Leviton D4DMX dimmer nightmares

[Benchtech]

I feel your pain. I have just gone through 8 of these this week, replacing the surface mounted DMX chips. If they are out of warranty, the factory recommends replacing them as they are too labor intensive to fix econimically and you can believe them. They are unless you have a lot of patience and spare time. As far as channels that stay on, the triacs usually short because of lamp failure when the filament breaks and spins around from the heat of the arc, causing it to hit the support posts and shorting them out drawing a lot of current. Sometimes this happens so fast that fuses or breakers can't catch it, taking out the semiconductors. Channels that don't come on usually are bad fuses/breakers, bad chokes (broken leads from vibration of moving) or optoisolators.
If you are going to take a hammer to them, drop me a note and I will gladly pay the freight and use them for spare parts, or if they are easily fixed, I'll let you know.
Benchtech

 

[JJWOODS]

Hi I'm new to this forum, I have several D4DMX-MD5 units that are in need of repair.
I would like to try and trouble shoot the triac's on these units however since Leviton can not provide me a schematic and I can't find any threads regarding this I was hoping someone on this forum might be able to provide some guidance?

I'm not sure where the triac's are located on the board or what they look like? Can someone provide some assistance or direct me to a photo of a triac? If I use a mulitmeter will a continuity function tell me if the triac's are good or bad?

One of the dimmer packs channels 1 & 3 Power on 100% and have no functionality, I think the triac's might be bad on these channels? Does any one know if there is 1 triac per channel or do they work in series?

Also a few of the units have bad led screens the numbers and letters are scribbled does anyone know where I can purchase a replacement screen?

Thank you in advance for any assistance you can provide. I spoke with Leviton tech support today after playing phone tag for three days to my dissapointment the tech support provided no help and told me they may be able to repair the units for $105 each which is about the price that we can purchase them for.

 

[JD]

Although there are many threads on this forum about repairing/ changing tricas on Shoebox dimmers (The term for a stand alone dimmer with a power cord, usually 4 channels) I must caution you:

Unless you are a qualified service technician, we could get in trouble for offering advice that may result in personal injury. This is why people here shy away from giving repair advice.

That being said, you may want to glance at this old thread: http://www.controlbooth.com/forums/lighting-electrics/7369-elation-dp-dmx20l-popped.html

 

[BillESC]

FWIW, the D4DMX packs are assembled in Mexico with parts supplied by Leviton/NSI, not manufactured in China.

 

[DimmerDotCom]

...I was using a voltage meter to check one of the non-working dimmers. My voltage meter dutifully told me that there was a strong 110v coming out of the dimmer. "So why won't my light turn on!!!", I cursed and fumed at the dimmer. Eventually I realized that although there was a nice strong flow of volts, there were unfortunetly no amps coming out! Doh!

Always measure dimmers with a load applied. Triacs are basically diodes. As such voltage seeks to go across the highest resistance, in this case the open plug which has near infinite "air gap" resistance. Thus the AC on the input side of the triac go to the load side. EVEN IF THE DIMMER IS TOLD TO BE AT ZERO PERCENT.

As for load current and dimming, here's an waveform of the current draw of an expression 3 console (uses a switching power supply which is more efficient than a linear power supply).

 

[hemi392]

Hi I'm new to this forum, I have several D4DMX-MD5 units that are in need of repair.
I would like to try and trouble shoot the triac's on these units however since Leviton can not provide me a schematic and I can't find any threads regarding this I was hoping someone on this forum might be able to provide some guidance?

I'm not sure where the triac's are located on the board or what they look like? Can someone provide some assistance or direct me to a photo of a triac? If I use a mulitmeter will a continuity function tell me if the triac's are good or bad?

One of the dimmer packs channels 1 & 3 Power on 100% and have no functionality, I think the triac's might be bad on these channels? Does any one know if there is 1 triac per channel or do they work in series?

Also a few of the units have bad led screens the numbers and letters are scribbled does anyone know where I can purchase a replacement screen?

Thank you in advance for any assistance you can provide. I spoke with Leviton tech support today after playing phone tag for three days to my dissapointment the tech support provided no help and told me they may be able to repair the units for $105 each which is about the price that we can purchase them for.

There are 4 triacs, one for each channel. If you remove the cover they are mounted to a heat sink and the edge of the circuit board opposite the circuit breakers. They are common triacs available from any electronic parts supplier. They are type BTA 16 600B and cost about $1.50 each. Make sure you get the BTA version and NOT the BTB version. You can use a multimeter or continuity checker to check for shorts. The circuit board MUST be removed for repair. This is NOT for the faint of heart. The good news is that the triac legs are mounted to the circuit board in slots so unsoldering and resoldering is really easy. BTW, the on/off switch only controls the DMX circuitry, so a shorted triac will be full on as long as the dimmer is plugged in. Good luck.

 

[JD]

A little bit more about BTA vs BTB triacs. The BTA triacs are electrically isolated devices that can be mounted directly to the heat sink. The BTB triacs have their mounting tab attached to one of their main terminals. If you were to use the BTB, the unit would short out the second you plugged it in. (best case) In the worst case, if your ground was not connected correctly, the entire case would be at line voltage potential.

As you can see, a simple part substitution can create a dangerous situation.

 

[NeroCaesar]

So I have read everything from the start and I have to ask.

In a perfect exact world Could you...

Plug 1000w into a 600w channel of the leviton packs and only bring it up to 50% with no problems (not that I'm trying to... well kinda... but I want to know the physics behind it)

or will it blow because it will draw the 1000w anyway
or will it not matter because the Triac is fluctuating so much that the actual power draw is not enough to blow 600w


I also ultimately ask because I have a EDI 96 dimmer rack where I converted some cards to nondims to run a bunch of leviton packs in the house. As I have been reading and tested if I have 4 units plugged in at about 20% each the pack it drawing about 2400w anyway. I am worried that if I have a bunch of Levi packs, a little more than enough to over draw the power to the EDI Rack, that I will blow something even if every light is at 20%

Does that make sense?
The EDI rack has 3 legs of 200 amps as I have been told
If I have 32 Levi packs and 128 S4 units at 575w each that's 613 amps about
Not to mention any other lights on the rack

 

[derekleffew]

...Plug 1000w into a 600w channel of the leviton packs and only bring it up to 50% with no problems (not that I'm trying to... well kinda... but I want to know the physics behind it) ...

See the thread http://www.controlbooth.com/forums/...61-overloading-dimmer-without-blowing-up.html :

...The same line of thinking applies to the dimmer branch circuit as to the main feed. This is a matter of good practice, not NEC rules. Intentionally overloading a dimmer is not a "good practice" solution to the problem. If you don't mind a tripped breaker in the middle of a performance, or a failed dimmer due to overheating--have at it. But if you want to implement a robust system that will be reliable, find another solution. And there are many--smaller lamp wattages, less fixtures, more dimmers, etc. ...

Especially bad practice on a shoebox dimmer like the Leviton D4DMX which uses fuse s instead of circuit breaker s.
EDIT: Incorrect statement; the D4DMX uses "Resettable circuit breakers in lieu of fuses." One would still have to traverse to the, likely inconveniently-located, satellite pack to reset a trip however.

As for the physics behind it, see http://www.controlbooth.com/forums/question-day/9903-dimmer-voltage-experiment.html and dimmer curve. At a DMX setting of 50%, voltage was 83%

... As I have been reading and tested, if I have 4 units plugged in at about 20% each, the pack is drawing about 2400w anyway. ...

Are you measuring the current with a true RMS responding ammeter?

 

[JD]

Some thoughts to clinch it:

When do these Triacs blow?

Top three reasons:
1) Lamp flash-over, which draws many times the running current.
2) Shorted cable, which draws many times the running current.
3) Cold lamp inrush, which draws many times the running current.

See a pattern? Number three is the one that will get you if you overload the dimmer. To get to 20%, you still have to come off of that cold filament. By some estimates, cold inrush can be ten times the running current of the lamp. It doesn't last long, but it's there. A properly designed 2400 watt dimmer will be designed with devices that can handle the cold inrush on a 2400 watt load. A 600 watt dimmer is not designed to handle the cold inrush of a 1000 watt load.

As has already been covered, Max current is achieved in the first 50% of the linear dimmer curve.
At 50%, your triac is actually switching on at the peak of the ac waveform! Other non-linear curves achieve this at even lower settings. Even the cold inrush current bringing the lamp to 5% idle is rather amazing to watch on the scope.

 

[NeroCaesar]

Hmm I'm testing with a $24 homedepot AMWatt appliance load tester I figure its only taking out the average. (plug in the dimmer to the amp meter and then plug amp meter into the wall)

Thanks JD That filled in some missing parts in my logic


I dont think I'm asking the question right though...

I have the same amount of units with or without the Levi Packs and with the EDI MX System rack I have never tripped anything before. I wanted the Levi Packs for individual control. The heart of the question is:

Has the EDI Rack only been drawing as much amps as needed and therefore not over loaded, and now if I have the Levi packs that Draw full amount every time a light comes on at 20% will my total amp draw be greatly increased during shows...

or has the EDI rack been pulling the full amount of amps when at 20% like the Levi packs do...

or do the Levi packs not actually Draw the full amount of amps and my meter has been deceived because its reading the average and everything is normal?

 

[derekleffew]

Your THP103 The Home Protectors | Product Details | Reliance Controls Corporation is giving erroneous results. Measuring current or voltage of thyristor dimmers is tricky; RMS, non-sinusoidal waveform, power factor, triplen harmonics, etc. Whether 20A load on the EDI dimmer, or 4 5A loads on the Leviton pack, at 100% your draw will be 20A (assuming 120V input and 120V lamps). At less than 100%, your draw will be less than 20A, but exactly how much is difficult to predict. Again, see the other thread where I measured voltage output vs. control input. Your dimmers (either of them) will be similar but not exactly the same. Also remember--the dimmers themselves don't draw power (in any significant amount anyway), it's the loads (lamps) connected that matter.

Oh, and there's a fatal flaw in your original question :

In a perfect exact world Could you...
Plug 1000w into a 600w channel of the leviton packs and only bring it up to 50% with no problems (not that I'm trying to... well kinda... but I want to know the physics behind it)

Although it was stated

Using a 1000w lamp on a D4DMX unit should be ok even at full. Its not a 600w dimmer. Its a 2400w max dimmer with a per channel capacity of 1100w. You can use 4 x 1000w lamps, just not all at once. ...

most assume that any 4-channel satellite pack will be 4x600W dimmers. (Is there/has there ever been, a variant of the D4DMX that's only 4x600?) But there's that other issue: even if the pack contains 4x1200W dimmers, doesn't mean you can safely put one 1000W lamp on each channel AND run them all at full.

From D4DMX-MD5 > D4DMX > Satellite Dimmer Packs > Dimmers > Lighting Controls > Products from Leviton Electrical and Electronic Products :

Item Description
4-Channel Programmable Dimmer Pack integrating stand-alone, 5-PIN DMX, & NSI Microplex, operates as 1, 2, or 4 channel with pre-programmed chases, 1200 Watts/Channel, 2400 Watts/Maximum, 15 Amp Power Supply Cord, 120V.

To me, 2400W Max and 15A supply cord appear contradictory, but maybe it's just me. And given the very likely Edison 5-15 connector, I'd be hesitant about putting 2400W or even 2000W on the pack. I've melted plenty (too many) of 5-15 plugs on 2000W Fresnels. 16.6>15, every day and twice on Sundays.

 

[65535]

Since a dimmer works by chopping the incoming power you can expect to measure peak currents at full load, however the average power used is a fraction of that. Still furthermore the triac or SSR units in the dimmers will see that full current even if the average consumption is below the dimmer rating, it may well not blow a fuse or a breaker but it can do damage to the dimmer.

Because of this cheap non RMS meters will register peak current which will give you a power consumption of full load, even if you're using far less.

Due to the filament resistance vs. temperature curve I could imagine seeing much higher peak currents at lower dimming levels.

 

[JD]

Ok, here is what is happening any why the readings are wrong-
1) In the top image, the full ac waveform is passing
2) With the dimmer at 75% (liner) the bulk of the waveform is still passing, including the point of peak current draw.
3) Even at 50%, you are switching the triac on at the peak current draw of the waveform.
4) At 25%, you still have a pretty good current spike being drawn.

What is happening is the duration of the waveform is being changed. Most meters are only measuring the amplitude of the waveform, thus the bad readings.
Until you get below 50%, the amplitude does not start to decrease.

(Picture credit goes to a quick google and to this website where I found the image: http://computerchristmas.com/christmas/link-how_to/HowToId-92/FullSize-true/LowLimit-10/StepId-12 )

To me, 2400W Max and 15A supply cord appear contradictory, but maybe it's just me.

When you see something like that on the cut sheet, it makes me question everything written on the page!

 

[NeroCaesar]

Ok excellent, that's what I thought, I just needed a bit of confirmation.

And luckily (kind of but not really) we are a L5-20 twist lock house so at least the plug I put on there can handle the amps.

I won't be trying to over load any channels any time soon. I've used these boxes for almost a decade and know they blow up with a strong breeze.

Thanks for the help and links,

 

[pkooistra]

I am wondering if someone can help me in my ignorance. We are trying to get a D4DMX dimmer working, but it always resets itself to 0 dim when the power is turned off. Is there a way to lock it so we can mount it some where and not have to worry about manually increasing the dim to 100 on the dimmer? I have looked around and read through the 4 page manual and can not seem to find an answer to this. Thanks for helping a newb at lighting.

 

[derekleffew]

The address reverts to 0 when power is cycled, or the intensity goes to 0%? Are you using a controller /console?

 

[pkooistra]

Intensity goes to 0. Simple 16 channel levitron controller, but it does not seem to act different if the controller is plugged in or not.

 

[Lotos]

Intensity goes to 0. Simple 16 channel levitron controller, but it does not seem to act different if the controller is plugged in or not.

Yeah... These packs aren't designed to hold their level indefinitely... Especially not once power has been cycled.
(Ironically, derek's question was a blast from the past, as I've seen D4DMX's "forget" their DMX address quite frequently when power cycled... Seems to be a semi-common issue with packs I've run into, both in my venue and elsewhere... I much prefer my older NSI packs to the newer Leviton ones.)

 

[Lightguy48]

I stumbled across this thread looking for technical documentation on one of these dimmers because I bought it used and like many other folks it had several shorted channels.

I happen to own some really old NSI equipment and it is sad to see how they have 'cheapened' their product offering over the years. I have never been a big fan of triacs for dimmers. I always preferred a dual-scr design. Sure a triac is nothing more than two SCR's packaged into one case but if you look at the thermal capacity of a dual-scr design vs a triac the dual-scr design always wins out even with the same current rating.

I have an NRD-8000 and an ND4600, both of them a virtually indestructible. Each individual SCR has a large heat sink per device, then you look at the D4DMX it has one single heat sink shared by 4 devices and then also relies on the heat to be distributed to the back mounting cover yet there is no heat sink grease to help conduct the heat to the rear panel.... very poor design indeed...

If there is only one good thing about the design it's that they still included chokes on each channel. I've seen really cheap Chinese crap with no filters at all or a single filter for all channels... those things put so much garbage on the line I've seen some of them actually generate enough noise to throw the gating/ramp signal into a tither within the dimmer pack itself.