Flicker-Free Medical Displays: DC Dimming, PWM and Why It Matters at the Bedside

Flicker is one of the few display characteristics that a specification can define completely and still miss in practice, because the failure appears at the dimmest setting, in peripheral vision, at night – three conditions a bench measurement rarely reproduces. For medical devices this matters twice over: clinicians work under these conditions, and cameras in endoscopy and microscopy see modulation that the eye does not.

How PWM dimming creates perceived flicker

Pulse-width dimming turns the light on and off quickly.

To reduce average brightness, a pulse-width driver shortens the on-time while keeping the current constant. At high brightness the pulses are long and closely spaced and the eye integrates them; at very low brightness the duty cycle becomes small, and both the modulation depth and any variation in pulse timing become perceptible, especially in peripheral vision where the eye is more sensitive to temporal change. If the pulse frequency also drifts with load or temperature, the modulation becomes irregular, which is more noticeable than regular modulation of the same depth.

The failure mode in practice is a bedside device that is perfectly comfortable at 30 per cent brightness and irritating at 3 per cent – and 3 per cent is where it sits all night.

Where the problem shows up: camera-based systems and sensitive users

Cameras see what the eye integrates away.

A rolling-shutter camera exposed for a short interval captures one part of the modulation cycle, producing banding that appears in recorded procedures or in image-guided systems. This is a functional problem rather than a comfort one: an endoscopy or microscopy image with banding loses detail, and in a system where the display is part of the imaging chain the artefact appears in documentation and in recordings. Users with heightened sensitivity to flicker also report headaches and eye strain, which becomes an occupational-health matter in a ward where the same screen is watched for a whole shift.

Both cases argue for specifying dimming behaviour rather than assuming that a “flicker-free” label covers the low end.

Flicker metrics that matter in a specification

Specify depth and frequency, not a marketing term.

The useful metrics are the modulation depth – the percentage variation of luminance around its mean – and the frequency content of that variation, measured at the brightness levels actually used. A specification can state a maximum modulation depth across a defined dimming range, and require that no prominent frequency component falls in a range that is visible or that interacts with the camera frame rate. Percent flicker and flicker index are the two conventions most often used, and the specification should name which one it means.

Whichever metric is chosen, the measurement conditions must travel with the number: brightness setting, temperature, and the instrument’s bandwidth. Flicker measured with a slow meter understates the problem.

DC dimming limits: colour shift and low-brightness behaviour

DC dimming removes modulation and introduces other trade-offs.

Dimming by reducing LED current eliminates the modulation entirely, but LED colour shifts as drive current changes, so a display dimmed far down can become warmer or cooler than it is at full brightness. At very low currents the driver’s own accuracy and noise dominate, and some panels show uneven luminance or a slight flicker from the driver’s control loop rather than from PWM. The practical limit of DC dimming is therefore set by the acceptable colour shift and by how low the driver can accurately regulate.

Where colour consistency at dimmed levels matters, the specification should include a colour-drift limit across the dimming range, not only a luminance limit.

Hybrid dimming strategies

Use each method where it works best.

A hybrid approach uses DC dimming in the low range, where modulation would be visible, and PWM at higher brightness where the pulses are short relative to the eye’s integration and colour accuracy is easier to hold. A variant keeps PWM but raises the frequency well above the range where it interacts with cameras, accepting a small efficiency cost. Both approaches can satisfy a demanding specification, and both need their own measurement to prove it.

CDTech’s notes on achieving flicker-free dimming describe the implementation options, and the medical-specific trade-offs are summarised in flicker-free displays for medical equipment.

Measuring flicker with a photodiode or light meter

The instrument’s bandwidth decides whether you see the problem.

A photodiode with a fast amplifier and an oscilloscope shows the waveform directly, which is the most informative measurement: depth, shape and any irregularity are all visible. A flicker meter gives a single comparable number but only if its bandwidth is high enough for the frequencies involved. Placing the sensor at a defined distance and angle from the screen, with the surrounding light controlled, makes the result repeatable.

Record the measurement at several brightness levels, including the lowest setting the device uses in service, and at both temperature extremes if the device operates across a range. CDTech’s material on building a display optical test setup covers the equipment and procedure side.

Interaction with camera exposure in endoscopy and microscopy

Frame rate and pulse frequency can beat against each other.

When a camera records a screen, or when a display is part of an imaging chain, the relationship between the camera’s exposure interval and the display’s modulation determines whether banding appears. Where the display is dimmed with PWM, the risk rises at low duty cycles; where the camera’s frame rate is close to the pulse frequency, a slow beat appears that changes over time. The engineering answers are to keep modulation shallow, to keep pulse frequency well above the camera’s frame rate or synchronised with it, or to use DC dimming in the recording modes used for documentation.

Test with the actual camera and the actual display settings rather than in isolation, because the interaction is what matters.

Night mode requirements in patient rooms

The night setting is the hardest operating point.

At night the device is dimmed to a level that does not disturb the patient, which is exactly where PWM modulation is most visible and where colour shift from DC dimming is most noticeable. A night-mode requirement should state the target luminance, the maximum modulation depth at that level, the acceptable colour shift and the transition behaviour when the room lighting changes. Smooth transitions matter as much as the steady state: a sudden step in brightness at 3 a.m. is noticed and remembered.

Validate the night setting in a dark room with the screen at the position it occupies in service, and include the effect of any front surface treatment on perceived brightness.

Test protocol to include in a purchase specification

A short protocol prevents a long argument.

1. Measure at the lowest, a mid and the maximum brightness setting used in service.
2. Use a fast photodiode or a flicker meter with stated bandwidth, at a fixed distance and angle.
3. Report modulation depth, dominant frequency and waveform character at each setting.
4. Repeat at the temperature extremes if the device operates over a range.
5. Record colour drift across the dimming range where colour consistency matters.
6. Record the camera interaction test result for any recording or imaging use.

The supporting measurement discipline – luminance, uniformity and stability – is described in CDTech’s notes on measuring brightness, contrast and uniformity, and quality-system expectations for medical suppliers appear through bodies such as AAMI.

Supplier questions about dimming implementation

Ask how it is done, not whether it is done.

• Which dimming method is used, and does it change across the brightness range?
• What is the minimum duty cycle or current at the lowest service brightness?
• What is the measured modulation depth at that level?
• How does colour shift across the dimming range?
• Does the design change with temperature or input voltage?
• What measurement was performed, with which instrument and at which settings?

A supplier that answers these with numbers has designed the dimming rather than inherited it. Where the device presents information used clinically, the manufacturer’s obligations sit within frameworks such as IEC 60601-1 and the EU medical devices framework.