Control/Dimming Line dimming LED tape

[TheTheaterGeek]

So this all started when my super and I were chatting about led dimming.

Normally, we use a repurposed 12V power supply from a server rack, and a little DMX decoder. It allows us to do 8 outputs of RGB and it works just fine.

We started discussing dimming white led tape however, and we stumbled on an idea.

"Hey...do you think we could do this with a dimmer?"

My super mentioned that you dim LED's by switching the power on and off really quickly(PWM), and that is fundamentally what an SCR does. So we gave it a shot.

We plugged a wall wart into a dimmer and plugged it into some tape a voila it sort of worked. There were some issues with curve however. The tape dimmed from 0-100 on dimmer output 0-30 it was smooth but didn't allow for much wiggle room. I mentioned that you can create your own curve and proportion in the ION to help, but he thought that he could mitigate the problem.

He decided to stop by RadioShack and pick up the components of a wall wart, minus the voltage regulator. He said that could be what was causing the problem. So I threw together the board in the photo.

120-12v transformer, bridge rectifier, capacitor, and inductor, out to the tape.

We prototyped it because it meant we could play around with different capacitors and inductors to see what happened.

It worked great. We hooked up our decrepit O-Scope.

What we saw was promising. Weird wonky sawtooth but overall it worked.

We left it dimmed for a while to see if we were burning anything up, and we weren't.

Between my super and I this seemed like a good solution.

Anybody else have any thoughts?




Sent from my iPhone using Tapatalk

 

[TheTheaterGeek]

One more thing, like 2 led's in the tape were always on at +- 10%. Discharging the capacitor fixed the problem, but it wasn't enough for us to say it had failed. Any ideas on how to avoid this?


Sent from my iPhone using Tapatalk

 

[FMEng]

You probably don't need the inductor. The always on glow is because the dimmer thyristors "leak" a bit when they are off. You can avoid it by adding some load across the transformer secondary. Something like a 100 ohm, 2 Watt resistor should do.

 

[TheTheaterGeek]

You probably don't need the inductor. The always on glow is because the dimmer thyristors "leak" a bit when they are off. You can avoid it by adding some load across the transformer secondary. Something like a 100 ohm, 2 Watt resistor should do.

I'll try that thanks! Why would the inductor be unnecessary? I am admittedly unsure of its role in this case.


Sent from my iPad using Tapatalk

 

[TheTheaterGeek]

I'll try that thanks! Why would the inductor be unnecessary? I am admittedly unsure of its role in this case.


Sent from my iPad using Tapatalk

No dice. It seems like the capacitor is the culprit.


Sent from my iPad using Tapatalk

 

[porkchop]

No dice. It seems like the capacitor is the culprit.


Sent from my iPad using Tapatalk

Try putting the resistor in parallel with the cap? Not exactly the most energy efficient solution but seems like it would discharge the cap while dimming.

 

[FMEng]

Try putting the resistor in parallel with the cap? Not exactly the most energy efficient solution but seems like it would discharge the cap while dimming.

Maybe I'm misinterpreting the description. Are the two LEDs glowing at 10% brightness, when the dimmer is at zero? If so, the culprit is not the capacitor holding a charge, and I stand by my previous comment.

If, on the other hand, just two LEDs are lighting with the dimmer at 10%, that's due to slight variations in the Vf of the individual LEDs. No way to resolve that.

As for the inductor, it doesn't hurt anything, but the additional DC smoothing it provides is not necessary. The cap can do the job on its own. A little ripple doesn't bother LEDs in the slightest.

 

[JD]

Maybe I'm misinterpreting the description. Are the two LEDs glowing at 10% brightness, when the dimmer is at zero? If so, the culprit is not the capacitor holding a charge, and I stand by my previous comment.

If, on the other hand, just two LEDs are lighting with the dimmer at 10%, that's due to slight variations in the Vf of the individual LEDs. No way to resolve that.

As for the inductor, it doesn't hurt anything, but the additional DC smoothing it provides is not necessary. The cap can do the job on its own. A little ripple doesn't bother LEDs in the slightest.

Exactly! Even at 0 you will see a waveform hitting the transformer. It's the nature of dimmer firing circuits. Result is you will have some DC on the output that will accumulate in the capacitor. The lowest breakover LED's (they vary) will light a bit. Removing the cap will only remove the accumulator (think of it as a storage tank.) Putting a bleeder resistor across the cap will extinguish them. Play around with the value and use the highest value (lowest conduction) that will shut them down.

 

[dmoes]

Looks good to me Just a bleed resistor across the capacitor should solve the glowing lights 100-200 ohm 2W would be fine they are firing at a fairly low current and the bleed from the dimmer would also be fairly low current the resistor would bleed off enough that the voltage would be below the LEDs threshold

A bleed resistor is a good idea in any power supply and should be there regardless of its purpose.


in a low voltage supply the inductor probably is not doing much and typically would have a second capacitor on the output side of the inductor. I have built a few power supplies simple as well as regulated in my day. I only chokes chokes in high voltage supplies for tube transmitters (800V and up) as far as ripple its really not important as the frequency would be 120hz (im assuming your in North America) because you are using a bridge rectifier and it would flash if at all faster than one can see anyway. Most flurecent and to some level low wattage incandescents flicker at this rate as well and we dont know it.