Forward makes sense. And after all it IS a diode and is blocking the current on one direction and doesn't have to work too hard at it since the reverse leakage current is probably enough that the resistor is soaking up / limiting the effective reverse voltage to a possibly tolerable level. Resistor only takes a beating - er a - heating at 50% duty cycle and at a halfasine duration only.

At first it does It does conjure up an instant vision of :

vvvfffvvvFFFRAP---POP--- sizzzzzzle.

Hence, my signature phrase

Originally Posted by shingadaddy

Forward makes sense. And after all it IS a diode and is blocking the current on one direction and doesn't have to work too hard at it since the reverse leakage current is probably enough that the resistor is soaking up / limiting the effective reverse voltage to a possibly tolerable level. Resistor only takes a beating - er a - heating at 50% duty cycle and at a halfasine duration only.

Take a look at any discrete LED spec sheet. "Absolute Maximum" Vrev=5V. (Ok I did find one that had Vr=9V) The Irev at Vrev is usually 100µA. So using your concept: 10K x 100µA would drop 1.0V across the resistor. That leaves 119V (actually 169Vpeak) reverse voltage across the LED.

Still doesn't make sense.

Ken

LED - 10K 1/2 watt resistor and a suicide cord and you have the test set.

Give it a whirl.

Thats why I wanted the LED part number so I could check to see if its one with a built in resistor and a non-typical voltage rating.

FLASH! Maybe its one of those fancy smancy LED's that actually have 2 diodes on the die. One in each direction.

If it had a built-in diode, with its polarity opposite the LED, it would work. I don't recall seeing ever seeing one like that. But then, I haven't seen everything...yet.

Ken

OK I did a bench test. But, no way I'm going to do the 120VAC/resistor/LED trick.

Normal forward bias connection of a red LED with a current-limited power supply: Vf=1.9v@If=20mA

Reverse Connection: Ir=~0 mA @ Vr <24V. As Vr was increased above 25V, Ir rapidly increased to 30mA. The LED was getting hot!

A combination of the 10K resistor limiting the reverse current to 12mA (120v/10K) would keep the LED from going Poof. But this cannot be doing the LED any good in the long run. A single diode, connect with opposite polarity across the LED would make the LED operate within its specified limits.

be80be,

Do you have any examples of this configuration used in commercial, or hobby, devices?

Ken

I'm done with the Led and 120 volt's now i don't no what you tired a diode blocks half of the 120 volts because it's AC so it only using 60 volts now the led will not burn up because I limited the Current to about 6mA. So it can't be getting hot. It's the current that burns the led out and most any small diode can handle 120 volts at less then 20 mA. And a led is a diode. Oh by the way if you have not seen them They are selling light bulbs with led in them $35.00 each lol and the one's I put in couldn't have more then a resistor and some led's in it. Oh what I showed you was real a cord plugged in to the wall outlet with a led hooked to a 10k resistor and and hooked to the cord and plugged in to the AC 120 volts
LED; Red, T- 1 3/4 (5mm); Super Brite, 6.3 mcd, Diffused Lens
$0.12 (Each) that's what it is

__________________________________________________ __
Linux help by be80be

The double direction (or dislecsic) diode is a " you can't get it backwards" design and glowage is ensured no matter which way it gets stuck in the circuit.

Ken I figured you tried the 120 volt test and that was what brought the electro-tech website down around 4ish central time today. I didn't hear a big bang though

lol It just a plain red led like you get from radio shack. no tricks It just works. Here there using a 6 watt lamp for the resistor Ultra-simple Voltage Probe - RED - Page25

Just remember that this non-linear relationship between voltage and current means that Ohm's Law doesn't work for LEDs.

Thanks!

Ken

Yes, a relay will work in this instance.
Sometimes size is a concederation and relays/sockets may be larger than acceptable.
Relays are also mechanical, some designs may call for solid state.
Cost may also be an issue, relays and sockets run around $5-$7 for the set, the components I listed will be around $1.
Wiring may also be a constraint on design, with relays, switched power needs to be wired, where the design I posted supply the power (2 wires vs 4)

Don't beat yourself up too bad for reading the whole thread.

Originally Posted by mike2545

Y
Don't beat yourself up too bad for reading the whole thread.

Are you implying that there was just toooooo much discussion?

ken

Originally Posted by KMoffett

Are you implying that there was just toooooo much discussion?

ken

Not at all, just that umer27 seemed a little exasperated, and I didn't want him to be that extreme, shooting himself and all.

I was going to suggest an optocoupler but...Meh