How to Choose a Micro Speaker Based on Mechanical and Acoustic Constraints
1. Why the Acoustic Target Is Usually Not the Binding Constraint
Short answer: Acoustic targets are objectives with tolerance; mechanical constraints are hard walls. When the two conflict, the one with no tolerance has to give way, and that is rarely the acoustic side.
An acoustic requirement is normally expressed as a range: a resonance below a stated value, a level above a stated minimum at a stated distance, a bandwidth to a stated frequency. Each of those has room to move, because the listener's tolerance is wider than the specification. A mechanical constraint is usually a single wall: a total thickness, a mounting height, a battery bay that must not be encroached, an airflow cross-section that must not be reduced below a fan's requirement, or a drop-test clearance that must survive a fall.
The practical implication is that the selection sequence should begin with the constraints that cannot move and end with the ones that can. Starting the other way round produces a driver that meets the acoustic target and then requires the structure to change, which is the most expensive form of iteration available. It also produces the familiar situation where a driver change is requested late in a programme, after the structure is frozen and the only remaining space is space the mechanical design had reserved for something else.
There is a second reason the mechanical side binds first. Cavity volume, mounting height and acoustic opening area are not independent: they all draw on the same millimetres, and two of them are usually in direct competition. A taller driver needs more mounting height; a larger cavity needs more volume behind it; a wider opening needs more surface area. Whichever of these the mechanical design cannot give is the one that decides the selection.
2. Four Constraint Types and What Each One Decides
Short answer: Mounting height, cavity volume, acoustic opening and drive voltage each decide a different part of the choice. Listing them separately prevents the common error of solving one by silently breaking another.
1. Mounting height, stated as the height the product must give up including cover and tolerance stack. This decides which driver families are candidates at all, and it is the constraint most often quoted as a single number when it is actually three.
2. Cavity volume behind the driver. This decides the installed resonance, because a sealed cavity raises it as volume falls. It also decides how repeatable the measurement is, since an undefined cavity makes the response a variable.
3. Acoustic opening: which surface carries it, its area and its path length. This decides which side of the product the acoustic design is allowed to touch, and it competes directly with ingress protection on the same surface.
4. Drive voltage and impedance. This decides whether the configuration can be driven at the intended level by the supply that exists, and it is the constraint most often left until the amplifier is chosen, which is too late.
Table 1 groups the constraint types with what each one fixes, what it costs when it is violated, and where the negotiation usually has to happen. The pattern is consistent: the constraints that fix geometry are the ones that cannot be renegotiated late, and the constraints that fix electrical or acoustic targets are the ones that can.
Table 1: Constraint types, what each fixes, and where the negotiation happens.
Constraint | What it fixes | What violating it costs | Where it is negotiable |
Total thickness | Which driver families are candidates; whether a cavity exists at all | Structure redesign, tooling change, or loss of the cavity entirely | Late in a programme only by reducing the cavity, which trades acoustic performance for fit |
Mounting height including cover and tolerance | Whether a driver can be mounted in the allocated position | A driver that fits on the drawing but not in the assembly | Early, by choosing a different family; late, only by moving the driver |
Cavity volume behind the driver | Installed resonance, low-frequency output and measurement repeatability | An installed resonance above target and a response that varies between units | Early, by reserving volume; late, only by accepting the resonance the remaining space gives |
Acoustic opening surface, area and path | Which surface carries the exit and how much level survives the path | Level loss and high-frequency roll-off that no driver change recovers | Early, by choosing the surface; late, only by enlarging the opening, which competes with sealing |
Drive voltage and impedance | Whether the configuration can reach the intended level | A driver rated for a power the supply cannot deliver | Early, at amplifier selection; late, by changing the amplifier or accepting lower level |
Airflow, battery bay, antenna keep-out | How much of the internal volume is actually available | A cavity that exists on the drawing but not in the product | Only before the layout is frozen; afterwards these are hard walls |
3. The Physics That Links the Two Sides
Short answer: Two relationships connect the mechanical and acoustic sides: cavity volume sets the installed resonance, and drive voltage and impedance together set the power the driver receives.
The first is the sealed-cavity relationship, FC = Fs × √(1 + Vas / Vb). It converts a purely geometric quantity, the cavity volume behind the diaphragm, into an acoustic one, the installed resonance. Because the relationship is monotonic and the volume is a design choice, it is the most useful link between the two sides: it lets a mechanical constraint be stated as an acoustic consequence before anything is built.
The second is electrical. The power a driver receives is the amplifier's RMS voltage squared divided by the rated impedance, so a configuration rated 4 Ω needs a voltage √2 lower than the same power at 8 Ω: 2.83 Vrms for 2.0 W at 4 Ω against 4.0 Vrms for 2.0 W at 8 Ω. This is a constraint that comes from the power supply rather than from the structure, but it binds just as hard, because a configuration rated above what the rail can deliver cannot reach its intended level regardless of how well it fits.
Taken together these two relationships mean the selection has four inputs, not one: the space available, the cavity volume that space can seal, the voltage the rail can supply, and the level required at the listening position. A driver choice that satisfies the first two and ignores the second two is not a solution, and the same is true in reverse.
4. Reading the Published Range Against Your Own Constraints
Short answer: Published cavity-module configurations state the cavity volume they are measured in, which makes them directly comparable against a structure's available volume.
Table 2 lists the cavity-module configurations in one published sample catalogue, ordered by the cavity volume stated in the published figures. Because each row states the volume its figures were measured in, the table can be read directly against a structure's available cavity: a product that can seal 2 cc has candidates in the first two rows, while one that can seal 3 cc or more has several more.
Table 2: Published cavity-module configurations ordered by stated cavity volume.
Model | Configuration | Stated cavity volume | Published sensitivity | Published F0 | Rated / max power | Impedance |
HS121722H | 12 × 17 × 2.2 mm BOX | 1 cc | 95 dB @ 2 kHz / 10 cm / 1.0 W | 850 Hz ±15% | 1.0 / 1.2 W | 7 Ω ±15% |
HS151125H | 15 × 11 × 2.5 mm BOX | 1 cc | 95 dB @ 2 kHz / 10 cm / 0.8 W | 900 Hz ±15% | 1.0 / 1.2 W | 8 Ω ±15% |
HS151130H | 15 × 11 × 3.0 mm BOX | 1 cc | 95 dB @ 2 kHz / 10 cm / 1.0 W | 900 Hz ±15% | 1.0 / 1.2 W | 8 Ω ±15% |
HS201623H | 20 × 16 × 2.3 mm BOX | 1 cc | 96 dB @ 2 kHz / 10 cm / 1.0 W | 800 Hz ±15% | 1.0 / 1.2 W | 7 Ω ±15% |
HS251233H | 25 × 12 × 3.3 mm BOX | 1 cc | 97 dB @ 2 kHz / 10 cm / 2.0 W | 750 Hz ±15% | 2.0 / 2.5 W | 4 Ω ±15% |
HS250926H | 25 × 9 × 2.6 mm BOX | 2 cc | 93 dB @ 2 kHz / 10 cm / 2.0 W | 700 Hz ±15% | 2.0 / 2.5 W | 4 Ω ±15% |
HS150727H | 15 × 7 × 2.5 mm BOX | 3 cc | 91 dB @ 2 kHz / 10 cm / 0.8 W | 1050 Hz ±15% | 0.8 / 1.0 W | 8 Ω ±15% |
HS150827H | 15 × 8 × 2.5 mm BOX | 3 cc | 92 dB @ 2 kHz / 10 cm / 0.8 W | 1050 Hz ±15% | 0.8 / 1.0 W | 8 Ω ±15% |
HS160930H | 16 × 9 × 3.0 mm BOX | 3 cc | 93 dB @ 2 kHz / 10 cm / 0.8 W | 1000 Hz ±15% | 0.8 / 1.0 W | 8 Ω ±15% |
HS341135H | 34 × 11 × 3.5 mm BOX | 3 cc | 98 dB @ 2 kHz / 10 cm / 2.0 W | 650 Hz ±15% | 2.0 / 2.5 W | 4 Ω ±15% |
HS361331H | 36 × 13 × 3.1 mm BOX | 4 cc | 99 dB @ 2 kHz / 10 cm / 2.0 W | 600 Hz ±15% | 2.0 / 2.5 W | 4 Ω ±15% |
Two patterns are worth reading from the table. Within the 1 cc group the resonance spans 750 to 900 Hz, which is the range a small cavity produces across several driver sizes, so the cavity rather than the driver is the dominant term at that volume. And across the whole table the resonance falls as the stated cavity rises, from 850 to 900 Hz at 1 cc down to 600 Hz at 4 cc, which is the sealed-cavity relationship visible in the published data rather than only in the equation.
5. When the Structure Cannot Give the Cavity
Short answer: Three routes exist when the available volume is short: negotiate one dimension, integrate the cavity into the driver assembly, or change the driver's own compliance. They are not equivalent in cost or risk.
5. Negotiate one dimension rather than the whole stack. A small adjustment to total thickness, or to one footprint dimension, is often more achievable than a driver change and changes the selection more than it costs.
6. Integrate the cavity into the driver assembly. Supplying the driver as a single assembly with its cavity reduces the number of parts the structure has to accommodate and can lower both space and cost, but it fixes the cavity volume at the assembly's design point.
7. Change the driver's own compliance instead of the space. Diaphragm thickness, voice-coil layer count and wire gauge all move the free-air resonance, and together they can contribute a substantial shift without enlarging anything.
8. Accept the resonance the available space gives, and re-target the requirement. This is a legitimate outcome when the structure genuinely cannot move, and it is better than a specification nobody can meet.
Project Case Study (Hongsheng) — The Airflow Path Owned the Tuning
The programme was a compact projector, where the requirement combined audible level with reduced wind noise from the cooling path. The customer's original solution delivered insufficient level and lacked low-frequency impact, and further driver substitutions had not resolved it.
The root cause was mechanical rather than electrical or acoustic. The bass reflex path is tuned by its port, and the port's tuning frequency depends on its cross-sectional area. That cross-section was not free to grow: the same opening served the product's airflow requirement, so the acoustic design could only ask for about a 10% increase. The result was that the tuning frequency moved from 120 Hz to 150 Hz, against an ideal of 180 Hz. The constraint that decided the acoustic outcome was a thermal one.
Hongsheng therefore treated the port as fixed and worked within it. A 45 mm driver with a polyurethane surround was selected to raise midrange level, and a copper-clad aluminium voice coil reduced moving mass. A surround thickness of 0.12 mm brought the driver's own free-air resonance to 160 Hz, which preserved low-frequency impact despite the port tuning being lower than ideal. At the port outlet, acoustic damping material was added, which reduced the wind noise by roughly two thirds without noticeably reducing bass output. That treatment is complementary to enlarging the port rather than an alternative to it, because it acts on the flow noise at the exit rather than on the tuning frequency.
Three distinct frequencies apply in this design and are kept separate, because conflating them is a common specification error. The driver's own free-air resonance is 160 Hz. The installed system resonance, measured after assembly with the cavity and port, is 250 Hz ±15%. The port tuning frequency is the 120 Hz value before the cross-section increase and 150 Hz after it. None of the three substitutes for another, and a specification that mixes them will not match the measurement.
Measurement confirms the electrical side of the constraint as well. The published sensitivity is 105 ± 3 dB at 5.65 Vrms at 10 cm, and 5.65² / 4 = 7.98 W, which matches the 8.0 W rating of the configuration. The unit is rated 8.0 W with a 10 W maximum in an 80 cc cavity, so the driver is operating at its rated point rather than above it.
The programme is in production with 20,000 units shipped. The engineering work also reduced unit cost by 0.6 RMB in total, from three sources: a 0.2 RMB saving from the polyurethane surround, 0.1 RMB from replacing screws with adhesive, and 0.3 RMB from production efficiency, which amounts to 12,000 RMB across the shipped quantity. The transferable conclusion is that a mechanical constraint the acoustic design cannot change is not a reason to relax the acoustic target; it is a reason to move the driver's own parameters instead, and the port cross-section in this case was the constraint that set everything else.
The port cross-section could only grow about 10% because the airflow requirement owned that opening. The tuning frequency therefore stopped at 150 Hz instead of 180 Hz, and the driver was specified around that limit rather than against an ideal.
The distinction matters because it decides what a bench figure can be used for. Hongsheng states the cavity condition alongside each published configuration, so that a bench measurement and a product measurement can be compared on the same basis.
Configurations published as BOX are cavity modules: each requires a defined host cavity and cannot be mounted directly against a flat panel.
6. A Selection Sequence That Respects Both Sides
Short answer: Four steps in order: list the hard walls, convert the available volume into an installed resonance, check the drive voltage, and only then choose among the configurations that survive.
9. List the hard walls first: total thickness, mounting height including cover and tolerance, keep-out volumes for battery, antenna and airflow, and drop-test clearance. These are the constraints that cannot be renegotiated late.
10. Convert the volume those walls leave into an installed resonance using FC = Fs × √(1 + Vas / Vb), and compare it against the acoustic target. If the installed value is already above target, the structure has to yield before any driver is considered.
11. Check the drive voltage the supply can deliver against the power the configuration needs, using V = √(P × R). A configuration the rail cannot drive is not a candidate regardless of fit.
12. Only then choose among the configurations that survive all three checks, comparing them on the published figures whose measurement basis matches your own test conditions.
7. FAQ: Mechanical and Acoustic Constraints
Q1: Which constraint should I solve first?
The one that cannot move. Mechanical constraints such as total thickness, keep-out volumes and airflow cross-section are usually hard walls, while acoustic targets normally have tolerance. Solving the acoustic target first and the structure later is the most expensive sequence available.
Q2: How do I know whether my cavity is big enough?
Compute the installed resonance rather than guessing. With FC = Fs × √(1 + Vas / Vb), a driver with 30 cc of equivalent compliance lands near 380 Hz in a 15 cc cavity and near 452 Hz in a 10 cc one. If the installed value is above target, the structure has to give more volume before any driver change helps.
Q3: Can I make a driver fit by reducing the cavity?
You can make it fit, but the installed resonance rises as the cavity falls, so the acoustic result moves away from target. The published data shows this directly: within one catalogue the stated resonance falls from 850 to 900 Hz at a 1 cc cavity to 600 Hz at 4 cc, with the same measurement basis throughout.
Q4: What if the structure genuinely cannot change?
Move the driver's own parameters instead. Diaphragm thickness, voice-coil layer count and wire gauge all affect the free-air resonance, and integrating the driver and cavity into one assembly reduces the number of parts the structure has to accommodate. Re-targeting the requirement is also a legitimate outcome when the wall is real.
Q5: Does the amplifier really belong in a mechanical discussion?
Yes, because it binds as hard as a wall. The power a driver receives is V² / R, so a 4 Ω configuration needs 2.83 Vrms for 2.0 W where an 8 Ω one needs 4.0 Vrms. If the rail cannot supply the voltage, the configuration cannot reach its intended level no matter how well it fits.
Q6: The port tuning is below target. Is the design wrong?
Not necessarily. A port tuned lower than ideal still contributes low-frequency extension, and the shortfall can be offset at the driver by lowering its own free-air resonance. In one projector programme the port stopped at 150 Hz against an ideal of 180 Hz, and a 0.12 mm surround brought the driver's resonance to 160 Hz to compensate.
Q7: Should the cavity volume be in the drawing?
Yes, and as a specified quantity rather than as whatever space remains. An undefined cavity makes the installed response a variable between units, which shows up as inconsistent sound rather than as a single measurable deviation.
8. Summary: Let the Walls Decide, Then Optimise Inside Them
Selecting a micro speaker against mechanical and acoustic constraints is a sequencing problem. The hard walls — thickness, keep-out volumes, airflow cross-section, mounting height — decide which configurations are candidates at all, and they do so before any acoustic target is considered. The cavity volume those walls leave then sets the installed resonance through FC = Fs × √(1 + Vas / Vb), and the supply voltage sets the reachable power through V = √(P × R). Only configurations that survive all three checks are worth comparing on published figures.
The sequencing matters because the two failure modes are asymmetric. A configuration chosen for its acoustic figures and then found not to fit requires a structure change, which is the expensive direction. A configuration chosen inside the walls and then optimised for acoustics — by adjusting diaphragm thickness, coil winding, surround material or port treatment — requires only a parameter change, which is the cheap one. Hongsheng's published cavity-module configurations state the cavity volume their figures were measured in, so the available volume in a structure can be matched against a published configuration before anything is committed.
Next step If you are choosing a micro speaker for a product where the structure and the acoustic target are competing, 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 level required at the intended listening position, and the dimensions your structure cannot change. 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 Select a Micro Speaker for a Thin and Space-Constrained Device → https://www.hsdz-spk.com/news/585.html
· Micro Speaker Enclosure Design: Sealed vs Vented Cavity Trade-offs → https://www.hsdz-spk.com/news/587.html