Replacing bad triacs

[Darrell]

Now that the show is over, it's time for some repairs.  I have 10 older LOR AC controllers (8 Gen-2 and 2 Gen-3).  At least 5 of the controllers have bad triacs.  I had enough empty channels spread around that I was able to move that string to a new channel and keep everything running but have now run out of spare channels.  I've gotten the email from the LOR help desk with the needed part numbers for the 2 different model numbers of the triacs.  I am confident in my soldering skills that I can do the repairs myself vs shipping them in.  Before I order the replacement triacs, I have a couple questions.  Is there anyway to check if a triac is good in addition to simply plugging in a string and see if it dims and turns off properly?  I'm wondering if there is some valid check with a multimeter that can show if the triac is still good.  I have several channels that did not fully turn off unless I added a "snubber" to the string.  Then, it turns off completely.  Is that triac going bad and need to be replaced as well?   The ones I am currently calling "bad" will not turn-off even if a snubber is attached.  I know that I will also need some new thermal paste when reattaching the triacs to the heat sink.  Anything else that normally needs to be changed or replaced when doing the work?  The help desk email stated they don't provide any instructions if doing the work yourself.  I'm just trying to get everything ready before I start a repair line for all 5 controllers.  Thanks.

[k6ccc]

For the most part, a bad triac will result in a channel that either is 100% full on, or maybe 50% on.

Snubber is usually only required on strings that draw VERY little current - LED strings are typical for that issue.

 

[Tim Fischer]

I've never seen a snubber needed for lights to turn off... I use them on a handful of circuits where dimming is affected without them (dimming will be 'jerky' instead of smooth).

Are you certain that your hot and neutrals are not swapped, either on the outlet/power source itself or within the controller? With LEDs, having the triacs switch the neutral vs the hot can lead to what you are seeing.

[tlogan]

I had one channel on an AC Gen 3 that wouldn't light up at all (it's 6 or 7 years old and it's the first problem I've had, so I really can't complain!). Is that also likely a bad triac? And like the OP, I have a few spare channels so was able to move it very quickly and I'm thanking S6 for that! Took less than 5 minutes to move it to a different channel. It probably took me longer, to go out to the controller, unplug, replug and walk in and back out to test it. And I must say that I didn't know I had any use for the S6 app for testing, until now.

[Steven]

I had the same situation this year in which channel 10 wasn't working so I moved the plug to channel 12 and changed the Preview. Now I have to repair the controller. In this case the channel won't turn on at all. I'm going to assume it's the triac, although it could also be the opto-isolator or a bad solder joint In that area. Just to make sure I'm going to probe it with an oscilloscope first.

I have another controller on which if either channel 12 or channel 13 is turned on, then both props plugged into channels 12 and 13 light. That's going to be more interesting.

[rmturner54]

In my experience, be it as it may. I find over the years dealing with triacs and opto-isolators. I have changed more opto's than triacs. Not to say triacs dont go bad as often. If youdont have a way to test with say an oscope> jsut change both. Your already there. Not much more work to do both.

[Darrell]

More info and questions - I've now put each of the suspected bad controllers on the bench with the following results.

5 Gen-2 controllers.  1 has a single channel that an LED string will stay on at 100% from the instant it is plugged-in.  It does not matter if the controller has a signal from a controlling PC or not.  By the prior definitions, I take it this is definitely a "bad" triac and needs to be replaced. 

3 Gen-2 controllers have a single channel each that a single LED string will light and stay on at about 10% brightness.  The 10% value is my judgement comparing to a similar string on the next channel of the same controller.  If I run the HU test, the channels will dim up and down but will not go completely off.  Just down to the 10% level and stay there.  This is also true if run via the sequencer.  If I put a snubber on these strings the brightness will go down from about 10% to about 5%.  This is the effect I was describing in the initial post.  At night and in the yard display, the lower brightness made it seem as if the snubber fixed the issue.  In a dark room, I can see the string stays on but at a reduced level.  Are these "bad" triac's or something else?  The last Gen-2 controller has 3 channels with this same effect.  Always on at about 10% but will still dim up or down but never off.  The rest of the channels work just fine.  Is this 3 "bad" triac's?

2 Gen-3 controllers have a single channel that seemed to have issues.  Using the HU, both of the suspect channels seem to work properly.  The single LED string will dim up and down and is completely off when it is suppose to be.  On the bench, I can not seem to find any issue.  However, both of the channels on both controllers had issues during the show runs.  They looked ok during the shows but after the timed shows ended, these channels had the "bad" strings partially lit during the middle of the night.  After 1 or 2 nights of the partially lit string in the display, I thought the channel was bad and shifted it to an empty channel.  Can the show player leave a string "on" when it stops on the scheduled time?

[Steven]

Yes, the show player can leave a channel on. I had that happen a couple of times even though I have a shutdown sequence. It's also possible to do this on purpose.

Not on the bench, but in a yard, especially if it's wet, if the hot and neutral to the controller are swapped, then it will appear mostly normal, except that LED strings will light even when turned off. This is caused by the hot, that is going through the neutral connection, is always supplying 120v to the string, and it flows to ground via capacitive coupling. That's never enough current to light an incandescent string, but LED strings don't take very much.

[Darrell]

So down to the last question - if the string will still dim up and down but only goes down to 10% on and never goes off, is the triac bad or is this something else?

[k6ccc]

1 hour ago, Darrell said:

So down to the last question - if the string will still dim up and down but only goes down to 10% on and never goes off, is the triac bad or is this something else?

If this is a pre Gen 3 controller and the string is LED (particularly short ones), that is pretty normal.  A snubber is likely the answer.  Gen 3 controllers have snubbers built in.

Usually when a triac fails it is either 100% full on (both directions failed), or 50% on (if only one direction in the triac failed).

 

[Darrell]

This are Gen-2 controllers and the channels in question runs a 50 or 100ct LED string.  Adding a snubber does not stop the string from always staying on at about 10%.  Will changing the triac fix this issue?

[Tim Fischer]

1 minute ago, Darrell said:

This are Gen-2 controllers and the channels in question runs a 50 or 100ct LED string.  Adding a snubber does not stop the string from always staying on at about 10%.  Will changing the triac fix this issue?

As I said above, I've never seen a bad traic do this. I have, however, seen swapped neutral and hot do this. I will ask again what I asked earlier in this thread:

Are you certain that your hot and neutrals are not swapped, either on the outlet/power source itself or within the controller? With LEDs, having the triacs switch the neutral vs the hot can lead to what you are seeing.

[rmturner54]

I had a similar thing happen way back. Get your hots and neutrales in order. If your using zip cord it can be essy to make a mistake. Black white is pretty easy

[Darrell]

I'm absolutely sure the outlets are wired correctly.  I tested them with an outlet tester which verifies the polarity.  The same issue follows the same channel/controller as it is moved from outside to inside outlets.  The Gen-2 controllers were bought about 12-14 years ago as the "kit" with the pre-assembled board and provided pigtails/outlet cords.  The hot outlet wire is definitely going to the "hot" connector on the board as it is labeled.  Same with white neutral wire.  These controllers worked perfectly for many years.  There was the dimming issue for these Gen-2 controllers that caused me to make and use snubbers for the LED strings shortly after getting them.  With the snubbers, everything worked well for 10+ years.  It's only in the last 2-3 years that some strings/channels would only dim down to the 10% level but never go off entirely.  I was usually able to move the effected string to an unused channel and keep the show going.  I always assumed that some aging or a surge had affected an individual triac and it now allows some amount of trickle charge to bleed through and keep the string lit at this 10% level.  Can a triac "bleed" a charge or is it always all good or all bad?

[Darrell]

Is the hot/neutral issue strictly internal to the controller?  Are LED strings sensitive to which way they are plugged in?  The LED strings don't have plugs that are polarity different like the zip cord outlets.

[TheDucks]

Just now, Darrell said:

Is the hot/neutral issue strictly internal to the controller?  Are LED strings sensitive to which way they are plugged in?  The LED strings don't have plugs that are polarity different like the zip cord outlets.

Yes. Big box store versions are half wave.

Test the port with a old school 25W-60W incan bulb

[Tim Fischer]

2 minutes ago, Darrell said:

Is the hot/neutral issue strictly internal to the controller?  Are LED strings sensitive to which way they are plugged in?  The LED strings don't have plugs that are polarity different like the zip cord outlets.

Yes this would be a controller issue, or the power source leading to the controller.

[Darrell]

I just tested the power source outlet again (indoor outlet in my office) and it shows as a wired correctly.  Here's a photo of the hot/neutral connection on the Gen-2 board.  The led is a full-wave "pro-grade" 5mm led string. All the other channels work properly.

[Darrell]

Here's a photo showing the lit string at about 10%.  Again, with the HU or S6 tester, the string/channel will dim all the way up to 100% and back down.  It just won't go below the 10% level and always stays on.  There are 3 other Gen-2 controllers that have this same issue on one or more channels but all the other channels work properly.

[Darrell]

I just tested with an "old school" 40W incan bulb and the channel worked perfectly.  The bulb went to 100% and dimmed all the way off.  I guess the triac is not bad (?).  I plugged the LED string back it and it instantly lit back at 10%.  However, I then plugged in a different snubber and the string went out completely.  Seems I may have some bad snubbers.  While this can solve one issue, it still doesn't answer why there is a "bleed" on the channel.  All the prior comments indicate that snubbers should only help with jerky dimming.  I'm still left with controllers that have channels with a constant "bleed" that will light an LED string at 10% which is enough to be visible in the yard display when that string is suppose to be off.  Can triacs "age" or degrade with time/use to allow some bleed?