light bulb still glows when dimmer fully off



We're working on a show that requires an exposed light bulb (regular 60-watt home use type bule) within the set to be conrolled from our lighting board. We're having difficulty because even when the dimmer is all the way down, there is a faint glow visible from the bulb when we are (should be) in complete blackout. I don't know what information is relevant but I'll share as much as I can. We use a 3-pin multiplex XLR system to control a portable dimmer pack from our board. The XLR cable run to the dimmer pack is fairly long (about 150-200 feet). Any ideas on why the bulb glows or what can be done to stop it?

There are three things that can be the problem:

1. With todays solid state dimming systems, there is always a small amount of leakage current going through the dimmers. Normally with the standard lighting fixtures this is not noticable. But when you connect a small wattage light bulb, this becomes noticable. What can be done in this case is to connect an additional light bulb to the dimmer that can be hidden somewhere. This light will increase the load on the dimmer which should solve the problem. The extra light connected to the dimmer is sometimes referred to as a ghost load, (not ghost light). edit-DL.

2. Try a shorter control cable and see if the problem goes away. With the 3-pin connector you are using it sounds like you are using microplex as your control signal type. Microplex uses an analog signal to set the levels of each channel. It is possible that there is some electrical interference that can be inducing some noise in that long a run of cable. Highly unlikely.

3. There is a problem with the dimmer. The dimmer could be "out of trim," the solution for which is beyond the scope of ControlBooth-contact the dimmer manufacturer or authorized service center. (Derek just realized this post was from 2003!)
Last edited by a moderator:
Thanks for this advice! I'll try these suggestions and let you know what happens. The control is indeed Microplex. Does Microplex run the XLR balanced to reduce interference like audio runs or is it different?
A fourth thing that could be the case is the dimmers can be trimmed for a filament warming current that is not noticed on higher wattage lamps but for something that's only 60 watts, it's enough to observe due to the size of the filament. Such a current is normal and protects lamp life when it jumps from zero to full - that's also why you kill the breaker on the dimmers before opening up the plug even if only a low voltage.

The need for "ghost loading" (solution one) is the easiest to detect and correct for. Try putting something on the dimmer in addition to the 60w lamp in another room. Say something 500 watt. Don't just set the ghost loading fixture on the floor expecially pointed down, much less pointed at anything because you can realistically have a fire in an unoccupied room. Perhaps put it in the dressing room to keep it watched even than as long as someone a trusted actor can be responsible for it or more importantly trusted to keep others from playing with it.

Next you can try swapping cable as per step two with a quick direct run light board to dimmers. Even if Microplex, you still need to be using DMX or at least digital quality XLR cable for this purpose. It's a computer signal that is going thru the wires and needs a cable with appropriate resistance and shielding. Normal microphone cable while using similar plugs don't do this well especially in long runs. The details on how the signal works is not in my expertise but doing this as advised would lead to a quick answer.

My step four would be next in detecting if it's a problem with the dimmer or something that's normal. Try the lamp on another dimmer channel, than another control channel, than try it on both that are different. This will tell you if it's a dimmer problem be it trim or dimmer itself.

Given it's still not the problem, and with the lamp plugged in, meter the voltage to the lamp. Than meter it on a dimmer that's part of the same phase and power source. All voltage testing needs to be done with a minimum of a 75 watt load.

Packs especially microplex dimmer packs frequently will be powering up more than one dimmer per power leg. The adjustments for how much voltage goes to each of those controlled dimmers is usually modifyable by a screw driver and located in the dimmer somewhere between the main breaker/power source and where the individual breakers take over. Otherwise it's for individual breakers depending upon brand and type. In any case, it's a simple modification to the "trim" that's necessary. If you are absolutely not trained for testing the voltage, much less adjusting the trim, get someone that is. For the trim on a Micropex dimmer, it's possible that the trim for it requires going into the inside of the pack thus the qualified tech person to do so. If it's exposed, than adjusting for the trim is really easy.

Get the manual for the dimmers and it should tell how to adjust this. Just don't forget to tune the voltage back up after you are done with the show. Otherwise you will possibly be blowing more lamps on the dimmer plus they will not go up to full. Trim the lamp down to a level that's just below what you can observe to incandess while on that 60w load plus the 15 watts making it become at least 75 watts. Say a 15 watt work light somewhere off stage. Don't trim it all the way down. Plus any other dimmers that are on the trim will become dimmer so it is a factor in design and possibly re-patching to account for this. Say a dimmer instead of going from 0 to 100 with a trim or warming voltage instead has a range of 10 to 100 percent. By trimming the dimmer to zero, you are now limiting it to 90% of full when you send it to full from the control signal. This might become a factor.

After all that's checked out, and especially if trimming the dimmer is not an option, plus the bringing the dimmer up to 75 or more watts in load did not solve it, it could be a bad dimmer given especially the lamp went out on another dimmer. But remember that the new dimmer you put it in could also have had a differing trim setting which is where my line of thought lies. Only metering the voltage while under load will be able to tell that.

How long has it been since the dimmers were maintained much less trim levels adjusted? Could be a part of a larger problem as trimming the dimmers is normal for a service call to them.
Most dimmers have a "keep-warm" setting where zero should be - for good reason. The filament in any lightbulb has a very low resistance until it gets hot. This low resistance causes an "inrush current" for the first few milliseconds after it's turned on that can be thirty or more times what the lamp will draw at full brightness. Especially if a lamp is being used in a chase, where it's being turned on and off frequently, this high current can pop a breaker or worse, damage the dimmer. To avoid that, most dimmers never go completely off... a fader position of zero will still allow enough current to keep the filament warm. The keep-warm current is usually set for the wattage the dimmer is designed to handle - with a typical portable dimmer pack of 600 watts per channel, your 60 watt bulb is getting enough current to keep a 600 watt PAR64 warm.

Some dimmer packs, geared toward DJ and concert lighting, have one or more channels where you are allowed to turn off this "keep warm" current. It's usually called something like "strobe compensation." Many strobe lights will continue to flash, though with an extended period between flashes, if they see "keep-warm" current. "Strobe compensation" is simply the ability to turn the "keep-warm" current off. Check the documentation on your dimmer packs for something along those lines. As long as you're just using a 60 watt bulb, the inrush current won't be enough to cause a problem, especially if you're not flashing it on and off in a chase.

If your dimmer packs don't have any way to turn the strobe compensation off, where are you and when is the show? If you're somewhere near Chicago (like so many on this forum), e-mail me ([email protected]). I have a James Lighting DR-2400 you could borrow. It's a 4-channel, 600W per channel pack, about the size of an NSI ND-4600, with switchable "strobe compensation" on channel 4. It does Microplex, but only with boards of 64 channels or less (the James multiplex protocol is identical to microplex except for the 64 channel limit).

I've also got a reasonably understandable explanation of these old multiplex systems on my website. Got to Multiplex.html.

Ok. I tried plugging another (higher wattage) ghost lamp into the same channel on the dimmer pack, and the glowing bulb got considerably dimmer, barely visible now. But I also made another discovery that has me confused. When equipment that is on the same circuit as the dimmer pack (but before the dimmer) is activated the glowing completely stops. Is this just because I'm drawing more power on the circuit? Is it anything I should be concerned about (the wiring in this old buildiing is pretty scary).

Thanks everyone for the wealth of information you've provided. It has been very helpful and educational (I feel like I should be paying tuition... :)

Yes probably. Get the school's maintinence head in to see what you are telling us. Short of metering out what the added load is doing or seeing the actual system, it's hard to say for sure.

Possibly the gauge of wire feeding the dimmers is too small for the load at the distance traveled and it's causing a voltage drop as what I think the most likely answer. That further dimming you are observing means something is very likely turining the voltage into heat. Heat can be a bad thing.

I would be open to other ideas on the cause given a very broad understanding of your system, but unless it's part of how your dimmers work and that would not seem normal, than it's probably something to do with the building and something to be concerned about now before it becomes a bigger problem.
scotru said:
When equipment that is on the same circuit as the dimmer pack (but before the dimmer) is activated the glowing completely stops. Is this just because I'm drawing more power on the circuit? Is it anything I should be concerned about (the wiring in this old buildiing is pretty scary).

I agree with Ship - sounds like something a licensed electrician should look at.

Solid state light dimmers are especially hard on old wiring because of the way they work. They use an SCR or triac - a very fast solid-state switch - to turn the light on and off very fast. A 1000 watt lamp dialed down to 10% doesn't draw a steady 100 watts: it draws nothing 90% of the time and 1000 watts 10% of the time. While the average power is 100 watts, the wiring and circuit breakers are seeing these little blips at full power, 120 times a second, that can cause the wiring to heat up over time. This isn't much of a problem for the actual wires, but over a period of years the repeated heating and cooling can cause the connections- where wires are attached to screw-terminals in the outlets or on circuit breakers, or twisted together in junction boxes - to loosen up a little. Loose connections have higher resistance and tend to heat up even more. They also cause a voltage drop - which may account for the pesky lamp going out when something else on the same circuit is turned on. So, like Ship said, point it out to maintenance - it's not a serious problem yet, but could be a disaster waiting to happen.

I'd like to also suggest the testing of the wiring, socket and plug on a normal instrument and the fixture in question with a volt meter and see if you are getting a "trickle" of electricity. Similar to dimmer creep that Cruiser said. I might be wrong, but if he has a lower wattage fixture that is say getting 10% voltage, and he gets a glow, and then plugs in a higher wattage fixture, the "glow" would be much less on that fixture because of the resistance of the higher wattate lamp, wouldn't it? My suggestion in addition to Ships and DMX tools would be to test the socket and the wiring with a AC volt meter and see if you are in fact getting a trickle of power...

Multiplex systems--most run power for the control boards via the packs thru the cable along with the could try changing the XLR to that pack but if you are in a rack of many you may just have a dimmer going bad...

my thoughts fwiw....
Yeah, just as I was reading what you said Wolf, before I got to the end the thing about the 10v line on some racks sprang too mind.

I know alot of old strand stuff used to power the desk off the racks, with 1,000ohm resistor in the line somewhere, and when you run a digital desk with a multiplexer you run the desk at like 80% and your stack at 100% which then stops pushing the racks, as theat equates to 100% voltage overall....

meh, forget that.... just think on your digital desk that the Grandmaster 0 to 100% as 0 to 10v.

On old desks, an extra 10v was on the line to power the console. So, when you use your multiplexer and put your desk to full, that 10v will still be in the line, which could be causing extra voltage to the regular lamp, as it is such a low wattage compared to a regular fixture.

Try running your desk at anywhere between 65 and 80% and see if that makes any differences....
Just an idea, ya never know!
Correct me if im wrong anyone, but I think I have the basic gist of things right =)
I know....late to the game.

I went through this with an NSI pack. After having a firing board replaced by them for an unrelated problem, cost almost $400 with shipping, the pack came back with all 8 dimmers exhibiting this glow - on the set and on the bench, control board or not. Tried all the reset, pot adjustments, and other stuff that the NSI techies suggested. I asked for a new board. Sent it back and came back again the same. Furious, I had to order another pack just to bring up the next show. They picked up the shipping of the old pack and tested it and could not replicate the problem. Management refused to install a new board. Until I sent them a photo of my bench test environment (replicates the power connections down the the correct phase) and the glowing light bulbs did they agree to replace the control board.

Works fine ever since.

Now I service them myself.

NSI service has improved since then.

Items in this thread contain some weird and wonderful ideas, a great many of which are so wrong that they will confuse students and should be corrected by a respected figure, so could we ask Steve Terry to go through it and explain the misconceptions, if I do it there'll be another bun-fight.
How about if you try a bigger lightbulb, like a 100W or 150W? It may be just enough not to glow. You do not have to turn it on all the way, i dont think the audience would notice it being slighty more yellow than a normal light bulb.
how about a 7.2k dimmer, I'm using a single s4 36* hpl 575 on it and the lamp still glows. I have an identical load and dimmer next to it without a glow, any tips? It doesn't glow bright enough to be onstage, so i haven't tuned it down, but it just annoys me that its on.

Users who are viewing this thread