Control/Dimming ETC D20 Dimmer Failure

[MNicolai]

One of my D20's failed today. We had the lights on for focusing, and then what looked like a lamp failing as it became really bright and then shut off, turned out to actually trip the breaker. Upon turning the breaker back on, the dimmer acts as a CC20 now instead of a D20. Not useless, but not acting as it's preferred to behave.

I've seen a lot of discussion on here about SCR's and triacs, and I'd like to better narrow down which part of the dimmer failed before I hand it to my electrician, who will probably want to inspect if before we contact tech support. I don't know that contacting tech support will do much though, as the dimmer was manufactured in Oct. 2003 and is almost certainly beyond warranty.

 

[Sony]

If I recall correctly, I remember hearing about how a lamp that blows can very rarely take a SCR Dimmer with it like that. Other than that I don't know, odds are it's just a fluke, they happen. They easiest thing to do would be to just get another D20 module, they are cheap enough if you buy them online, especially used. We always keep 2 or 3 extra modules around just for this reason.

 

[MNicolai]

It appears that the lamp failed, but I was too busy preparing for a tech rehearsal to perform a follow-up. I'll do that tomorrow morning.

Luckily, we have multiple performance spaces and a number of Sensor and Unison racks, so we have a few spare modules here or there. Even if we don't replace it in the long run, I'm certain we can put a CC20/D20 hybrid module to use here or there.

 

[Les]

From what I've heard, the SSR is a fairly cheap and easy part to replace. I don't know if UL would cover your new relay module ;-)

 

[Sony]

It appears that the lamp failed, but I was too busy preparing for a tech rehearsal to perform a follow-up. I'll do that tomorrow morning.

Luckily, we have multiple performance spaces and a number of Sensor and Unison racks, so we have a few spare modules here or there. Even if we don't replace it in the long run, I'm certain we can put a CC20/D20 hybrid module to use here or there.

Just be careful, before you use it I would suggest putting an Oscilloscope to the output of it. Even though it has become for lack of a better description a CC20/D20 unit, it still may not be outputting a true sine wave, so it still wouldn't be suitable for powering movers or stuff like that.

 

[derekleffew]

...and I'd like to better narrow down which part of the dimmer failed before I hand it to my electrician...

Each D20 module consists of (from left to right): two circuit breakers, two chokes, and one SSR.

See the glossary entry http://www.controlbooth.com/forums/glossary-m-z/9027-ssr.html, and this thread: http://www.controlbooth.com/forums/lighting/6987-three-dimmers-wont-turn-off.html.

See also this document, IF QUALIFIED: http://www.crydom.com/en/Tech/Tips/How%20to%20test%20an%20AC%20solid%20satate%20relay.pdf.

 

[Slitterst]

Upon turning the breaker back on, the dimmer acts as a CC20 now instead of a D20. Not useless, but not acting as it's preferred to behave.

Sounds like a blown SCR/power cube. SCRs fail on, either as a design feature, or just by nature, I've never figured out which. But it makes such failures easy to detect because it behaves just as you've described.

Replacing the power cube in a D20 is pretty easy, and generally cheaper than buying a whole new module.

 

[JD]

As Derek pointed out, they use a SSR (Solid State Relay) as compared to discrete SCRs. These tend to cost a lot more, and yes, it is the SSR that went. The lamp may have developed a filament short from what you described. (Very bright, unusually high color temp.) When this happens, the draw current skyrockets. Although not common, the SSR failed in a shorted state prior to the breaker kicking out.

 

[church]

the usual failure mode for a TRIAC or SCR is a short circuit as indeed it is for a bipolar transistor. a transistor is three junction device manufactured from doped semiconductor material, an SCR is a four junction device made from the doped semiconductor material and a TRIAC is basically two SCRs arranged back to back. So the failure mode is not really by design rather it is implied by the device physics of semiconductors and the effect on the crystal lattice.

Having said that when a lamp fails in a short circuit mode it often causes the TRIAC, SCR or SSR to fail. A trained person, technologist, technician etc.can easily confirm what has failed and replace the faulty part. Note it is unlikely that a fuse or breaker can protect the TRIAC/SCR/SSR because these devices cannot interrupt the fault current fast enough.

 

[STEVETERRY]

One of my D20's failed today. We had the lights on for focusing, and then what looked like a lamp failing as it became really bright and then shut off, turned out to actually trip the breaker. Upon turning the breaker back on, the dimmer acts as a CC20 now instead of a D20. Not useless, but not acting as it's preferred to behave.

I've seen a lot of discussion on here about SCR's and triacs, and I'd like to better narrow down which part of the dimmer failed before I hand it to my electrician, who will probably want to inspect if before we contact tech support. I don't know that contacting tech support will do much though, as the dimmer was manufactured in Oct. 2003 and is almost certainly beyond warranty.

Mike, congratulations.

You are one of the select fraternity of .0001% of Sensor users that have experienced a failed dimmer module.

Seriously, this is a very, very rare occurrence. You need a new power cube, which can be ordered from your dealer or ETC. Or, perhaps it's under warranty!

ST

 

[derekleffew]

...of .0001% of Sensor users that have experienced a failed dimmer module. ...

Shenanigans! One ten-thousandth of one percent would be one in a million users.

I could maybe believe a failure rate of 1/1,000,000 dimmers, but have no way of determining the number of users per dimmer. Assuming an average installation size of 163 dimmers, and 3 users (LD, ME, AME) per installation, that's a failure rate of one every 54 million dimmers!

Granted it's not common, but not as uncommon as Mr. Terry would lead us to believe. Although Sensor dimmers do seem less susceptible to SSR failure than other makes/models.

 

[Thranduil]

The powerbricks fail, however, they are not that difficult to replace.
I also refuse to believe that they have a failure rate of one in every 54 million. Out of 600 dimmers, in the last three years we have had 10 power brick failures.

 

[STEVETERRY]

The powerbricks fail, however, they are not that difficult to replace.
I also refuse to believe that they have a failure rate of one in every 54 million. Out of 600 dimmers, in the last three years we have had 10 power brick failures.

My comment might have contained some hyperbole, but used poetic license to illustrate a point. That's why it included a smiley.

However, power cube failures on this product are, in reality, still a startlingly small number.

ST

 

[derekleffew]

...However, power cube failures on this product are, in reality, still a startlingly small number.

Well they'd better be, at a replacement cost of almost $86!

One of many fine sources for replacement parts:

Power Cube, Dual, D20, ETC

And I suppoe it beats spending $296.15 on a new module. But it's good to have spares!

 

[DimmerDotCom]

See also this document, IF QUALIFIED: http://www.crydom.com/en/Tech/Tips/How%20to%20test%20an%20AC%20solid%20satate%20relay.pdf.

This will check SSR's for switching on and off. For dimming, the SSR used should be phase controllable (aka non zero crossing or random firing). To test these SSR's fully you need to apply a pulse width modulated signal. I've seen some SSR's that flicker at the lower end or at some range probably because of a defective/damaged opto circuit or internal SCR. If you do not have a test jig install them in a dimmer and use your rack as your tester. Be warn also that higher power SSR's may work with lower loads but fail at high current since inherently the SCR's in the SSR are current devices (i.e. 125A SSR works with a 575W test load but flickers with a 10kW load.

Maurice Garcia
dimmer.com Home Page

 

[revmark]

I agree. .000001% I have a 10 year old Sensor. 48 D20's in the last year I have replaced 5 Q141's modules and 4 in the past 4 months.

 

[JimHaver]

Greetings all.
My local high school is gearing up for their fall production.
They have a Sensor Rack, 96 channels, CEM+ module, balanced 3-phase feed, ETC 48/96 console.
On Saturday, I went down to inventory fixtures, pull gels from last year, etc.; essentially reset to zero for the fall show.
Noticed a number of dimmer channels were not responding.
Mentioned to theater director, he said "a number" of the dimmers "weren't working".
What I did:
Turned on channels 1&2, 2 fixtures on.
Started swapping dimmer modules (D20).
worked my way through every module, removing from current position, installing into 1/2 position.
Discovered a total of 29 "dead" channels; some modules had only 1 functional channel, some had none, some had both.
29 channels? Holy Guacamole!
Checked breakers: OK. Verified continuity from power in to connector on power cube: OK.
Seems like power cube problem. No budget to replace the cubes.
What am I missing?

 

[derekleffew]

...Seems like power cube problem. No budget to replace the cubes.
What am I missing?

You're not missing anything, other than someone is going to have to spend ~$85.58 +labor per dual module to fix the problem. How long has anyone known "a number" of the dimmers "weren't working"? Are the venue's fixtures, cables, two-fers, adapters in good repair or are there sometimes shorts/sparks? Twenty-nine out of ninety-six dimmer failures couldn't have occurred all at once; more likely over a long period.

Just an aside, what's done is done, but ETC now recommends, as a requirement of NFPA 70E, that the dimmer rack be powered down (Lock-Out/Tag-Out) before swapping modules.

...Because of NFPA 70E, and its regulations on arc flash potential, we [ETC] cannot recommend hot-swapping any dimmer module while the rack is powered on. We strongly recommend that you power off the rack before swapping modules to prevent the potential for an arc-flash incident.

 

[Les]

Did you notice any dust build up while looking at the modules? That's a high percentage of failures, which I guess could be indicative of airflow/fan problems.

ETC does have safeguards against this (the rack will shut down in an extreme overtemp event), but this has got me thinking that there's more than meets the eye.

 

[JimHaver]

Thanks for the LOTO note; I had just turned off the breakers on the module before pulling them. This seems more of a personal safety note, rather than an equipment damage prevention. Which is fine, I'd just as soon stick around for a while!

All the dimmer modules were dust-free.
They actually remove all the modules before each show run (3 per year) and either shop-vac them or take them outside and hit them with the (oil-free) air compressor.
This install is only about 3 years old.

They rarely use 2-fers or extension cables; I checked every cable they used last spring
just as a general safety thing - showed a couple students how to put a stagepin connector on a cable, how to test the cables, etc.
Fixtures are a mixed bag; some scoops older than me, a few fresnels, mostly Source4 instruments, but no sparky-sparky.

I can install the cubes; that'll take all of 10 minutes each (once they're out of the rack), so at least they won't have to pay for labor.

As far as I know, all the channels worked last fall.
Seriously, though, 29 die in less than a year? Doesn't seem reasonable.
I've got portable racks at work that just hum along for years, despite being moved all over the place on a regular basis.
Based on that experience, I couldn't believe so many had died, and was hoping there was some other cause - you know, something I could fix for less than $86 per module.

I can buy a couple for them, but all in, they need 21 modules!
Oh, well. Guess it's time for the Theater Director to start writing grant applications..

Thanks for your help, guys.