Traics continue to FAIL, why?

[JeffF]

Finally tally for me this year was 6 failed triacs with 3 on one controller and 3 on the another.  In total we have 19 LOR controllers and these were some of the newest(I think they *are* the two newest tbh and were bought in the last two years).

 

Replaced the failed triacs with eBay bought triacs and kept the 3 free replacement triacs LOR sent as back ups.  The markings on the LOR provided and the eBay bought triacs are almost identical except for the lot number.  The triacs that failed have strange markings compared to these, even though they are supposedly the same brand?

 

Just funny that I didn't have this problem the previous 5 years and once I switched off the defective channels all was good.  I'm chalking it up as defective triacs.

[zman]

Yes. Triac tester here: http://www.ebay.com/itm/Atlas-SCR100-Thyristor-and-Triac-Tester-/161061126371?pt=LH_DefaultDomain_0&hash=item257ffdc0e3

 

I have one and it works very well.

 

This triac is about the biggest you can get at 24 amps without outgrowing the circuit board.

 

Keep spares if you need to.

 

I have theatre type of dimmers that can use the bigger 40 amp versions.

But they have a bigger spread on the terminals and won't fit into LOR devices.

Ken that Triac link is bad or at least I am not able to get Mouser to bring it up. Can you verify it, thanks.

Edited January 13, 2014 by zman

[Ken Benedict]

Ken that Triac link is bad or at least I am not able to get Mouser to bring it up. Can you verify it, thanks.

 

Try this on Ebay: http://www.ebay.com/itm/New-SCR100-Thyristor-and-Triac-Tester-Analyzer-Peak-Atlas-SCR100-/121252779668?pt=LH_DefaultDomain_0&hash=item1c3b3a9694

 

or this on Amazon: http://www.amazon.com/Atlas-SCR100-Thyristor-Triac-Tester/dp/B005NIT5EQ

 

and a review: http://www.peakelec.co.uk/downloads/scr_review_sc_apr05.pdf

 

the datasheet: http://www.wattbits.com/media/datasheets/Atlas_SCR_Datasheet.pdf

 

the User Guide: http://www.peakelec.co.uk/resources/scr100_userguide_en.pdf

Edited January 13, 2014 by Ken Benedict

[a31ford]

Just a note, about triac's (there's that word again !)

 

Heat sinks, the size of heat sink (and the traces on the PC board) is/are going to determine what amperage a channel can handle (NOT just the Fuse).

 

IF the channel is ONLY on/off (no dimming used) then it will be the copper traces on the board that will be the limit, If DIMMING is used then it's the size of the heat sink, so here's the thing, the triac displaces the other portion of the AC waveform, into the heat sink (instead of the bulbs)  when dimming, if you are 50% dimmed, then the other 50% of the load goes into the heat sink.

 

Example: 1000 watt load at 50% brightness will cause 500 watts to be "sinked" into the heat sink. Have you ever felt a wall dimmer with it being dim for a while >?  they get quite warm.

 

The LOR boxes (and heat sinks) are designed for only so much amperage or current, you could get quite costly if you start installing bigger triac's into them.(blowing traces off the board).

 

HOWEVER, being a person that used to build theatrical stage dimmers,  one could use the LOR board (new case or enclosure) and different heat sinks (and a modified wiring system) to used HUGE 40 amp heat sinks on ALL the channels of a LOR board, right up to their rated output !!!

 

40 amps at 120 volts = about 4600 watts per channel times 16 = over 72 kilowatts of light !

 

Greg

[-klb-]

Half power on a 1,000 watt load does not put 500 watts into the heat sinks. The LOR heat sinks are not rated anywhere close to dissipating 1800 watts that a dual inlet controller would have to at 50% under your assertion. Do you really think they could handle the output of a portable space heater into those sinks, not having as much area, not the fans?

Your statement is closer to true for an analog dimmer using transistors, but I don't think I have ever seen that design, only phase angle, where the switching losses are much smaller.

[RayKob]

i've had three channels fail me this year on a controller that was a year old (g3), two in the beginning of the Christmas season and one in the middle.  The first two i was able to swap the channels with another controller the third i just had to suck it up and let it stay on all the time.  I reported it to LOR and they sent me replacement triac's no questions asked.  Things fail that's a fact of life but good service is sometimes hard to come by and i have to say the response i got from LOR was second to none... Thank you LOR.... 

[a31ford]

KLB, I'll agree that the 1/2 load thing is not truly accurate, but the idea was to let people understand that there is a penalty to pay if using larger triacs...

 

Since we have an edit button again, I'll correct my statement.

 

From a textbook I have.

 

Calculation of heatsink size

To calculate the required heatsink thermal resistance

for a given triac and load current, we must first calculate

the power dissipation in the triac using the following

equation:

P = Vo x IT(AVE) + Rs x IT(RMS)

2

.

Knee voltage Vo and slope resistance Rs are obtained

from the relevant VT graph in data book of factsheet. If the

values are not already provided, they can be obtained

from the graph by drawing a tangent to the max VT curve.

The point on the VT axis where the tangent crosses gives

Vo, while the slope of the tangent (VT/IT) gives Rs.

Using the thermal resistance equation given above:

Rth j-a = T/P.

 

Greg

Edited January 13, 2014 by a31ford

[zman]

Try this on Ebay: http://www.ebay.com/itm/New-SCR100-Thyristor-and-Triac-Tester-Analyzer-Peak-Atlas-SCR100-/121252779668?pt=LH_DefaultDomain_0&hash=item1c3b3a9694

 

or this on Amazon: http://www.amazon.com/Atlas-SCR100-Thyristor-Triac-Tester/dp/B005NIT5EQ

 

and a review: http://www.peakelec.co.uk/downloads/scr_review_sc_apr05.pdf

 

the datasheet: http://www.wattbits.com/media/datasheets/Atlas_SCR_Datasheet.pdf

 

the User Guide: http://www.peakelec.co.uk/resources/scr100_userguide_en.pdf

Dang I hate it when you can't read my mind and my posts are not explicit. I meant the link to the Triac was not working, but it seems to be now. Thanks.

[Max-Paul]

I own 10 of the PC units and 2 DC units. Just did my 4th year. Just had my first bad triac. Part of it is cause I must have missed installing the screw and nut for this channels. Although 1, 100 cnt LED string should not kill it. I think that the ice must have added to the load, enough to damage the Triac. So, I am feeling fairly lucky compared to what I am seeing others users are having.

[plasmadrive]

OK.. reality check on triacs and heat.  Assuming normal operation for the following.

 

When a triac is in the off state it dissipates virtually no heat.  When in the on or in the conduction state, they dissipate about 1 watt per amp, (round numbers).  They drop approx 1 volt when in the conduction state so 1 amp x 1 volt equals 1 watt. (again, round numbers). 

 

Whether you use 12 amp or 30 amp triac is pretty much not relevant heat wise.  They will dissipate about the same heat given the same load. 

 

There is no in between state, either they are on or they are off, (actually there is a rise time of about 3-5us and there is heat developed during that time, but that averages very little in the scheme of things).  That is why phase angle control is so inexpensive to make and use.

[k6ccc]

Plasmadrive - I think you've got it there.

[stanward]

I own 20 of the CTB16PC units purchased over a course of 6 years. I had triacs go out more during periods of large rain. And as noted before, the triacs would fail between the end of a show and the next evening (before the show).

I had a bunch of blown triacs in 2011 because that was the rainiest season for me in the 6 years of using LOR. And I had many triacs fail on channels that only had a single strand of LED lights.

On a side note, a lightning storm this past Dec 30, 2013 caused a huge amount of triacs to blow, but this is not a normal case (of a lightning storm that is).

[rwertz]

On a side note, a lightning storm this past Dec 30, 2013 caused a huge amount of triacs to blow, but this is not a normal case (of a lightning storm that is).

 

That's interesting. There has been speculation before that some of the TRIAC issues were related to ESD hits. If they were it could explain why the occurrences seem somewhat inconsistent. I also think other issues sometimes get misdiagnosed as bad TRIACs because they get discussed so much.

 

I'm up to 15 controllers after six years and had my first bad TRIAC last year. I'm using all older controllers.

[Greg.Ca]

Isn't the real problem is that LOR designed the output triacs without fuse protection? There would be NO blown triacs if LOR had incorporated current protection (fuses) in their designs. --Greg--

[a31ford]

Actually, no, and for one simple reason.

 

the fuse requires an over amperage condition for a long enough time that the triac would pop the junction (silicon junction internally) would pop, faster than the melt-able link within the fuse (I'm talking fast action type fuses).

 

ALSO, it would give a person the chance to install higher current fuses (defeating the purpose in the first place), AND, add a large cost to the controllers for nothing.

 

Better wiring practice AND testing of light strings BEFORE they are installed, greatly reduces the shorting out of triacs in the first place.

 

Greg

[plasmadrive]

Isn't the real problem is that LOR designed the output triacs without fuse protection? There would be NO blown triacs if LOR had incorporated current protection (fuses) in their designs. --Greg--

That is just blatantly false!  As an engineer that has designed lots and lots of triac circuits over the last 30 years, I can tell you there are all kinds of reasons why you are wrong, but suffice to tell you that no fuse is fast enough to always protect a triac.  I doubt seriously more than .1 - 1% of the blown triacs are due to load current, (that is a guess judging by what has been posted here).  If however the failures are from ESD hits, (which I would not bet the farm on, but is an interesting thought), there are no fuses that will protect them. 

 

I am sorry, but spouting off like that without any facts to back up your statement is just plain wrong.  I don't often make statements like this, but yours is right up there with the tin foil hat group.   There is nothing wrong with LOR's design regarding the triacs and the price point the market can stand.  Also keep in mind that LOR as already told us that there were a bad batch of triacs from the manufacturer.  Hardly LOR's fault, but they made them good. 

[a31ford]

Gawd !! I hit "F5" (refresh page for those that don't know)  Saw plasma's post saw "Greg" LOL, and thought he was talking about my post.....

 

LMAO I can be wrong, or tainted sometimes, but TG it wasn't this time.

 

Greg Manning (NOT Greg.ca, even though I live in Canada)

[a31ford]

Sorry all. just had to.... "300 and counting" now if I only had that many channels.

 

Greg

 

edit... let me guess, this one will cost me some rep....

Edited January 16, 2014 by a31ford

[RayKob]

I see several people had problems with water getting into the controllers, this is what i did this year.  I purchased the large christmas gift bags, cut a section of the bottom out to fit over my controllers and tied the top like you normally would a gift.  looked better than just having the controllers exposed and kept the rain out.   

[RayKob]

Well I've replaced the three bad triac's and tested my controller,  Christmas 2014 here I come.......