No I am calling out the least common denominator for a reliability example.

My FACTS are based on a typical triac, say a MAC8N.

At 4A you dissipate 4W per triac min. The die temperature has to be between 40C and 280C with a max of 125C (2.2C/W j-c or 62.5C/W j-a times 4W plus 30C max ambient) meaning you need heat sinking of 20C/W with proper ventilation.

"And you're basing your idea" entirely on untested theory of a single app that is notorious for abuse. The fact is that for your theory to be correct the single bulb would have to conduct 100A for 2.6mS in a properly designed unit in order to have a chance of blowing the triac. (26A²S fusing per the mentioned spec sheet)
It does not matter the application. It is either properly speced or it isn't. It is either properly heat sunk or it isn't.

It would be much more common failure in an app like that to be either overloaded, which a club is likely to do, or the have design corners cut in the heat sinking, which substandard engineers are likely to do. If properly designed and loaded, I would still not put it past a club to prevent air circulation past the heat sink.

You should probably clarify to the less knowledgeable that the 3000W is the circuit capability and not the unit capability.

 

Your 'facts' have no relevance to the thread at all - I'm not posting an opinion, I'm posting what happens, and what everyone who's ever been involved in effects lighting systems knows - when a bulb goes it commonly takes the TRIAC out with it. Most go S/C (so that channel is permanently ON), but occasionally they go O/C instead.

I've done disco's, I've done PA and lights for bands - replacing failed lamps and TRIAC's is a regular part of doing so.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Yes and that is called FUSING, and my facts are quite relevant.

It is highly doubtful that all things being proper even a single 1KW bulb could draw 22KW for 2.6mS before going open circuit. It follows, therefore, that you're blaming the failure of the triacs entirely on the triacs is unfounded.

It is FAR more likely that the triacs are overloaded and overheated when the bulb blows. Come to think of it, another DESIGN issue is trying to flash a cold filament from peak line voltage, bad for the bulb and would not sit well with an overheated triac.

 

Sorry?, I've never blamed the TRIAC's for anything.

Which is what I've been saying all along - the bulb blowing often takes the TRIAC out.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Yes... which you proceeded to blame on the triacs to the extent of recommending that poor guy not use them.

How is that not blaming the triacs?

 

The exact quote is below - perhaps you might read it again in context?.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Was not really needed... the context of my reply to that was then the triacs are not being used properly.

The old computer saying: "garbage in, garbage out" ... or in your case "garbage out, cash in" ... if they were properly designed and utilized the triacs would not be blowing. Period. Which is what I said in the beginning.

 

So almost all professional and semi-professional light units are poorly designed?, interesting idea

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

I wouldn't know... I do know that it is my experience there are many wanna be engineers in niche markets that are less than well qualified, and that your experience supports mine, though you might not like admitting it.

 

Ok, I'm new here, and this is my first post, so I don't want to step on anyone's toes, but I've read this thread, and have a few points I'd like to make.

1. I'd say any piece of equipment that takes out a triac (and possible other components) because a light globe fails is poorly designed. Light globes are user-serviceable pieces of equipment with a limited life expectancy, and are expected to fail. If they take out semiconductors when they do, then it says to me that the designer didn't take this into account.

2. I can't see why a light globe would go short circuit when the filament breaks. How can there be a low impedance between the 2 contacts if the only part that could possibly cause a short is the filament, which is resistive. Nigel, if you could explain the mechanism that causes the light globe to become a short circuit, it could well clear up a lot of confusion.

3. Is it possible that the reason you see a lot of professional lighting equipment with a blown globe and a dead triac is because the globe failed and was never replace, and sometime later the triac, which is driving several other globes, failed for a completely unrelated reason? From what I've seen, as the number of lights in a lighting system increases, so too does the probability of having a blown globe, so I think this is possibly the reason why you see lots of dead triacs connected to blown globes. But that's really nothing more than an educated guess.

4. If this thread is any kind of measuring stick, I'm gonna like this forum, A LOT

 

What happens when the power line gets hit by lightning?

I used to work on SM power supplies, and every single model that was designed to be connected to the mains had to pass a Hipot test. > 2kV between the AC input and the DC output/case. If the triacs weren't driven by an opto coupler, then the circuit would surely fail this test. The test was designed to ensure that if the power line was hit with lightning, and you were touching the unit at the time, you would be somewhat protected from a good zapping.

 

All down to cost - they generally fit TRIAC's rated at least double the maximum load rating, I've seen then fitted with five times the rating, and they still blow (but probably slightly less?).

No idea what the mechanism is, it's just a well known commen occurance, presumably it's either the filament shorting out the contacts, or some kind of arc as it fails.

Perhaps it's a characteristic of PAR36 bulbs (and similar), although normal bulbs cause it as well, but not so often - but you generally get far more PAR38's than normal bulbs fed from light systems, so it's not a fair comparison.

No, the TRIAC's fail immediately the bulb goes - generally (always?) you have multiple bulbs per channel - you will see one bulb on that channel go out (sometimes with a brighter flash), and the rest go permanently ON. If the fuse fails as well, all bulbs obviously go out, but this is less common, the fuse usually survives.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

I've repaired (a few) and examined (a lot) of lightning damaged equipment, 2KV insulation is nothing to lightning - it's presumably just the standard test voltage used for insulation testing.

Incidently, the old saying 'lightning never strikes twice in the same place' is completely untrue, we had the exact same houses struck almost exactly a year apart - bad luck or what?.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Ive experienced that too, first time the lightning struck my house - it blew my computer up (dead power pack, although the rest of it survived), a year and 2 days later the house was struck again (we were in the garden watching this wall of rain come up the street - was really freaky) and our 3 day old 52" plasma screen got zapped good and proper - never to work again, and left a big 'Y' shaped scorched mark etched into the screen.

After reading this thread, i think I need to go through all my old disco gear and I have loads of effects that are no longer working - need to check the triacs on them.

__________________
There will come a day when PICs will rule the world!

 

My younger brothers house was struck the other year, it blew a big hole in his roof (through his £22 each handmade tiles ) and vapourised a number of rafters. Not a single electrical item in his house survived, and he had a LOT of gear - included FOUR - yes FOUR - surround sound systems.

All my old mates who used to run disco's seem to have retired from it now, so I don't get loads of repair work any more

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk