logged on the inspection sheet so any problem can be traced to a specific time window and machine.

What Designers Can Take Away

Ask about distribution before measuring numbers. The pattern of variation points at tool, process or incoming material.

Masterbatch is usually the easiest link to prove or clear. Ask for carrier, ratio and batch, with a plaque and retained sample per batch.

Thermal history is where light parts go yellow. Keep regrind below 15 to 20 percent and shot size between 20 and 75 percent of barrel capacity.

Agree the ΔE criteria in writing before mass production. Fixed light source, fixed observer angle, sealed limit plaques and a visual criterion.

FAQ: Injection Molding Color Questions

Q1: What ΔE counts as acceptable for injection molded color?

A: Most projects work to ΔE of 1.0 or less, 1.0 to 2.0 requires customer agreement, and above 2.0 the difference is clearly visible. Light and white cosmetic parts are usually held tighter. What matters is written confirmation before production, sealed upper and lower limit plaques, and an agreed visual criterion.

Q2: A single batch has mismatched colors. What do you check first?

A: The distribution pattern, not the numbers. Cavity-to-cavity differences point to runners and mold temperature; drift over time points to barrel residue and residence time; a jump when the batch changes points to masterbatch and base resin batches; a gradient within one part points to fill and shear. Only if none applies do you suspect the measurement.

Q3: The masterbatch ratio has been adjusted and the color is still unstable.

A: Look at the blending method and the carrier first, then the regrind ratio and residence time. Replace hand blending with a gravimetric dosing unit, keep the carrier in the same family as the base resin, hold regrind low on cosmetic parts, and remember that a shot size below 20 percent of barrel capacity means long residence and drifting color.

Q4: How do you purge cleanly when changing from black to white?

A: Use a purging compound rather than expecting plain resin to clear it, then run a dozen transition shots. After the compound, continue with the natural-color material about to be produced and check every hot runner nozzle and barrel dead spot. The first dozen shots after start-up have inconsistent thermal history and should be discarded.

Originally published on the Hertz Industrial Design website (hezidesign.com), Materials and Manufacturing column.

Counterweight Design: Three Steps to Tune the Perceived Weight of a Handheld Device

Asher Hu — Fri, 25 Sep 2026 03:33:25 +0000

Perceived weight matters far more than actual weight. A product that feels hollow reads as cheap and unreliable the moment a user picks it up. Counterweight design is how we make a device feel solid and stable in the hand, and it is not a matter of dropping an iron block inside the housing. It is a process: set the target, optimise the layout, then tune the feel by hand.

Why designers should care: perceived weight is a perceived-value decision made on the drawing, and once the tool is cut, the options collapse to adhesive and hope.

The core method, in three steps: set the target feel weight for the category; maximise the mass you already have in the PCB, battery, motor and shields; add dedicated counterweight material only where those two are not enough.

Inventory existing mass before buying any counterweight.
Tungsten carbide, cast iron and lead cover most applications; density drives volume.
The centre of gravity belongs near the web of the hand, not at the centre of the product.

Step 1: Before Adding Iron, Count the Cards You Already Hold

The core answer: every product already carries mass in its printed circuit board, battery, motor, speaker and metal shields. The first job is not choosing counterweight material but rearranging those parts so the overall centre of gravity matches how the product is held. The design parameter that matters is the target feel weight: for consumer electronics held directly in the palm, a comfortable perceived weight usually falls between 180 and 350 grams.

On a handheld medical test device we worked on last year, the requirement was comfortable, stable grip. Early in the project the structure engineer clustered all the electronics and the battery at the top of the handle, which made the device top-heavy and tiring to hold. We replanned the layout: moved a larger lithium battery (about 45 grams) downward, centred the main control PCB (about 12 grams), and reserved a symmetrical screw boss area at the base of the handle for a possible future counterweight. Moving existing parts alone shifted the centre of gravity down by roughly 15 millimetres, and the balance improved immediately.

Too light reads as a toy, too heavy causes fatigue. The band is not a law, it is the pattern from user feedback on products already in the market, and where a specific product lands depends on its category and its user.

Step 2: When Layout Runs Out, Add Dedicated Counterweight

The core answer: when internal space is tight or the existing parts cannot be arranged to meet the feel requirement, dedicated counterweight material comes in. Density decides volume, and volume is usually the binding constraint.

Tungsten carbide: about 19.3 g/cm³, smallest volume for a given mass, highest cost. For high-end products where space is at a premium.
Cast iron: about 7.2 g/cm³, lowest cost, the value choice, but occupies the most volume.
Lead: about 11.3 g/cm³, in between, but environmental rules and sealing against oxidation complicate production.

Whatever the material, the housing needs a designed mounting position, on the PCB or on the plastic shell.

Last month we revised a Bluetooth speaker whose owners said it was so light that it slid across the table when nudged. We added two symmetrical cast iron blocks at the bottom, 15 grams each. The drawing mistake was not designing raised locating bosses on the bottom shell, only flat mounting pockets. In trial production the blocks shifted during the vibration test, which moved the centre of gravity and produced a rattle. The fix added four small screw bosses and locked the blocks down with self-tapping screws. Counterweight mounting has to be reliable: clips and adhesive do not survive vibration and drop.

Step 3: Tune the Feel by Hand, Not on the Screen

The core answer: every mass calculation and layout simulation ends in physical feel testing. We build prototypes with several counterweight schemes and have people with different hand sizes take turns holding them. A handheld device's centre of gravity should sit near the web between thumb and index finger. Too far forward, toward the fingertips, and the product feels like it is falling out of the hand; too far back, toward the palm, and it feels clumsy.

Our internal method is crude and effective: put the product on a balanced ruler to find the centre of gravity, then simulate the grip by hand and feel whether the pressure at both ends of the ruler is even. We also test several grip styles: tight, light, one-handed and two-handed.

One handheld barcode scanner had all its counterweight placed at the rear in pursuit of light weight. Scanning items high on a shelf, the front probe was too light and users had to work to control the direction. Moving 20 grams of counterweight from the rear to the middle of the body solved it. Total mass met the target the whole time; the balance was wrong.

What Designers Can Take Away

Set the number before the geometry. Decide the target feel weight for the category, then let layout serve it.

Move what you already have. Repositioning a battery and a PCB shifted the centre of gravity 15 millimetres at no material cost.

Lock the counterweight down properly. Screw bosses and self-tapping screws, not clips or adhesive.

Feel is decided by hand. Multiple prototypes, multiple hand sizes, multiple grip styles, and a centre of gravity near the web of the hand.

FAQ: Counterweight and Hand-Feel Questions

Q1: How do you decide how heavy a new product should be to feel premium?

A: Benchmark the competition, then set a target band by category. Consumer electronics are usually comfortable between 180 and 350 grams. We ask clients for three to five competitor units early on and run a blind grip test, recording the team's subjective sense of weight and balance as the starting point.

Q2: Do cast iron counterweights rust, and how is that handled?

A: Yes. The usual answer is an electrophoretic or spray anti-rust coating, which is inexpensive, or galvanised cast iron. We have seen a batch of untreated blocks develop rust spots in a damp warehouse before assembly and stain the inside of the housing, so suppliers now have to deliver blocks already treated.

Q3: How much does a counterweight add to cost?

A: The material itself is a small share; the mould features that locate it dominate. A simple cast iron block costs cents, but adding the structure to the shell can require another slide, which puts tooling in the tens of thousands of yuan. We first check whether rearranging existing parts can remove the need entirely.

Q4: The tool is already cut and the product feels too light. Can it be fixed?

A: Yes, with limits. The usual route is finding space inside the existing shell and fixing small blocks with double-sided or hot-melt adhesive, or adding mass to a removable part such as the battery cover. It is a stopgap: reliability and appearance are both worse than a structure reserved at the design stage. It is the reason counterweight planning belongs in the early design review.

Originally published on the Hertz Industrial Design website (hezidesign.com), Design Experience column.

Soft-Touch Paint Durability: Thickness, Substrate, and Abrasion Limits

Asher Hu — Thu, 24 Sep 2026 04:37:40 +0000

Soft-touch paint gets its feel from an elastic resin with low crosslink density, elastic microspheres a few microns across, and a heavy load of matting agent. Film thickness normally runs 25 to 50 microns; below 20 microns there is barely any feel, above 60 microns blocking and scratch resistance start to fail. The system is mainly two-component polyurethane, with 60-degree gloss pressed to 2 to 5 units.

Why designers should care: soft-touch film is easy to specify and impossible to touch up, so thickness and annealing are fixed before production.

The core fact: softness comes from the formulation, not the film build. Thickness only masks substrate roughness and lowers gloss; below 20 microns the mold grain shows through, above 60 microns the film turns tacky.

Ask the coating supplier for the microsphere ratio and resin system, not just a softer quote.
25 to 50 microns is the working band; below 20 and above 60 both fail.
Heavy matting agent improves feel but leaves a less dense film.

Point 1: The Hand Feel Comes From the Formulation, Thickness Only Amplifies It

The core answer: the mainstream system is two-component polyurethane (2K PU), a hydroxy acrylic or polyester polyol with an HDI trimer as curing agent, plus silicone or PU elastic microspheres and matting agent. Softness comes from a low-modulus film deflecting under fingertip pressure; the microspheres supply the rebound.

Projects often ask whether spraying thicker makes it softer. It does not: the hand feel is the elastic modulus of the resin. In our projects the window is 25 to 50 microns: below 20 microns the mold grain shows through and the surface reads hard and dry, above 60 microns the rebound slows, the surface turns tacky, and stacked packaging sticks. When a client wants more rubber feel we adjust the microsphere ratio, not the film build, because thickening brings blocking and scratch problems.

More matting agent improves the hand feel but leaves a less dense film, so a dozen wipes with an alcohol pad start to polish the surface.

Point 2: Cracking, Tackiness and Gloss-Up - The Substrate and Cure Hurdles

The core answer: PC and PC/ABS parts sprayed with soft-touch paint most often fail by stress cracking, from a few hours to a day or two after spraying, as fine cracks along weld lines, near gates, or where wall thickness changes abruptly. The cause is not the paint but molding stress plus the ester and ketone solvents in the coating; PC is inherently sensitive to them.

There are three responses. Anneal the molded part: ABS grades go through roughly 2 hours at 70 to 80 degrees Celsius. Have the coating supplier switch to a weak-solvent system, since changing paint unilaterally can change the hand feel. Or reduce wall steps and keep gates off the loaded cosmetic face. Annealing is cheapest, but production often skips it and parts crack in batches.

Cure temperature is the second trap. A 2K PU system needs 70 to 80 degrees Celsius for 30 to 60 minutes, and ABS softens at 80 to 90 degrees Celsius, so thin-wall parts commonly warp after an hour at 80 degrees. The options are a low-temperature catalyzed 2K system cured around 60 degrees Celsius, or UV-cured soft touch, which is fast but feels a step below 2K PU.

How far soil and abrasion resistance go: RCA tape abrasion (ASTM F2357) lands in the hundreds of cycles versus a thousand for ordinary matte paint. DEET turns the coating tacky and hand sanitizer polishes it; on handheld products a spot of sunscreen or repellent sits on a sample for 24 hours.

Point 3: What Has to Change on the Drawing Before Painting

The core answer: radius every sharp edge on the cosmetic surface to R0.3 or more; avoid slots narrower than 2 mm and deeper than 3 mm; choose a fine matte texture over a deep grain; anneal PC and PC/ABS parts before painting; replace silk screening with laser engraving; keep snaps and slides on hard surfaces.

A sharp edge pulls the film thin, so it chips through and leaves a mismatched colour. Narrow slots and deep pockets should be ruled out early: a slot under 2 mm wide and over 3 mm deep cannot be reached by atomized paint, leaving dry spray and pitting at the bottom. Mold texture should not be cut too deep either, because a granular grain makes the film expose the peaks and fill the valleys.

Secondary operations need deciding early. Silk screening adheres poorly on soft-touch coatings, and a fingernail lifts it. Where laser engraving is possible, switch to it, because the laser removes microspheres and matting agent and the marking reads bright against the matte surround. If silk screening is unavoidable, it needs a dedicated primer, or printing moves ahead of painting.

Assembly comes last. Two soft-touch surfaces rubbing together shed powder and polish up, so snaps and rails should contact on hard surfaces. The coating also cannot be touched up locally, only stripped and resprayed whole, so spare parts belong in the quotation.

What Designers Can Take Away

Hand feel is a formulation decision, not a thickness decision. Past 60 microns, thickness only buys blocking and scratch risk.

Annealing is cheap insurance. Two hours at 70 to 80 degrees Celsius before painting prevents stress cracking along weld lines and gates.

Fix the geometry before the paint. R0.3 radii, no slots under 2 mm wide or over 3 mm deep, fine matte texture, hard-surface contact at snaps.

FAQ: Soft-Touch Paint Questions

Q1: Are soft-touch paint and rubber paint the same thing?

A: Interchangeable in everyday use; both mean a two-component elastic polyurethane coating, with rubber paint implying higher elasticity.

Q2: How thick is soft-touch paint normally sprayed?

A: 25 to 50 microns. Below 20 microns the feel reads hard; above 60 microns blocking and scratch risk rise. Confirm with a gauge.

Q3: Why do PC parts crack after painting?

A: Molding stress plus the ester and ketone solvents in the paint; annealing and a weak-solvent system resolve most cases.

Originally published on the Hertz Industrial Design website (hezidesign.com), Materials and Manufacturing column.

Plastic Part Design: Wall Thickness, Ribs, Texture Draft, and Glass-Filled Traps

Asher Hu — Wed, 23 Sep 2026 04:32:10 +0000

For consumer electronics and appliance enclosures, wall thickness lands between 1.8 and 2.5 mm, and high-gloss ABS or PC+ABS covers eight of ten requirements. Parts that must survive drop tests step up to PC or glass-filled PA. Ribs run 0.5 to 0.6 times the main wall. On textured surfaces, every 0.025 mm of extra texture depth needs one more degree of draft, or mold trial day shows ejection whitening and scuffing.

Why designers should care: wall thickness and draft angle are nearly impossible to reverse once the mold is cut. Getting them wrong costs a new cavity insert, not a drawing change.

The core fact: ABS shrinks 0.4% to 0.7%, PC 0.5% to 0.7%, POM as much as 1.8% to 2.5%. Ribs run 0.5 to 0.6 times the main wall, and texture depth is linked to draft angle, so changing one forces you back to the other.

Fix wall thickness first, material second. Read shrinkage off the resin datasheet, never from memory.
Cosmetic parts in order of preference: high-gloss ABS, then PC+ABS, then glass-filled grades.
Texture depth and draft angle are a linked pair: change one, revisit the other.

Point 1: Wall Thickness and Ribs - Keep Ribs at 0.5 to 0.6 of the Main Wall

The core answer: ABS shrinkage generally sits at 0.4% to 0.7%, PC at 0.5% to 0.7%, POM as high as 1.8% to 2.5%. Rib thickness is taken at 0.5 to 0.6 times the main wall; past 0.7 times, a sink mark appears on the back of the rib, and uneven wall thickness turns the shrinkage difference into assembly interference.

A desktop appliance enclosure we built in PC+ABS started with a uniform 3.2 mm wall, assuming thicker would feel more solid. Mold trial produced a shadow line along the back of all four ribs under directional light: rib thickness sat at 2.4 mm, exactly 0.75 times the main wall. Pressing the ribs to 1.8 mm, or 0.56 times the main wall, made the sink marks essentially invisible, with nothing else changed in the tool. Past 0.7 times, added cooling and a tuned packing profile only reduce the sink mark.

Shrinkage is not a single value: within one shot, the gap between flow and perpendicular directions reaches 0.2% to 0.3%, which on a 200 mm part is 0.4 to 0.6 mm. Model both rows of snap fits at nominal size and assembly runs tight on one side and loose on the other, so long cosmetic parts get roughly 0.3 mm per side.

Point 2: Texture and Draft Angle - Lock the Texture Plate Before the Outer Structure

The core answer: each 0.025 mm of texture depth needs at least 1 more degree of draft. A standard etched finish at SPI #3 to #5 wants 2 degrees or more, and deep grain or leather grain 3 degrees. A glossy part releases at 1 degree; a deep-grain part at 1 degree usually ejects with whitening or scuffing.

A client brought a leather-grain plaque from a cosmetics package as the target feel: an MT-series deep grain measuring about 0.08 mm. Sidewall draft was set at a conventional 1.5 degrees, since the previous revision had been glossy. Mold trial gave a white line at the top of the sidewall that survived wiping, and slowing ejection plus raising mold temperature to 80 degrees Celsius never removed it. Taking the draft from 1.5 to 3.5 degrees with a new cavity insert solved it, at a tooling cost in the tens of thousands of RMB, still cheaper than a three-week delay.

Point 3: Material Pairing and Glass-Filled Grades - Never Run POM Against POM

The core answer: on mating surfaces with relative motion or repeated friction, the same material on both sides causes galling, and a POM-on-POM sliding pair has a shorter life than POM on PA. Glass-filled grades lose weld line strength and show floating fiber, so cosmetic parts should avoid them; where they are unavoidable, hold glass content between 10% and 15%.

On a handheld device we first specified PA6 with 30% glass fiber for the cover. Two problems appeared: floating fiber streaking under directional light, and four snap features on weld lines, two of which broke in a 1.2 m drop test. A weld line is where two flow fronts meet, glass orientation is disrupted there, and strength drops to roughly 60% of the bulk material. Switching to PC+ABS and moving the snaps about 15 mm clear of the weld line passed the same drop height.

Two internal sliders ran against each other, both in POM, assuming POM self-lubricates well. Of 200 units assembled for testing, a batch seized about half a month later with visible wear on the mating surfaces; one side POM and one side PA ended it. Above 20% glass content floating fiber cannot be controlled, and if strength falls short we would rather thicken the wall than move the glass ratio.

What Designers Can Take Away

Wall thickness is a material decision, not a styling one. The shrinkage figure is on the datasheet, and a thick wall chosen for hand feel is how sink marks get designed in.

Texture depth and draft angle travel together. Ask for the plate number and depth at the cosmetic approval gate; reserving draft for deep grain costs nothing, while finding out at mold trial costs a cavity insert.

Move load-bearing features off weld lines. Snaps, bosses and hooks follow flow analysis, not the visual layout.

FAQ: Plastic Material Selection and Texture Draft

Q1: Should an enclosure use ABS or PC+ABS?

A: Without a hard strength requirement, high-gloss ABS, which is cheaper and molds well; with a drop test or assembly stress, PC+ABS. ABS notched impact strength runs in the teens to low twenties in kJ/m2 and PC+ABS reaches 40 and above.

Q2: How does texture depth map to draft angle?

A: Each 0.025 mm of texture depth adds 1 degree. Etched finishes start at 2 degrees, leather or deep grain at 3 degrees. A tall thin sidewall is better given 2.5 degrees or more even with shallow texture.

Q3: How thick should a reinforcing rib be?

A: 0.5 to 0.6 times the main wall, and 0.7 times at the very most. Rib height should stay under 3 times the main wall, and rib spacing at least 2 times the main wall so cooling channels can be routed between them.

Originally published on the Hertz Industrial Design website (hezidesign.com), Materials and Manufacturing column.

iFixit Tears Down the iPhone 18 Pro: How a Variable Aperture and a Vapor Chamber Fit in One Corner

Asher Hu — Tue, 22 Sep 2026 04:37:41 +0000

The iPhone 18 Pro went on sale, and iFixit's teardown landed the next day with a repairability score of 7 out of 10. Two things in the teardown photos held my attention: the variable-aperture mechanism inside the rear camera module, and the new-generation vapor chamber covering the mainboard. Both are fighting for the same real estate in the upper-left corner of the chassis.

Why designers should care: a mechanical aperture is a moving part, and a vapor chamber is a large thin-wall part. Their assembly order and mounting method decide whether this phone earns its 7 — and they also decide how far the camera bump has to stick out.

The core fact: iFixit gave the iPhone 18 Pro a 7/10 repairability score, and the teardown shows two design elements — a variable-aperture mechanical structure in the rear camera module and a new-generation VC vapor chamber.

What the Teardown Actually Shows

Two parts deserve a closer look. One is the variable aperture on the main rear camera, which means Apple has put a moving mechanical assembly inside a phone camera module. Huawei's Mate 50 series used an f/1.4-f/4.0 variable aperture back in 2022, so the Android camp knows this road well — Apple is following for the first time. The other is the new vapor chamber, a routine generational update to the thermal solution with a revised coverage area.

On the 7/10 score: iFixit's rating covers four things — teardown steps, fastening methods, parts availability, and repair documentation — with the weights published on their site. Phones have never scored well because of waterproofing and thinness requirements.

The key design shift: a positively driven aperture blade mechanism is added inside the camera module while the vapor chamber area grows. The two share the upper-left corner, which makes assembly order and mounting method the core structural constraints.

Point 1: Putting a Mechanical Aperture in a Phone — Blades and Actuation Are the Hard Part

Building an aperture into a camera lens is routine: typically 6 to 9 blades, plenty of room, and if it breaks you open it up and repair it. A phone module is different. The outer diameter is constrained, and along the thickness axis it competes with the sensor, the lens stack, and the OIS motor. Add another blade group and a micro actuator, and the module's axial height has to be re-budgeted.

The pitfalls we hit on handheld devices are much the same: moving parts fail in drop tests far more often than fixed ones. Blade anchor points and the return travel of the drive lever all look adequate on a drawing; one round on the drop rig settles it, and the fix usually ends up being thicker joints and added cushioning ribs.

Phones also have to deal with dust ingress. The gaps between blades are a natural dust channel, so where the sealing goes and what material it uses is a decision CMF and structure have to make together.

Point 2: The Vapor Chamber Is Turning From an Accessory Into a Structural Part

Ultra-thin VC vapor chambers are now down to fractions of a millimeter in thickness while covering most of the mainboard. They move heat by phase change inside a sealed cavity, and their effective conductivity is an order of magnitude above graphite sheet — one reason flagship phones have dared to push SoC clocks higher in the last two years.

At that thinness, the chamber's own stiffness is worth using. Bonded to the mainboard, it can share some of the bending load and reduce the number of stiffening ribs. The price is a tougher edge-sealing process: crush it in a drop, lose the cavity, and the thermal performance is gone.

Point 3: The 7 Points Come From Teardown Order, Not Materials

Phones fail to score high for a very concrete reason: IP68 waterproofing relies on gaskets and dispensing, so one teardown destroys the seal and reassembly leaves waterproofing degraded. That is a physical limit no structure can design around.

What can change is the path. Put high-failure-rate parts — battery, screen, camera — earlier in the teardown chain so a few screws get you a replacement, and the score rises on its own. Apple's software-side "repair mode" is the other half of the same idea: let the user put the machine into a serviceable state before touching it.

What it means for structural design: repairability is decided by teardown order, not by any single material. Moving wear items to the front, reducing dispensing area, and consolidating screw types are the three actionable levers.

What Designers Can Take Away

Don't leave micro-mechanisms until the structural phase. By the time industrial design fixes the camera bump height, whether the module contains a moving part is already driving thickness. Hand it to a structural engineer to pull back later, and you are changing the ID.

Plan thermal parts as structural parts. The vapor chamber's footprint should be reserved during ID. Wait until thermal simulation says the area is insufficient, and what you are moving is the midframe.

Repair order is a design output. Which screw comes out first and which flex cable disconnects first — whether that is written into the drawings decides after-sales cost. On small appliances for clients, simply compressing the battery-compartment removal procedure from five steps to three saved a meaningful slice of repair labor.

FAQ: Variable Aperture, Vapor Chambers, and Repairability

Q1: What does a variable aperture on a phone actually do for photos?

A: Mainly it controls light intake and depth of field. In bright daylight a small aperture holds back highlights and prevents blown-out images; at night you want a wide aperture for more light. The blade count also shapes the bokeh highlights — more blades means rounder blobs. Phone modules are tight on space, so they usually use fewer blades than camera lenses.

Q2: How is a vapor chamber different from a graphite sheet?

A: Different heat-transfer principles. Graphite spreads heat through the material's own high conductivity — passive diffusion. A VC chamber is a phase-change loop inside a sealed cavity: evaporation absorbs heat, condensation releases it, and effective conductivity is an order of magnitude higher, which suits the SoC's instantaneous power spikes. The price is process complexity — a bad edge seal leaks and fails.

Q3: What does a 7/10 repairability score mean?

A: Middle-to-upper for the phone category. iFixit scores out of 10 across teardown steps, fastening methods, parts availability, and repair documentation, with published weights. Phones are held back by waterproofing and thinness, so keeping a 7 while adding a mechanical assembly means the teardown path was designed in advance.

Originally published on the Hezi Industrial Design website (hezidesign.com), Industry Watch column.

Plastic Parts Sinking, Drag Marks, Cracking? A 10-Year Structural Designer on Material and Process Selection

Asher Hu — Mon, 21 Sep 2026 10:22:26 +0000

The conclusion first: 80% of the appearance and assembly problems that show up late in a product program are not drawing errors — they are material and process decisions made wrong at the selection stage. Get the shrinkage rate wrong and a 0.3mm wall difference sinks the surface. Pair a texture depth with the wrong draft angle and the first mass-production batch comes out with drag marks. Swap electroplating-grade ABS for ordinary ABS and the housing cracks the day after assembly. This article skips theory and covers only the parameters and pitfalls we paid for across 320+ products in Dongguan.

Core answer: Sink marks come from uneven wall thickness, not from the "wrong material." Keep the main wall at 1.8-3.0mm for ABS and 2.0-3.5mm for PC, make ribs 0.5-0.6x the main wall with an R0.25-R0.5 root fillet, and about 90% of sink marks can be prevented before the mold is cut.

1. Don't Just Read the Shrinkage Table — Anisotropy Is the Real Trap

Most newcomers grab a supplier's shrinkage table, enter 0.5%, and cut the mold. The problem is that the table gives a typical value, while in real molding the shrinkage along the flow direction and perpendicular to it can differ by a factor of two.

We built a handheld barcode scanner housing in PA6 + 30% glass fiber. In glass-filled material the flow-direction shrinkage is about 0.3%, but the cross-flow direction reaches 0.7%-0.9%. The drawing placed the button holes on a uniform 0.5% shrinkage, and the samples came out with oval holes — the buttons jammed during assembly. The fix was to lay out the holes along the glass-fiber flow direction and align the slot orientation with the melt flow.

Keep one habit: for any glass-filled material, dimension hole diameters and center distances on the mold drawing separately according to anisotropy. One coefficient for everything does not work. For high-shrinkage materials like PP on large thin-wall parts, run a mold-flow analysis instead of trusting experience — warpage is PP's regular guest.

2. Texture Depth Determines Draft Angle — Decide Them Together

Texturing (etching) is the most common surface treatment for cosmetic parts. The trap: the structural engineer dimensions 1° draft for a polished surface, the mold shop etches to VDI 3400 grain 27 (Ra about 1.6μm, depth about 25μm), and at ejection the texture grips the side wall — producing alternating light and dark drag marks.

Core answer: 0.5-1° is enough on polished surfaces, but on a textured surface you add 1° of draft for every 0.025mm of texture depth. VDI 27 grain calls for 2-3°, coarse VDI 33 grain (about 50μm deep) calls for 4-5°, and deep-draw areas need one more degree.

We made the upper and lower housings of a desktop speaker in PC+ABS with VDI 27 grain. The first drawing specified 1° draft all around the sides, the mold shop did not flag it, and the tool was cut. At trial molding the upper housing had vertical drag marks all over the side wall — rough enough to feel with a finger and impossible to hide under paint. The rework changed the sides to 2.5°, re-routed the mating surfaces, and re-etched the texture: over 30,000 RMB extra on one mold.

Our process now is: fix the grain number at the structural review, then work backward to the draft angle. One more note: if one mold etches several different grains, the transition between depths needs a step of at least 0.3mm, or the boundary line smears.

3. Surface Treatment Changes Your Structure — Two Iron Rules for Plating and Two-Shot Molding

Many people treat surface finishing as the last step: finish the structure, then pick a process. In reality plating, two-shot molding, and IMD directly change the wall thickness, fillets, and material grade you can use.

Rule one: plating requires dedicated plating-grade ABS. A client once asked for a bright silver plated button. The engineer used ordinary ABS from stock. Samples looked fine, but two days after assembly the button surface showed fine cracks. The reason: the butadiene rubber phase in ordinary ABS is unevenly dispersed, the plating pretreatment chemicals penetrate it, and once the coating stress releases, it cracks. Switching to plating-grade ABS (Chimei PA-727, LG HI-121H and similar) and holding the coating at 15-25μm solved it completely. Structurally, every sharp corner must be rounded to R0.5 or more — plating builds up at sharp corners and burns.

Rule two: never guess the overmold thickness in two-shot molding. For a soft material over a rigid one (TPE over ABS/PC), keep the overmold layer at 0.8-1.5mm. Too thin and it short-shoots; too thick and shrinkage stress lifts and delaminates the edge. On a waterproof flashlight grip, our first version used only 0.6mm of soft material. Trial molding produced a white, detached ring around the edge that lifted with a fingernail. Going to 1.2mm plus a 0.5mm-deep dovetail groove on the rigid side for mechanical interlock finally made it hold.

Core answer: Overmold thickness 0.8-1.5mm, a 0.5mm-deep dovetail groove or through-hole on the rigid side for mechanical interlock, and a 0.3mm tapered fade-out at the soft-material edge to avoid stress concentration and delamination. Bonding also depends on material compatibility — ABS/PC with TPE usually requires the right SEBS base grade.

Frankly, there is no shortcut in materials and processes. You learn them one project at a time. Our team now keeps an internal material and process library: after every product we record the grade used, shrinkage, grain number, draft angle, and the pitfalls we hit. By the three-hundredth product, a glance at a structural drawing avoids most of them.

FAQ: Material and Process Selection

Q: ABS or PC+ABS for a housing — how do I choose?

A: Look at strength and cost. ABS is cheap and easy to process; PC+ABS has better toughness and impact resistance, so outdoor products or ones with drop requirements should prefer PC+ABS.

ABS runs 1.8-3.0mm main wall and suits indoor consumer electronics housings. PC+ABS shrinks about 0.4%-0.6% and has noticeably higher impact strength — products with a 1.2m drop test requirement basically all use it. Note that PC+ABS flows worse than ABS, so deep-draw parts need larger gates and better venting.

Q: How thick can a rib be without causing sink?

A: Rib thickness is 0.5-0.6x the main wall, with an R0.25-R0.5 root fillet and rib height no more than 3x the main wall.

For a 2.0mm main wall, the rib is 1.0-1.2mm. Too thick and the root cools slowly, leaving a sink mark; too tall and it ejects with white marks. If the structure needs more height, use several short ribs rather than one tall rib.

Q: What wall thickness should a die-cast aluminum part have?

A: ADC12 die castings normally run 1.5-4.0mm wall with 1-3° draft. Thick sections over 4mm tend to produce porosity and shrinkage voids.

Transitions between thick and thin must be gradual — avoid jumping from 1.5mm to 5mm. Where threads are needed, leave at least 0.5mm of machining allowance; as-cast threads are not strong enough and are usually tapped afterward. Deburring and shot blasting are mandatory before surface treatment, or powder coating adhesion suffers.

Q: How do I know whether the draft angle on a drawing is enough?

A: Work backward from the surface state: 0.5-1° polished, 2-3° fine grain, 4-5° coarse grain, plus 1° for deep draws.

The safest practice is to confirm grain number, texture depth, and draft angle together at the mold review, and have the mold shop confirm drawability in writing. After one bad experience, our drawings now specify the grain number and its corresponding angle for every surface.

Originally published on the Hezi Industrial Design website (hezidesign.com), Structure Design Field Notes column.

How Much Does Structural Design Cost? A 10-Year Designer Breaks Down the 10,000-100,000 RMB Pricing Logic

Asher Hu — Sun, 20 Sep 2026 10:13:32 +0000

Direct answer: Structural design fees in China's industrial design market typically run from 10,000 to 100,000 RMB, layered by part complexity: 10,000-25,000 RMB for five parts or fewer, 25,000-50,000 RMB for 5-15 parts, 50,000-80,000 RMB for 15-30 parts, and 80,000-150,000 RMB for 30+ parts with transmission or sealing structures. There are three pricing models: fixed project price, hourly rate (1,200-3,000 RMB per person-day), and a 1%-5% royalty on product sales price.

How Much Does Structural Design Usually Cost? Start With a Benchmark Range

Conclusion: Quotes for small and mid-sized consumer products cluster between 15,000 and 50,000 RMB. That range covers the majority of plastic-housing-plus-internal-bracket structural design work on the market. If you are searching for "how structural design fees are calculated" or "structural design quotation standards," use that number as your anchor, then see which tier your own part count falls into.

Breaking down real quotations into three tiers: the first is simple structure — a single plastic housing, five parts or fewer, common in night lights, humidifiers, and Bluetooth speakers. Structural design fee: 10,000-25,000 RMB, design cycle 5-10 working days. The second tier is medium complexity, 10-30 parts including plastic parts, sheet metal, screw bosses, and snap-fit assemblies: 25,000-50,000 RMB, cycle 15-25 days. The third tier is complex electromechanical products, 30+ parts with transmission, sealing, heat dissipation, and waterproofing requirements: 50,000-100,000 RMB or more, cycle 30-60 days. This is also the point most clients miss when they ask about "structural design pricing standards": the price difference is not in the designer's hourly rate, it is in the product's part count and the number of validation rounds.

Direct answer: Structural design fees are not charged as a percentage of product sale price. They are priced on three dimensions: part count, structural complexity, and validation rounds. For the same exterior, doubling the internal part count typically raises the structural design fee by 60%-120%.

Project-Based, Hourly, or Royalty: Which Pricing Model Should You Choose?

Conclusion: If requirements are frozen and the drawing scope is clear, choose a fixed project price. If requirements are still moving and the design must be adjusted as you go, choose hourly. If mass-production volume is predictable and you want to reduce upfront cash outlay, you can negotiate a royalty on sales price. Choosing the wrong model loses more money than the price level itself.

Project-based pricing is the most common approach: the client gets a fixed price plus a deliverables list. The advantage is a predictable total that is easy to push through procurement. The risk is that requirement changes are billed separately — the industry convention is that going beyond the original scope by more than 20% triggers a change order, and revising a structural design once typically adds 2,000-8,000 RMB. Best suited to clients whose product definition is clear and whose exterior is already finalized.

Hourly pricing: the industry rate for a structural design engineer is 150-400 RMB per hour, roughly 1,200-3,000 RMB per person-day, with senior engineers typically at 2,000-2,500 RMB per person-day. The advantage is that neither side loses: you pay for exactly the time spent. The risk is that the total is uncontrollable — if the project drags on for three months, the fee can roll from 30,000 to 80,000 RMB. Best suited to R&D-heavy projects, non-standard equipment, and teams whose requirements are still iterating.

Royalty on sales price: the common ratio is 1%-5%, and it can be tiered — 5% under 10,000 units shipped annually, dropping to 1%-2% above 50,000 units. The advantage is low upfront cash pressure, aligning the design firm with sales performance. The risk is that sales and financial data are opaque; if the client sells well but cannot clearly account for it, the design firm has little way to verify. This model suits partnerships with an existing trust base, a mature category, and a clear path to mass production.

Where the Money Actually Goes in a 10,000-100,000 RMB Structural Design Fee

Conclusion: The composition is roughly 30% concept design, 35% detailed 3D modeling, 20% prototype assembly validation, and 15% mold follow-up and trial-mold revisions. A 50,000 RMB structural design fee typically means 15,000 RMB for concept development, 17,500 RMB for modeling and drawings, 10,000 RMB for prototype validation, and 7,500 RMB for mold follow-up.

Many clients assume structural design is just "drawing a 3D model." What actually consumes the hours is the last three blocks. Concept design fixes the part-splitting logic, snap-fit form, assembly sequence, and ejection direction — this step decides whether the product can be manufactured smoothly, and it carries the highest rework cost. Detailed 3D modeling delivers complete part drawings, exploded views, a BOM, and 2D engineering drawings; this determines whether the mold shop can actually cut the tool from your files. Prototype validation usually means one or two rounds of prototypes, with outsourced prototype cost of 3,000-15,000 RMB per round depending on part count and process. Mold follow-up covers T1 and T2 trial-mold support and revision recommendations; changing a steel mold structure once typically costs 2,000-8,000 RMB.

Direct answer: A structural design cycle typically runs 5-60 working days. A simple single-shell part takes 5-10 days, medium complexity 15-25 days, and complex products with transmission or sealing structures 30-60 days. Rush jobs can compress the schedule by around 30%, but quotes usually rise 20%-40%.

What Pushes the Price Up — and What Can Bring It Down

Conclusion: Five factors drive pricing, in order of weight: part count, moving mechanisms, ingress protection rating, revision count, and delivery schedule. For the same exterior, a product with waterproof structure typically costs 30%-50% more, and one with gears or linkage transmission costs 40%-80% more.

In real quotations from Dongguan product design firms, a few points are commonly underestimated. First, IP waterproof rating: reaching IPX7 requires designing seal grooves, ultrasonic energy directors, or dispensing paths, and that doubles the validation rounds. Second, drop test requirements: a 1.5-meter six-face drop forces thicker walls and revised ribs, pushing both mold and material costs up. Third, safety and certification: products seeking CCC, CE, or UL certification must reserve creepage distance and flame-retardant structure. Fourth, revision count: every time a client's internal review changes the exterior or functional definition, the structure basically has to be redone for a full round.

The room to reduce cost sits in the same places. Starting structural design after the exterior is finalized saves 15%-25% of rework. Reducing part count is the most cost-effective saving of all — every 3-5 parts removed typically cuts the structural design fee by about 10%. Writing safety and test standards into the design input up front avoids a second round of mold changes. And allowing a schedule of 30 days or more usually earns a quote about 20% lower than a rush job.

FAQ: Common Questions About Structural Design Fees

Q: How much does structural design generally cost?

A: A simple single-shell part runs 10,000-25,000 RMB, medium complexity 25,000-50,000 RMB, and products with transmission or sealing structures 50,000-100,000 RMB.

The quote depends mainly on part count, structural complexity, and validation rounds — not on where the designer is located. The reason quotes are lower in Dongguan for the same product is usually a mature supply chain and cheaper prototype and mold revision costs.

Q: Is project-based or hourly pricing more cost-effective?

A: If requirements are frozen, choose project-based pricing — typically 10%-20% cheaper than hourly. If requirements are undecided, choose hourly to avoid disputes over changes.

With hourly pricing, agree on a person-day cap, for example no more than 30 person-days, with anything beyond that requiring written confirmation from both sides. That keeps the risk bounded.

Q: How long does structural design take, and can it be expedited?

A: Simple parts take 5-10 working days, medium complexity 15-25 days, and complex products 30-60 days. Rush jobs can compress the timeline by about 30%, with quotes 20%-40% higher.

Expediting relies on parallel workstreams and extra staffing, but outsourced prototyping and trial-mold scheduling depend on suppliers, so not every step can be compressed.

Q: Does the structural design fee include prototyping and mold costs?

A: No. Outsourced prototypes cost 3,000-15,000 RMB per round, and mold costs are separate. The structural design fee covers drawings and mold follow-up service only.

Before signing, always confirm the quotation boundary: does it include 2D engineering drawings, a BOM, how many revision rounds, and how many mold follow-up visits. Those are where the extra charges pile up later.

Q: Can I pay the structural design fee as a royalty on product sales?

A: Yes. The common ratio is 1%-5%, and tiered royalties work too. It suits long-term partnerships with predictable mass-production volume and a solid trust base.

For a royalty model, agree on the settlement basis up front: factory price or retail price, units shipped or cash collected, and how often settlement happens. An unclear basis is the most common source of disputes.

Originally published on the Hezi Industrial Design website (hezidesign.com), Company/Service column.