Lots of failing outdoor LED lamps...very long overhead mains cable.

[ocset]

Hello,
We have installed  one hundred  25W outdoor LED lamps in a town in Bosnia. Sixty failed in just a few weeks.       

They all are fed from a single 240VAC generator which delivers power to the 2500 metres  of mains cable. A contactor near the generator is used to switch them all on/off. The  lamps are fitted every 25 metres along this cable. The cable is overhead (not buried). It’s aluminium  cable. Its twisted, though the installers tell us that in places the twisting is “not so tight”. We suspect there is an awful lot of mains stray  wiring inductance.
The strange thing is that they have another run of one hundred of the exact same  lamps, but in this case the 2500 metre mains cable is buried in the ground. –None of these lamps with the buried cable have failed despite being installed months ago.
The lamps  are controlled by microcontrollers in them which slowly dim the lamps as the night goes past the small hours, so as to save energy when  few people are really outside.

We suspect that maybe there is some kind of transmission line effect going on, with the long mains cable. We have advised the installers to fit an RC snubber across the contactor, and another across Live/neutral, just downstream of the contactor. We also wonder that with the “not so tightly twisted” mains cable, maybe noise is getting into the long overhead mains cable, and disrupting the microcontroller.
The lamps are not SMPS based, but instead are sucessively switched linear  current regulators. There is no mains input filter in each lamp, and the input stage is as in the attached. (ltspice sim also attached, including ZR431 model)

May I request what are your thoughts on why they have failed?
 

[Gyro]

Not another treez faulty LED lamp thread. 

[CopperCone]

a rival company could be using a injection transformer on the overhead lines to destroy them. Is living in Bosnia like grand theft auto?

I wonder if a compulsator could be used for this purpose.

Do you do failure analysis on these lights? That could tell you alot.

[nfmax]

Lightning-induced transients?

[Andy Watson]

... We suspect there is an awful ....
We suspect that maybe ...
We also wonder that ...

But what evidence do you have?
Why don't you install some mains monitoring equipment into one (or more) of the lamps. Until you have some data you're not going to know how to cure the problem. Essentially you're shooting in the dark.

Can you rule out a faulty/sub-standard batch of components?

May I request what are your thoughts on why they have failed?
 

Aliens - for sure.

[sleemanj]

You need to go and investigate in perwon, or at the VERY  least get a failed example back for analysis, and photos of the installation.  It sounds like all the information you have currently amounts to "light no work no more".

Could be as simple as water ingress running down the overhead wires and not up from the underground ones.

[Niklas]

There is no input filtering visible in the schematics. Is that true or is it just omitted?

[dmills]

I thought these pieces of shit had a 22nF cap in them (Not shown in your spice model)?

Get a power analyser out there (Fluke make the good stuff) and actually measure what happens, then  you will be able to work out what is failing and more importantly why.

2,500m is short for transmission line effects at the power frequency fundamental, but a arcing contractor could pump a lot of RF into the line and transmission line effects at ~300kHz (1/4 wave unterminated line) and 900kHz  (3/4 wave line) are possible. Also, lightning produces large low frequency components, and an overhead line that long looks a lot like an aerial, but until you actually run a logger on the circuit it is all wild arsed guessing.

In something like this it is very much worthwhile to spend whatever you have to to get some of the duds back to your lab (ideally together with information about where in the line they were and those power analyser logs).

Regards, Dan.

[james_s]

They're running off a generator? Have you tried a different generator? I've seen some that were not at all well regulated, and some produce a really nasty output. I've also seen strange interactions between certain types of LED drivers and a transformer product I was involved with. LED drivers particularly with active PFC had a strange interaction where the voltage out of the transformer was boosted up beyond 150V.

[Kjelt]

Could be tons of reasons.
Are your drivers potted, do they have Movs, sparkgaps and other protection?
Can they survive brownouts, dips and peaks upto 277V ? No? Redesign your product.

You can add something like this and see if it makes a difference
http://www.lighting.philips.co.uk/oem-emea/products/xitanium_led_xtreme_drivers/philips-surge-protection-device

[ocset]

We are going to ask the installer to put small boxes across live/neutral. In these will be a series combo of TVS and 32mA quick blow fuse.
The TVS will be SMCJ400CA. 
In another box we will have two series SMCJ250CA TVS's  in series with a fuse. (this gives a bit higher voltage detection than a single SMCJ400CA)

....If the fuse blows, then we at least know there was an overvoltage transient      

Are your drivers potted, do they have Movs, sparkgaps and other protection?
Thanks , Not potted, there is an LSP05 MOV module upstream of the product.
 

I thought these pieces of shit had a 22nF cap in them (Not shown in your spice model)?

Thanks , The 22n  goes in the updated version. Not this version.

There is no input filtering visible in the schematics. Is that true or is it just omitted?

Thanks  , Yes theres no input filter. We don’t need it, we aren’t using SMPS.

Lightning-induced transients?

Thanks, do you think a lighting strike to one bit of the overhead cable would take out say 30 lamps either side of the strike point?.....i am guessing it would?     
 

[Jeroen3]

There is no input filtering visible in the schematics. Is that true or is it just omitted?

Thanks  , Yes theres no input filter. We don’t need it, we aren’t using SMPS.

Then what is going to protect your rectifiers?

[Towger]

It is not clear if the same generator is running both strings of lights?

Did they fail after a thunder storm?

The problem is you appear to be dealing with second hand information (Chinese Whispers), so you cannot fully trust the information.

As others have said you need to get some back.

Edit:
A direct lightning strike would take out the lot and the generator.   You do not need a strike when you have more than a few meters of wire in the air.

[Wolfram]

Has the design passed mains surge immunity testing to the applicable standards? Would your proposed "TVS + fast fuse" transient detectors survive such immunity testing?

[Ice-Tea]

Has the design passed mains surge immunity testing to the applicable standards? Would your proposed "TVS + fast fuse" transient detectors survive such immunity testing?

^
|

This.

1) Test your device against surges from an applicable standard (K21 comes to mind)
2) Use a datalogger to monitor the mains like http://www.fluke.com/fluke/uken/power-quality-tools/logging-power-meters/fluke-1740-series.htm?pid=56029

Everything you are doing right now is guesswork.

[tggzzz]

We have installed  one hundred  25W outdoor LED lamps in a town in Bosnia. Sixty failed in just a few weeks.
They all are fed from a single 240VAC generator which delivers power to the 2500 metres  of mains cable.

Are these the same failed lights that you have raised in your numerous other threads, or are they different failed lights? That information is important.

In your previous threads you have always referred to assumed (but not verified) "problems" with the UK mains. This thread is about a completely different setup.

Generators are very different to the UK mains supply, and are far less tightly regulated. For example, if you simply remove the load the output voltage will rise significantly until less fuel is fed to the motor.

Ditto the "contactor" you mention; in the absence of contrary information you should assume it doesn't cleanly interrupt the current.

[tggzzz]

We also wonder that with the “not so tightly twisted” mains cable, maybe noise is getting into the long overhead mains cable, and disrupting the microcontroller.

That should be easy to simulate in your spice model.

Have a look at professional equipment to see how such interference is prevented. The necessary components are in any of the big disties' catalogues.

[mikeselectricstuff]

Thanks, do you think a lighting strike to one bit of the overhead cable would take out say 30 lamps either side of the strike point?

If you had a strike to the cable you would know about it by looking at what, if anything, is left.

A nearby strike could easily introduce a big transient though, as could things like flashovers on nearby HV power lines.
Also things like nearby radio transmitters could induce significant voltages on the line.

[mikeselectricstuff]

And of course you need to get hold of a couple of failed units for analysis as quickly as possible - as others have said it could be something as stupid as water ingress due to bad installation or case design.

[Kalvin]

Where did you get the design in the first place? Any references to an application notes, existing reference designs, reverse-engineered existing products etc.?

[floobydust]

AC generators are terrible and will make huge transients when switched in, and while running.
I encounter very old, sluggish field-regulator designs that overshoot and glitch. They're pretty much best suited for running AC motors as a load, not electronics with PFC. The genset might have trouble regulating without some "normal" load present.

Also check how the generator is earth-grounded, I've seen many 3-phase units that just float. It could be using the feeder wiring/LED lights as (capacitive) ground.

You could check the (load) start-up sequence, if one LED string/feed is coming up before the other, it may be the sacrificial lamb getting the generator to settle.

Transients can be very fast and surges very slow- using a clamp for transients, would quickly fail if it's the genset outputting a gross surge.

These kind of problems are difficult to troubleshoot and I'd suggest renting a power analyzer, it will pay for itself.

[Jeroen3]

It's only 2500 watts, so the generator is probably tiny as well. Meaning there can very well be severe distortion from the generator just being small.
Think about triangle instead of since waves.

[nfmax]

as others have said it could be something as stupid as water ingress due to bad installation or case design.

That might fit: with overhead wiring the water runs down the cable, past the less-than-perfect cable gland, and into the luminaire. With underground wiring the cable (probably) enters the light from below, so water runs away from the luminaire. This is why you need drip loops...

[Zero999]

Hello,
We have installed  one hundred  25W outdoor LED lamps in a town in Bosnia. Sixty failed in just a few weeks.       

They all are fed from a single 240VAC generator which delivers power to the 2500 metres  of mains cable. A contactor near the generator is used to switch them all on/off. The  lamps are fitted every 25 metres along this cable. The cable is overhead (not buried). It’s aluminium  cable. Its twisted, though the installers tell us that in places the twisting is “not so tight”. We suspect there is an awful lot of mains stray  wiring inductance.
The strange thing is that they have another run of one hundred of the exact same  lamps, but in this case the 2500 metre mains cable is buried in the ground. –None of these lamps with the buried cable have failed despite being installed months ago.
The lamps  are controlled by microcontrollers in them which slowly dim the lamps as the night goes past the small hours, so as to save energy when  few people are really outside.

We suspect that maybe there is some kind of transmission line effect going on, with the long mains cable. We have advised the installers to fit an RC snubber across the contactor, and another across Live/neutral, just downstream of the contactor. We also wonder that with the “not so tightly twisted” mains cable, maybe noise is getting into the long overhead mains cable, and disrupting the microcontroller.
The lamps are not SMPS based, but instead are sucessively switched linear  current regulators. There is no mains input filter in each lamp, and the input stage is as in the attached. (ltspice sim also attached, including ZR431 model)

May I request what are your thoughts on why they have failed?
 

You do realise that you can't just model the cable with lumped components, such as resistors and capacitors?

What components are failing?

[floobydust]

yeah, 'fess up - what components are failing? Where are your autopsy results?