How to Select a Micro Speaker for a Thin and Space-Constrained Device
1. The Problem With Specifying Thickness Alone
Short answer: Thickness is a total budget, not a single specification. In a thin housing the driver height, the cavity volume and the acoustic opening path all compete for the same millimetres, so satisfying one constraint can silently remove another.
The common request is precise and incomplete: fit a speaker into 2.5 mm. What that number does not say is whether 2.5 mm is the driver's own height, the total stack including the cavity, or the distance from the front face of the product to the nearest structural rib. Each of those three readings leads to a completely different component.
There is also a second assumption embedded in the request: that the remaining space can be filled with sound. In practice the acoustic opening, the cavity and the driver's own magnetic assembly are the three consumers of that space, and the cavity is usually the one that gets cut when a programme runs late. A driver that fits beautifully into 2.3 mm but leaves no volume behind it produces a different result from a slightly taller driver with a defined cavity behind it, and the datasheet will not tell you which one your product will get.
2. How to Read the Five Specs That Decide Fit
Short answer: Fit is decided by mounting height, footprint, cavity volume, sound output direction and acoustic opening area. A datasheet that lists any four of these without the fifth is describing a component, not a fit.
1. Mounting height, not overall height: the number that matters is the height the product must give up. Cover, flange and any tolerance stack sit on top of it.
2. Footprint in two dimensions, and whether the mounting is adhesive, screw or solder terminal. Adhesive mounting saves height but fixes the acoustic position permanently.
3. Required cavity volume behind the driver. A sealed cavity raises the system resonance; a driver specified free-air will not measure the same once installed.
4. Sound output direction: front-firing, side-firing or rear-firing changes which surface of the product has to carry the opening and therefore how much of the budget the opening consumes.
5. Acoustic opening area and path length. A short, wide opening is easier to keep free of grille cloth resistance than a long, narrow one, and it costs less of the remaining height.
3. The Physics Behind the Thin-Device Trade
Short answer: Two numbers move in opposite directions when you thin a driver: resonant frequency tends to rise as height falls, while radiation area tends to shrink. Both directions reduce low-frequency output, so thinning costs output twice.
The relevant relationships are worth stating plainly, because they are what makes the trade-off a real trade rather than a preference. For a sealed cavity the loaded resonance follows the familiar form, where the free-air resonance of the driver and the compliance of its suspension set the nominal value, and the cavity compliance pulls the installed value upward as the cavity volume shrinks: FC = Fs × √(1 + Vas / Vb). A smaller cavity therefore raises the installed resonance, not lowers it.
The second effect is geometric. As a driver's height is reduced to fit a thinner housing, the usable cone area usually shrinks with it, because area and height are coupled by the magnetic and suspension architecture. Output scales with radiating area at a given drive level, so a smaller diaphragm delivers less acoustic output at the same voltage. These two effects point the same way: thinning the driver reduces output and raises resonance at the same time. That is why a thin device rarely recovers its lost low end by picking a slightly thinner driver — the trade does not reverse.
There is a related point about thin construction and durability that is often stated too strongly. A thin driver is not automatically a short-lived one. Long-term stability depends on the adhesive joint, the diaphragm and surround material, the magnet, the voice coil, the suspension, the potting and sealing, and the assembly process. Thin construction can make material selection and joint design more sensitive to the application's temperature and humidity conditions, which is a statement about design discipline rather than a statement about service life.
4. Thin Configurations in a Published Sample Range
Short answer: Configurations down to 2.2 mm mounting height are available, and within one power class the range spans from 0.72 cm² to 4.68 cm² of diaphragm area — a difference far larger than the height difference.
Table 1 groups published configurations with a mounting height of 3.6 mm or less by height, then by rated power within each height band. It is worth reading down the power columns rather than across them. In the 1.0 W group the mounting height varies only from 2.2 mm to 3.4 mm, but the diaphragm area ranges from 1.65 cm² to 3.60 cm². In the 2.0 W group the same pattern is stronger: heights from 2.6 mm to 3.5 mm, areas from 2.25 cm² to 4.68 cm². Height is a coarse constraint; area is where the output actually lives.
Table 1: Configurations with mounting height of 3.6 mm or less, grouped by rated power.
Model | Configuration | Height | Diaphragm area | Rated / max power | SPL (basis as published) | F0 | Impedance |
HS121722H | 12 × 17 mm BOX, 1 cc | 2.2 mm | 2.04 cm² | 1.0 / 1.2 W | 95 dB @ 2 kHz / 10 cm / 1.0 W / 1 cc BOX | 850 Hz ±15% | 7 Ω ±15% |
HS201623H | 20 × 16 mm BOX, 1 cc | 2.3 mm | 3.20 cm² | 1.0 / 1.2 W | 96 dB @ 2 kHz / 10 cm / 1.0 W / 1 cc BOX | 800 Hz ±15% | 7 Ω ±15% |
HS250926H | 25 × 9 mm BOX, 2 cc | 2.6 mm | 2.25 cm² | 2.0 / 2.5 W | 93 dB @ 2 kHz / 10 cm / 2.0 W / 2 cc BOX | 700 Hz ±15% | 4 Ω ±15% |
HS361331H | 36 × 13 mm BOX, 4 cc | 3.1 mm | 4.68 cm² | 2.0 / 2.5 W | 99 dB @ 2 kHz / 10 cm / 2.0 W / 4 cc BOX | 600 Hz ±15% | 4 Ω ±15% |
HS251233H | 25 × 12 mm BOX, 1 cc | 3.3 mm | 3.00 cm² | 2.0 / 2.5 W | 97 dB @ 2 kHz / 10 cm / 2.0 W / 1 cc BOX | 750 Hz ±15% | 4 Ω ±15% |
HS341135H | 34 × 11 mm BOX, 3 cc | 3.5 mm | 3.74 cm² | 2.0 / 2.5 W | 98 dB @ 2 kHz / 10 cm / 2.0 W / 3 cc BOX | 650 Hz ±15% | 4 Ω ±15% |
HS001534H39 | φ15 × 3.4 mm round | 3.4 mm | 1.77 cm² | 0.8 / 1.0 W | 90 dB @ 2 kHz / 10 cm / 0.8 W | 800 Hz ±15% | 8 Ω ±15% |
HS241534H34 | 24 × 15 mm, solder tag | 3.4 mm | 3.60 cm² | 1.0 / 1.2 W | 95 dB @ 2 kHz / 10 cm / 1.0 W | 800 Hz ±15% | 8 Ω ±15% |
HS150727H | 15 × 7 mm, 3 cc BOX | 2.5 mm | 1.05 cm² | 0.8 / 1.0 W | 91 dB @ 2 kHz / 10 cm / 0.8 W / 3 cc BOX | 1050 Hz ±15% | 8 Ω ±15% |
HS151125H | 15 × 11 mm, 1 cc BOX | 2.5 mm | 1.65 cm² | 1.0 / 1.2 W | 95 dB @ 2 kHz / 10 cm / 1.0 W / 1 cc BOX | 900 Hz ±15% | 8 Ω ±15% |
HS160930H | 16 × 9 mm, 3 cc BOX | 3.0 mm | 1.44 cm² | 0.8 / 1.0 W | 93 dB @ 2 kHz / 10 cm / 0.8 W / 3 cc BOX | 1000 Hz ±15% | 8 Ω ±15% |
HS151130H | 15 × 11 mm, 1 cc BOX | 3.0 mm | 1.65 cm² | 1.0 / 1.2 W | 95 dB @ 2 kHz / 10 cm / 1.0 W / 1 cc BOX | 900 Hz ±15% | 8 Ω ±15% |
HS204130H30 | 20 × 14 mm, round magnetic | 3.0 mm | 2.80 cm² | 1.0 / 1.2 W | 90 dB @ 2 kHz / 10 cm / 1.0 W | 900 Hz ±15% | 8 Ω ±15% |
HS280935H35 | 28 × 9 mm track magnetic | 3.5 mm | 2.52 cm² | 1.0 / 1.2 W | 91 dB @ 2 kHz / 10 cm / 1.0 W | 900 Hz ±15% | 8 Ω ±15% |
5. Four Ways to Buy Space Back
Short answer: When the height budget is fixed, the recoverable space usually comes from the footprint and the opening, not from the driver. Side-firing routes, wider footprints, defined cavity modules and a single thin dimension each return a different amount.
6. Change the output direction rather than the driver. A side-firing configuration moves the acoustic exit from the front face to a side surface, which often has more available length than the face the product designer wanted to keep unbroken.
7. Trade footprint for height. Within a fixed height band the published range above shows area varying by a factor of roughly two, so giving the driver a longer footprint is usually the cheapest way to recover output.
8. Use a cavity module instead of a bare driver. A BOX configuration carries its own defined volume, which converts an undefined space in the product into a specified one — often the only way to get repeatable measurement conditions.
9. Negotiate one dimension, not all three. In projects where total thickness is fixed by an industrial design target, a small adjustment to one dimension is frequently more achievable than a driver change, and it changes the selection more than it costs.
Project Case Study (Hongsheng) — A 19.9 cc Cavy That Started at 3.5 cc
The requirement came from a desktop AI companion robot with a defined acoustic target set against a reference product: rounded low end, non-diffuse sound, clear articulation. The available space was 41 × 27 × 18 mm, roughly 19.9 cc in total. Two structural facts shaped everything that followed. First, the product's acoustic cavity could not be sealed in its original structure, and an unsealed cavity let bass energy disperse into the housing, mixing with the midrange and reducing clarity. Second, the general-purpose configurations small enough to fit offered a free-air resonance around 800 Hz, far above the roughly 450 Hz the target required, and no substitution among the available 28 mm drivers, 27 mm multimedia drivers or cavity variants of the same family changed that.
The root cause was therefore not the driver model. It was that the original structure had not reserved a controllable acoustic volume. Hongsheng handled the two steps in order, because either one alone leaves the problem in place. The product structure was first adjusted to give the cavity a defined space, taking the volume from 3.5 cc to 12 cc — about 60% of what the available space could hold — which required a structural tooling change. Only after the cavity existed was the driver itself optimised.
Three material and winding changes brought the driver's free-air resonance down from 570 Hz to 420 Hz without enlarging the driver at all. The diaphragm thickness was reduced from 0.7 mm to 0.6 mm; for a flat diaphragm whose bending mode governs, frequency scales with thickness, so the reduction moves the resonance in the required direction. The voice coil was rewound from two layers to four, and the winding wire gauge increased from 0.11 mm to 0.12 mm; both increase moving mass, which lowers resonance through mass loading. The three changes together account for the full 150 Hz.
Measurement on the assembled module confirms the outcome. The test bench reports a resonance of 429.49 Hz with an impedance value of 5 Ω, and the impedance curve shows its peak at the same point near 430 Hz, so the value is confirmed by two independent measurements. The frequency response reaches a 95–100 dB plateau from roughly 500 Hz and stays near 100–105 dB through 8 to 10 kHz before rolling off, which is how the required bandwidth was verified. Total harmonic distortion falls below 5% above about 500 Hz and stays under 3% through the 1 to 10 kHz band. The delivered part is rated 2.0 W with 2.5 W maximum, driven at 2.83 Vrms.
The project reached design freeze, and the customer confirmed the result met the requirement. The tooling change cost 5,000 RMB, against an approximate 300,000 RMB tooling cost for the complete product, and the housing material cost rose by 0.1 RMB. In a project where the total available volume was under 20 cc, the conclusion that carried forward was that cavity first and driver second is the order that works — and that a driver's own materials can contribute roughly 150 Hz of resonance adjustment before any larger driver is considered.
In under 20 cc of available space, three material and winding changes gave 150 Hz of resonance adjustment, but only after the structure reserved a defined cavity.
The two-step approach is also what Hongsheng applies as standard on space-constrained programmes: the cavity is defined against the product structure first, and the driver is then specified for that cavity rather than in isolation.
6. What to Confirm With the Supplier Before Freezing the Structure
Short answer: Eight items should be settled with the supplier before the product's tooling is committed, and the cavity volume must be one of them rather than a consequence of whatever space is left over.
10. The exact mounting height, stated as a number the product must give up, including cover and tolerance stack.
11. The required cavity volume behind the driver, and whether the driver is specified free-air or in a defined cavity — the measured resonance differs between the two.
12. Whether the supplied configuration is a bare driver or a cavity module. A configuration published as BOX use only requires a defined host cavity and cannot be mounted against a flat panel.
13. The sound output direction, and which surface of the product must therefore carry the acoustic opening.
14. The acoustic opening area, path length and any grille material in the path, since grille cloth and mesh resistance act on high frequencies first.
15. The measurement basis for the published sensitivity, including distance, test voltage and cavity condition, because these are not interchangeable between suppliers.
16. Whether the driver can be supplied with an output pattern or mounting arrangement suited to the acoustic opening, rather than relying on a general-purpose front-firing position.
17. The cost of the structural tooling change, if the cavity volume cannot be met in the original structure. This is usually known earlier than programmes expect.
7. FAQ: Micro Speakers for Thin and Space-Constrained Devices
Q1: What is the thinnest micro speaker that still produces usable output?
Published configurations reach a mounting height of 2.2 mm with a 1.0 W rating and a 95 dB sensitivity figure. At that height the constraint that emerges is not output but resonance: the smallest configurations in the published range sit at 850 Hz or above, which is fine for voice announcement and unsuitable where low-frequency content is expected.
Q2: Is a thinner driver always quieter?
Within a power class, the published range shows the opposite relationship in places. In the 2.0 W group, a 3.1 mm configuration publishes 99 dB while a 2.6 mm configuration publishes 93 dB, because the taller one has roughly twice the diaphragm area. Height constrains the choice; area determines the output.
Q3: Do I need to specify the cavity volume?
Yes. Once a driver is installed in a sealed cavity its resonance shifts according to the cavity volume, with smaller volumes pushing the installed resonance upward. A driver specified as free-air will not measure the same as the assembled product, and the datasheet value will not match what the customer hears.
Q4: What is the difference between a BOX configuration and a bare driver?
A BOX configuration is supplied as a cavity module with a defined volume. It cannot be mounted directly against a flat panel because its acoustic behaviour depends on the cavity it is built around. Several published configurations are marked for cavity use only and should be treated accordingly in a specification.
Q5: Can a side-firing driver solve a thin-front-face problem?
Often, yes. Moving the acoustic exit from the front face to a side surface converts a face-length problem into a side-length problem, and side surfaces in thin products are frequently less constrained by the industrial design target. The trade is that the opening now has to be designed into a structural surface, including its water and dust protection.
Q6: Does a thin driver fail sooner?
Not inherently. Service life depends on the adhesive joint, diaphragm and surround material, magnet, voice coil, suspension, potting and sealing, and the assembly process. Thin construction makes material selection and joint design more sensitive to the application's temperature and humidity conditions, which is a design-discipline point rather than a service-life limit.
Q7: How much can be recovered by enlarging the footprint instead of the height?
Within the sub-3.6 mm group in Table 1, the 1.0 W class spans 1.65 to 3.60 cm² of diaphragm area while heights differ by only 1.2 mm. The area range is the larger and usually cheaper lever, because footprint length is often available where total thickness is not.
8. Summary: Allocate the Millimetres Before Choosing the Driver
Thinness is a budget with three consumers, and the sequence that works is to allocate it before selecting anything. Decide how much height the product gives up, how much volume goes behind the driver, which surface carries the acoustic opening, and only then look at candidate configurations. When that order is followed, the published range shows that height is the coarse constraint and diaphragm area is the fine one, and the choice within a height band is wider than the datasheet summary suggests.
The harder lesson from small-volume projects is about ordering. A cavity that is defined first and a driver optimised second produces a measurable result; a driver optimised inside an undefined cavity produces a result nobody can repeat, because the cavity becomes a variable. Hongsheng can evaluate a required mounting height against a published configuration before a product structure is committed, and its engineering team can supply cavity modules and low-profile drivers down to 2.2 mm mounting height for voice announcement, portable listening and industrial housings.
Next step If you are evaluating a micro speaker for a thin or space-constrained device, the shortest route to a configuration worth testing is to state five things: the space available for the driver and its cavity, the impedance the amplifier will drive, the power the rail can supply, the target level at the intended listening position, and the surface that has to carry the acoustic opening. With those specified, our engineering team can recommend a suitable configuration for evaluation, or state plainly which part of the acoustic design has to change first.
More in This Series
· How to Compare Micro Speaker SPL, Power and Impedance Correctly →https://www.hsdz-spk.com/news/582.html
· Why Does My Thin-Device Speaker Sound Weak? Miniature Speaker FAQ →https://www.hsdz-spk.com/news/583.html