LED lights not completely turning off when connected to dmx

So I have 24 cans, 10 par 46, 14 par 38, split 16 on a truss, 4 each on two stands. Incandescent bulbs with gels on the 16 on the truss, LED bulbs (Emerson screw in sockets) on the two stands. Solid color LED bulbs. All are connected DMX to Elation controller, not sure of the model at this time. All of my incandescent bulbs completely turn off when commanded, the LED bulbs in the 38's on stands, do not... Frustrating, because I can't completey isolate individual colors on stage, because the LED lights won't turn off. Any help?
 
Does the Elation have a 'lamp warmer' setting? That is, rather than sending a dmx 0, it sends a 1 or 2 to keep the filaments of an incandescent 'warm'. You can't see it on stage, maybe staring close to the bulb you can. But, in an LED you will be able see it clearly. The instrument profile should also keep track of which channel is intensity vs. those for color, so check that that is correct too.
 
i bet that a ghost load will solve the problem becauce there is a small amount of leakage from dimmers that will cause the leds to be on alittle
 
At first read I though you meant DMX controlled LED units, but you are running LED replacement lamps off of a dimmer pack which can cause several problems. The most notable one is the one addressed above. All dimmers (SCR or Triac) leak some current through their own firing circuits. An incandescent lamps supplies enough of load that this current will result in very little voltage across the lamp. LED lamps however contain a small solid-state power supply (switch-mode) and will allow this leakage current to accumulate in the circuits capacitor enough to keep the lamp lit. Ghost loading is the solution, although Elation makes a plug adapter that I am told eliminates this problem.
The second problem is going to be that even when working correctly, there is going to be a jump-in and Jump out point on the dimmer curve that you will not see on the incandescent lamps. In other words, they will go right from 0% to about 10% in both directions. There is no good work-around for this which is why dedicated LED DMX pars are preferred.
 
I, too, didn't consider you were trying to run plain LEDs from a dimmer pack. It's not the best idea, for the reasons explained - the 10% jump in/out is unavoidable even if you work around the final turn-off.
The plug adapter does work, JD. We used it with some Christmas light strings for some color dots behind a scrim once. (the snap off wasn't an issue)
 
I, too, didn't consider you were trying to run plain LEDs from a dimmer pack. It's not the best idea, for the reasons explained - the 10% jump in/out is unavoidable even if you work around the final turn-off.
The plug adapter does work, JD. We used it with some Christmas light strings for some color dots behind a scrim once. (the snap off wasn't an issue)
Glad to hear it works! Here is what we are talking about:
http://elationlighting.com/ProductDetails.aspx?ItemNumber=1718&MainId=1&Category=Dimmers and Relays
proxy.php
 
Yeah, I think it's a neat device as well, but I have never figured out it has a max 0.5W draw with a 2000 ohm resistor. My math shows much more than that.

David
 
Yeah, I think it's a neat device as well, but I have never figured out it has a max 0.5W draw with a 2000 ohm resistor. My math shows much more than that.

David
Have not had the opportunity to smash one open and find out what's inside, but I suspect it is more of a type of snubber containing containing resistor, capacitor, inductor, etc. so it may present an impedance of 2k but only be active during part of the waveform, thus the low draw. I have tried a high wattage 2k resistor and it is not enough to quench the output, so there's something else inside. Too cheap to buy one to cut it open ;)
 
Have not had the opportunity to smash one open and find out what's inside, but I suspect it is more of a type of snubber containing containing resistor, capacitor, inductor, etc. so it may present an impedance of 2k but only be active during part of the waveform, thus the low draw. I have tried a high wattage 2k resistor and it is not enough to quench the output, so there's something else inside. Too cheap to buy one to cut it open ;)

That could be but my guess is that it's purely resistive. There are really two firing issues, right? One is too low of a load to maintain conduction and the other is keeping the load close enough to resistive to keep from false firing across the zero cross mark. Most LED lamps are capacitive, so if it was a load device and a power factor corrector, it would have to be resistive and inductive.

Now I want to cut one open. Time to get out the paypal account.

David
 
That could be but my guess is that it's purely resistive. There are really two firing issues, right? One is too low of a load to maintain conduction and the other is keeping the load close enough to resistive to keep from false firing across the zero cross mark. Most LED lamps are capacitive, so if it was a load device and a power factor corrector, it would have to be resistive and inductive.

Now I want to cut one open. Time to get out the paypal account.

David
Tell me what you find! (Hopefully it's not potted.) My math says 2k would be 7.2 watts at 120 volts, so something's keeping that resistor company!
 
Ok, so I didn't mention it the other day, but the only way I could think that this would really work is if the device was electronic. There is some IP regarding such solutions so I don't tend to talk about them. Indeed, the Elation unit is electronic and I have no doubt that it creates a resistive load near zero-crossing, which is where triacs and SCRs need a load to fire correctly, and to turn off correctly. However, I'm not sure that these plugs will help when there is a false firing to full, on a dimmer, when the load is more capacitive than the zx off timing can handle.

I did not power this up and map the result with a scope nor will I publish a schematic. This exercise was more for curiosity. If you need a dummy load on a dimmer to get it to turn completely off, act like a dimmer instead of a relay or get a better bottom end curve, this could very well do it. It does not cost much and does not waste much energy. This will not solve all LED problems, but if I were doing shows, I would carry a few of these with me.

A warning: these are not UL listed, they are thermally protected but you should avoid putting too much load on them, and they ship with this annoying paper sticker over the socket, which should be fully removed for safety's sake.

Questions?

David

photo 1.JPG
photo 2.JPG
photo 4.JPG
 
Neat! I see a couple of zener diodes and Q2 might actually be a triac! I suspect you are right that this thing becomes active right before ZVC. As for loading, shouldn't be a problem since it would be used to fix low load LED circuits. Agreed they would be handy to have in the tool box. I am impressed! I was not expecting it to be that sophisticated. Even if one were to draw the schematic out, it would be impossible to beat the price trying to build one. Kudos to Elation!

Of course, with the full wave bridge in there, Q2 may just be a transistor.....
 

Users who are viewing this thread

Back