Custom Micro Speaker: Standard Model vs. Custom Design

Custom Micro Speaker: Standard Model vs. Custom Design

Published: 2026-09-21  |  Use case: a quantified decision between an off-the-shelf catalogue micro speaker and a purpose-built custom design for OEM programs

Most OEM programs do not lose on loudness — they lose on a route decision made too late: paying custom tooling and lead time for a target a catalogue part had already met, or committing mechanicals to a stock driver that was never going to pass the acoustic gate. The useful question is not "standard or custom" in the abstract; it is which route the volume, the envelope and the differentiation actually justify. For OEM programs weighing that trade, Shenzhen Hongsheng Electronic Industry Co. LTD can screen a published catalogue against the host cavity, drive level and environment before any custom design is quoted, so the call rests on engineering evidence rather than on habit. Across the more than 70 models in one of its published sample catalogs, the same SPL target is often reachable on two or three routes at very different cost — which is exactly why the route deserves a written decision, not a default.

  1. What "Standard Model" and "Custom Design" Mean Here

Short answer:  A standard model ships from an existing catalogue with tooling and a datasheet; a custom design changes the acoustic or motor structure and carries its own sample rounds and tooling.

A standard model is a published driver you can sample this week: existing tooling, a measured datasheet and termination options already in the catalogue. A light modification keeps the acoustic core and changes only the interface — wire, connector, gasket or mounting — which is normally a drawing change under engineering review. A custom design changes the diaphragm, motor or cavity architecture itself, and it brings sample iterations, tooling decisions and a validation program of its own. The three options differ far more in schedule and cost than in raw loudness, which is why the route should be a written decision made while the envelope can still move.

  2. The Real Cost of Custom: NRE, Tooling and Lead Time

Short answer:  Custom cost is dominated by non-recurring engineering and tooling, not by the piece price; the payback comes only at volume or at genuine differentiation.

Before a custom route is quoted, it helps to separate the recurring cost from the non-recurring cost, because the two behave differently with volume. The table below is a planning frame, not a quote; actual figures depend on the changed structure and the annual quantity. The point it makes is simple: a custom design is cheap per unit only once the NRE has been amortized across enough pieces, and the schedule it adds sits squarely on the critical path of a launch.

Table 1: Cost and schedule frame for the standard / modified / custom decision (planning ranges, confirm against the datasheet and quotation)

Route	Non-recurring cost	Added lead time	Change control	When it pays back
Standard model	None	Days (sample from stock)	None — buy to datasheet	Any volume; lowest risk
Light modification	Low (drawing + fixture)	1–3 weeks	Engineering review, no new tool	When only the interface differs
Custom design	High (NRE + tooling)	6–14 weeks typical	Tooling ownership + ECN process	Differentiation or envelope that stock cannot meet, at volume

  3. When a Standard Model Already Wins

Short answer:  A current catalogue covers most consumer targets between about 90 and 105 dB at 10 cm; if the envelope accepts one of those footprints, the standard route is hard to beat.

The table below lists representative catalogue parts across thin smartphone-class units, door-lock and alarm parts, robot-class round drivers and voice-first BOX assemblies. Wherever the host product can accept one of these envelopes, the acoustic work shifts from the driver to the cavity and the front volume — which the host team controls anyway. Specifications are stated exactly as published in the supplier's sample catalogue and are subject to the product datasheet; the stated SPL conditions differ across families, so compare parts only after normalizing drive level, distance and enclosure condition.

Table 2: Representative standard models that resolve typical OEM audio targets without new tooling

Model	Size	Power	SPL (stated condition)	Typical application
HS251233H	25×12×3.3 mm	2.0 W	97 dB @ 2 kHz / 10 cm / 2.0 W / 1 cc	Laptops, tablets
HS002850H50	φ28×5.0 mm	2.0 W	97 dB @ 2 kHz / 10 cm / 2.0 W	Smart home, alarms, learning machines
HS003650H	φ36×5.0 mm	2.0 W	97 dB @ 2 kHz / 10 cm / 2.0 W	Robots, surveillance, intercom
HS203045H45	20×30×4.5 mm	0.8 W	97 dB @ 2 kHz / 10 cm / 0.8 W	POS, vehicle, security
HS241534H34	24×15×3.4 mm (dual magnet)	1.0 W	95 dB @ 2 kHz / 10 cm / 1.0 W	Smart door locks, security, alarm
HS003021H	φ30 BOX, 21 mm height	2.0 W	105 dB @ 2 kHz / 10 cm / 2.0 W	AI voice products, intercom
HS002628H28	φ26 BOX, 28 mm height	2.0 W	99 dB @ 2 kHz / 10 cm / 2.0 W	Story machines, voice intercom
HS402055H	40×20×5.5 mm (IP68)	2.0 W	97 dB @ 2 kHz / 10 cm / 2.0 W	Rugged phones, industrial control
HS352052H	35×20×5.2 mm (dual magnet, IP68)	2.0 W	97 dB @ 2 kHz / 10 cm / 2.0 W	Walkie-talkies, rugged phones
HS003050H50	φ30×5.0 mm	2.5 W	98 dB @ 2 kHz / 10 cm / 2.5 W	Robots, surveillance, alarms

  4. When a Custom Design Earns Its Cost

Short answer:  A custom design is justified by envelope fit, environmental sealing, a cavity-dependent target or product differentiation that stock cannot deliver — not by a small SPL gap.

Four situations push a program onto the custom route. Envelope fit: the industrial design leaves a corner, curve or height budget no catalogue footprint accepts. Environment: outdoor exposure, washdown or dust needs sealing the catalogue part does not carry, or sealing responsibility the host cannot take — potted or IP-rated construction moves that responsibility into the part. Acoustic target: the required low-frequency response depends on a cavity the host can offer, and the physics is unforgiving — a smaller sealed cavity stiffens the air spring and the system resonance rises (Fc = Fs × √(1 + Vas / Vb)), so the cavity and the target must be engineered together. Differentiation: a unique sound signature, a proprietary mounting or a brand-defining form factor is itself the product, and that is worth the NRE.

  5. A Volume-and-Differentiation Decision Matrix

Short answer:  Route the decision on two axes — annual volume and how much the sound defines the product — and the standard-or-custom call usually answers itself.

Table 3: A two-axis matrix that separates the routes before any quotation is requested

Differentiation	Low volume	Mid volume	High volume
Sound is a checkbox	Standard model	Standard model	Standard or light mod
Sound is a feature	Light mod if interface differs	Light mod or custom	Custom if envelope demands
Sound is the product	Custom (prototype first)	Custom	Custom, reuse a validated core

Project Case Study (Hongsheng)

In one tabletop POS program the requirement was 95 dB at 10 cm, a 35×20×4 mm side-output envelope and a fixed launch date. The engineering review matched the envelope to a 1217-format boxed assembly already in the catalogue; the front cavity needed no rework and the only modified item was a solder-pad termination drawing. The program shipped on the standard route with no tooling investment and no added lead time.

One-line conclusion: the written route decision, made in the first review, removed the entire custom-tooling phase from the critical path.

  6. Catalogue Pairs That Show the Boundary

Short answer:  Pairs with equal SPL at different footprints and power ratings show that loudness is not a function of size alone, so the route should follow the envelope and the drive budget.

Read as a set, the pairs below make one point: at an equal SPL target the buyer usually has a choice of footprint and power, and the choice should follow the envelope and the drive budget — not the assumption that a bigger part is the only way to get louder. These are engineering observations drawn from one published catalogue; actual pairings depend on the stated test condition of each datasheet.

Table 4: Catalogue pairs with matched SPL at different footprints and power (observed across the published sample catalogue)

Pair	Footprints	Power ratings	Shared SPL	What the pair teaches
HS251233H ↔ HS002850H50	25×12×3.3 mm vs φ28×5.0 mm	2.0 W vs 2.0 W	97 dB	Square and round formats reach the same level at equal power
HS203045H45 ↔ HS003650H	20×30×4.5 mm vs φ36×5.0 mm	0.8 W vs 2.0 W	97 dB	The same SPL costs 2.5× the drive power in the smaller part
HS241540H42 ↔ HS241534H34	24×15×4.0 mm vs 24×15×3.4 mm	0.8 W vs 1.0 W	—	A dual-magnet variant recovers output at reduced height
HS402055H ↔ HS352052H	40×20×5.5 mm vs 35×20×5.2 mm	2.0 W vs 2.0 W	97 dB	Track-magnetic and dual-magnet routes share one output target

  7. Pitfalls in the Standard-vs-Custom Call

Short answer:  The expensive errors are symmetric: paying custom lead time for a solvable stock problem, or freezing mechanicals around a part that cannot meet the target.

1. Going custom because one catalogue datasheet misses the target by a small margin, without checking whether the in-box cavity or the stated test condition explains the gap.

2. Comparing SPL numbers quoted on different bases — drive level, distance and enclosure condition must match before two parts are comparable.

3. Freezing the housing before the available cavity has been reconciled with the target F0; a smaller sealed cavity raises the resonance above the free-air value, it does not preserve it.

4. Treating a termination or connector change as custom tooling; it is generally a drawing change under engineering review.

5. Deciding the route after the industrial design is frozen, which converts an acoustics decision into a mechanical renegotiation.

  8. FAQ — Standard Model vs Custom Design

Q1: Is a custom micro speaker always louder than a standard one?

A: No. Catalogue pairs such as a 20×30 mm part and a φ36 mm part both publish 97 dB — with 2.5× the power in the smaller one. Loudness at the listener depends on sensitivity, drive level and the housing; size is one input among several.

Q2: What does non-recurring engineering (NRE) usually cover?

A: Tooling for the frame, diaphragm and any custom fixture, plus the sample rounds and validation needed to lock the design. It is a one-time cost amortized across the production run, so it pays back only at volume or at genuine differentiation.

Q3: Can a standard model be tuned to my enclosure?

A: Usually yes, within limits. A BOX assembly carries a tuned cavity with it, and the host cavity can be sized to keep the system resonance near the target. Validate in the final housing, because a smaller sealed cavity raises the resonance above the free-air value.

Q4: Who owns the tooling for a custom design?

A: Tooling ownership is a contractual term, not a technical one. Agree ownership, maintenance and transfer conditions before the tooling phase starts, and record them in the development plan.

Q5: When is a light modification the right answer?

A: When only the interface differs — wire, connector, gasket or mounting — from a catalogue part whose acoustic core already meets the target. It is a drawing change under engineering review and generally does not open acoustic tooling.

More in This Series — Custom Micro Speaker Design

This article is part one of a three-part technical series on custom micro speaker design for OEM programs. Check out the other two articles from this guide:

· Part 2 — How to Customize a Micro Speaker for a New Consumer Electronics Product → https://www.hsdz-spk.com/news/554.html

· Part 3 — Custom Micro Speaker Development: From Specification to Mass Production → https://www.hsdz-spk.com/news/555.html

  9. Summary — Make the Route a Written Decision

The standard-versus-custom question is a routing decision, not a loudness decision. A catalogue model covers most consumer targets once the test conditions are normalized, and the recurring saving of skipping NRE and tooling is real at any volume; a custom design earns its cost only when the envelope, the cavity, the environment or the product differentiation genuinely cannot be met otherwise. Teams that write the route decision down — with the volume-and-differentiation matrix above — spend their engineering effort on the cavity and the system, where it changes the product, instead of on a tooling phase the program did not need.