Flickering light (LED) circuit

[Tipiford]

They are, about $50 apiece. And, as we start replacing bulbs, we'll be looking to replace the incandscents with LEDs--which would be very expensive, and I don't want to go to analog switches because our lighting system is a part of our security system.

 

[Tipiford]

Even if they're less expensive (the prices I've seen are $50 apiece), our entire house has these dimmers on our security system, so I'm not wont to replace them, especially not in the number we would be exchanging for our incandescents as they burn out.

 

[Arouse1973]

Well I don't know a work around. All I will say that if you use a dimmer that is not suitable for the LEDs then you don't have any comeback if one fails. So it's up to you what you do, you may be able to find a box of tricks that makes them compatible but I have not looked for one. You could try placing a load across the dimmer switch output to simulate a bulb, this might help but seem pointless in the fact your replacing the bulbs with energy efficient alternatives.
Thanks
Adam

 

[KrisBlueNZ]

Where is the dimmer in your diagram in post #1?

What was the reason that you connected a 0.47 µF capacitor across each group of LEDs? Where did you get the idea from?

What is the brand and model number of the dimmer?

What is the brand and model number of the lamps? Are they all the same? Do you have any technical information on them? A link to a data sheet?

 

[Tipiford]

Adam,
The RC circuit is somewhat taking the load from the circuit without, I'm thinking, burning energy (I have not tested for current, but the circuits themselves are at ambient temperature--and, in fact, the RC circuits are replacing a 4w lamp I started out with).

KrisBlue,
The dimmer in the circuit is represented by the on/off switches. There is no issue in the dimming feature, just on and off.
I figured the .47uf would be sufficient--which it is, or at least, seems to be for the random flashing issue when the dimmers are turned off--but, perhaps, I need to look at other capacitance values to see if a different one might work, As far as where the idea came from, it was from my own thought processes.
The brand of dimmers is X10, models, W467 and W477 (I think).
I mentioned in an earlier post that the bulbs were all the same: Wrong. They're both Utilitech brand lamps, the six are reflector style LBR30HO, 20W, 1150 lumens, model not apparent; the two are A15 style LA100/830, 12W, 1100 lumens, Model UL#E330072 (I could find no spec sheets or links to same).

One other note, when I pulled out one of each light to get the answers to KrisBlue's questions, the on state of both lamp circuits, five and one, instead of the blinking you see on the video, the blinking on the single bulb became rather rapid, as if it were taking up all the trace voltage in the circuit. And that the six lamps do not blink at all, not even when matched in number with the other circuit, perhaps trying a different bulb would give a different result--maybe even correct the issue.

 

[BobK]

The brand of dimmers is X10, models, W467 and W477 (I think).

Aha, finally a clue.

X10 uses communication over the power lines. I suspect that the noise created by one dimmer is interfering with the other.

Bob

 

[Tipiford]

There should be no noise except when a signal is sent, ON/OFF/DIM/BRIGHT. The issue with this dimmer--and most others--is that when turned off, there's still minute current present, unless of course, the master switch of the dimmer is off.
What I'm seeing is random flashing of the lights when they're off (off, but with that trace current present), and connecting a small load to the circuit, i.e., a 4W lamp, the random flashing ceases. That works, but adding extra usage of energy; so I tried the RC network on each circuit, and that works perfectly--until both circuits are energized, and then I get the flickering on the two-lamp circuit (not the six lamp). I think my next part of this experiment will be to try a different brand of bulb for the 2-lamp circuit, and then try a larger capacitance, unless of course, one of you comes up with a solution. Thanx for all the attention on this.

 

[BobK]

Here is a thread about it on an X10 forum:

http://forums.x10.com/index.php?topic=25646.0

They say noise created by the LED bulbs might be the problem.

Bob

 

[Tipiford]

"the switch steals power through the load..." They studied a different electronics than I did.
I've looked at that site and they recommend what the manufacturers claim their limits are, which is kind of a pull and plug mentality; I was looking for work-arounds to those limitations. Guess I'm too old school.

 

[BobK]

They were talking about WS467, same as one of them that you are using.

When they say it steals power from the light, I assume they mean it is installed in a switch box with no neutral wire, which means the power to it comes through the light. I can see why this might not work with an LED light, since it is not a simple resistor like an incandescent. And I can see why putting a small bulb in parallel would fix your problem.

Bob

 

[Tipiford]

Oh, yeah, the low wattage bulb works perfectly, circuits energized or not, and replacing the switches with SPST & DPST also would work, but I'm still hoping for a work-around, which would be a more perfect inexpensive solution.

 

[Tipiford]

GOT IT! I wish to thank everyone who participated in this discussion. I had planned first to buy a different brand of bulb to try, but it didn't make sense not to first try the larger capacitance, as i had those lying around, well not larger, but I paralleled two .47uf, making .94uf and now it's working as it should. Thanks each and all. Tip.

 

[BobK]

Alright, you just need to pass a little more current when the bulb is off to make the controller happy. Sounds good.

Bob