Help needed with LED House light dimming.

Wireless LED lamps - look at http://www.ketra.com/. Pretty cool stuff. Also good dimming on some mains dimmers. They are known for their high CRI - and retrofit retail is a primary market. I'm trying them for a house light system that was low efficiency incandescent and has been replaced with compact fluorescent and never dimmed (or not in a long time - hard to tell - might have been auto-transformer when built in 50's). I'd like to experiment more but I think it could be a good cue light - fast and multiple colors - and/or running light - dim blue or full white at a button press - all in existing lamp holders ad no wiring change.

Otherwise google "wireless dimmable light bulbs" - lots out there.

My uneducated take on mains dimming of retrofit LEDs - it will never be without defects. A retrofit lamp has a solid state driver that needs power to work and when the power drops below a certain level, the driver ceases to work. Holding a charge - like in a capacitor - might allow the driver to continue to function for a while (I think the principle behind the R-LED and presumably others) but that's fine for an uncontrolled 10% to 0% but doesn't do as much for on. Some also won't "bump" out but must fade. The wireless may be a better solution but replacing fixtures will always be better light, better control, and likely more energy efficient and longer lived.
 
How is Compact Flourescent technology doing these days? I had CFL house lights at my old theater, it was a joke because they blinked on and off and came on a dim pink, slowly warming up to 2700K and full brightness over about a minute. Has that technology improved?
 
My uneducated take on mains dimming of retrofit LEDs - it will never be without defects. A retrofit lamp has a solid state driver that needs power to work and when the power drops below a certain level, the driver ceases to work. Holding a charge - like in a capacitor - might allow the driver to continue to function for a while (I think the principle behind the R-LED and presumably others) but that's fine for an uncontrolled 10% to 0% but doesn't do as much for on. Some also won't "bump" out but must fade. The wireless may be a better solution but replacing fixtures will always be better light, better control, and likely more energy efficient and longer lived.

I wonder whether a better result for mains dimming might come from losing the 1:1 equivalence of the chopped waveform input to light level output. Even in residential type applications, would anyone care that what used to be 15% on is now 5% and anything below the old 10% say is off? I'd assume some scaling so that the remaining 90% was scaled out to the full range of output brightnesses...
 
My uneducated take on mains dimming of retrofit LEDs - it will never be without defects. A retrofit lamp has a solid state driver that needs power to work and when the power drops below a certain level, the driver ceases to work. Holding a charge - like in a capacitor - might allow the driver to continue to function for a while (I think the principle behind the R-LED and presumably others) but that's fine for an uncontrolled 10% to 0% but doesn't do as much for on. Some also won't "bump" out but must fade. The wireless may be a better solution but replacing fixtures will always be better light, better control, and likely more energy efficient and longer lived.

There is a way, but it would require a dedicated product. The Cree lamp actually looks at the sawtooth waveform to adjust its output. What would be required (in theory) is a lamp that had an output of 0 to 100% proportional to a 10 to 100% voltage input. Basically. the lamp would always have power, but be programed to interpret reaching 0 at a 10% dimmer setting. The dimmer would then be profiled to output 10 to 100% over the 0 to 100% fader span.

EDIT: Well, I see I was a little slow on the post, but basically the same idea as the above post! ;)
 
Yes, I've considered same idea, but I think it's a market issue. Too many different lamp and socket forms and too many different dimmers and too many different control systems, that all have to cooperate on something they were not designed to do and then asking that the users of each understand profiles.
 
The Ketra's look pretty, but I assume their control is proprietary, and getting DMX to them is a pricey option?

No idea on pricing, but their N3 interface has a DMX input. The wireless signal is proprietary and there's one interesting catch in converting to it from DMX: it works best when you use a profile that includes a fadetime parameter (RGBF, RGBIF, and similar).

The reason for this is that the wireless connection is fairly slow and there's a lag in converting from DMX. If you try to send a 5s fade as a series of in-between values sent over a period of 5s (ie the normal way for DMX dimmers), then you'll see individual lamps changing level at different times along the way. To avoid that you would want to set the fade time parameter to 5s and then snap to the new level in 0s (letting the fixture calculate the actual fade).
 
The Ketra's look pretty, but I assume their control is proprietary, and getting DMX to them is a pricey option?
I though the box with dmx in to speak to Ketra was not a pricey option - but define pricey. I thought like $150-200? I'll try to follow up with sales rep but you could find a dealer and ask. Barbizon is listed as a dealer - guessing hey have a presence in Florida.
 
No idea on pricing, but their N3 interface has a DMX input. The wireless signal is proprietary and there's one interesting catch in converting to it from DMX: it works best when you use a profile that includes a fadetime parameter (RGBF, RGBIF, and similar).

The reason for this is that the wireless connection is fairly slow and there's a lag in converting from DMX. If you try to send a 5s fade as a series of in-between values sent over a period of 5s (ie the normal way for DMX dimmers), then you'll see individual lamps changing level at different times along the way. To avoid that you would want to set the fade time parameter to 5s and then snap to the new level in 0s (letting the fixture calculate the actual fade).

So you could still do that by DMX, you're saying, simply by lying to the board about the actual performance, and then making sure the fixtures match your lie?
 
Jay - you need to arrange a demo and hands on. I did - about a year ago - and had one opportunity to recommend - freebie local project consulting - and don't recall all the details. Pretty sure I saw a console (ColorSource 40) running the Ketra lights through the interface, including dialing in color, and doing bumps that <might> be fast enough for a cue light. I came away understanding their emphasis was on great color and retail - a much bigger market than performing arts - but it has a place in some theatres.
 
So you could still do that by DMX, you're saying, simply by lying to the board about the actual performance, and then making sure the fixtures match your lie?

It's like saying, "Walk 100 steps north in 5 minutes" instead of, "Take a step north... Take a step north... Take a step north..." Ketra's protocol uses the former while DMX is generally more like the latter. It's entirely possible to get very good results as long as you're programming it right.

If you've done much moving light programming, then thinking of it as a motor-speed parameter might help.
 
On dimming retrofits, recall that the control range is not the same as the voltage.

Incandescents usually take 15% of their voltage to preheat. So even on a 120V system minimum dimmer output will be far higher than the typical 5V LED or driver requirement. Worst case we lose a bit of the bottom range of a dimmer, but could still dim smoothly from 0-100% brightness. The theory is sound, the engineering and marketing is still developing.
 
On dimming retrofits, recall that the control range is not the same as the voltage.

Incandescents usually take 15% of their voltage to preheat. So even on a 120V system minimum dimmer output will be far higher than the typical 5V LED or driver requirement. Worst case we lose a bit of the bottom range of a dimmer, but could still dim smoothly from 0-100% brightness. The theory is sound, the engineering and marketing is still developing.

It may be that the technology for better mains dimming LEDs is developing, but tempus fugit. There is no good reason to plan on it in new build, and existing installs will be fewer and fewer as they are renovated properly of demolished. Hard to develop that market. Even the RLED seems to have a short marketing life, as its cost is not much less than replacing the fixture, which I believe is a better investment for longer life and higher efficiency. Kind of reminds me of the device with a bunch of solenoids that sat on a Selectric keyboard and was driven by the computer - an early high quality computer printer - sort of.
 

Users who are viewing this thread

Back