Triac issues (lots of them)

[Jerry]

Wow Dan,  That was quite a read with all that HTML junk. 

 

I think that answers all my questions... So great to have a company be up front about things that just happen.. My hat is off to LOR for being upfront and concise...

 

I judge a company not by whether they have a problem or issue, but rather by how they handle that problem or issue..  LOR Rocks!!

 

The bad thing is now I have no puzzle to solve.. sniff sniff..

 

Craig 

I'm sure there will be more "puzzles" for ya.

[Jerry]

Got you covered, Dan --> Cleaned up:

-----------------------

There was a question "was there a manufacturing issue with triacs? is that real or a myth?"... In 2011 we did have a bad batch of triacs. We learned a lot from that experience and have implemented procedures to test batches. We built a 240 amp testing rig that can push triacs through an extreamly harsh test and is used to verify batches of triacs. </span>

We are constantly looking to make the controllers more robust. We use multiple manufactures with duplication so in the event there is an issue with one of the manufactures we have backup. Most of the cards we sell are made in the USA. Most of the parts for LOR products are purchased directly by LOR and we always use highly recognized distributors and in some cases we purchase directly from the component manufacture. All of our manufactures are certified and we do not see any significant difference in quality from the cards delivered from these manufactures. 

Triac failure has been the number one issue with LOR controllers since day one. About half the time we can determine that the failure was caused by an external problem and the remainder of the issues we are not sure. For example we had one fellow who was complaining that triacs were failing all over his display, no special lights, no heavy loads, etc... The problem was determined to be little worms that were crawling up into the warm controllers and shorting out the circuits. 

Worms and insects, little lizards are commonly found fried in controllers that are sent back for repair. But with all that said, there are cases were we simply are not sure why the triacs failed. As far as bad batches of triacs, as mentioned earlier, that did happen one time and we have testing procedures in place now to test the batches, look for bad quality, counterfeit parts, etc. 

When we designed the Gen3 cards, one of the criteria was to increase robustness, we looked at the most common failures and attempted to reduce the chance of failure at those points. One point was the triacs. We moved from a 16amp rated triac to a much more expensive 25 amp rated triac. The result is that we have seen a reduction in triac failures with G3 controllers. The second most common failure was RJ45 connectors that were pulled loose. By reversing the mounting orientation of these connectors on the Gen3 controllers there has been a reduction in those failures as well. 

We are constantly looking to make the controllers more robust. We use multiple manufactures with duplication in who manufactures individual components so in the event one manufacture has an issue we have backup. Most of the cards we sell are made in the USA. Most of the parts for LOR products are purchased directly by LOR and we always use highly recognized distributors and in some cases we purchase directly with the component manufacture. All of our manufactures are certified and we do not see any significant difference in quality from these manufactures.

Regardless of where a component is manufactured, when it is delivered to LOR it is tested / packaged / re-tested... For example for a CTPC16P-ReadyToGO, We get the CTB16PC card from one of the two manufactures that build them for us. We program and test the card to ensure all 16 channels are operating. We then install the card into the plastic enclosure and connect all the wires. Then the controller is plugged into a test fixture where it can be load tested. We verify all labeled outputs (1 - 16) are on the correct channel. Each channel is load tested and full banks of 8 are load tested. We test the unit with a high pot tester to look for shorts between high/low voltage and ground. 

Back to triac failure, we do not know of any particular issue that causes triacs to fail.with common incandescent and led Christmas lighting. We do know that there is a certain load that appear to damage triacs. These lighting products contain a type of switching power supply and are most commonly used in 240 volt countries where line voltage is not used on led and most incandescent Christmas lights. Note: (we became aware of this this issue this season and are not sure of the cause yet or if it is the triacs being damaged)

------------ When a triac is damaged -------

We will of course fix controllers under warranty at no charge including shipping,. We will repair controllers that are no longer under warranty for free with the user covering the shipping charges. Regardless of the age of a controller, we will send triacs at no charge to a domestic user who needs them.

Thanks Dan, Question is,,,,can we replace our damaged triac's , one at a time , as needed, without doing any other modifications to the boards with the 25 amp ones ? (GEN 1 & GEN 2 )

[Ken Benedict]

On a related note, Elation made a single channel DMX light dimmer a few years ago with a 16 amp triac, and had so many problems that they redesigned the dimmer to use a 40 amp version, with excellent success. I've heard the failure rates with the 40 amp version are almost nil.

 

But the 40 amp version is physically larger and more expensive than the 25 amp version.

 

I've used these DMX dimmers on larger indoor displays that need 800-1000 watts per channel.

 

And use the LOR iDMX1000 quite successfully to drive them.

[plasmadrive]

You should be able to swap 25amp for 16 amp.  From the specs I saw they are physically the same and the drive specs are not an issue. You should have no issue replacing them one at a time. 

 

But.... I am not LOR.. so my info is unofficial.

I'm sure there will be more "puzzles" for ya.

I think I must have watched too many episodes of "House". 

[LightORamaDan]

^{We would like to be in a position where there were no failures and the equipment was still affordable. Although there are many folks that have encountered one or more triacs failures, there are many more who have never encountered a triac failure}
 

^{We have noted that failures sometimes cluster and we do not understand why. We will see a user with 20 controllers run a year with no triacs failures. We will see another customer with 20 controllers see 10 triac failures in a year. Both users seem to be using the same type of lights and same loads on the lights (on average) and neither seems to be treating the controllers better or worse than the others. .}
 

^{We do know that people living in warmer climate where there are lighing storms have a higher equipment failure rate as the long extension cords can pick up electrical charge even with fairly remote lighting strikes. That is the only environmental relationship we have seen to failures.}
 

^{As far as replacing all the 16 amp triacs for 25 amp tiracs on a card, I would be hesitant as there is a risk of damaging the card during the process and that risk may outweigh any benefit. The 16 amp triacs have worked reliably for up to 10 years on cards and are not them selves a weak choice. .}
 

^{Believe it or not there was a group (not a private user) who mounted their controllers upside down in a mega tree. That way the dangle cords were pointing up and made it easier to plug things in. Their failure rate was rather high and this created a cluster of failures. Now I am highly doubtful that anyone here has done that but there may be some more subtle issues that even I would do that could cause problems.}
 

^{For the record we saw triac failure on controllers sold this year less than  0.01%. As mentioned we see clusters of failures with where the local failure rate is higher. We will continue to improve things and to get a better understanding of the clustering phenomenon, which certainly could be a manufacturing problem or a user sentimental problem.}

[David Rise]

Haven't had any issues with my Gen 3 boards (knock on wood).  3 of my Gen2 boards have/had triac issues.  No biggie.  i bought extra triacs and can replace them easily enough.

 

I'm slowly replacing my Gen2 boards anyhow.  Can't say enough good things about the Gen3 boards.  not sure what more improvements can be made.  At least not for the average user.

[mrduncan]

I have a total of 8 LOR controllers on which 5 are Gen2 and 3 are Gen3. I have had no problems to date except for a erratic channel on a Gen2 controller which was promptly repaired by LOR. However, if I controller is installed outside, it is protected against direct rain and snow. And, all circuits are GFCI protected.

I saw a couple of links for Gen2 Triacs earlier in this series of threads. Can someone send out a link to the Gen3 Triacs? I want to stock up with a couple of each just in case I need to perform a quick fix.

 

[Robert Burton]

Ok,

FINALLY have mine on the work bench. Before I tear into it...

With nothing being sent to the controller....

1 Channel flickers a little

1 Channel is stuck at 50%

Sounds like triacs????

[Orville]

Have 4 2010 CTB16PC Controllers and have only had one Traic failure in one of them.   And they have been left ON in rain, cold {no snow in Florida} and freezing temps when we've had them, and they were all on a GFCI circuit.  Had a power glitch from the power company one night, tripped the GFCI and shut everything down.    And I believe that is what may have blown the triac on my controller.   Just lucky it was only one that blew and not any others.

[Robert Burton]

Ok,

FINALLY have mine on the work bench. Before I tear into it...

With nothing being sent to the controller....

1 Channel flickers a little

1 Channel is stuck at 50%

Sounds like triacs????

Hello???

Buehler????

[Santas Helper]

Hello???

Buehler????

Yes, sounds like triacs.

[-klb-]

Stuck at 50 sounds like a triac.  Flickers a little is not one of the common documented failure modes for the triacs. It might also be the opto triac. But once you are in there, you might as well change both triacs, then go from there if the flickering one still has issues.  But always start with a reset...

[Robert Burton]

Ok. I will try another reset. Did reset it several times during the season, but honestly...not sure if the reset took. Wouldn't that be nice if it fixed it!

 

There is a radio shop in downtown ATL. Probably will pick up a hand full of optos and the triacs.

[-klb-]

Reset will retain unit ID.  For a Gen2 controllers, the key thing is to see the fast blink when you power up without the jumper, or with the unit ID dials set to 00.  Not sure the key thing to look for on Gen3.  Just want to be careful that we aren't talking people into changing triacs without having tried the reset.

[rwertz]

I had a flickering issue on one controller that I resolved by reinstalling the firmware (same version). At the time, I didn’t think to try a reset first. So I can’t say if a simple reset would have resolved the issue.

[LOR Staff]

In general, channels stuck at 50% with an INCANDESCENT load attached are dead triacs.  Flickering channels are usually not bad triacs (but could be).  The best thing to test with is a regular old normal 60W incan light bulb.  That eliminates all the variables of LED lights, etc....

 

Here is what we send for troubleshooting a bad channel.  Follow these instructions and see what happens.  If you still have problems, you can always open a trouble ticket, and we can RMA the device.

 

--------------------

 

 

How odd... it's either a suspicious string of lights or the triac controlling that channel is in the process of going bad. Let's hope it's the first because it's easy to fix.

 

Some quick tests.  Power off the controller and plug a different light load into the stuck channel.  Power up and run your tests again.  We've seen too many lighting products that behave out-of-the-normal because they weren't designed for animation environments.

 

Another trick is to 'reverse the polarity' of the lights.  Where the lights plug into the extension cord or power dongle, turn the light plug around 180 degrees and plug it back in.  AC power is non-polarized but some light strings will act normal again by turning the plug around (we know it makes no sense... but it sometimes works.)

 

One more trick used by the pros is to put a small incandescent light load on the troublesome light channels, usually at the end of the light string.  Typically one incandescent C7 bulb does the trick... LEDs will then take on most of the dimming characteristics of the incandescent.

 

If the problem is still there, let's do a board reset:

 

If it's a 1600 series Generation 1 or 2 (metal enclosure and no LED display inside)), disconnect the power, set the unit ID switches to 0-0, and reapply power.  The status light will blink very fast.  Wait a few seconds, remove power, change the Unit ID switches back to where they were and reapply power.  You've reset the board.  

 

If it's a 1600 series Generation 3 (metal enclosure with LED display inside), disconnect the power, press the up and down buttons simultaneously and reapply power. The display will show '0000.'  Wait a few seconds, release the up and down buttons, the controller will be reset and return to normal operation in about two seconds. 

 

If it's a CTB-16PC series Generation 1 or 2 (plastic enclosure and has a jumper block to the right of the data cable connectors), disconnect the power, remove the jumper next to the data connectors (usually in the second position (or J2)) and reapply power.  The status light will blink very fast.  Wait a few seconds, remove power, replace the jumper on the J2 position and reapply power.  You've reset the board.  Go into the Hardware Utility and make sure the board is assigned to the correct physical location where it was before the reset.  

 

If it's a CTB-16PC series Generation 3 (plastic enclosure and has a jumper block to the left of the data cable connectors), disconnect the power.  On JP3, counting from the left, move the jumper on pins 4 and 5 over one notch to pins 5 and 6.  Reapply power.  The status light will blink very fast.  Wait a few seconds, remove power, move the jumper pack to pins 4 and 5.  Reapply power.  You've reset the board.  Go into the Hardware Utility and make sure the board is assigned to the correct physical location where it was before the reset.  

 

Now, run your tests again.

[Robert Burton]

Ok Mike,

Have done the above. I have a Gen 1 PC board in plastic box. 4.32v .  The flickering is not all the time. I have just noticed it every once in a while. The other channel, is definitely on at 50%. I reset it and we will see if the flickering comes back. Have a 60 watt bulb on it.

 

We will see how this turns out. I know as generations go forward, sometimes better parts have been found...such as in the Gen 3 board. What opto chip is being used now?

[Max-Paul]

Just one channel Robert? If so, are we sure that all of your wiring for that channel are good clean and tight at each connection.

[rwertz]

Also, if the flickering channel only occurs when you are playing a sequence you might want to make sure you don’t have a channel config issue in your sequences. Duplicate Unit-Circuit settings between channels (or a main sequence and a background sequence) can cause some strange effects. You can run LOR’s Verifier program to look for issues like that.

[Robert Burton]

Max-Paul - Just checked...yep. Connections are tight. :-)

 

rwertz - Yep. Been there done that! That is not the case here. I had it on the Hardware LOR module when I saw it flicker. I did a reset and we will let it sit on the bench with this 60watt bulb on it...to see if it will flicker again.

 

Fingers crossed.