Control/Dimming - Dimmer Fail

[LavaASU]

So tonight we were getting ready for show when one of out crew went to focus a light that had slipped out of focus. When he touches the light, there's a flash and it goes out. After a brief scramble a replacement lamp is installed and it's plugged back in (the dimmers were on the whole time, just the fixture unplugged). All seems well up until we start doing a final channel check when that channel won't go off. We decide there's something up with the dimmer (as everything else checks out) and start doing some basic troubleshooting. We flip the pack off and the dimmer channel STAYS ON. Eventually we unplugged it and took out the fuse to kill that channel.

So-- my GUESS as to what happened is that the lamp failed into a dead short (bright flash) which somehow took out the triac before the fuse could go (and also eventually burned up the shorting bit within the lamp). What I can't figure out is how is it on when the master power switch to the pack is off??? We did open the pack up and didn't see anything obviously wrong (mostly to see if there was a big burned section and/or anything dangerously wrong).

Am I right? And any ideas why it's staying on even with the pack off?

[derekleffew]

Bad day for triac s. See this thread: Chauvet DMX-4 Non Dim Problem . What make and model of pack is it? (On higher quality dimmers, sometimes but seldom I've had a lamp failure trip a circuit breaker, but never case damage.)

[MNicolai]

Bad day for triac s. See this thread: Chauvet DMX-4 Non Dim Problem . What make and model of pack is it? (On higher quality dimmers, sometimes but seldom I've had a lamp failure trip a circuit breaker, but never case damage.)

I have had an HPL brightly flash, fail, and then take out half of an ETC D20 module, causing it to fail in the On position.

[dramatech]

...What I can't figure out is how is it on when the master power switch to the pack is off??? We did open the pack up and didn't see anything obviously wrong (mostly to see if there was a big burned section and/or anything dangerously wrong).

Am I right? And any ideas why it's staying on even with the pack off?

You must have one of the Chinese shoebox dimmers. In the past we have had posts from folks that have noted that the shoebox dimmers are dangerous because the main power switch on the pack is only rated for a few amps. Truth be told, in those dimmer packs the power switch only controls the circuit board and has nothing to do with the triacs. The triacs are wired direct to the incoming power, through the fuses.

In your case the triac popped into the always on, or shorted mode when the lamp shorted internally. This is pretty common in the small dimmer packs with fuses and triacs, rather than the rack modules with breakers and solid state relays, although they also die of shorted filaments, it just doesn't seem to happen as often. Of course they can also die in the open position, where the light won't come on at all.


The fact that shoebox dimmers have the triacs wired directly to the incoming power, means that they are powered all of the time, when power is present at the outlets that they are plugged into. One might question if they could be a fire hazard if left on when the building is not occupied. When asking this question, realize that the SSRs in a dimmer rack are also powered all of the time, unless the main disconnect is turned off. In most School auditoriums, and many theatres, the dimmer rack is turned on 24/7. In theory, both dimmer racks and dimmer packs have either fuses or breakers on each load, that should prevent any fire hazard problems. I quess it is just the nature of knowing how most of the shoebox dimmers are constructed as compared to dimmer racks, that makes me less trusting of the dimmer packs.

To replace the triacs in a dimmer pack is pretty easy to do if you have any electo/mechanical skills. I have repaired quite a few for the various theatres in my area. I now keep a supply of 20amp triacs on my work bench. The real weakness of most of the lesser expensive dimmer packs is the poor quality and low amperage of the triacs. If these are replaced with a high quality and much larger amperage triac, they tend to last longer.
Speaking of the less expensive Chinese "Shoebox dimmers", there is quite a large disparity in quality between the various packs. Most of the packs are manufactured by Irradiant, a sub company of NEO-Neon. They manufacture many different qualities based on who is relabeling the product.
A few years ago, I purchased 3 Matrix dimmers from bulb america. They advertised as BulbAmerica dimmers, but they are famous for sending you whatever they have the most of. The initial 3 Matrix dimmers, are great. They are pretty well made, with a decent thickness of metal and with chokes on each channel. Based on the success of these 3 packs, we ordered 5 more. In this case we were sent the Matrix Pro. With the company suggesting that we got a better deal, because of added features. The Matrix Pro is pure crap. There are no chokes, and the metal is much thinner and the dimming curve is a real mess.
Lightonics, NSI and Leprecon, are American companies that have made reasonable if not great quality dimmer packs for many years. They are American made and have most of the right stuff. Although lately, NSI packs appear to be relabeled Chinese packs. Lightronics is offering both their own American made packs and the Chinese imports.
While I would prefer to have nice dimmer racks with many circuits, the "shoebox dimmer" really has a very usefull purpose as well. It has just become difficult to know in advance when purchasing small dimmer packs what you are buying. Price isn't always the answer as to the best quality. But the bargains are becoming more difficult to identify.

Tom Johnson

[LavaASU]

Hey Tom, any idea where to get a half dozen or so triacs cheaply?

And yes, these are the WONDERFUL Matrix Pros-- just in case one would like rope light connectors?!?! Yes, we also ordered the standard Matrix and got "upgraded".

[headcrab]

Mouser Electronics - Electronic Component Distributor should have your triacs. Make sure the new triac has an equal or greater current/voltage rating. Obviously, the pinout must be the same.

[JD]

The standard part on low end packs is a 15 amp triac fused at 10 amps. Not enough headroom. If you go to Mouser, you can get a 25 amp version of the same triac in the same case for about 50 cents more. I forget the part numbers right off hand, but they have been posted on CB before.

EDIT by Moderator:

BTA24 is a direct drop in. (Mouser 511-BTA24-600BW) I've swapped them out every time I work on such a pack. At 25 amps, at least you have some headroom. These packs are rated at 10 amps per circuit, 2400 watts total. I like to think of them more as 600 X 4 packs. ...

EDIT:
Ahhh... Now is a good time to remind everyone that unless you are qualified to do such work, we recommend you take the unit to a qualified service center.

[dramatech]

In choosing a replacement triac, there are several things to consider. In dimmer modules on major manufactures dimmer racks, the solid state relays are 40 amp capacity for 20 amp dimmers. Doubling the fused current, allows for in rush current and helps protect the Solid state device. In triacs, it is probably agood idea to do the same thing. As most "shoebox dimmers" are fused for a maximum of 10 amps per channel, it would make sense to have 20 amp or greater triacs in the dimmer packs.
The standard package for a triac is called a TO220 or some deriviative of that number. While there are triacs much larger than 25 amps, that is the largest that I have found in a TO220 package. It is also possible to get triacs with the mounting tab isolated from any of the three leads, and ones where the mounting tab is connected to one of the leads. Using one of the triacs with the lead and mounting tab connected, could possibly put life threatening voltages on the dimmer case. Or could cause a short to case that might cause damage, or even a slight possiblity of a fire. Therefore it is important that triacs with isolated mounting tabs be used.
As was edited in one of the other posts on this thread, Mouser has good quality triacs with isolated tabs. The part # given in the other post, was for a 25 amp triac. At the time I purchased my last quantity from mouser, they were out of the 25 amp triacs. I purchased 20 amp triacs that are still double the rated channel. The part number for those triacs is Mouser part# 511-BTA20-600BW. The price is low enough, that it is worth purchasing a pretty good quantity, as the shipping will be about $7 regardless of the quantity of triacs purchased.

Tom Johnson

[LavaASU]

Hey Tom-- time is rather an issue in getting this thing back up. Would these work 3 pcs. Q2025L6 TRIAC 25 Amp 200v Triac TO-220 25A - eBay (item 290284121860 end time Mar-09-11 12:49:51 PST) ? Also, is it critical for cooling that the mounting tab is in contact with the metal frame, or would it be possible to use a non-conductive method of mounting (most of our senior crew are actually engineers so I doubt devising a method would take much effort ). Also, as you are familiar with these packs (Matrix Pro), is the triac on the middle right edge of the circuit board (looking at the back of the pack with the mounting bar up) an identical one for the rope lights? Basically is that one the same and could we swap it with the dead one without killing anything other than the rope light stuff (which we're never going to use)?

Thanks!!!

[JD]

No-Go on the Ebay triac above. Needs to be a minimum of 400v to work in a 120v circuit. (even if the non-iso tab was not a problem.) Remember 120 is a RMS number, peak is much higher even if surges did not exist.

(120 RMS = 169.7 peak = 339.5 P-P. On pure resistive loads, p-p would not matter. )

[dramatech]

Hey Tom-- time is rather an issue in getting this thing back up. Would these work 3 pcs. Q2025L6 TRIAC 25 Amp 200v Triac TO-220 25A - eBay (item 290284121860 end time Mar-09-11 12:49:51 PST) ? Also, is it critical for cooling that the mounting tab is in contact with the metal frame, or would it be possible to use a non-conductive method of mounting (most of our senior crew are actually engineers so I doubt devising a method would take much effort ). Also, as you are familiar with these packs (Matrix Pro), is the triac on the middle right edge of the circuit board (looking at the back of the pack with the mounting bar up) an identical one for the rope lights? Basically is that one the same and could we swap it with the dead one without killing anything other than the rope light stuff (which we're never going to use)?

Thanks!!!

Acccording to the data sheet, the "L" in the part number indicates that it is an isolated tab. Yes, that triac will work, but as noted in another post, this is a 200 volt triac, and not a 600 as the one that we have been discussing in this thread. Having said that, it is doubtful that the original triac that has now failed, was any better than 200 volts. Everything should be fine, but lightning strikes or large electric motors such as AC, may put spikes on the AC power lines that exceed the 200 volts.
I find it difficult to believe that you can get some of those triacs on ebay shipped quicker than Mouser. If time is of the essence, Radio Shack use to sell a 6 amp triac as part of their stocked parts. Don't know if that is true any longer. BEWARE, that a 6 amp triac is not going to last very long with anything other than a 300 watt parcan or used as a dimmer for "Practicals" of low wattage.
On the Matrix Pro, the triacs are mounted on the metal bar that is between the edge of the circuit board and the power recepticles. The proper way to replace it requires removing the entire circuit board with the triacs and their heatsink bar intact. unscrew the screw holding the errant triac, and desolder the three leads of the triac from the board.
Circuit boards of Chinese manufactures are generally not of super quality, and can be damaged while desoldering and soldering. My method is to cut the leads of the triac as close to the body of the triac as possible. Then attach a hemostat or heavy clip lead to the leads one at a time and apply heat from a soldering iron, while wiggling the board. Usually the weight of the hemostat will pull the lead free. If it doesn't happen all at once and quickly, then remove the heat, let it cool for a few seconds and start again. The objective is to keep the circuit board as cool as possible during the operation. Too much heat applied to the board will make the copper traces become delaminated from the glass epoxy or phenolic, depending on the construction of the board. Once the leads are removed, you will need to clean the three holes so that the leads to the new triac can be installed. This can be done with a desoldering tool or solder wick, both available at Radio Shack. If in a real hurry, you can cut the leads of the triac and solder the leads of the new triac to the old ones coming from the board. This is a temporary "crap" fix, and should be done properly as soon as possible, when the dimmer pack is not in use.

Having repaired several of the Matrix Pros, may I advise that you take pix or make drawings of where every connection goes to the circuit board so that you can get them back where they belong. The power switch and transformer connections appear the same and could cause some serious damage if swapped. Also realize that every connection to that board has life threatening voltages. MAKE SURE THAT THE POWER IS REMOVED WHEN WORKING ON IT.
The several that I have worked on, I removed the ropelight connector, jumpered the power switch connector, and made the tranformer only work on 120 volts. Removed the 120/240 selector switch and changed the connectors for the DMX to 5 pin.

Good luck and if you have questions contact me direct at

Tom Johnson

[church]

the challange with TRIACs is the di/dt response - under a short circuit which occaisonally happens when a lamp fails the rapid current increase/decrease exceeds the maximum di/dt of the device resulting in overstressing of the semiconductor junctions and device failure. You can get the same problem with a higher current device.

Dimmers are phase-control circuits where the triac istriggered into conduction part way into each half cycle. At the end of each half cycle, the on-state current in one direction must drop to zero and not resume in the other direction until the device is triggered again. This ‘commutation’ turn-off capability is fundamental to triac power control applications. A TRIAC is essentially two Silicon Controlled Rectifiers or thyristors back to back on a single piece of silicon in a single package. As one SCR turns off, there is a possibility that its reverse recovery current (due to unrecombined charge carriers) may act as gate current triggering the other thyristor as the voltage rises in

the opposite direction. This is described as a ‘commutation failure’ forcing the triac to continue to conduct in the opposite direction instead of blocking the circuit current. The probability of a TRIAC failing commutation is dependent on the rate of rise of reverse voltage (dV/dt) and the rate of decrease of conduction current (dI/dt).

The higher the dI/dt the moreunrecombined charge carriers are left at the instant of turn-off. The higher the dV/dt the more probable it is that some of these carriers will act as gate current. The commutation capability of any TRIAC is specified in terms of the turn-off dI/dt and the reapplied dV/dt it can withstand, at any particular junction
temperature.

To overcome this problem snubber circuits using Rc networks for and inductive circuit and LC circuits for resitive circuits are used. Unfortunately this adds complexity - read cost - and inductors able to handle the current in dimmer circuits without saturation are expensive.

This is a particular issue with TRIACs - SCRs are seperate devices