The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: firstname.lastname@example.org (N. Thornton)
Subject: Re: TRIAC "wears out" on dimmer circuit
Date: 28 Oct 2002 03:43:45 -0800
NNTP-Posting-Date: 28 Oct 2002 11:43:46 GMT
email@example.com (John Harris) wrote in message news:...
> I built a circuit to fade a string of twelve 20W tungsten filament
> light bulbs connected in parallel (for outdoor use). This translates
> to a peak current of less than 2A at 240Vac (after the initial
> switch-on surge). I used a TIC236D triac, which is rated for 12A at
> 400V. The circuit worked perfectly for a few days (around 60 hours of
> operation), but then a fault developed, whereby the string would only
> fade from fully on to around half brightness, instead of fully off.
> When the opto-transistor that I used to trigger the triac (via a
> capacitor) was disconnected from the driving signal, the lights
> remained at half brightness instead of remaining off. When the circuit
> was switched on from cold, it would work fine for a few minutes and
> then the problem would re-appear, even though everything was well
> heatsinked and the triac remained barely warm to the touch anyway. I
> replaced the offending triac and the fault disappeared for another few
> days before returning again. The string is around 10 metres long and
> no bulbs have blown. There is heavy interference on the AM radio band.
> I have a 240V 0.47uF capacitor connected across the triac output, but
> no inductor in series. I tried using inductors before and after the
> triac stage with no noticeable reduction in EMI. There is obviously a
> design flaw on my part and any suggestions would be appreciated.
There are a few problems here. As has been pointed out, the cap has to
go. LC filtering is best done on the input side. If you need
capacitance on the output, use a smaller cap with a good series R.
Secondly the bulb surge currents. 12x20w = 240 watts at 240v, 1A rms
running. With bulbs at a tenth their R at switch on, that will be 10A.
With mains at Vpeak in the cycle, that will be 14A. With noise on the
mains as well it will be higher still at times.
If you pick a 143 triac it should cope with that easily. But that
isn't the end of the story.
Triac voltage: 240v supplies see occasional spikes of way above 240 x
1.4. In fact if you dont have a filter on the input I would go for a
1kV rated part for reliability. Filtering is very desirable. A 400v
triac with no filter is a no-no really.
Thirdly I assume you want this thing to survive. Dimmers are famous
for dying when a bulb blows. When a filament bulb blows it sometimes
produces an arc across the filament, and i can effortlessly exceed 60A
until the bulb fuse blows. So I would want a triac that can handle
more than just the running load.
Moral of the story: avoid capacitor surges, use a bigger triac, and
fuse carefully to catch bulb failures.
I've assumed this is a one off: if its a commercial run things might
be a cut closer.
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup