The short answer
First decide whether the complaint exists to the eye, only in recorded images, or both. Keep the scene and LED operating condition fixed, turn off automatic camera controls, and repeat a short matrix of mechanical or electronic shutter, frame rate and shutter-speed settings that the camera manufacturer permits. Record the exact LED driver, dimmer/controller and dimming level separately. If the bands change with camera settings while the lighting condition stays fixed, you have demonstrated a camera-light interaction, not proved that the LED is fault-free. If people can see modulation, or the symptom persists across controlled camera settings, stop guessing and obtain product-specific driver, dimmer and temporal-light data from the manufacturer.
1. Separate what the eye sees from what the camera records
Interview the person who reported the problem before touching settings. Ask whether they see brightness pulsing while looking at the sign, streaks when moving their eyes, instability only through a phone preview, horizontal bands in stills, rolling bars in video, colour shifts or exposure changes. Record the viewing distance, ambient light, time of day and whether the effect appears at full output or only while dimmed.
Use two independent observations: a direct visual check by more than one observer, and a camera record with settings written down. The US Department of Energy distinguishes the time-varying light stimulus from the visible or perceptual artefact it can produce. A camera artefact is another measurement path; it is not a substitute for what a person reports.
Treat health or safety complaints seriously. Do not ask someone who experiences discomfort, migraine symptoms or photosensitive reactions to keep repeating the test. Isolate the area where appropriate and refer the system evidence to a competent lighting professional or manufacturer.
| Observed symptom | What it establishes | Next controlled check |
|---|---|---|
| Bands only in one camera or mode | A camera-light interaction is possible | Repeat with fixed exposure and documented shutter modes |
| Rolling bars in several videos | Temporal interaction is likely | Compare frame rate and shutter while the LED setting stays fixed |
| Brightness change visible to observers | A visual temporal artefact may be present | Record operating point and obtain manufacturer data |
| Symptom only at reduced output | The dimming condition is material | Repeat full-output and each affected dimming level |
| Colour or brightness drifts over minutes | Not necessarily a flicker problem | Check supply, thermal and control stability separately |
2. Freeze and record the lighting configuration
Identify the exact LED product, power supply or driver, controller or dimmer, firmware where relevant, control method, supply voltage and connected load. Photograph labels and wiring. Note whether the system is constant voltage or constant current, whether any colour-mixing channels are active and the percentage or scene selected on the controller.
Do not change the driver, dimmer, cable or LED product during the camera matrix. A useful test changes one variable at a time. If the installation has several zones, select one representative zone and record which one. Check whether nearby mains-powered lights, displays or illuminated equipment are also in frame because they can create their own bands.
Repeat at full output and at the exact dimming levels that trigger the complaint. A percentage shown in an app is a control command, not a measured light-output waveform. Do not invent a PWM frequency, modulation depth, PstLM or SVM value when the exact product documentation does not provide it.
3. Put the camera in a repeatable manual state
Record camera or phone model, lens, still or video mode, frame rate, shutter speed, aperture, ISO, white balance, exposure compensation, stabilisation, anti-flicker or variable-shutter setting, and whether the shutter is mechanical, electronic first curtain or fully electronic. Disable automatic exposure and automatic white balance where the equipment allows so the camera does not hide or create differences between runs.
Use a tripod and keep framing, focus, distance and scene brightness constant. Include a neutral reference surface and a written slate identifying the run. For video, capture enough time to show whether a bar moves through the frame. For stills, take several frames at each setting because a single exposure can coincide with a favourable or unfavourable part of a lighting cycle.
Sony explains that a focal-plane shutter can record bright and dark phases at different sensor positions and recommends camera-model features such as mechanical shutter, anti-flicker or variable shutter where available. Those controls are evidence about the camera system; they do not prove a universal shutter value for every LED installation.
4. Run a one-variable camera comparison
Start with the lighting at a fixed, documented output. Use the camera manufacturer's supported settings and local mains guidance as a starting point, then change only one camera parameter per run. Sony notes that changing shutter speed can reduce captured flicker and gives 1/50 s or 1/100 s as common 50 Hz-region starting points, but the effective setting depends on the camera and scene.
Where the camera permits it, compare mechanical shutter with electronic front-curtain or fully electronic capture. For video, compare the project frame rates with several permitted shutter settings. Do not claim that a setting has repaired the lighting; record only whether the captured bands become stronger, weaker, stationary or moving.
Keep the original files and metadata. Social platforms, messaging apps and editing software can change frame rate, exposure or compression. A screen recording of a phone preview is not equivalent to the original camera file.
| Run field | Record exactly | Hold constant |
|---|---|---|
| Lighting | Zone, driver/controller, output level, supply condition | All camera comparisons in that set |
| Camera body | Make, model, firmware and lens | Position, focus and framing |
| Capture mode | Still/video, resolution and frame rate | Lighting and ambient scene |
| Exposure | Shutter, aperture, ISO and exposure mode | White balance and scene |
| Shutter system | Mechanical, electronic first curtain or electronic | Other exposure settings where possible |
| Result | Band direction, width, movement and visibility to eye | Original file and metadata retained |
5. Interpret patterns without overclaiming
If changing only shutter speed, frame rate or shutter type materially changes the bands, the result demonstrates sensitivity to the capture configuration. It does not establish the LED waveform, certify the driver or show that no observer could experience a temporal artefact.
If the direct visual symptom remains while the camera result changes, maintain two fault records: one for visible behaviour and one for imaging. The lighting manufacturer may need operating-point data or instrument measurements; the photographer may still be able to choose a camera configuration that produces clean footage.
If the effect appears only in automatic modes, repeat with fixed manual settings and check whether auto ISO, auto shutter or rolling exposure is changing between frames. If it appears across cameras and settings, check for common lighting controls, dimming, unstable supply or a product-specific incompatibility rather than treating every device as faulty.
6. Worked example: a 24 V sign looks steady but bands on video
A workshop films an illuminated sign at a reduced controller level. Two observers report that the sign looks steady, but dark bars move through 25 fps phone footage. The team records the LED, power supply and controller labels, the selected output level, the phone model and the original file. No product-specific temporal-light metric is available.
They place a camera on a tripod, fix white balance and exposure, and hold the sign at the same output. They capture short clips at the camera's supported 25 fps settings while changing only shutter speed, then repeat at full output. They also compare an available anti-flicker or variable-shutter function according to the camera guide.
The bars alter markedly with shutter configuration and are absent from one supported setting, while direct observers still report steady light. The record therefore says that the captured result depends on camera configuration. It does not say the LED has passed a flicker standard. The production team can use the clean camera setup for that shoot and ask the lighting supplier for exact driver, controller and temporal-light evidence if the project requires a performance limit.
7. Know when camera adjustment is not enough
Escalate when people can see brightness instability, when the symptom causes discomfort, when it occurs at normal full output, when several known-good camera configurations show severe artefacts, or when a client specification requires measured temporal-light performance. Retain the exact operating conditions so a manufacturer or lighting specialist can reproduce the case.
Do not bypass protective devices, open a sealed power supply or change mains wiring to chase a recording problem. Isolate unsafe equipment and use a competent person for electrical investigation. A camera setting is not a remedy for loose connections, overloaded supplies, incompatible controls or thermal faults.
Close the record with two conclusions: the lighting conclusion supported by visual or measured evidence, and the capture conclusion supported by the controlled files. Keeping them separate prevents a clean video setting being mistaken for a lighting certification.
Frequently asked questions
Do horizontal bands in a video prove the LED is visibly flickering?
No. Bands can result from the interaction between time-varying light and the camera's shutter, frame rate and sensor readout. Check direct visual observations and repeat a controlled camera-setting matrix before concluding what people see.
What shutter speed should I use for LED lighting in the UK?
Use the camera manufacturer's guidance for the exact model. Sony lists 1/50 s or 1/100 s as common starting points in 50 Hz regions, but the effective setting varies with the light source, camera mode and scene.
Can I use a phone slow-motion video as a flicker test?
It can reveal that a camera records temporal variation, but it is not a calibrated flicker measurement and does not establish PstLM, SVM, modulation depth or a product pass/fail result.
What should I send the lighting supplier?
Send the LED, driver, dimmer/controller and firmware identifiers; wiring and load; supply and dimming conditions; observer notes; camera model and full settings; and original files showing each controlled run.
Technical references
Use the current datasheet for the exact product you install. These manufacturer and standards resources explain the general principles.
Separate the lighting record from the camera setup
Share the exact LED, driver and control labels with the original files and full camera settings so Glowfy can help organise a reproducible review.