Create a baseline
Tap a key once. Its first event has no earlier signal to compare, so it becomes the baseline.
Keyboard chatter.
Test repeats.
Find suspicious duplicate key events that can turn a single press into two characters. Test repeatedly, tune the interval, and verify the pattern before blaming the switch.
Test for chatterRepeat the key.
Inspect the gap.
The panel focuses automatically. Tap normally, then repeat any suspicious key enough times to reveal a pattern.
Remember: a short interval is evidence to investigate, not proof that the switch is defective.
Chatter monitor
Keep this panel focused and tap one key repeatedly
Recent signals
Newest event first · latest 100 retained
- No signals yetPress a key to create its baseline.
A short gap
is a clue.
The tester compares each initial keydown with the previous event from the same physical key. If the gap is shorter than your selected threshold, that event is marked for investigation.
Repeat naturally
Tap the same key several times at your normal pace. Normal hold-to-repeat events are ignored.
Look for a pattern
One warning can come from deliberate fast tapping. Repeated short gaps on the same key are more useful evidence.
Verify elsewhere
Reset and compare another browser, computer, connection, or keyboard before deciding the hardware is at fault.
Bounce, debounce,
and duplicate input.
Mechanical contacts do not always settle into one clean electrical transition. Keyboard firmware normally filters that movement before it becomes a report.
Contact bounce
When metal contacts meet, they may make and break connection briefly before settling. Dust, wear, oxidation, or mechanical damage can make that behavior less stable.
Firmware debounce
The controller ignores rapid transitions for a short window. A longer window can reject more bounce, while an aggressively short window may expose unstable signals.
Browser evidence
This page sees keyboard events after firmware, the connection, the operating system, and the browser have processed them. It cannot inspect the switch contacts directly.
Choose a threshold
with context.
The default 80 ms is a practical screening value, not a universal definition of chatter. Faster typists and deliberate repeated taps may need a shorter threshold.
Strict screening
Highlights only very short same-key gaps. Useful when fast intentional tapping would otherwise create false positives.
General testing
A balanced range for repeated-key checks. The default starts here so suspicious bursts remain visible.
Sensitive observation
Flags a wider interval and can catch slower duplicates, but deliberate double taps are more likely to be marked.
Repeated characters
have more than one cause.
Use the event pattern to narrow the next check, not to skip directly to switch replacement.
- Switch contamination or wear. A mechanical contact can become unstable with debris, oxidation, liquid damage, or long-term use.
- Debounce configuration. Performance firmware or custom QMK/VIA settings may use a window too short for a particular switch.
- Macros and remapping. Keyboard software, accessibility tools, game overlays, and automation utilities can generate duplicate input.
- Connection or firmware faults. A hub, receiver, cable, outdated firmware, or controller fault can alter event delivery.
- Non-contact switches. Optical and Hall-effect boards avoid traditional metal-contact bounce, but sensor contamination, calibration, firmware, or software can still produce repeats.
Troubleshoot from
least risky to most.
Reset and retest after each step so you know which change mattered.
- 01
Rule out software
Disable macros, remappers, accessibility repeat tools, and keyboard utilities temporarily. Compare a plain text editor and another browser.
- 02
Change the connection
Try another direct USB port, cable, wireless receiver position, or computer. Avoid changing several variables at once.
- 03
Check supported settings
Review the manufacturer's firmware, debounce, rapid-trigger, and calibration guidance. Record the original setting before changing it.
- 04
Clean safely
Disconnect power and follow the manufacturer's cleaning instructions. Do not spray regular WD-40 or drip liquid into an assembled switch.
- 05
Replace or repair
On a hot-swap board, test a known-good compatible switch. Leave soldered-switch work to someone with suitable tools and experience.
Private by design.
Key labels and timing intervals remain in this tab's memory. They are not uploaded, saved to an account, or sent to a server.
0events transmittedEvidence, not a verdict.
A webpage cannot see switch contacts or firmware reports. Confirm a repeated pattern under controlled conditions before repairing hardware.
100recent events retainedFAQ
Clear answers about duplicate key events and browser-based chatter testing.
What does a suspected chatter warning mean?
The browser received two non-repeat events from the same physical code within the selected interval. That is worth repeating and comparing, but it does not by itself prove a defective switch.
Why is the first press marked as a baseline?
There is no earlier event for that physical code to compare. The next accepted event establishes the first measurable same-key interval.
Is holding a key counted as chatter?
No. Browsers mark normal operating-system hold-to-repeat keydowns with the repeat property, and this tester ignores those events.
What threshold should I use?
Start at 80 ms, tap naturally, then reduce the threshold if intentional fast double taps are being marked. Reset between formal comparisons so every item was classified under the same setting.
Can this identify the faulty switch?
It can show which physical key code produced repeated short intervals. It cannot determine whether the cause is the switch, firmware, keyboard software, connection, operating system, or browser.
Do optical or Hall-effect keyboards chatter?
They do not use traditional metal-contact closure in the same way, but contamination, calibration, firmware, controller, or software problems can still create duplicate events.
Why do system keys not appear?
The operating system or browser may reserve shortcuts such as the Windows key, screenshots, media controls, or security combinations before a webpage receives them.
Are my keystrokes saved?
No. The latest 100 event records and per-key timing baselines exist only in this page's memory and disappear when you reset or leave.
Method and
further reading.
The screening logic separates initial keydowns from operating-system repeat events before comparing intervals.
Repeat-event behavior
MDN explains the browser's KeyboardEvent.repeat flag used to ignore normal hold-to-repeat events.
Related guides
Browse the keyboard testing guides for controlled diagnostic procedures and clear limits on what browser evidence can prove.
Seen it twice?
Test it again.
Reset the run, keep the threshold fixed, and check whether the same key produces the same pattern.
Run another test