|
|
|
| | Energy Star proposes to revert to 120 Hz for LED lamps | | 22 Jan 2010 | | A change to the requirements for replacement LED lamps is proposed in response to industry feedback. | |
EPA is proposing to an amendment to the Energy Star program requirements for Integral LED Lamps, published on December 3, 2009, which will change the LED Operating Frequency requirement from ≥150 Hz to ≥120 Hz.
The ≥150 Hz requirement has been challenged by a number of companies and Energy Star stakeholders.
A letter from Alex Baker, EPA’s Lighting Program Manager for Energy Star, invites comments on this proposed change by Friday, February 12, 2010. Please send written comments to Kate Buck.
If accepted, the proposed change will result in a new version of the specification, Version 1.1. The effective date of the specification, August 31, 2010, would remain unchanged. The letter says:
“The performance requirements laid out in the Energy Star specification for Integral LED Lamps (V1.0) are intended to ensure that qualified products are energy efficient and meet user performance expectations. One of the requirements contained in the criteria document is for minimum LED operating frequency. This requirement was intended to mitigate potential health and safety problems caused by visible flicker. Feedback from industry stakeholders on earlier drafts of the integral LED lamp criteria was minimal and inconsistent, suggesting alternative frequency requirements ranging from 100 Hz to 500 Hz.
“In the final criteria published December 3, the minimum LED Operating Frequency requirement was increased from the previously proposed 120 Hz to 150 Hz in an effort to address stated industry concerns. However, subsequent industry feedback has indicated significant concerns exist about this higher requirement, citing cost and design implications, and the exclusion of products using AC-driven LEDs. Further, several stakeholders state the requirement as written is inadequate to prevent visible flicker and potentially should address modulation depth, waveforms, duty cycle, and other characteristics, in addition to frequency.
“EPA seeks to address this issue in close cooperation with the Department of Energy, industry and standard-setting organizations to achieve the following objective: Energy Star qualified lamps do not exhibit visible flicker during full output or dimmed operation.
“An IEEE working group including leading human factors experts is working to develop Recommended Practices of Modulating Current in High Brightness LEDs for Mitigating Health Risks to Viewers, but the results of this work likely will not be available until next year.
“To avoid delays in the implementation of the Energy Star specification for Integral LED Lamps, we propose reverting back to a ≥120 Hz requirement, pending the completion of further research on this topic.
“Stakeholders are advised that EPA intends to again address visible flicker concerns once industry consensus measurement methods and recommended performance thresholds are available. If additional changes are warranted, they will be proposed and finalized through a stakeholder process, allowing adequate time for transition.” | | COMMENTS | | Name: ron russell Posted: Mon, 25 Jan 2010 20:01 |
| LEDs have no persistence, meaning that when the input energy is removed, the light output instantly goes to zero making it a perfect strobe light !
If the EPA/Energy Star allows AC LED light sources into the market for general lighting, safety problems relating strobe effects in industrial settings, and other issues such as beat patterns on video displays with 120 Hz refresh rates will immerge and just upset the SSL market with another wave of discouraged customers.
SSL offers the promise of a perfect light source with no modulation, so why would an agency that’s job is to ensure the quality of SSL products allow such an inferior product as AC LEDs ?
I for one say “ NO “ to AC LEDs ! | | Name: alzie Posted: Wed, 27 Jan 2010 19:01 |
Magnetically ballasted fluorescent lights have flickered @ 120 Hz for decadez. Their "persistence" really wasnt all that much, as you could use them as a strobe source. They didnt seem to cause any health issues. Their main complaint was the harsh cool white color if anything.
TVs flickered at an honest 60 Hz with virtually no persistence, and
people have put in thousands of hours a year with no adverse effects.
I think a 120 Hz led flicker is a non issue. Let us have the relaxed requirements / lower price. | | Name: eric kinast Posted: Wed, 27 Jan 2010 19:01 |
Regarding the lack of persistence noted by the previous poster, it is important to realize that most light sources used today in industrial environments also lack persistence, and therefore exhibit the same stroboscopic events. These include HPS, MH, and similar discharge lamps operating from standard magnetic ballasts. The hazard of this effect is that rotating machinery can appear stationary, or moving very slowly, when illuminated by such lamps. A secondary concern is that at very high illumination levels the flicker may begin to become percepible and cause fatigue.
For traditional MH and HPS lamps, two classes of service have evolved. For low cost applications (i.e warehouse) standard ballasts, producing the stroboscopic effect are used. For applications where this is objectionable, the traditional solution has been to run arrays of lamps on all phases of 3 phase power, which works if the illuminated object receives light from multiple lamps. However, modern electronic ballasts can also provide flicker-free operation of single lamps, at the expense of higher cost and lower reliability.
Similarly, it is expected that LED lighting fixtures will evolve into two types: those which have flicker at twice the line frequency, and premium fixtures, which do not. Note that completely eliminating the 120 Hz flicker means that the fixture must have large energy storage capacitors. Such electrolytic capacitors tend to be the limiting factor in the relibility and lifetime of the fixture electronics at commercially acceptable cost points. Therefore, there are additional incentives besides cost to permit fixtures with 120 Hz flicker in applications where this is acceptable. | | Name: shoi_ Posted: Wed, 27 Jan 2010 19:01 |
White leds use phosphors whic do have a persistence. They are quite right to wait until they have an adequate specification covering modulation and so on.
There is a useful existing set of comments here. | | Name: uplight scandinavia Posted: Wed, 27 Jan 2010 20:01 |
I have to agree with ron russel. Why revert back to a lower goal just to end up with more and new problems later? We have several customer during 2009 that were intially very happy with their LED installations in private homes. After some time 1-4month they had problems with headaches and other health issues. Several had problems with their video camera. The products used were several different. Some were supposed to be over 175Hz according to the spec. Some were lower. Maybe below 120Hz but not according to their spec. I am no expert but my statitics tell me three things:
1. The human eye and brain is far more complexed that we want this time.
2. This issue is not only about the Hz. The maximum pulsewidth is also detectable by a sharp eye and fast brain.
3. The whole SSL industry will have a big problem within say five years from now if this is not solved soon. It can be a setback for some that will be hard to recover from. I know big companies pumping out those products I have tried and later replaced. They don't know what's coming...
I am sorry for being so negative! I actually love this business! :) | | Name: dh121 Posted: Wed, 27 Jan 2010 22:01 |
| Do sufficiently slow & energy efficient phosphors exist that can average out the light pulses ? | | Name: stevenp Posted: Thu, 28 Jan 2010 00:01 |
"ron russell Posted: Mon, 25 Jan 2010 20:01
LEDs have no persistence, meaning that when the input energy is removed, the light output instantly goes to zero making it a perfect strobe light !
- True for the Diode, but many of the emitters being used are UV and have phosphors to emit yellow and blue.
SSL offers the promise of a perfect light source with no modulation, so why would an agency that’s job is to ensure the quality of SSL products allow such an inferior product as AC LEDs ?
- Quality assurance is not the agency's job. They are trying to accelerate the use of SSL to save energy.
- By running the LED in a pulsed mode you may get more light and/or higher efficiency. (40% duty cycle gives 60% of constant on level) | | Name: lyeph Posted: Thu, 28 Jan 2010 03:01 |
| If the EPA is concern about flicker, then EPA should just specify the final minimum flicker requirements to eliminate this concern (eg. what is the minimum out variation required). This approach would be better than trying to specify the operating conditions required to meet the flicker requirement. This would allow the industry to solve the flicker conditions using whatever appropriate method (existing or to be developed). Specifying final outcome would lead to more innovation rather than trying to specify what should be done to achieve the outcome. As a quick pass, solving flicker from LEDs using normal AC input should not be a difficult issue from the design point of view. | | Name: marcus wu Posted: Thu, 28 Jan 2010 08:01 |
| Yes, LED has no persistence. However, the AC LED is not likely to connected to ac line directly. In fact, a sophisticated converter (an IC with some passive components for stabilization and protection) should be presented between line and LED cluster, which can eliminate the strobe effect. If LEDs is hooked direct to ac line which is not a correct way to drive LEDs.
| | Name: pmbpmb Posted: Thu, 28 Jan 2010 10:01 |
In Europe double mains frequency is 100Hz. No products will be Energy Star compliant if the 120Hz requirement is adopted.
However even higher frequency chopped light is annoying when seen as multiple lights while moving the eyes rapidly. I refer to e.g. break lights of cars.
Let's feed these LEDs with DC-current adding an inductor/capacitor to the converters. | | Name: kevin willmorth Posted: Thu, 28 Jan 2010 14:01 |
There is far too much unsupported opinion swirling around the issue of AC LEDs and flicker - basing policy decisions on such conjecture must be avoided. There are numerous strategies used in development of AC LED systems to mitigate perceived flicker. The factors involved include intensity, physical spacing of individual die, use of simple rectification to add an underlying DC component, duty cycle, and others - all producing excellent results with a simpler, lower cost total system. The work that needs to be done is to determine objectively what makes a good AC system and what constitutes a poor one, and lay those specifications out in standards development. Yes, there are some horrid examples of really bad AC lighting systems (LED and others), and these should be used in any study to define what makes them bad, in specific and supportable terms.
The body of research on flicker completed to date has addressed issues of health, such as migraine (frequencies in the 65Hz range), epilepsy (low 30-45Hz range), and human visual system integration performance - or fusion factor (CFF - 100Hz). However, none have specifically been focused on defining all of the factors that produce flicker perception in general illumination environments, where direct and indirect lighting from several sources - potentially operating out of phase with one another - combine to produce illumination of the environment. This is an important area to be investigated.
AC LED systems will have their place in the lighting market, along side PWM systems, and DC systems, based on definable performance, verifiable quality, and realized economic factors - not opinion and emotional conjecture and anecdotal information taken out of context or from obsolete past practice. If this were the case, then ALL LEDs are dead in the water, since the first versions of the technology, and the existing poor examples point to this technology being a wash out. We all know that is not true, so, let's see give those developing high performance LED systems the same chance to participate in what is going to be a huge opportunity for us all. | | Name: doug l Posted: Thu, 28 Jan 2010 19:01 |
Unlike HID, arc, flourescent, and similar sources, LEDs quite readily lend themselves to analog, or current dimming. In doing so, the LEDs increase in their efficientcies, tested numbers show 230% increase with some beands, but not OSRAM. So, a 135 lm/W power LED will see well over 200 lm/W when analog dimmed!
Yes, there is some tint shift. This can be minimzed by analog dimming part way, and then PWM'd for the rest. Any designer worth his beans can simply PWM at 5,000 Hz with no increase in cost. What the "designers" are really saying, is that there is no pre-designed reference modules for them to simply copy from companies like Future Electronics.
It is a pretty sad state of affairs, when designers loose their creativity, and start affecting policies.... | | Name: g wootton Posted: Sat, 30 Jan 2010 15:01 |
| Why allow AC LEDs? It's a matter of commercial fairness! If you're going to allow other devices that exhbit severe flicker, some worse than a phosphor converted led operating at line frequency and even allow them an energy star label, you've got to allow the same for LED lamps. Any other approach would create an unfair commercial advantage to some technologies - one has to wonder who the 'concerned' industry experts were that raised this issue.
From a technical perspective, it is entirely possible to eliminate the flicker from HPS, Halide, mercury and fluorescent lamps while saving energy by merely using a high-frequency electronic ballast. If the regulators were truly concerned about flicker, they would set a standard for flicker that applies uniformly to all lighting and display technologies. The fact that they didn't indicates their true level of concern andf suggests other motivations. In general, there seems to be a concerted effort to put LED lighting at a commercial disadvantage.
- For some reason, LED lights are rated by their efficiency with ballast included while other technologies are rated exclusive of the ballast - in some cases, the commonly used ballasts have efficiencies under 90% which would significantly reduce the efficiency rating if included. If the playing field were level, all devices would be rated with the ballast included. Additionally, to be fair, ballasts for all technologies should be held to the same standard for efficiency and power factor (why are they not?).
- For some reason, LEDs are required to meet a CRI spec that other rated devices are not. HPS lamps have a CRI near zero. There are even a number of studies that show that the reduced efficacy of peripheral vision is reduced under HPS lighting (due to poor spectrum) producing a measurable increase of accident frequency still they have done nothing to set a minimum CRI standard. Again, the 'fair' thing to do would be to adopt the same minimum standard for all lighting technologies.
One can but admire the ineptitude of the stated requirement. The frequency of the flicker is surely the least of the issue. As someone pointed out, we stared at CRT TV's that flickered at 30 Hz (it's would only be 60 Hz if the crt beam or your eyes are out of focus). We also seemed quite happy with projectors flickering at 18 Hz. The answer is that the ampiltude (peak to peak) and strength (AC / RMS) of the modulation is of great importance - this is easily measurable yet the standard ignored this (perhaps because HPS lamps exhibit a stronger flicker than white leds when line operated since the non-linearity of the former serves to increase flicker while the latter does the opposite). Inexpensive rectifier input DC ballasts will likely retain some line frequency ripple (I've personally observed this in mercury lamps operated from a high frequency DC ballast); consequently, the issue is not mainly one of frequency but strength of the flicker. But again, this requirement was not intended to advance the state of the art in lighting but merely to create a differential cost structure in favor of other lighting technologies.
BTW, the issue of capacitors in rectifier input power supplies is problematic: if the capacitor is larger, power supply ripple is reduced at the expense of reduced power factor. Depending on the lamp driving technology, the 'ideal' ballast would likely just keep the DC supply above the drop-out voltage of the lamp driver electronics. This appears to be a larger challenge for CFLs which exhibit a notoriously poor power factor. This issue is substantially larger than just lamp ballasts - millions of wall warts and appliances are on-line 24/7 reducing the efficiency of power distribution systems.
| | Name: g wootton Posted: Sat, 30 Jan 2010 15:01 |
| Doug .. you'r wrong. It is entirely possible to make HPS, mercury, fluorescent and other lamps dimmable using electronic ballasts with the same general benefits that would be obtained for LEDs. As a machine vision guy, I've used quite a few of these ballasts. | | Name: brianc Posted: Tue, 02 Feb 2010 15:02 |
| Time to dispel some 'myths' and 'beliefs' regarding flicker.
FIRST: "This requirement was intended to mitigate potential health and safety problems caused by visible flicker." This statement is misleading; flicker is not visible at 120Hz, let alone 150Hz. The desire to use 120Hz is obvious, at least in 60Hz countries.....cost.
SECOND: Broadcast TV is 30fps; interlaced to 60Hz (every other line transmitted every 1/60th of a sec). Why? This solved several problems; one, bandwidth; two, flicker. Many folks are/were susceptable to 30Hz flicker. Movies: all theaters (worldwide) project @ 48Hz (we're taliking film). Movies are shot @ 24fps, and projected @ 48Hz; each frame is shown twice by means of a 'butterfly' shutter. Why? To eliminate flicker. I never hear of folks complaining about flicker after watching a movie at the local screen. Persistance is not possible in this scenario, due to the shutter.
In a previous life (job), I worked for an audio equipment mfg; on most models, we designed in a large power "ON" LED straight off the mains. We never heard a complaint about flicker, and it was wired half-wave (60Hz), with only a resistor in series! No persistance, and no flicker.
My point is, shifting from 150Hz to 120Hz will have little or no ill effect, at least regarding flicker. However, the shift will enable less expensive modules, resulting in wider acceptance, which benefits all. (for what it's worth, my daughter has photosensitive epilepsy, so I'm very aware of flicker sensitivity) |
| | | |
