Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Mike Sokol]

One of my NSZ readers sent me a picture of a 12-volt LED overhead lighting fixture that was factory installed in his brand new RV. He claimed that he got up off of his couch, walked across the floor and touched the exterior of the light fixture. He felt a shock and  the bulb in the fixture dimmed, got really bright, then started smoking and burned up internally. He didn't understand how he was shocked by 12 volts DC, or how he burned up the fixture as you can see below.

I suspect that what he felt was a static shock like you might feel walking across the carpet and touching a doorknob. But in this case, by touching the outside of the plastic fixture he made a tiny spark jump from his hand to the electronics inside. That spark could have punched through a P-N junction in the LED current limiting circuitry, dumping a LOT of current from the 12-volt supply into the LED's. There is no individual fuse on these fixtures, just a 20-amp fuse on the battery feed wire. So that's a lot of amperage potential if the circuitry looses its current limiting.   

Doing a little more digging, another RV installer told me there's been a lot of these new imported LED lighting fixtures burning up for unknown reasons. So I wonder if perhaps they're being zapped by a static jolt from someone on carpet, which shorts out their current limiting circuitry, and allowing full battery current to flow through the LEDs, causing a burn-up.

Thoughts?

[Stephen Swaffer]

Typically what I have seen for current limiting is just a resistor-the LED is a steady known load-I would think in fact it could be designed to run without any current limiting and for efficiency in lighting less is more.  But others may be smarter on that side.  A static shock should just go to ground though, so I am not sure how it could have affected internal components?

I do know there is tremendous pricing pressure in LED lighting-have heard that some manufacturers sell their reject chips and they wind up in cheaper lighting fixtures.  My supplier carries 2 identical looking, same lumen output/wattage rating from a supplier-price is about 40% higher for one-but it has better color rendering, etc-they were forced to start offering the cheaper version to compete with big box pricing.  My guess is everyone in the supply chain for that light is getting beat up on price-and even the end user has no idea how to compare lighting so who cares-cheapest wins-and if we have to do a few warranty repairs oh well.  Price is more important than quality and the result is what you see.

[John Roberts {JR}]

+1 Static caused a fault in the drive electronics.

Is there UL or similar safety agency rating on those fixtures?

JR

[Jonathan Johnson]

Is there UL or similar safety agency rating on those fixtures?

Because vehicles typically aren't subject to building codes, including legally-enforceable electrical inspection, I doubt that they'd go to the effort and expense of having low-voltage devices tested and listed by a safety agency. Many of the 120V devices are UL listed, only because those same devices are supplied to installations that will be inspected.

The RVIA does have safety standards and does perform inspections of member manufacturers. Of course, a manufacturer that is not a member of the RVIA would not be subject to those programs. Whether or not they test individual components, I don't know.

[John Roberts {JR}]

Because vehicles typically aren't subject to building codes, including legally-enforceable electrical inspection, I doubt that they'd go to the effort and expense of having low-voltage devices tested and listed by a safety agency. Many of the 120V devices are UL listed, only because those same devices are supplied to installations that will be inspected.

The RVIA does have safety standards and does perform inspections of member manufacturers. Of course, a manufacturer that is not a member of the RVIA would not be subject to those programs. Whether or not they test individual components, I don't know.

I ask because UL performs a rigorous analysis of how products fail and how individual components fail. So this seems to fall in that wheel house.

The automotive industry generally uses modest sized fuses, perhaps 20A@12V is not excessive for incandescent lighting it seems a lot for modern high efficiency LEDs.

If it burned up internally but did not cause a larger fire, or danger from excessive smoke, that might be considered an acceptable failure mode. UL mainly dislikes starting house fires or killing people, but products self destructing is a market concern not human safety issue..

JR

[Mike Sokol]

Is there UL or similar safety agency rating on those fixtures?

I don't think so, but I've forwarded the fail pictures and email info to the US importer/distributor of these LED fixtures. I'll see what he says...

[Mike Sokol]

I heard back for the "manufacturer" of this LED module, and their engineer came up with what I consider to be a nonsense failure mechanism for it burning out. He proposed that the RV's 12-volt converter (essentially a 120-volt AC to 12-volt DC power supply that charges the 12-volt lead-acid batteries) must have had a momentary diode failure that dumped 120-volts AC into the battery system and burned out the LED module in addition to shocking the RV owner. The failed power supply diode then fixed itself before damaging any other electronics or the main storage battery. And, of course, this occurred at the exact second the RV owner touched the light bezel.

Yeah, and Martians ate my homework... So I'm calling BS on his proposed failure mechanism. After making my doubts known to the president of the company, he agreed to send me several LED fixtures to see if I can blow them up with a static discharge.

Should be fun... 

[Stephen Swaffer]

I heard back for the "manufacturer" of this LED module, and their engineer came up with what I consider to be a nonsense failure mechanism for it burning out. He proposed that the RV's 12-volt converter (essentially a 120-volt AC to 12-volt DC power supply that charges the 12-volt lead-acid batteries) must have had a momentary diode failure that dumped 120-volts AC into the battery system and burned out the LED module in addition to shocking the RV owner. The failed power supply diode then fixed itself before damaging any other electronics or the main storage battery. And, of course, this occurred at the exact second the RV owner touched the light bezel.

Yeah, and Martians ate my homework... So I'm calling BS on his proposed failure mechanism. After making my doubts known to the president of the company, he agreed to send me several LED fixtures to see if I can blow them up with a static discharge.

Should be fun... 

Not to spoil your fun-but the answer is yes you can!

I am guessing that dumping 120 VAC to a 12 V battery circuit would blow up more than an LED-though you might want to use care proving that theory.

[Mike Sokol]

I am guessing that dumping 120 VAC to a 12 V battery circuit would blow up more than an LED-though you might want to use care proving that theory.

I'm not going to try to prove HIS theory... I'm going to try to prove mine. So if he wants to rig up a test to dump 120-volts AC into a 12-volt battery, he's going to make a battery/bomb I think.

I'm going to hook up one of his LED fixtures to a 12-volt car battery with a 15-amp fuse and 14-gauge wire, establish a ground-plane for the negative line, then flash the fixture with a piezo barbeque lighter held to the outside of the plastic bezel.   

[John Roberts {JR}]

Yup that definitely sounds like a BS answer. No real engineer with half a clue would claim that.

JR

[Stephen Swaffer]

Perhaps a real engineer with a boss that hasn't a clue might-worked for engineers before that were happy to lie to whomever and go to confession on the weekend-and happy to admit it (behind closed doors).

[John Roberts {JR}]

Sometimes the customer service contact is wary about being blamed for the failure or incurring liability. I recall a similarly BS excuse from a car dealership service manager. Back a few decades ago I had the local car dealership replace my spark plugs and do an oil change. Within about 2 weeks of the service, my car developed a rough idle and a miss. Checking under the hood for something easy like a loose spark plug wire, instead I found a loose spark plug that I could easily remove by hand. Further the plug electrode was burned up. Apparently the loose attachment to the head, reduced heat flow away from the plug. I borrowed a plug wrench from a co-worker and replaced the bad plug with a new one. Then I drove over to the dealership to show him the burnt up plug. His explanation was even more fantastic than your LED engineer. He argued that the plug condition looked like water damage (WTF?). Water damage inside a combustion chamber? Then he offered to give me a new plug. :-(. I told him I didn't want anything from him, but to give him some corrective feedback so his mechanics don't make the same mistake again. Arghhh people...

JR

[Stephen Swaffer]

It can be hard to admit you made a bad decision or an honest mistake-but when troubleshooting it never pays to be too big to fail.  I no longer work for that engineer-got tired of BS being valued more than real solutions.  Had I been patient and stayed with that employer I would no longer work for him either-it caught up with him and instead of being one of the top brass at the facility like he used to be, he has a low level inspector job with no one under him anymore.

If they want their LEDs to be better quality, they will listen and learn from Mike's generosity in doing their R&D for them!

[Steve M Smith]

The failed power supply diode then fixed itself before damaging any other electronics or the main storage battery.

Whilst semiconductors like to upgrade themselves to full conductors, I have never known one to downgrade itself back to semi status!
 
 
Steve.

[Mike Sokol]

Time to initiate the TMG (Technobabble Malfunction Generator) at http://www.springhole.net/writing_roleplaying_randomators/malfunctionmaker.htm

When you click on it, you get randomly generated and really cool sounding explanations for system failures like the one below. 

The upper ionic flux hyperdrive regulator has short circuited, resulting in an explosion in the larboard sonic power field.

[Tim McCulloch]

Time to initiate the TMG (Technobabble Malfunction Generator) at http://www.springhole.net/writing_roleplaying_randomators/malfunctionmaker.htm

When you click on it, you get randomly generated and really cool sounding explanations for system failures like the one below. 

The upper ionic flux hyperdrive regulator has short circuited, resulting in an explosion in the larboard sonic power field.

That was my second guess...

[frank kayser]

Seems it's too expensive to analyze the problem and fix production in China. There's always another fool in line to buy the cr*p.  Remember way back when "Made in Japan" meant a cheap joke?  China may learn that lesson eventually.

If Mike were to come up with a definitive fix and served it to them in gold ink on a silver platter,I doubt any change  would be made. Hell, these are selling.   The saying "if it ain't broke, don't fix it" applies only to their business model, not the actual product anymore.

[John Roberts {JR}]

Seems it's too expensive to analyze the problem and fix production in China. There's always another fool in line to buy the cr*p.  Remember way back when "Made in Japan" meant a cheap joke?  China may learn that lesson eventually.

If Mike were to come up with a definitive fix and served it to them in gold ink on a silver platter,I doubt any change  would be made. Hell, these are selling.   The saying "if it ain't broke, don't fix it" applies only to their business model, not the actual product anymore.

If that is the only failure incident, and the vehicle did not burn down to the chassis, they do not have a problem. If it is a common failure and they are a real company they will hear about it and can address in in future production runs.

Sometimes products from China are not fully engineered but copied from some other similar product built in that same factory by some other real company. That's why I was asking about approvals, but sometimes they copy those agency marks too. These days some approval marks use technology for stickers that are hard to copy.

When you fish from the bottom of the barrel sometimes you hook a turd instead of a fish.

JR

[jasonfinnigan]

I heard back for the "manufacturer" of this LED module, and their engineer came up with what I consider to be a nonsense failure mechanism for it burning out. He proposed that the RV's 12-volt converter (essentially a 120-volt AC to 12-volt DC power supply that charges the 12-volt lead-acid batteries) must have had a momentary diode failure that dumped 120-volts AC into the battery system and burned out the LED module in addition to shocking the RV owner. The failed power supply diode then fixed itself before damaging any other electronics or the main storage battery. And, of course, this occurred at the exact second the RV owner touched the light bezel.

Yeah, and Martians ate my homework... So I'm calling BS on his proposed failure mechanism. After making my doubts known to the president of the company, he agreed to send me several LED fixtures to see if I can blow them up with a static discharge.

Should be fun... 

And How is everything else on the RV fine? hmm..

[Mike Sokol]

And How is everything else on the RV fine? hmm..

Yes, that was the only electrical failure. Suspicious, isn't it.

I'm a huge proponent of using the logic of Occam's Razor to help determine failure mechanisms. The simplest explanation is usually the correct one, and highly complex theories a generally BS. As noted earlier, there's not a high likelihood that Martian's in a spaceship actually landed in my yard and ate my homework. More likely I just lost it or didn't do it at all. Yes, at times there's the perfect storm of problems that comes together. But that's pretty rare.

This same idea of simple explanations happens in the medical health field all the time. For instance, before Pasteur discovered that bacteria caused spoilage there were dozens of theories as to why things rotted. But once you understand bacteria and the fact that you can kill them with heat, then you understand the process of canning. However, there were illnesses that didn't seem to be caused by bacteria, which had to wait until the invention of the electron microscopes to discover viruses. And so it goes with most fails. You have to force yourself to stand back from the problem and see the big picture, then find the seemingly unimportant detail that turns out to be the lynchpin of the problem. That's the really fun part.   

[Jonathan Johnson]

I'm a huge proponent of using the logic of Occam's Razor to help determine failure mechanisms. The simplest explanation is usually the correct one, and highly complex theories a generally BS. As noted earlier, there's not a high likelihood that Martian's in a spaceship actually landed in my yard and ate my homework. More likely I just lost it or didn't do it at all. Yes, at times there's the perfect storm of problems that comes together. But that's pretty rare.

This same idea of simple explanations happens in the medical health field all the time. For instance, before Pasteur discovered that bacteria caused spoilage there were dozens of theories as to why things rotted. But once you understand bacteria and the fact that you can kill them with heat, then you understand the process of canning. However, there were illnesses that didn't seem to be caused by bacteria, which had to wait until the invention of the electron microscopes to discover viruses. And so it goes with most fails. You have to force yourself to stand back from the problem and see the big picture, then find the seemingly unimportant detail that turns out to be the lynchpin of the problem. That's the really fun part.

As a general rule, I find that the problems that have the biggest impact often have the simplest solutions, and the problems that are mere annoyances are often the most difficult to solve.

[Scott Holtzman]

I was a contract engineer for Sprint PCS. In 99 as lead network engineer for the Motorola CDMA FoA (fist outside application).

Digital radio was new to cellular and not well understood by management.  As a CDMA expert I had wide latitude as a contractor. 

We had a pack of software guys from India analyzing stack traces and core dumps,to say this 50 million dollar switch was less than stable would be an understatement. 

However we had a test cell at the MTSO and several adjacent for basic functionality testing.  This however didn't stop corporate from parading visitors through my facility that was not inspection ready.  The engineers were not very presentable and some of the staff had deferred personal hygiene. 

So one afternoon a parade of suits were in the facility when we had the switch go offline.  Some guy in a cheap suit runs into the control room wanting to know what's wrong. I am standing by the alarm printer trying to find the root cause in the hundreds of alarms so I reply "it's broke" hoping he will apply.  He asks what O am going to do about it. I replied "fix it".  Instead of going away he asked me what kind of idiot I was and who I reported to.  When he saw my contractor badge he thought I was with the vendor.  He took my badge and had me escorted off the property.

My boss called a little while later, I thing I was already half lit up at the bar with a some key folks from the project.  After an exchange of expletives I asked if he had a better idea.  He suggested I make up a technical explanation. 

The moral is Mike is right.  This guy was a senior VP.  I could have explained real process issues, vendors forced into deadlines that doomed them to failure then blamed for the companies failure to plan.  They didn't want to hear that.

Sometimes plausible deniability requires daily bathing in ignorance.

[ProSoundWeb Community]