The hidden, extremely costly LED lighting maintenance

[gadfly56]

Only 6,690 posts to go... :lol::lol:

 

[Besoeker]

HPS can't strobe or spaz blink. The ignitor simply doesn't produce enough voltage to hot strike and if it did, it would arc over elsewhere in the fixture before re-striking a hot lamp. It's something only LEDs suffer due to design flaws. Just disallow it in specs and put in remedies so the cost of corrections get entirely picked up by the LED sales vendor or the coverage they get at their cost. Alongside this, prohibit silently letting "thermal management" from compromising performance during normal usage unless it is disclosed for consideration (i.e. thermal management for parking lot may go into protective dimming if they're left on with the sun baking on the fixture)

OK. Can you provide a detailed cost/benefits of HPS over LED for their life expectancy?

 

[tw1156]

I used to be the street lighting graduate engineer for a larger city in Texas where we were converting from HPS/Mercury Vapor/Metal Halid/Pulse Start MH to LED. By design, we had three different specifications for local streets, arterial/collector streets, and freeway and we followed the Model Lighting Ordinance(MLO) for most of the specifications. We were able to weed out a lot of junk products because of the MLO, which was meant to help municipalities.

We did not have issues with flickering as noted in this sub, but I can argue that we did note that LED street lighting's biggest issue to overcome is glare, especially as you are getting into the higher drive currents. We capped the driver current at 700mA (and even then I believe it to be too high, but money is always a factor in design considerations). I am not discounting the flickering, but when going with a solid manufacturer, these issues become non-issues.

LEDs remind me somewhat of Mercury Vapor lamps in that over a long period of time they won't just burn out (for the most part), but will get more dim over a longer period of time than Merc. Vapor. That street light pole will be hit or replaced by the time the luminaire will go out, at least this is what we observed as we did not have failure rates as high as they seem to be indicated in this sub.

There's a lot of issues that we observed with LED street lighting and the general public, but flickering was not one of them, at least for us.

 

[Electric-Light]

Model Lighting Ordinance(MLO) for most of the specifications.

Does it address anything about inherent protection against spaz blinking? (repetition of short pulse duration, repetitive flashing with some irregularity in rate, duration). There was a study done by a company on possible use of LEDs in place of xenon flash tube for fire alarm. It might have been Silent Knight. The finding was that mS of pulse duration was the key component of attention getting ability.

Limitations imposed by natural laws have avoided the need to talk about certain things until LEDs came along. The thermal inertia of filament, arc tube or luminescence of phosphors limit the modulation depth (contrast) and the rise and fall times. White LEDs are a half-ass existence between xenon arc lamps and metal halide lamps. They lack the ability to tolerate enough surge impulse to have value for freezing motion while lacking desirable inertia to dampen undesirable modulation.

Many flashlights have this gimmicky and extremely annoying flash mode with a seemingly irregular rates which is a close approximation to the symptom of spaz blink caused by a failing lighting LED ballast. Xenon lamps can repeatedly flash too but the flashing is usually intentionally controlled. This can not be done with ordinary HIDs or filament lamp due to thermal inertia. Fluorescent lamps have enough phosphor persistence to limit the fall rate.

We did not have issues with flickering as noted in this sub, but I can argue that we did note that LED street lighting's biggest issue to overcome is glare, especially as you are getting into the higher drive currents. We capped the driver current at 700mA (and even then I believe it to be too high, but money is always a factor in design considerations). I am not discounting the flickering, but when going with a solid manufacturer, these issues become non-issues.

Glare only affects imaging sensors. Light meters are blind to glare, because it's not a resolving sensor. wearing a pair of glasses made with frosted glass gives you a better idea of how light sensor sees the world.

http://www.cv.nrao.edu/course/astr534/Brightness.html

To demonstrate the concept of inferiority of high surface brightness LEDs, use a pair of same wattage ceiling fan lamps. One frosted and one clear. Set them side by side. Use a lens to focus the light bulbs on a piece of paper until you can see the bulbs sharply focused. Better yet if there is a place where there's an ordinary HPS and LED street light alongside, project them on the same sheet of paper.

The frosted bulb or HPS fixtures will image nicely and you will see the shape very clearly projected. You can see tiny extremely bright dot of each single LED element or the shape of the clear bulb filament. This is exactly how glare is projected onto our retinas. That's why street lights which are criticized of being inefficient are designed the way they are instead of cramming a double ended arc tube inside a fixture that looks like a halogen work site light. If they did that, it would glare like LED street lights !

LEDs remind me somewhat of Mercury Vapor lamps in that over a long period of time they won't just burn out (for the most part), but will get more dim over a longer period of time than Merc. Vapor. That street light pole will be hit or replaced by the time the luminaire will go out, at least this is what we observed as we did not have failure rates as high as they seem to be indicated in this sub. There's a lot of issues that we observed with LED street lighting and the general public, but flickering was not one of them, at least for us.

The critical difference is that LEDs are almost always driven with a transistorized ballast. LED stack suddenly increase and decrease in resistance without the huge cut-in and cut-out separation of discharge lamps. LED ballasts were not designed prudently enough to avoid load hunting and this is the fault of fixture integrators and LED ballast designers.

LED elements also have problems of their own. White LEDs are made with blue LED chip surrounded smothered in yellow glowing phosphor impregnated clear rubber or plastic. Problems can occur on the elements that usually have something to do with moisture or chemicals in atmosphere affecting plastics or rubbers, cracking, crazing, delamination (which affects diffraction and dispersion issues and affect the color) and yellowing. Sometimes it's caused by LED elements passing out from the stink of its own out-gassing in gas tight environment intended to protect the LEDs from chemical vapors.

 

[mbrooke]

My money is still on open phase/open neutral unless proven otherwise. Yes there are junk LEDs on the market, but if you spend money on quality LEDs I don't see the issue.

 

[Electric-Light]

My money is still on open phase/open neutral unless proven otherwise. Yes there are junk LEDs on the market, but if you spend money on quality LEDs I don't see the issue.

Isn't the onus on proving rather than disproving?

LEDs are susceptible to patent trolls.

One method of avoidance includes interruption of load upon detection of failure by any method that has been used before and not novel. Detection of current and voltage variation on load end can be read through a resistive shunt, a combination of coil, resistors and capacitors used to produce signal on flashing and use of methods such as magnetic coupling, feedback from a photosensitive component responding to LED output. Basically I am discussing things that have been used or talked about in the past and simply substituting the load with LEDs do not lend to unique or novel.

http://www.ledinside.com/news/2014/12/patent_trolls_rampant_in_chinas_led_lighting_industry

Just in case: there is an interest in stopping design pitfalls in the LED ballast from creating a nuisance such as spaz blinking. The use of common sense methods to allow LED ballast controller to become aware of spaz blinking by noticing the load change; then responding by shutting down the load by any method like shutting off trigger signal to output transistor like device does not lend itself as unique and novel by simply applying it for LED, or for LED used in lighting or LED used in street illumination. Flashing or spaz blinking can be detected by current shunt, magnetic coupling or capacitive coupling to only provide signal on presence load pulsation. Use of photo sensor, with or without narrow pass blue filter near the DWL of LEDs to optically detect the flashing. Use of simple analog filtration or software code. Just taking notes publicly to announce this to be prior art just in case some patent troll attempts to patent something dated past today with some electronics engineering common sense as novel by applying to something about LED or specific implementation of LED using products.

 

[mbrooke]

Isn't the onus on proving rather than disproving?

LEDs are susceptible to patent trolls.

One method of avoidance includes interruption of load upon detection of failure by any method that has been used before and not novel. Detection of current and voltage variation on load end can be read through a resistive shunt, a combination of coil, resistors and capacitors used to produce signal on flashing and use of methods such as magnetic coupling, feedback from a photosensitive component responding to LED output. Basically I am discussing things that have been used or talked about in the past and simply substituting the load with LEDs do not lend to unique or novel.

http://www.ledinside.com/news/2014/12/patent_trolls_rampant_in_chinas_led_lighting_industry

Just in case: The use of common sense methods to gain awareness of spaz blinking by noticing the load change and responding by shutting down the load does not lend itself as unique and novel by simply applying it for LED, or for LED used in lighting, LED used in street illumination. Flashing or spaz blinking can be detected by current shunt, magnetic coupling or capacitive coupling to only provide signal on presence load pulsation. Use of photo sensor, with or without narrow pass blue filter near the DWL of LEDs to optically detect the flashing. Use of simple analog filtration or software code. Just taking notes publicly to announce prior art just in case some patent troll attempts to patent some electronics engineering common sense as novel by applying to something about LED or specific implementation of LED.

But why would you need it? It just adds one more thing that can go wrong- and raises the overall cost of LED.


Even if those Detroit LEDs are flickering because they all failed, that does not nullify the fact hundreds of thousands of LED upgrades are working without issue. As with everything it all boils down to who can make a quality light.

 

[Electric-Light]

But why would you need it? It just adds one more thing that can go wrong- and raises the overall cost of LED.

well, i don't see many things discussed about how to deal with spaz blink. Detecting load acting up through load current pattern is obvious and just because it hasn't really been talked about for doing the same with LED doesn't make it unique but someone might go try to go file patents and say they invented the idea of trying to something obvious because it is used with LEDs... or LEDs in certain applications. This is harder for them to do when what they claim as unique is already talked about.

Even if those Detroit LEDs are flickering because they all failed, that does not nullify the fact hundreds of thousands of LED upgrades are working without issue. As with everything it all boils down to who can make a quality light.

Just suggesting writing specs language to say prohibit products that spaz blink upon failure and allow the end user to legally make the vendor furnishing and installing LEDs entirely responsible for future costs directly related to spaz blinking failure. If it happens, it happens. spaz blinks are extremely annoying, so it will get fixed. The difference is who pays. ok? It shouldn't affect the cost at all if this is all baseless fear.

 

[mbrooke]

well, i don't see many things discussed about how to deal with spaz blink. Detecting load acting up through load current pattern is obvious and just because it hasn't really been talked about for doing the same with LED doesn't make it unique but someone might go try to go file patents and say they invented the idea of trying to something obvious because it is used with LEDs... or LEDs in certain applications. This is harder for them to do when what they claim as unique is already talked about.



Just suggesting writing specs language to say prohibit products that spaz blink upon failure and allow the end user to legally make the vendor furnishing and installing LEDs entirely responsible for future costs directly related to spaz blinking failure. If it happens, it happens. spaz blinks are extremely annoying, so it will get fixed. The difference is who pays. ok? It shouldn't affect the cost at all if this is all baseless fear.

So what is poor quality LEDs spaz blink? So what if high quality LEDs spaz blink at end of life? HPS lamps cycle, MH lamp arc tubes may explode (taken care of by using "O" lamps) and florescent tubes do all sorts of funky stuff. As long as there is no fire hazard or release of heavy metals its fine with me.

 

[tw1156]

Electric,

Here are a few snippets from the LED specification using the MLO as a base template, but the specific word "Flicker" is not used in the entire document.

· A luminaire failure will be deemed to have occurred if any of the following conditions exists following the completion of testing:
a) A hard power reset is required to return to normal operation
b) A noticeable reduction in full light output (e.g. one or more LEDs fail to produce light, or become unstable) is observed (I take this to mean flickering under the unstable section of the LEDs)

There is a Q/C section as well that requires testing of the luminaires in the field prior to acceptance. I can send a link to download if you want to view the document in its entirety.

 

[Fulthrotl]

Only 6,690 posts to go... :lol::lol:

we are at 130 now, and still seem to be running with a full head
of steam....

the lights may flicker, but the posts do not, so "spaz posting" isn't
a problem....... yet.

 

[tw1156]

MSSLC - Model Specifications for LED Roadway Luminaires

MSSLC - Model Specifications for LED Roadway Luminaires

I incorrectly stated earlier by saying the MLO was the guide for the LED luminaire specification:ashamed1:; the document that was used was actually put out by the DOE and was labeled MSSLC (Model Specification for LED Roadway Luminaires). https://energy.gov/eere/ssl/downloads/model-specification-led-roadway-luminaires-v20 The DOE provided a lot of base information for Municipalities to incorporate into their specifications when evaluating LED luminaires.

The MLO was referenced as a guide for best practices in choosing luminaires, but is more of an "aesthetic" guideline (full cutoff, reduced glare, etc)

 

[gadfly56]

we are at 130 now, and still seem to be running with a full head
of steam....

the lights may flicker, but the posts do not, so "spaz posting" isn't
a problem....... yet.

I'm figuring a couple hundred more and he can collect all his posts and edit them down to a textbook. He does have a lot of interesting technical tidbits scattered about.

 

[Fulthrotl]

I'm figuring a couple hundred more and he can collect all his posts and edit them down to a textbook. He does have a lot of interesting technical tidbits scattered about.

+1 on that. an amazing level of detail.

 

[ActionDave]

His inability to be succinct is a big reason he looses credibility with me.

 

[mbrooke]

His inability to be succinct is a big reason he looses credibility with me.

To me its not so much the succinct aspect, I enjoy an in depth descriptive discussion, but rather that the conversation seems one sided. Any positives regarding LEDs are ignored, and there is a constant fixation on the occasional pitfalls of a technology still maturing. Further when posting facts which contradict or prove statements wrong (like instant re-strike igniters for 5kv pulse rated sockets) he just ignores them.

 

[Electric-Light]

I incorrectly stated earlier by saying the MLO was the guide for the LED luminaire specification:ashamed1:; the document that was used was actually put out by the DOE and was labeled MSSLC (Model Specification for LED Roadway Luminaires). https://energy.gov/eere/ssl/downloads/model-specification-led-roadway-luminaires-v20 The DOE provided a lot of base information for Municipalities to incorporate into their specifications when evaluating LED luminaires.

The MLO was referenced as a guide for best practices in choosing luminaires, but is more of an "aesthetic" guideline (full cutoff, reduced glare, etc)

I think spaz blinking is one of those things that did not cross their mind just like power factor and THD when standards were initially made for CFLs. Many CFLs actually have a bridge that plugs straight into the utility with a big capacitor across it which makes for great DC link regulation but power factor that is close 0.6 or so with 100%+ THD as well as very high inrush.

They put a restriction on power factor and essentially all EnergyStar lamps and fixtures have to have 0.9 of higher PF. Along with this is a meaningless guideline that allows flicker as long as it's 120Hz or greater but did not set a limit on how far it can notch down or a flicker index value. So there are LED products that simply use strings of multi-tapped LED strings that are paired with fast semiconductor tap changers that extends and shortens the length each half cycle to reduce cut-in voltage to lower harmonics. But these design have enough non-electrolytic capacitor to keep the controller energized and light drops to zero twice every cycle with the output pulsation closely following the stepped sinewave. The flicker is worse than fluorescent as there's no afterglow in LED phosphors to dampen the flicker. Since the restrictions weren't made on flicker index, this was a loophole used to meet the requirements of power factor and stay within the flicker limits of the rules but excessively flickery in application. In general, commercial application external fluorescent ballasts are very well regulated that convert AC into DC link with very little line frequency ripple and many often provide output with less flicker than high wattage incandescent that any sign of 120 Hz pulsation is hardly visible on light sensor fed into oscilloscope with some of well designed ballasts with flicker comparable to that of studio production flicker suppressed lights.

Despite the fact LEDs already having inferior flicker dampening on its own, many LED fixtures have LED ballasts with far inferior flicker regulation than common pin-based CFL linear lamps. Ballast is one of the places LED products often cut corners in order to be able to meet the price point while using LED elements that can meet specs. Although dedicated voltage ballasts have become nearly obsolete, they've made a come back with LEDs. While there are dedicated voltage ballasts wtih constant-wattage regulation, most single voltage ones are not. So keep this and the fact many brand name ballasts are built to commercial use durability when you wonder why T5/T8 fluorescent ballasts cost so much relative to cheap LED fixtures. Many LED ballasts are also incapable of isolating load power delivery with a +/- 10% line voltage variation which means. So like 1990s T8 ballasts, you can not ignore line voltage fluctuation unless you specify line regulation. High line voltage causes over illumination, acceleration of LED element degradation and wasted energy while low line voltage causes output reduction.

To me its not so much the succinct aspect, I enjoy an in depth descriptive discussion, but rather that the conversation seems one sided. Any positives regarding LEDs are ignored, and there is a constant fixation on the occasional pitfalls of a technology still maturing. Further when posting facts which contradict or prove statements wrong (like instant re-strike igniters for 5kv pulse rated sockets) he just ignores them.

I acknowledge LEDs have their place such as generic residential lighting application that has strong expectations of immediate start but tend not to accumulate that many hours per year on average.