Back to Silas S. Brown's home page

Lighting notes

This is posted in the hope that it is useful but without any warranty. LED | CFL | Halogen | Pygmy | ESL | Task lighting | Spectrum issues

LED room-light flicker

My notes on this were superseded by excellent discussions of flicker at LEDBenchmark Australia and German consultant Peter Erwin, and the IEEE took notice and published their recommendations in a 76-page standard: IEEE 1789-2015.

IEEE 1789 limits driver-circuitry's flicker modulation fraction to Hz/1250 of frequencies above 90Hz and Hz/4000 of those below, with an additional 2½× reduction recommended (and plenty of studies are cited to show that we sufferers are not just making this up---stroboscopic "phantom array" effects are particularly irritating if you have a condition like nystagmus to aggravate your saccade movements and V5 blindsight to make you more sensitive to peripheral movement at photopic light levels, but sales personnel tend to confidently deny any LED products flicker because they don't see it). Hopefully the existence of IEEE 1789 will now increase the number of off-the-shelf LED products with acceptably low flicker, but it's still rarely discussed on packaging.

B22, E27 etc LEDs

Integrated LED fixtures

LED tubes

In 2024 in Dorset we bought a V-Tac VT-151 "LED tube" designed to replace a 5-foot fluorescent strip light without requiring removal of old-style magnetic ballasts (it was supplied with an "LED starter"---a straight-through connection with fuse---to replace the fluorescent starter), and we found no noticeable or camera-detected flicker from it.

Other LED

Many LED lamps are also subject to phosphor degradation over time.

CFL phosphor leakage and degradation

21st-century versions of "energy saving" CFL bulbs (with high-frequency electronic ballasts) give reasonably good light but they have other issues, not least of which is decreased availability when more retailers prefer LED.

Due to concerns about UV leakage, CFLs should not be used at short distances (see task lighting).

Additionally, all CFLs (like any phosphor-based device) gradually dim due to phosphor degradation.

Details:

In areas with high humidity and/or short running times (e.g. small kitchens or bathrooms), CFL electronics can cut out early. If this happens often enough to make CFLs uneconomical and unenvironmental (because you "get through them" too quickly) then see below.

Incandescents for humid rooms

If you experience CFLs cutting out very quickly in humid rooms, and you cannot control the humidity, you might be stuck with bulbs that do not include PCBs, which means either flickering non-rectified LEDs or hot power-hungry incandescents (switch off whenever possible).

(These notes assume the room does not have fluorescent tube fittings. If it does then you might have a problem: some high-frequency flicker-free electrical ballasts can take only 85% RH and only for 30-60 days/year; others are more tolerant, but many fittings try to be robust by using a magnetic ballast, which flickers at twice the mains frequency and might also flicker at the mains frequency if the tube's electrodes are bad. You might at least be able to replace the starter with an electronic one for slightly nicer startups.)

ESL unsuitable for UK use?

ESL bulbs use an unfocused electron beam on phosphor. As of 2011 it's difficult to find 240V ones, and they might turn out to be too heavy for the UK---a B22-to-E27 adapter won't overcome the effective weight limit of a Bayonet socket. Also Vu1 haven't yet explained how they've avoided X-rays in their unshielded bulbs (does their phosphor allow the use of lower-energy electrons?)

Task lighting

Spectrum issues

Certain brain disorders, such as Irlen's syndrome (scotopic sensitivity, identifyable if high-contrast symbols on a printed page sometimes seem to move), and possibly some forms of autism, can result in sensitivity to certain wavelengths of light being overly prominent in the spectrum; the exact wavelengths depend on the individual and are typically addressed using customized eye or page filters, but lighting will obviously have an effect.

The spectral pattern (spectral power distribution) of non-incandescent lighting usually depends on the manufacturer's phosphor chemistry and is likely to have "spikes" at frequencies corresponding with each chemical. Higher CRI "soft white" bulbs typically use more chemicals to reduce the prominence of particular spikes, but it's hard to approach incandescent's CRI=100 without using halogen. (It's possible that phosphor-driven bulbs will irritate less if a "main" lamp is halogen and the others merely add "background" ambient light, but this again depends on the individual.)


Copyright and Trademarks: All material © Silas S. Brown unless otherwise stated.
Anwio is a trademark of Shanghai Denglian Trading Co., Ltd.
CorePro is a trademark of Philips Lighting Holding B.V.
DFS is a trademark of DFS Group Limited.
IEEE 1789 is a trademark of the IEEE.
Ledvance is a trademark of LEDVANCE GmbH.
Osram is a trademark of OSRAM GmbH.
Wi-Fi is a trademark of the Wi-Fi Alliance.
Any other trademarks I mentioned without realising are trademarks of their respective holders.