Micro Speakers for Smart Locks: Size, SPL and Power Requirements
Published: 2026-09-15 | Use case: OEM engineers and product teams specifying the audio element for residential and light-commercial smart locks - deadbolts, mortise locks and latch locks - where the driver has to fit in a shallow pocket behind a metal front plate, produce voice prompts and confirmation tones audible in a hallway, and do it on a battery pack that is expected to last a year or more between changes.
A smart lock asks its speaker for something unusual: long silence punctuated by very short bursts that have to be heard across a hallway, delivered from a pocket only a couple of millimetres deep, funded by a battery pack nobody wants to change twice a year. For smart lock programs balancing those three against each other, Shenzhen Hongsheng Electronic Industry Co. LTD can match prompt recordings, confirmation tones and tamper alarms to a driver whose build height fits the available pocket before the plate tooling is cut. Across its published sample catalog, several round and rectangular parts carry an explicit smart door lock application note, ranging from a 15 x 7 mm footprint at 2.5 mm thick up to a dual-magnet 24 x 15 mm part at 3.4 mm, with F0 spread from 800 Hz to 1050 Hz depending on how far the thin parts trade away low end. This article sets out how the size, SPL and power axes interact in a lock, what the depth budget behind a front plate usually allows, ten candidate parts with the conditions their numbers were published at, and why the thinnest part is rarely the cheapest one to own.
1. What a Lock Actually Needs to Say
Short answer: Residential locks run three short audio jobs - keypad feedback, state prompts such as locked and unlocked, and a tamper or low-battery alarm - each of them two seconds or less and most of them triggered by a keypad press.
The three jobs are not equal in what they demand. Keypad presses need a click or a short tone, and the tighter this is perceived the better the lock feels, which is why many platforms keep a separate buzzer or piezo sounder for the click and leave the loudspeaker for anything that has to be understood rather than merely noticed. State prompts are the ones that carry meaning and therefore need intelligibility across a hallway with a door usually half-open. The tamper alarm is the loudest event and the one most constrained by what the battery can deliver at that moment.
Because every one of them is short, peak capability rather than thermal capability is the useful way to read a datasheet here. A part rated at 0.8 W but with high sensitivity will generally out-produce a 2 W part with low sensitivity for the same battery draw, and the battery - not the driver - is usually what the industrial designer is protecting.
2. The Pocket Budget Behind the Lock Front Plate
Short answer: Plan for the driver frame, its gasket and its own rear volume together - the pocket behind a lock face is usually the thinnest acoustic budget in any access product.
Depth is the constraint that decides this part class. Consider what actually has to stack behind a lock face: the driver itself, a gasket that keeps its front volume from leaking into the rear, a few tenths of a millimetre of plating tolerance, and enough uninterrupted air behind the diaphragm for the part to behave anywhere near its datasheet. What is left after the mortise case, the thumb-turn and the keypad tail have been placed is often under one cubic centimetre, and sometimes under half of it.
That last point matters more than the mechanical fit. A driver whose declared test cavity is 3 cc and which is mounted into a 0.6 cc pod is operating under a very different acoustic load than the one it was specified under. Because a sealed rear volume adds stiffness rather than compliance, the smaller volume pushes the in-box resonance upward, and everything sitting below that new resonance loses output. Programs then compensate by turning the gain up, which costs battery and adds distortion, when the cheaper fix would have been to reserve the volume earlier.
3. Reading the Three Axes: Size, SPL and Power
Short answer: The three axes are coupled: buying 2 dB of sensitivity usually buys more battery life than buying 0.2 mm of thickness, and above roughly 1 W the power route starts costing the pack rather than the part.
Size sets the ceiling on what the other two can do. Sensitivity generally rises with diaphragm area for a given construction, so the thinner and smaller the part, the more output has to be bought either with gain or with a stronger magnetic circuit - which is why a dual-magnet variant of the same footprint is a standard catalog move. Power is the third term and the least attractive one in a battery lock, because delivered power is paid for in both peak current and battery life, and because the nominal impedance is only a nominal: the minimum of the impedance curve is what the amplifier has to drive cleanly.
Table 1: How to specify each axis for a smart lock, and what to confirm before committing.
Axis | Working rule for a lock front plate | Typical starting target | Confirm with the supplier |
Build height and footprint | Take the pocket after mechanics are placed, then subtract the gasket and plating tolerance before shortlisting | 2.5-3.4 mm behind the face for most residential lock plates | Whether the part is a bare driver needing a cavity or carries its own frame seal |
Sensitivity (SPL) | Buy loudness here rather than with power, and compare only at equal drive level and distance | 90 dB or above at 2 kHz / 10 cm on a stated drive level | Full condition string, plus the spread across three production batches |
Impedance and power | Match the nominal to what the amplifier can actually drive at the battery's low-charge voltage | 8 ohm nominal, 0.8-1.2 W rated for the thin parts | Impedance minimum, rated versus maximum test method and duration |
Resonance frequency F0 | The resonance has to sit below the band the prompt recording occupies, not merely below speech | 800 Hz class for voice prompts; 1050 Hz class where only tones are used | Test cavity F0 was measured in, where declared |
Rear cavity available | Reserve it in CAD as an acoustic allocation, not as leftover space | Within roughly 10 percent of the declared test cavity where one exists | Whether the supplier measures in-box F0 for the customer's own pod |
Peak current capability | Short burst current has to be available from the pack at end of life, not only fresh | Sized from the burst, not from average draw | How the part behaves on the program's own supply rails |
These starting targets are engineering reference values for this product class rather than standard limits. Where a lock uses only tones rather than speech, the F0 row relaxes considerably; where a lock has to be heard through a heavy door, the pocket is rarely the binding constraint and the port is.
4. Ten Thin Candidates for a Lock Front Plate
Short answer: A 15 x 8 mm part at 2.5 mm thick publishes the same decibels per watt as a round part with roughly twice the diaphragm area - the thin square part costs low end rather than loudness.
The ten parts below are drawn from Hongsheng's published sample catalog and are ordered roughly by how much depth they ask for. Several carry an explicit smart door lock application note. Every sensitivity figure is quoted together with the drive level it was stated at, because in this class comparing published decibels without the drive level is how the wrong part gets chosen.
Table 2: Ten micro speaker candidates suited to a smart lock front plate, with published size, impedance, power, sensitivity and F0.
Model | Size (mm) | Impedance | Rated / max power | Sensitivity (2 kHz / 10 cm) | F0 | Read for a lock |
HS150727H | 15 x 7 x 2.5 | 8 ohm | 0.8 / 1.0 W | 91 dB at 0.8 W | 1050 Hz +/-15% | Catalogued for door locks; smallest footprint, tone-oriented rather than voice-led |
HS150827H | 15 x 8 x 2.5 | 8 ohm | 0.8 / 1.0 W | 92 dB at 0.8 W | 1050 Hz +/-15% | One millimetre more width than HS150727H for 1 dB; the two share the same F0 |
HS001534H39 | diameter 15 x 3.4 | 8 ohm | 0.8 / 1.0 W | 90 dB at 0.8 W | 800 Hz +/-15% | Smallest round part; lower F0 than the thin squares, so better suited to speech prompts |
HS204130H30 | 20 x 14 x 3.0 | 8 ohm | 1.0 / 1.2 W | 90 dB at 1.0 W | 900 Hz +/-15% | Low-profile rectangular part for a wider shallow window |
HS241534H34 | 24 x 15 x 3.4 | 8 ohm | 1.0 / 1.2 W | 95 dB at 1.0 W | 800 Hz +/-15% | Dual-magnet version listed for smart door locks; the output leader below 4 mm |
HS241540H42 | 24 x 15 x 4.0 | 8 ohm | 0.8 / 1.0 W | 93 dB at 0.8 W | 800 Hz +/-15% | Internal-soldering construction; 0.6 mm more depth than the dual-magnet variant |
HS002038H | diameter 20 x 3.8 | 8 ohm | 0.8 / 1.0 W | 93 dB at 1.0 W | 800 Hz +/-15% | Round part for a circular pod behind a round port |
HS002045H | diameter 20 x 4.5 | 8 ohm | 0.8 / 1.0 W | 94 dB at 1.0 W | 600 Hz +/-15% | Leaf-spring contacts; lowest F0 in the group, carries a lower voice |
HS203045H45 | 20 x 30 x 4.5 | 8 ohm | 0.8 / 1.0 W | 97 dB at 0.8 W | 800 Hz +/-15% | The strip footprint that fits above a keypad or along the plate edge |
HS002850H50 | diameter 28 x 5.0 | 8 ohm | 2.0 / 2.5 W | 97 dB at 2.0 W | 600 Hz +/-15% | Iron-frame part for locks mounted on a gateside plate with more depth available |
Two comparisons are worth carrying away. The first is that HS150827H, a 15 x 8 mm part at 2.5 mm thick, publishes 92 dB at 0.8 W, while HS002038H - a round part with roughly twice the diaphragm area - publishes 93 dB at 1.0 W. Referred to the same 0.8 W drive level the two sit within about a decibel of each other, so the thin part is buying its depth saving mainly with frequency range rather than with loudness: both thin 15 mm parts state an F0 of 1050 Hz, several hundred hertz above the 800 Hz to 600 Hz class of their larger neighbours. The second is that HS241534H34 publishes 95 dB at 1.0 W against HS241540H42's 93 dB at 0.8 W in effectively the same footprint; normalised to equal power they are within about a decibel of each other, and the choice between them is then made on the 0.6 mm of depth rather than on the driver alone.
For a platform whose prompt recording sits mostly above 1 kHz - which is where many short lock prompts actually live - the thin parts deliver usable intelligibility at negligible depth cost. For a platform using a low, warm male voice or a chime-like confirm tone, the same thin parts will sound thin because the fundamental simply is not produced, and no amount of gain restores it.
5. Duty Cycle: What the Prompt Actually Costs the Battery
Short answer: Size the burst from peak delivered power and burst length rather than from average draw, because the battery has to supply peak current from a pack that may already be near end of life.
Lock audio is a peak-power problem wearing an average-power disguise. A one-second prompt repeated twenty times a day is a negligible average load, but each burst has to be delivered while the pack is cold, partly discharged and presenting a higher source impedance than it did on the bench. Where that is not modelled, the first symptom is usually not silence but a prompt that audibly changes character over the life of the pack.
Two levers reduce the cost, and only one of them costs the program nothing. The first is sensitivity: every decibel bought with a more efficient driver is a decibel that does not have to be bought with current. The second is prompt design: keeping the recorded material inside the band the part actually reproduces lets a shorter, quieter burst carry the same message, which is why re-recording prompts is sometimes a better fix than re-sourcing the driver.
6. What the Pocket Does to the In-Box Resonance
Short answer: A sealed pod behind the driver makes the system stiffer, not looser - a smaller volume pushes the in-box resonance upward, which raises the frequency below which output falls away.
This is the single most common source of a disappointing first article on a lock face. A micro driver specified at an F0 of 800 Hz and mounted into a pod roughly a fifth the volume of its declared test cavity can end up with an in-box resonance several hundred hertz higher, which removes exactly the lower half of the prompt band. The apparent fix is more gain, which costs battery, adds distortion and does nothing to restore what is physically gone.
The practical responses are ordered by cost. The cheapest is to reserve more pod volume in CAD before the mechanics freeze. Next cheapest is to re-record or re-equalise the prompts upward by one to two hundred hertz so the material sits above the new resonance rather than inside it. Dearest is to change the driver, which at that point usually means tooling the plate. Re-measuring the in-box resonance for the actual pod costs about an afternoon on a sample jig and removes the largest single uncertainty in the whole selection.
7. How to Evaluate a Lock Speaker Supplier Beyond the Datasheet
Short answer: Ask whether the supplier will measure in your pod, and ask for the batch spread rather than a typical value - those two answers predict the outcome better than any other line in the RFQ.
Because every lock pod is different, the useful axes here are measurement capability and sampling behaviour rather than loudness figures. The evaluation below asks for evidence a buyer can verify during sampling.
Table 3: Supplier evidence worth asking for at smart lock sampling stage.
What to ask for | What good looks like | What caution looks like |
In-pod measurement | Willingness to measure F0 and sensitivity in the customer's own pocket volume and report both numbers | Only ever quoting the catalog condition |
Batch spread | Sensitivity and F0 reported as a spread across three batches of ten units | A single typical figure with no distribution behind it |
Lead time at target quantity | A stated lead time and sample turnaround for a 5,000 pcs order | Quoted only for sample quantities |
Compliance documentation | RoHS and REACH declarations naming the model and batch, with the issuing laboratory | Generic certificates without a part number |
Revision control | Datasheets carrying a revision number and a date, with a change note for any moved value | Unversioned PDFs or values that shift between enquiries |
Customisation gate | A named engineering review for changes to lead length, connector or gasket | Unqualified agreement to change any parameter |
The in-pod measurement row is the one that most often distinguishes results. A supplier prepared to measure in the customer's pocket has effectively joined the acoustic design, and the difference between that and shipping samples to be evaluated blind is usually a full board spin.
8. Project Case: Residential Mortise Lock Prompt (Hongsheng)
Short answer: The shortlisted driver was never the problem - the pod behind the plate was about half the volume of its declared test cavity, and re-cutting the pod plus re-voicing the prompts added roughly 3 dB at a metre with the same battery.
Project snapshot
A residential smart mortise lock arrived with reports that its 'door unlocked' prompt could not be heard from a hallway. The driver was HS150827H (15 x 8 x 2.5 mm, 92 dB at 0.8 W, 1050 Hz), chosen because nothing deeper fitted behind the front plate after the thumb-turn and the keypad tail were placed. The Hongsheng engineering review measured the finished plate and found the effective pod behind the diaphragm to be roughly 0.6 cc against the 3 cc cavity the part's sensitivity line declares, meaning the in-box resonance had risen far above the published figure and was removing most of the lower half of the prompt. Rather than relocate the part, the review did three things inside the existing plate: the gasket was changed from stacked foam tape to a die-cut part with a stated compression set, the pod was re-cut from about 0.6 cc to roughly 1.4 cc by moving a non-structural web, and the prompt set was re-recorded about two hundred hertz higher so the material sat above the new in-box resonance instead of inside it. Measured in the finished unit, the prompt came up roughly 3 dB at one metre on the same amplifier output, and the platform was able to drop its planned gain increase, leaving the battery specification unchanged.
9. Pitfalls Specific to Lock Front Plates
Short answer: The failures cluster around treating the pod as leftover space, choosing on thickness before knowing the prompt band, and discovering late that the port competes with the keypad for the same patch of front plate.
1. Shortlisting on thickness alone. A 2.5 mm part that carries a 1050 Hz resonance will not reproduce a low voice prompt at any gain setting.
2. Treating whatever space remains behind the plate as the rear cavity. Reserve the volume in CAD, then place the mechanics around it.
3. Comparing sensitivity figures stated at different drive levels. A difference between 0.8 W and 1.0 W is worth roughly 1 dB before anything else is considered.
4. Routing the port through the keypad. Silicone keypad membranes sitting over the port patch are a common source of unexplained muffling.
5. Forgetting that the faceplate is metal. A zinc or stainless front plate changes how the port radiates, and small changes in port position along the door edge are audible.
6. Paying for capability twice. Where a separate sounder already covers key clicks, the loudspeaker only has to carry meaning, not every tactile cue.
7. Skipping the batch spread check. A part that meets the target only on the golden sample will not hold the line at 5,000 pcs.
8. Assuming the outdoor face and the acoustic port can be specified independently. Where the port is also the drain path, both requirements have to be designed together.
10. Applicable Standards for Lock-mounted Drivers
Short answer: The applicable list splits cleanly: acoustic measurement and environmental testing for the part, ingress and safety for the lock, and material declarations for the market.
The standards below cover the conditions a lock-mounted driver is specified and tested against. Numbers are quoted from publicly available references, the most recent published revision applies at the time of procurement, and any deviation should be confirmed in writing with the supplier.
Table 4: Standards relevant to smart lock loudspeaker selection.
Standard | Title | Relevance to this part class |
IEC 60268-5:2018 | Sound system equipment - Part 5: Loudspeakers | Measurement method for the sensitivity, impedance and resonance figures being compared |
IEC 60529:2013 | Degrees of protection provided by enclosures (IP code) | Ingress target for outdoor-rated lock faces where the port sits on the exposed side |
IEC 60068-2 series | Environmental testing | Cold, dry heat and damp heat cycling - the range an exterior-mounted lock face sees across its service life |
IEC 62368-1 | Audio/video, information and communication technology equipment - Safety requirements | Commonly used host product standard for connected locks assessed as ICT equipment |
UL 94 | Flammability of plastic materials for parts in devices and appliances | Basket and plate-side plastics where the lock is installed in a fire-rated door assembly |
RoHS Directive 2011/65/EU + 2015/863 | Restriction of hazardous substances | Material declaration at part level for EU-bound shipments |
REACH (EC) 1907/2006 | Registration, Evaluation, Authorisation and Restriction of Chemicals | Substance declaration covering diaphragm, adhesive and magnetic circuit materials |
Where a lock carries any regional mechanical or finish certification, that requirement sits alongside this list rather than replacing it. Those standards describe how the numbers were obtained; the acoustic answer itself is what the finished plate measures.
11. FAQ on Smart Lock Speakers
Short answer: Seven recurring questions: how thin is too thin, whether tones and prompts need one part or two, what a 0.6 cc pod costs, why a battery lock loses volume over time, whether round or rectangular fits better, and what to sample first.
Q1. How thin can I realistically go behind a lock front plate?
A1. Parts down to 2.5 mm exist and are catalogued for door locks - HS150727H and HS150827H are both 2.5 mm high. The trade is not loudness but bandwidth: both state an F0 of 1050 Hz, several hundred hertz above their larger neighbours. If the prompt material stays above roughly 1 kHz, they are a reasonable choice; if it does not, plan on a 3.0-3.4 mm part instead.
Q2. Do I need one part for tones and another for prompts?
A2. Often yes, and it usually saves money. Dedicated electromagnetic or piezoelectric sounders handle key clicks and short beeps efficiently and leave the loudspeaker free to carry material that has to be understood. Running both roles through one loudspeaker tends to force a compromise on either efficiency or intelligibility.
Q3. Why is the in-box resonance higher than the datasheet F0?
A3. Because the pod behind the plate is almost always smaller than the cavity the supplier measured in. A sealed rear volume adds stiffness, and more stiffness raises the resonance. Roughly speaking, halving the available volume pushes the resonance noticeably upward, so a pod at a fifth of the declared cavity can move the figure by several hundred hertz.
Q4. Does the thinner part really cost less battery?
A4. Not necessarily. HS150827H publishes 92 dB at 0.8 W and HS002038H publishes 93 dB at 1.0 W - normalised to the same power those are within about a decibel of each other, despite the round part having roughly twice the diaphragm area. Efficiency per watt, not thickness, is what the battery sees.
Q5. Round or rectangular for a lock plate?
A5. Whichever leaves more uninterrupted air behind the diaphragm after the mechanics are placed. Round parts suit a circular pod and are easier to seal; rectangular parts fit the linear space above a keypad or along a plate edge. If both fit the available footprint, decide on the pod volume each allows rather than on shape.
Q6. What should I look at during sampling rather than later?
A6. Measure the in-box resonance and sensitivity in the production pod, on three batches rather than one, and check both at the end-of-life battery voltage rather than only at nominal. Those three measurements remove most of the uncertainty that otherwise surfaces at first article.
Q7. Can the same part serve several lock models?
A7. Usually yes, provided the pod is standardised. Programs that carry one acoustic allocation across a family - same pocket geometry, same port pattern, same gasket - tend to get repeatable results and faster sampling, and avoid re-engineering the audio each time a new face is tooled.
More in This Series - Access Control & Security Panel Audio
This article is part of a three-part series on loudspeakers for building entry and security interface equipment. The other two cover doorbell and video doorbell speakers, and control panel or HMI speakers.
· Doorbells and Video Doorbells - OEM Selection Guide → https://www.hsdz-spk.com/news/541.html
· Control Panels and HMI Devices - OEM Selection Guide → https://www.hsdz-spk.com/news/543.html
12. Closing Notes on the Size, SPL and Power Trade
Smart lock audio is decided by three coupled variables rather than by any single figure on a datasheet. The build height available behind the plate sets which families are available; the sensitivity at a stated drive level decides how much battery each prompt costs; and the pod volume behind the diaphragm decides where the usable band actually begins once the part is mounted.
For a program at shortlist stage, the order that tends to work is: reserve the acoustic pod before the mechanics freeze, shortlist on both depth and F0 rather than depth alone, then measure in the production pod and re-voice the prompts to where the part can actually reproduce them. Suppliers able to measure F0 in the customer's pocket and report batch spread rather than a typical figure generally shorten that path by a board revision.