staged dimming

[tom4bren]

Remember when it was OK to have your dome light come on when you opened your door and turned off when you closed it - AAHH the good ol' days.

I guess you've already realized that LEDs don't dim (that's the Diode part of LED) and that you're OK with that. You are only concerned with the buzzing of the relay.

Your dweebish EE friend (I can say that cause I are one) was on the right track with removing the capacitor. Before you remove it though, you might just try putting a BFC (Big F%%%ing Capacitor) across the power leads of your relay. It should hold the switched voltage long enough (in theory) for the dimming action to finish and then bleed off fast enough so that the switch in the relay doesn't 'bounce'.

[etlsport]

hehehehehehe

[WyrreJ]

I guess you've already realized that LEDs don't dim (that's the Diode part of LED) and that you're OK with that.

What are you talking about? Of course LEDs dim. In fact, it takes special circuitry (voltage regulator) to prevent them from dimming when running from a battery - as voltage decreases so does brightness.

[tom4bren]

WyrreJ,

"Because the voltage versus current characteristics of an LED are much like any diode (that is, current approximately an exponential function of voltage), a small voltage change results in a huge change in current."

It can be done but shouldn't be done. All of the benefits of an LED go down the toilet when you try to dim them. If you feel the need to dim LEDs then you should seriously consider PWM (Pulse Width Modulation) as they react very well to that scenario. Of course the effectiveness of PWM is very dependent on the color of the LED you are using as your eye will integrate the different colors at different efficiencies.

OOOPPSS - Just looked in a mirror and saw a dweeb. Sorry 'bout that.

Tom

BTW ETL - the pix look great

[WyrreJ]

PWM requires a much more complicated circuit. If you want to run at peak efficiency than PWM may be the way to go, but not always. But in applications like this, there is hardly any need for it - reduced efficiency for a few seconds isn't going to hurt anyone. Which is all really beyond the point, as an engineer I'm sure you can appreciate my desire for precision -- "LEDs don't dim" is just simply incorrect.

[tom4bren]

PWM is definitely too complicated for this application.

LEDs can dim - a little. If you want to retain the dimming though, keep the incandescant setup (I meant to make that statement in my earlier comment but forgot). If you try to dim LEDs (especially in an array like ETL is using) you'll end up frying it or the circuit. LEDs try to be constant power so the current will increase dramatically as the voltage drops ... and the LEDs are NOT protected by the fuse.

The more pertinent question is: can ETL debounce his relay so that it doesn't buzz?

[Lizardmen3477]

Wow where did u get that and how hard is it to install i would love to buy that please send me website and how it was to install. Thanks Lot.

[WyrreJ]

LEDs can dim - a little.

No, they can dim a lot.

Here is just one of many graphs of the output of an (unregulated) LED flashlight brightness versus time starting with a fresh set of batteries:



They pretty much all look like that - 50-75% dimming.

Here's a ton of discussion on PWM versus current-control for LED lighting.
The general consensus seems to be that current-control is preferred.
http://www.candlepowerforums.com/vb/...d.php?t=152185