LEDs ... The Good, The Bad, and The Ugly

I just did that (in the original post to 'lose the incorrect version), then scrolled down and read your post.

It's a bit clumsy as I didn't save the original in Eagle.

Neil

P.S. the extra junctions are the result of me being over-enthusiastic, not Eagle, I wanted to be sure it was clear..

Edited By Neil Wyatt on 26/09/2014 17:41:34

Thanks, Neil ... Much appreciated.

MichaelG.

Edited By Neil Wyatt on 26/09/2014 17:41:39

Hi Neil,
I think your schematic is still wrong as you have the resistor and capacitor connected across the mains. The two components (In parallel.) should be in series with the mains input to the bridge rectifier.

Les.

Edited By Neil Wyatt on 26/09/2014 17:41:52

..frequently

Edited By Neil Wyatt on 26/09/2014 17:42:00

.

[With apologies to HansR.]

The Quality/Safety situation with "Mains Voltage" LED lamp units continues to worry me.

Have a look at this video about Corn-Cob lamps.

Please ... seriously consider using only low-voltage LED units, especially in the workshop.

There are some excellent G5.3 and G4 fitting lamps available, and a multitude of Car bulbs; so a 12volt circuit is a realistic option.

[let us live to tell the tale]

MichaelG.

Edited By Neil Wyatt on 26/09/2014 17:42:07

Michael,

While I agree about bare-LED lamps, how is a fully-enclosed LED lamp more dangerous than a glass envelope filament bulb or a fluorescent?

All of them will expose live contacts if broken, but only the LED lamp would have anything in series to limit the current.

Neil

Edited By Neil Wyatt on 26/09/2014 09:38:27

Edited By Neil Wyatt on 26/09/2014 17:40:34

Neil

Fair point ... my answer would be the concern about build-quality.

If I recall correctly; Machine Lighting was supposed to be either SELV or driven from a Safety Isolating Transformer.

There have been several suggestions; on this forum and elsewhere, that GU10 desk lamps might make cost-effective machine lights.

All I ask is that people consider the implications.

MichaelG.

Edited By Neil Wyatt on 26/09/2014 17:40:30

On the subject of build-quality:

I was reluctant to pst this one, because the author apparently made an error when he "reverse engineered" the schematic ... but the salient points regarding Quality/Safety are very well illustrated, so look here.

MichaelG.

Edited By Neil Wyatt on 26/09/2014 17:40:19

Not that it seems I'm not interested in that LED discussion. In fact I am, as a couple of these outdoor floodlights are under way to me. I hope they are bang good! Link seems to work very slowly, so have patience!

So may I propose that these LED discussions ace collected into a dedicated new thread? Nobody will find them again in this one...

Kind regards, HansR.

Edited By Neil Wyatt on 26/09/2014 17:40:25

Following recent discussion on the "What did you do today" thread, I am starting this one as a general repository for comments and links about LEDs.

Please contribute; but please try to keep it relevant.

MichaelG.

.

An excellent point, HansR.

I have just started a new thread, and shall be happy for one of our Moderators to move [or copy] my recent posts on LEDs to that thread.

MichaelG.

Edited By Neil Wyatt on 26/09/2014 17:38:22

.

As a licensed radio amateur I consider them to be nothing short of the work of the Devil himself.!

Many of them from dubious geographical origins (that's most of them then) scatter RFI to infinity an beyond. There is no way if the import laws that are in place regarding such were even half enforced they would be allowed into the country.

Many have a badly designed and manufactured switch mode power supply that is the problem. This also extends to many other items sold.

Regards, Nick-Victor-Meldrew

As another radio amateur I can confirm that. the cheap ones are the worst, as the manufacturers leave out the interference suppression components (sometimes even the safety bits) after the things have been CE tested, or even just mark them CE without authority, as there is no regulation on the use of the mark. Good ones might be OK, but a quality manufacturer will be able to show you the test results. Well-known names in the industry might be OK, but check, and be prepared to pay!

Geoff

We and other members of this forum may well be prepared.

Unfortunately people living a few doors away probably will not. RFI is no respecter of fence lines.

Nick

To put things in perspective - there is is nothing much wrong with LEDs. They have come along way from being (red) low level indicators to now providing satisfactory levels of illumination with a spectrum that suits many applications, use less power and run relatively cool.

I do agree though that the way they are powered can be an issue. When 240v LED lamps first became available, I seem to remember that the interference from from the in built power supply in some of them wiped out DAB radio reception in the home. Now, in my office I have a reading lamp with an LED lamp close to a DAB radio and there is no problem. If one buys lamps from established reputable manufacturers that have a history of supplying the UK, one would hope that the safety issues highlighted (that are a serious concern) are not present.

If we are considering low voltage (12v) LED arrays, then we have a choice when selecting a suitable and safe power source that is up to the job. The problem of interference and safety becomes more of an issue when we try to do things 'on the cheap'. I know it is sometimes difficult and for instance I would not have a 'wall wart' plug in power supply in the house if I had a choice. My main aversion to the older ones is that the weight tends to cause them to not stay secure in a wall socket, the newer lightweight switch mode types overcome this but can cause interference, suffer poor reliability and I believe carry a higher fire risk if they fail.

It is doubtful if we can have much influence on preventing people from buying rubbish but if we do it right, there is no reason not to take advantage of LEDs in the home and in the workshop when there use is appropriate.

John

I have a PIR LED Floodlight on the rear wall of my house, it's been working for several months, reading this thread has got me worried now.

Edited By V8Eng on 26/09/2014 19:35:38

Hi all,
One thing I noticed on a 10 Watt mains voltage floodlight that I bought as that it was only fitted with 2 core cable. I would recommend fitting 3 core cable and connecting earth to the metal case. As the PCB was just floating loose and just insulated with heatshrink sleeving a short to the case would make it live. Better for it to trip the RCD or MCB rather than electrocuting someone. The units that versaboss posted a link to also appear to be fitted with 2 core cable.

Les.

.

John,

That's an excellent summary of the situation as I see it.

I do have some LED "household lightbulb" replacements [from IKEA and from B&Q], all of which are performing admirably, so far.

However: My main interest in LEDs is for Microscope Illumination and I have a very strong preference for battery powered devices in that situation.

I have done some preliminary testing on a 20W LED, with a view to using one or more of them as a Fibre Optic Illuminator ... Given that the light output from LEDs is essentially cold; it's surprising how much heat is generated from the back of the device [effectively, of course, it's the vast majority of the input power]. This thing will need a serious heatsink! ... This is also true of the GU10 lamps that we have been discussing, and I suspect that the heat build-up within typical domestic lamp-holders may be greater than users expect. Add that to "economy" construction, and the prognosis is not good.

I have a small "non-contact" thermometer on order, so should be able to do some quick'n'dirty experiments.

MichaelG.

.

Sorry, just spotted my silly typo. The LED under test is 10W not 20W

... and, for completenes, I should mention that it is actually an array of nine LED elements arranged in a 3x3 grid within a large "puddle" of phosphor.

MichaelG.

Les, I wrote earlier of my mains powered 10 Watt flood light with RFI problems, I cured it with a small isolating transformer.

Ian S C